diff --git a/search/search_index.json b/search/search_index.json
index 15946ea..1ab8ede 100644
--- a/search/search_index.json
+++ b/search/search_index.json
@@ -1 +1 @@
-{"config":{"indexing":"full","lang":["en"],"min_search_length":3,"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) introduces IBM watsonx Code Assistant's generative AI capabilities and lays the groundwork for the hands-on training that will follow. [6 min] i. Automation is indispensable to modern IT strategy Despite the innovations and advancements made in the domain of automation, IBM sellers and partners know first-hand from discussions with clients that many businesses are still struggling to keep up with their IT operations. The rapid pace of technological innovation\u2014 in particular, areas such as AI and machine learning \u2014are obviously challenging for any organization to strategize and plan around. But smaller, more practical challenges also stand in the way of these businesses. The fact remains that IT operations, and wrangling those operations in an efficient and streamlined manner, remains a difficult problem to solve. Three primary pain points that IBM consistently hears from the marketplace include: an ever-increasing skills gap in IT management; that Day 2 operations continue to be labor-intensive, mostly manual endeavors; and that the complexity of the systems needing to be managed are out-pacing many organization\u2019s ability to adapt. All of these pain points are potential automation challenges to be solved. Each of them impedes a company's ability to move quickly and adapt for the future. And as such, for many IBM clients, solving these automation challenges have become an indispensable element in their strategy to modernize IT. UNPRECEDENTED RATE OF GENERATIVE AI ADOPTION Even though generative AI is relatively new, the widespread popularity of ChatGPT has created significant interest in the notion of large language models (LLMs) and foundation models (FMs) \u2014 and what they can do for business. It took quite some time for enterprises to start moving toward traditional AI. In contrast, generative AI has experienced massive early adoption: 80% of enterprises are already working with, or planning to leverage FMs, and plan to adopt generative AI in their use cases and workflow. Moreover, the following data points to an ever-growing adoption trend for generative AI: Scale Zeitgeist 2023 AI Readiness Report notes that with the companies they reviewed, 21% have generative AI models in production; 29% are experimenting with generative AI and another 31% are planning to work with generative AI models; a total of 81% are either working with or planning to work with generative AI models Goldman Sachs has estimated that generative AI will have a very deep economic impact \u2013 raising global Gross Domestic Product (GDP) by 7% within 10 years, reflecting the technology\u2019s huge potential. Boston Consulting Group (BCG) noted that generative AI is expected to represent 30% of the overall market by 2025 ii. Generative AI-assisted code lifecycle management achieves what LLMs alone cannot Following the debut of OpenAI's ChatGPT, the marketplace has been awash with competing large language model (LLM) and generative AI-based assistants. It's one thing to train and deploy an LLM; it's another thing entirely to make it applicable and tangibly beneficial for business. What separates IBM watsonx Code Assistant (WCA) offerings from competing vendors in the marketplace? The design and implementation of WCA is purposely built to assist, using generative AI, software and code lifecycle management. In short, generative AI-assisted code lifecycle management helps to achieve what large language models cannot achieve on their own. It is what distinguished WCA from other code assistants in the marketplace today. Code lifecycle management begins with understanding client code, through training across a myriad of programming languages and specializations in paradigms such as Ansible Automation Platform, and applies that understanding across a client\u2019s application and runtime environments. Users are able to plan next steps based on generative AI analysis of their existing application code. Operations teams can rapidly transform their codebases with optimized design and architecture that is recommended according to IBM Granite's best-practice models. Administrators can validate the outcomes with automatically generated unit tests. Afterwards, they can deploy those services and applications using automated processes like Ansible's automation engine. Over the course of that application or code's lifecycle, generative AI can maintain healthy operations with runtime insights. iii. Introducing the IBM watsonx Code Assistant product family IBM watsonx Code Assistant is the flagship offering in a suite of generative AI code assistant products, which also include offerings for Ansible Automation Platform (IBM watsonx Code Assistant for Red Hat Ansible Lightspeed) and IBM Systems modernization (IBM watsonx Code Assistant for Z). These solutions accelerate software development tasks with AI-powered capabilities including context-aware code generation, explanation, documentation, translation, and unit test generation. It does so while maintaining the principles of trust, security, and compliance with regards to IBM client's data and intellectual property. Developers and IT Operators can utilize WCA to speed up application modernization efforts and generate Ansible-based automation jobs to rapidly scale out (or scale up) IT environments. IBM watsonx Code Assistant products are powered by IBM Granite foundation models that include state-of-the-art large language models designed for code. For offerings such as WCA for Ansible Lightspeed and WCA for Z, bespoke code models\u2014 tailored to working with Ansible Automation Platform and COBOL-to-Z use cases, respectively \u2014are invoked. Universally true for all of the watsonx Code Assistant offerings is that they are geared towards helping IT teams create high-quality code using AI-generated recommendations, based on natural language requests or existing source code. These AI models, and the recommendations they generate, are seamlessly integrated via extensions with the world's most popular developer integrated development environments (IDE), including Visual Studio Code and Eclipse. Granite is IBM\u2019s flagship brand of open and proprietary LLMs, spanning multiple modalities. Granite models exist for code, languages, time series, and GeoSpatial \u2014 with additional modalities expected in future. IBM Granite code models are a series of decoder-only models for code generative tasks, trained with code written in 116 different programming languages. The Granite code models family consists of models ranging in size from 3 to 34 billion parameters, in both a base model and instruction-following model variants. These models have a range of uses, from complex application modernization tasks to on-device memory-constrained use cases. The larger the block size for a particular language on this chart, the larger percentage of training corpus data of that language was used to train the Granite code model. Languages and formats such as Java, C, JSON, JavaScript, HTML, and PHP are subjects in which the model \u201cMajors\u201d and excels. Other languages such as Ruby, SQL, and Swift could be considered \u201cMinors\u201d where the generalized code model can work with the language, but has less training data to base those recommendations on. These percentages and training data volumes will continue to evolve as the Granite code models mature. WATSONX CODE ASSISTANT vs. WCA FOR ANSIBLE LIGHTSPEED? For those familiar with other IBM watsonx Code Assistant offerings\u2014 such as WCA for Red Hat Ansible Lightspeed and WCA for Z \u2014the generalized code model approach, as seen here, differs from the specialized code model approach of those two aforementioned offerings. The WCA for Ansible Lightspeed flavor of IBM Granite code models specializes (Majors) only in Ansible Playbooks and YAML Similarly, the IBM Granite code model used by WCA for Z specializes in transforming COBOL mainframe code into modernized Java code Ansible Playbooks (YAML) and mainframe (COBOL) code are both supported (Minor) languages for the generalized IBM Granite code models\u2014 and therefore are supported by IBM watsonx Code Assistant \u2014but if a client wishes to specialize in those particular languages and frameworks, they would be well advised to utilize the bespoke WCA for Ansible Lightspeed and WCA for Z offerings, respectively, to do so. iv. Solution architecture of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed IBM watsonx Code Assistant for Red Hat Ansible Lightspeed meets developers where they are: with a rich plugin via VS Code extensions, where developers input their prompts directly in the code editor. Prompts are sent to the Ansible Lightspeed service, and the service sends a suggestion back (a completion ) that\u2019s powered by IBM Granite LLMs for code. It is important to note that all data in transit is encrypted and ephemeral so users can be confident and have trust in the security of the service during this exchange. In terms of data security, client Ansible playbooks and customized models that they may potentially have are stored in client-owned Cloud Object Storage and are not shared with IBM, Red Hat, or any other clients. In order to utilize IBM watsonx Code Assistant for Red Hat Ansible Lightspeed, a client must have an existing license for Red Hat Ansible Automation Platform (the \u201dred tile\u201d component in the center of the diagram), as well as a license for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (the \u201cblue tile\u201d on the right of the diagram). Generative AI has recently demonstrated proficiency in creating syntactically correct and contextually relevant application code in a variety of programming languages. For example, if trained on a large dataset of Ansible Playbooks, generative AI models can be fine-tuned to understand the nuances of Playbook syntax and structure. An enterprise organization with dozens or hundreds of Playbooks within their IT estate today would have a rich corpus of training data on-hand that could be used to fine-tune AI models that are tailored to the automation needs and programming style or standards of that particular company. WHAT ARE PLAYBOOKS? Ansible Playbooks instruct Ansible\u2019s automation engine on how to execute tasks in a step-by-step manner. Playbooks defines roles, tasks, handlers, and other configurations; in turn, these attributes allow developers and users to codify complex orchestration scenarios. Conceptually, think of a Playbook as a recipe book for system administration: each recipe (or Playbook) spells out the steps required to achieve a particular system state or to complete a given operation. One of the standout features of Ansible Playbooks is that they are idempotent : executing Playbooks multiple times on the same system won't create additional \"side effects\" (unintended operations or creation of unwanted artifacts) after the first successful run. This ensures consistency and reliability across deployments of the Red Hat Ansible Automation Platform ( AAP ). As you will see throughout the hands-on training material, generative AI models provide a natural language prompt to users which in turn is understood and translated by the AI models into the necessary Ansible Task code. For example, a user might describe a desired system state in plain language ( \"I want a Playbook to install and start an Apache web server\" ) and the model will generate the appropriate Ansible Tasks for a Playbook. All of this is achieved without physically writing code or requiring much programming expertise. Not only does this speed up the automation process by cutting the time needed to author Playbooks, but it also democratizes access to automation in general. Even those within a company with limited Ansible or programming expertise will be able to produce effective Playbooks. There are plenty of caveats of course, and thorough validation and testing of AI-generated code will be needed before being put into production. However, the productivity gains and broadening of skillsets within an organization can be tremendous. And as a whole, generative AI brings the original goals of Red Hat Ansible Automation Platform (the democratization of automation for everything) that much closer to a reality. v. Lab objectives The material covered for this hands-on training is intended to prepare IBM sellers and business partners with the skills necessary to create Ansible automation tasks using the generative AI capabilities of WCA. The curriculum will leverage WCA's generative AI code recommendations for automating cloud-based and infrastructure-based automation tasks. In-depth explanations accompanying Ansible Playbook templates will also explain: How WCA uses natural language prompts , as well as Ansible Playbook contents, to generate contextually-aware Task code recommendations Post-processing capabilities that refine the generative AI suggestions into syntactically correct code (adherent to best practices) How WCA provides content source matching attribution and \"explainability\" for all AI-generated content Leveraging WCA's model tuning capabilities to tailor content and code recommendations to an organization's standards, best practices, and programming styles vi. Next steps The module ahead will outline the evaluation criteria for IBM sellers and business partners. Afterwards, you will setup your local environment with the necessary pre-requisites for getting started with the hands-on material.","title":"Introduction"},{"location":"#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) introduces IBM watsonx Code Assistant's generative AI capabilities and lays the groundwork for the hands-on training that will follow. [6 min]","title":""},{"location":"#i-automation-is-indispensable-to-modern-it-strategy","text":"Despite the innovations and advancements made in the domain of automation, IBM sellers and partners know first-hand from discussions with clients that many businesses are still struggling to keep up with their IT operations. The rapid pace of technological innovation\u2014 in particular, areas such as AI and machine learning \u2014are obviously challenging for any organization to strategize and plan around. But smaller, more practical challenges also stand in the way of these businesses. The fact remains that IT operations, and wrangling those operations in an efficient and streamlined manner, remains a difficult problem to solve. Three primary pain points that IBM consistently hears from the marketplace include: an ever-increasing skills gap in IT management; that Day 2 operations continue to be labor-intensive, mostly manual endeavors; and that the complexity of the systems needing to be managed are out-pacing many organization\u2019s ability to adapt. All of these pain points are potential automation challenges to be solved. Each of them impedes a company's ability to move quickly and adapt for the future. And as such, for many IBM clients, solving these automation challenges have become an indispensable element in their strategy to modernize IT. UNPRECEDENTED RATE OF GENERATIVE AI ADOPTION Even though generative AI is relatively new, the widespread popularity of ChatGPT has created significant interest in the notion of large language models (LLMs) and foundation models (FMs) \u2014 and what they can do for business. It took quite some time for enterprises to start moving toward traditional AI. In contrast, generative AI has experienced massive early adoption: 80% of enterprises are already working with, or planning to leverage FMs, and plan to adopt generative AI in their use cases and workflow. Moreover, the following data points to an ever-growing adoption trend for generative AI: Scale Zeitgeist 2023 AI Readiness Report notes that with the companies they reviewed, 21% have generative AI models in production; 29% are experimenting with generative AI and another 31% are planning to work with generative AI models; a total of 81% are either working with or planning to work with generative AI models Goldman Sachs has estimated that generative AI will have a very deep economic impact \u2013 raising global Gross Domestic Product (GDP) by 7% within 10 years, reflecting the technology\u2019s huge potential. Boston Consulting Group (BCG) noted that generative AI is expected to represent 30% of the overall market by 2025","title":"i. Automation is indispensable to modern IT strategy"},{"location":"#ii-generative-ai-assisted-code-lifecycle-management-achieves-what-llms-alone-cannot","text":"Following the debut of OpenAI's ChatGPT, the marketplace has been awash with competing large language model (LLM) and generative AI-based assistants. It's one thing to train and deploy an LLM; it's another thing entirely to make it applicable and tangibly beneficial for business. What separates IBM watsonx Code Assistant (WCA) offerings from competing vendors in the marketplace? The design and implementation of WCA is purposely built to assist, using generative AI, software and code lifecycle management. In short, generative AI-assisted code lifecycle management helps to achieve what large language models cannot achieve on their own. It is what distinguished WCA from other code assistants in the marketplace today. Code lifecycle management begins with understanding client code, through training across a myriad of programming languages and specializations in paradigms such as Ansible Automation Platform, and applies that understanding across a client\u2019s application and runtime environments. Users are able to plan next steps based on generative AI analysis of their existing application code. Operations teams can rapidly transform their codebases with optimized design and architecture that is recommended according to IBM Granite's best-practice models. Administrators can validate the outcomes with automatically generated unit tests. Afterwards, they can deploy those services and applications using automated processes like Ansible's automation engine. Over the course of that application or code's lifecycle, generative AI can maintain healthy operations with runtime insights.","title":"ii. Generative AI-assisted code lifecycle management achieves what LLMs alone cannot"},{"location":"#iii-introducing-the-ibm-watsonx-code-assistant-product-family","text":"IBM watsonx Code Assistant is the flagship offering in a suite of generative AI code assistant products, which also include offerings for Ansible Automation Platform (IBM watsonx Code Assistant for Red Hat Ansible Lightspeed) and IBM Systems modernization (IBM watsonx Code Assistant for Z). These solutions accelerate software development tasks with AI-powered capabilities including context-aware code generation, explanation, documentation, translation, and unit test generation. It does so while maintaining the principles of trust, security, and compliance with regards to IBM client's data and intellectual property. Developers and IT Operators can utilize WCA to speed up application modernization efforts and generate Ansible-based automation jobs to rapidly scale out (or scale up) IT environments. IBM watsonx Code Assistant products are powered by IBM Granite foundation models that include state-of-the-art large language models designed for code. For offerings such as WCA for Ansible Lightspeed and WCA for Z, bespoke code models\u2014 tailored to working with Ansible Automation Platform and COBOL-to-Z use cases, respectively \u2014are invoked. Universally true for all of the watsonx Code Assistant offerings is that they are geared towards helping IT teams create high-quality code using AI-generated recommendations, based on natural language requests or existing source code. These AI models, and the recommendations they generate, are seamlessly integrated via extensions with the world's most popular developer integrated development environments (IDE), including Visual Studio Code and Eclipse. Granite is IBM\u2019s flagship brand of open and proprietary LLMs, spanning multiple modalities. Granite models exist for code, languages, time series, and GeoSpatial \u2014 with additional modalities expected in future. IBM Granite code models are a series of decoder-only models for code generative tasks, trained with code written in 116 different programming languages. The Granite code models family consists of models ranging in size from 3 to 34 billion parameters, in both a base model and instruction-following model variants. These models have a range of uses, from complex application modernization tasks to on-device memory-constrained use cases. The larger the block size for a particular language on this chart, the larger percentage of training corpus data of that language was used to train the Granite code model. Languages and formats such as Java, C, JSON, JavaScript, HTML, and PHP are subjects in which the model \u201cMajors\u201d and excels. Other languages such as Ruby, SQL, and Swift could be considered \u201cMinors\u201d where the generalized code model can work with the language, but has less training data to base those recommendations on. These percentages and training data volumes will continue to evolve as the Granite code models mature. WATSONX CODE ASSISTANT vs. WCA FOR ANSIBLE LIGHTSPEED? For those familiar with other IBM watsonx Code Assistant offerings\u2014 such as WCA for Red Hat Ansible Lightspeed and WCA for Z \u2014the generalized code model approach, as seen here, differs from the specialized code model approach of those two aforementioned offerings. The WCA for Ansible Lightspeed flavor of IBM Granite code models specializes (Majors) only in Ansible Playbooks and YAML Similarly, the IBM Granite code model used by WCA for Z specializes in transforming COBOL mainframe code into modernized Java code Ansible Playbooks (YAML) and mainframe (COBOL) code are both supported (Minor) languages for the generalized IBM Granite code models\u2014 and therefore are supported by IBM watsonx Code Assistant \u2014but if a client wishes to specialize in those particular languages and frameworks, they would be well advised to utilize the bespoke WCA for Ansible Lightspeed and WCA for Z offerings, respectively, to do so.","title":"iii. Introducing the IBM watsonx Code Assistant product family"},{"location":"#iv-solution-architecture-of-ibm-watsonx-code-assistant-for-red-hat-ansible-lightspeed","text":"IBM watsonx Code Assistant for Red Hat Ansible Lightspeed meets developers where they are: with a rich plugin via VS Code extensions, where developers input their prompts directly in the code editor. Prompts are sent to the Ansible Lightspeed service, and the service sends a suggestion back (a completion ) that\u2019s powered by IBM Granite LLMs for code. It is important to note that all data in transit is encrypted and ephemeral so users can be confident and have trust in the security of the service during this exchange. In terms of data security, client Ansible playbooks and customized models that they may potentially have are stored in client-owned Cloud Object Storage and are not shared with IBM, Red Hat, or any other clients. In order to utilize IBM watsonx Code Assistant for Red Hat Ansible Lightspeed, a client must have an existing license for Red Hat Ansible Automation Platform (the \u201dred tile\u201d component in the center of the diagram), as well as a license for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (the \u201cblue tile\u201d on the right of the diagram). Generative AI has recently demonstrated proficiency in creating syntactically correct and contextually relevant application code in a variety of programming languages. For example, if trained on a large dataset of Ansible Playbooks, generative AI models can be fine-tuned to understand the nuances of Playbook syntax and structure. An enterprise organization with dozens or hundreds of Playbooks within their IT estate today would have a rich corpus of training data on-hand that could be used to fine-tune AI models that are tailored to the automation needs and programming style or standards of that particular company. WHAT ARE PLAYBOOKS? Ansible Playbooks instruct Ansible\u2019s automation engine on how to execute tasks in a step-by-step manner. Playbooks defines roles, tasks, handlers, and other configurations; in turn, these attributes allow developers and users to codify complex orchestration scenarios. Conceptually, think of a Playbook as a recipe book for system administration: each recipe (or Playbook) spells out the steps required to achieve a particular system state or to complete a given operation. One of the standout features of Ansible Playbooks is that they are idempotent : executing Playbooks multiple times on the same system won't create additional \"side effects\" (unintended operations or creation of unwanted artifacts) after the first successful run. This ensures consistency and reliability across deployments of the Red Hat Ansible Automation Platform ( AAP ). As you will see throughout the hands-on training material, generative AI models provide a natural language prompt to users which in turn is understood and translated by the AI models into the necessary Ansible Task code. For example, a user might describe a desired system state in plain language ( \"I want a Playbook to install and start an Apache web server\" ) and the model will generate the appropriate Ansible Tasks for a Playbook. All of this is achieved without physically writing code or requiring much programming expertise. Not only does this speed up the automation process by cutting the time needed to author Playbooks, but it also democratizes access to automation in general. Even those within a company with limited Ansible or programming expertise will be able to produce effective Playbooks. There are plenty of caveats of course, and thorough validation and testing of AI-generated code will be needed before being put into production. However, the productivity gains and broadening of skillsets within an organization can be tremendous. And as a whole, generative AI brings the original goals of Red Hat Ansible Automation Platform (the democratization of automation for everything) that much closer to a reality.","title":"iv. Solution architecture of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed"},{"location":"#v-lab-objectives","text":"The material covered for this hands-on training is intended to prepare IBM sellers and business partners with the skills necessary to create Ansible automation tasks using the generative AI capabilities of WCA. The curriculum will leverage WCA's generative AI code recommendations for automating cloud-based and infrastructure-based automation tasks. In-depth explanations accompanying Ansible Playbook templates will also explain: How WCA uses natural language prompts , as well as Ansible Playbook contents, to generate contextually-aware Task code recommendations Post-processing capabilities that refine the generative AI suggestions into syntactically correct code (adherent to best practices) How WCA provides content source matching attribution and \"explainability\" for all AI-generated content Leveraging WCA's model tuning capabilities to tailor content and code recommendations to an organization's standards, best practices, and programming styles","title":"v. Lab objectives"},{"location":"#vi-next-steps","text":"The module ahead will outline the evaluation criteria for IBM sellers and business partners. Afterwards, you will setup your local environment with the necessary pre-requisites for getting started with the hands-on material.","title":"vi. Next steps"},{"location":"advanced/","text":"OPTIONAL SECTION \u2014 NOT REQUIRED FOR LEVEL 3 CREDIT IBM technical sellers and business partners are NOT required to complete this section to receive Level 3 credit. However, it is highly recommended by the authors that you attempt the material, given the tremendous demonstration value it provides for clients. COSTS & BILLING Be aware that deploying a live Amazon EC2 instance on AWS does incur real-world charges which are the responsibility of the individual to pay for. If you wish to go ahead with deploying the live environment\u2014 which you are encouraged, but NOT REQUIRED, to complete \u2014you will need to do so on your personal AWS account, with any potential billings and charges incurred to your personal credit card. The billing associated with an Amazon EC2 t2.small instance and other services needed for the demo is minimal : an on-demand hourly rate of $0.023 USD at the time of publication. Running the fully-deployed WordPress and EC2 environments for 24 hours cost USD $2.03 in billing when testing for publication. Additional AWS pricing details and plans are available online. To conclude the Level 3 Technical Sales education on IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will be provided with a complete set of instructions for deploying a live Amazon Web Services (AWS) environment with Ansible automation, consisting of a WordPress web application running atop an Amazon EC2 VPC . The deployment of the environments will be conducted entirely using Red Hat Ansible automation, executed on your local machine, and deployed to AWS cloud. The generation of the Ansible Tasks and code required to automate these deployments will be created using the generative AI capabilities of the WCA extension for VS Code. Ansible Playbooks to support this demo were prepared ahead of time in a public GitHub repository and are available for your use at the links below. TEMPLATE ( .yml ) : https://github.com/bienko/WCA-Lightspeed-L3/blob/main/TEMPLATE.yml The TEMPLATE.yml Playbook contains the skeleton of the application you will ultimately build. Playbook structure, key variables, Task descriptions, and comments are provided. Use the generative AI capabilities of WCA to fill out the details of the various Ansible Tasks, then compare the results with the SOLUTION.yml file below. SOLUTION ( .yml ) : https://github.com/bienko/WCA-Lightspeed-L3/blob/main/SOLUTION.yml The SOLUTION.yml Playbook contains the complete set of Ansible Playbook instructions needed to fully deploy a live EC2 VPC instance on AWS and host a WordPress application within that environment. You will need to replace certain variables (indicated by comments in the code) with details specific to your personal AWS account and environment. Afterwards, you can\u2014 and are encouraged to \u2014execute the Playbook with Ansible and deploy a live , at-cost instance to your personal AWS account. Once the two YAML files have been downloaded to your local machine, open them with VS Code and follow along with the instructions below. DOWNLOAD .yml FILES TO DESKTOP You can easily downlod the TEMPLATE and SOLUTION .yml files to your local machine. Navigate to the GitHub links provided (above) and click the Download raw file button. HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible Use WCA to generate code for the TEMPLATE.yml Playbook, then compare against the SOLUTION.yml Playbook to evaluate the results In the steps ahead, you will use WCA's AI-generated code suggestions to populate the contents of Tasks within the TEMPLATE.yml Playbook. There are 14 Tasks in total and we will examine them each in sequence. For each Task, a tabbed code snippet is provided within this lab documentation to make it easy for you to toggle between the TEMPLATE (the Task prior to AI-generated code suggestions) and the proper SOLUTION (the Task as it should be written for a successful deployment to AWS). Compare the AI-generated code suggestions you receive in VS Code with the expected SOLUTION code. You will find that there are occassional discrepancies and variables that need to be adjusted. Generative AI can save users tremendous amounts of time and effort by automating the creation of powerful blocks of code\u2014 as you will observe in generating these tasks \u2014but it still requires a degree of human supervision and double-checking. However, the time saved with hands on keyboard is enormous and the potential for the future is boundless. Open the TEMPLATE.yml Playbook in VS Code and ensure that the Ansible Lightspeed extension is active. Scroll down MODULE 1 ( Line 4 ) and take note of the comments, as well as the <PLACEHOLDER> values. If you wish to fully deploy this environment to AWS, you will need to register for an AWS account, create an IAM (Identity Access Management) user, and associate an SSH key with the necessary permissions to access the account. These values will then be substituted in your code for the aws_access_key and aws_secret_key variables. Instructions on how to sign up for AWS and create those keys will be included at the end of this module, and you will be reminded at that time to update these variables. However, for now, let's focus on generating the Ansible Tasks. Continue scrolling down until you reach TASK 1 ( Line 18 ) for the creation of a virtual private cloud (VPC) named 'wordpress'. Generate the Task code within VS Code and compare your results with the SOLUTION tab. TEMPLATE SOLUTION # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress amazon.aws.ec2_vpc_net : name : wordpress cidr_block : 10.0.0.0/16 state : present register : wordpress_vpc TAGS One element of AI-generated Task code that you may find yourself frequently needing to update are tags and other naming conventions, such as the register: wordpress_vpc pair in TASK 1 . While the Ansible Lightspeed extension will remember the context of variables that are already named within the Playbook, it will struggle with how to name new variables \u2014 like wordpress_vpc . Take note, however, that as you generate code for the other Tasks in the Playbook that wordpress_vpc will be recalled and referenced. After validating the previous code block, continue scrolling down until you reach TASK 2 ( Line 23 ) for the creation of a security group which allows network traffic over SSH and HTTP/s. Generate, evaluate, and refine the Task code as needed. Notice that the wordpress_vpc attribute which was registered in TASK 2 has been referenced in the creation of the vpc_id in TASK 3 Ignore the additional space between adjacent {``{ characters that are included in SOLUTION code tabs \u2014 these are included only to allow rendering of the code block in GitHub and are not intended as part of the finalized code TEMPLATE SOLUTION # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP amazon.aws.ec2_security_group : name : wordpress description : Allow SSH and HTTP/HTTPS vpc_id : \"{ { wordpress_vpc.vpc.id } }\" # Ignore extra space between { { cbaracters. rules : - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 443 to_port : 443 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 register : wordpress_sg Continue scrolling down until you reach TASK 3 ( Line 28 ) for the creation of an internet gateway for the wordpress VPC. Generate, evaluate, and refine the Task code as needed. TEMPLATE SOLUTION # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ { wordpress_vpc.vpc.id } }\" state : present register : wordpress_igw Continue down to TASK 4 ( Line 32 ) for the creation of an AWS network subnet for the WordPress VPC. TEMPLATE SOLUTION # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc amazon.aws.ec2_vpc_subnet : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" cidr : 10.0.0.0/16 register : wordpress_subnet Continue down to TASK 5 ( Line 36 ) for the creation of a routing table associated with wordpress VPC's subnet and internet gateway. TEMPLATE SOLUTION # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw amazon.aws.ec2_vpc_route_table : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" tags : Name : sg_public Project : phoenix subnets : - \"{ { wordpress_subnet.subnet.id } }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ { wordpress_igw.gateway_id } }\" register : wordpress_route_table WCA can be \"creative\" with its choice of tags and identifiers. You are welcome to adjust these to your liking, or even eliminate them from the Task code altogether. In the majority of cases these tags and identifiers are used for organization and tracking purposes, but not necessarily anything critical to the functioning of the Ansible Playbook or the infrastructure it is automating. Continue further down until you reach TASK 6 ( Line 40 ), which is one of the more consequential steps in the Playbook as it involves the creation of the Amazon EC2 t2.small instance on AWS according to the settings defined in TASKS 1-5 . Note that the image_id and ami values correspond to a specific Amazon Machine Image (ami) for Red Hat Enterprise Linux 9. Hundreds of alternative operating system images are available to select from on AWS, so you may modify this if you so choose; however, for the purposes of deploying WordPress it will make no meaningful difference. The pair key_name: \"bienko-key\" relates to an EC2 instance key (SSH) that you will need to create before attempting to deploy an EC2 instance via Ansible. Instructions for how to create that key will be given at the end of this module for those who wish to deploy a live environment. TEMPLATE SOLUTION # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip amazon.aws.ec2_instance : name : wordpress image_id : \"ami-026ebd4cfe2c043b2\" instance_type : t2.small vpc_subnet_id : \"{ { wordpress_subnet.subnet.id } }\" security_groups : \"{ { wordpress_sg.group_id } }\" network : assign_public_ip : true key_name : \"bienko-key\" state : running register : wordpress_server Continue down to TASK 7 ( Line 48 ). You will notice that in both the TEMPLATE and SOLUTION versions of the Playbook that this Task is fully defined already. For now, leave this as a hard-coded element within your Playbook and do not attempt to recreate the task with code generation. TEMPLATE SOLUTION # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" PLAYBOOK MODIFICATIONS REQUIRED Although TASK 7 does not require WCA to generate any code suggestions, it does require the user to modify the value of the ansible_ssh_private_key_file attribute. Set this attribute to the directory path where the private SSH key ( .pem file) is located on your local machine. Be sure to include the .pem filetype as part of the value and also ensure you have changed the permissions on the file (using Terminal and executing the chmod 400 bienko-key.pem instruction) before attempting to run the Playbook. Scrolling further down the Playbook you will find MODULE B ( Line 60 ), made up Ansible Tasks concerned with installing and configuring WordPress on the newly-provisioned AWS EC2 (VPC) instance. Locate TASK 8 ( Line 67 ) which waits for a connection to be established to the EC2 instance before advancing the sequence of automation tasks. TEMPLATE SOLUTION # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection ansible.builtin.wait_for_connection : delay : 10 timeout : 30 Next, continue further down the Playbook and locate TASK 9 ( Line 71 ) which commences installation of the necessary drivers and services on Amazon EC2 once a connection to the instance has been established. TEMPLATE SOLUTION # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server ansible.builtin.package : name : - httpd - php - php-mysqlnd - mariadb-server state : present Scroll down further until you locate TASK 10 ( Line 75 ) which is responsible for downloading and decompressing (unarchiving) WordPress installation contents onto the EC2 instance. TEMPLATE SOLUTION # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress ansible.builtin.unarchive : src : https://wordpress.org/latest.tar.gz dest : /var/www/html remote_src : true creates : /var/www/html/wordpress Afterwards, locate TASK 11 ( Line 79 ) which sets the ownership attributes for the WordPress environment on Amazon EC2 and prepares it for hosting via Apache webserver. TEMPLATE SOLUTION # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache ansible.builtin.file : path : /var/www/html owner : apache group : apache recurse : true Further down the Playbook, locate TASK 12 ( Line 83 ) which deploys the services installed in TASK 9 . TEMPLATE SOLUTION # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services ansible.builtin.service : name : \"{ { item } }\" state : started enabled : true loop : - httpd - php-fpm - mariadb MODULE C ( Line 87 ) of the Playbook covers the final set of operations. It is responsible for ensuring that WordPress is deployed on the newly-provisioned AWS EC2 (VPC) instance and that the WordPress web page is accessible to the user. As you did previously in Step 2 , you will need to replace the <PLACEHOLDER> values and region details on Lines 96-98 if you want to fully deploy this application into a live AWS environment. Locate TASK 13 ( Line 101 ) which is responsible for gathering facts about AWS EC2 (VPC) instance and the deployed WordPress application that is running atop of it. TEMPLATE SOLUTION # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress amazon.aws.ec2_instance_info : filters : tag:Name : wordpress instance-state-name : running register : ec2_facts Finally, you have arrived at the final step in the Playbook configuration: TASK 14 ( Line 105 ), which is responsible for hosting the WordPress web app and making it accessible to users. TEMPLATE SOLUTION # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name ansible.builtin.debug : msg : \"{ { ec2_facts.instances[0].public_dns_name } }/wordpress/readme.html\" To deploy a live Amazon EC2 instance and WordPress application on AWS, follow along with the steps below. Congratulations on making it this far! At this stage, your Playbook is ready for execution. However, there are still some preparation that you need to do before this Playbook can be used to automate deployment of the WordPress application into a live Amazon EC2 environment. In order to execute the Ansible Playbook on your local machine, you will need to install Ansible locally. Red Hat provides extensive online documentation for how to go about installing Ansible The authors of your lab guide recommend doing so by executing the following instructions in a Terminal window (macOS), or equivalent steps using PowerShell (Windows) Follow the prompts until you receive confirmation that Ansible is installed brew install ansible HOMEBREW IS MISSING \u2014 CLICK TO EXPAND If you do not have the Homebrew package manager (\"brew\") installed in a macOS environment, execute the following instruction inside a Terminal console: /bin/bash -c \"$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)\" Wait for the installation to conclude and then retry the instructions from Step 18. If you have not registered for an AWS account previously, you need to do so now. You will need to supply a personal credit card for billing expenses If you have an existing Amazon account, your AWS account can be linked to that profile if you so choose AWS Registration : https://aws.amazon.com/resources/create-account/ With a registered AWS account, you next need to navigate from the AWS dashboard into the Identity and Access Management (IAM) tool. You can search for this using the Search bar at the top of the screen and entering IAM . Once inside the IAM tool, look along the left-side of the interface and drill down into Users . Click the Create User orange button located in the top-right corner of the interface. Within the Create User panel, specify your user details and permissions. Set Username to a name of your choosing. Click Next to advance the configuration tool. Under Set Permissions > Permission Options , select the Add user to group tile. Click the Set permissions boundary - optional tile to expand permission options. Toggle the checkmark next to the Use a permissions boundary to control the maximum permissions option. Under Permissions Policies , search the table for the AdministratorAccess policy and enable that policy. You may receive warnings against doing so \u2014 ignore those for now. Continue following the prompts until the User has been created. Once the User has been created, your web browser will reopen the User panel of the IAM tool. Here you can see various details about the User, including their Security Credentials and any keys associated with the User. This section will be empty given that you have only just now created the User. Drill down into the Access Keys category and click the Create Access Key button located below the (empty) table. A secret access key will be created and the web browser will reload to show you details about that key. Record this information to a notepad. Details about the secret access key can only be viewed or downloaded at the time the key is created (now). It cannot be recovered afterwards. However, if you lose your key details, all is not lost \u2014 you can always create a new secret access key at a later time. Record the Access Key , which will be used to replace the <PLACEHOLDER> value assigned to aws_access_key in Step 2 and Step 15 of this lab guide. Record the Secret Access Key , which will replace the <PLACEHOLDER> value assigned to aws_secret_key in Step 2 and Step 15 of this lab guide. Take note at this time as well about the Region in which your AWS account is located. You can easily look up this information by examining the URL in your web browser address bar, as shown below ( us-east-1 ). Record the region's name, as it will replace the region value assigned to the Playbook. PLAYBOOK MODIFICATIONS REQUIRED Replace the <PLACEHOLDER> values for variables aws_access_key and aws_secret_key with the Access Key and Secret Access Key , respectively. Lines 13, 14, 96, and 97 of the TEMPLATE.yml Playbook Lines 13, 14, 179, and 180 of the SOLUTION.yml Playbook Replace the value for variable region with the region assigned to your unique AWS account. Lines 15 and 98 of the TEMPLATE.yml Playbook Lines 15 and 181 of the SOLUTION.yml Playbook Next, you'll need to create an SSH Key Pair with authorization to interact with (and ultimately deploy) Amazon EC2 instances. Return to the search bar at the top of the AWS interface and enter EC2 . Drill down into the Amazon EC2 service page from the results. From the left-hand navigation bar, drill down into the Network & Security > Key Pairs tab. Click the orange Create key pair button in the top-right of the page. A configuration tool will ask for additional information about the Key Pair: Name can be set to a value of your choosing Key pair type set to RSA Private key file format set to .pem When satisfied, click the orange Create key pair button The web browser will redirect to a page where you can view details about the newly-created Key Pair for EC2. Download the .pem file to your local device, preferably to the same directory that your TEMPLATE.yml and SOLUTION.yml Playbooks are located within. PLAYBOOK MODIFICATIONS REQUIRED The value of ansible_ssh_private_key_file must be updated within your Playbooks to account for the full path to the .pem file on your local machine. If you save the Key Pair file to the same directory that your Ansible Playbooks are located within, you do not need to qualify the value with anything more than the name of the file ( bienko-key.pem ). If it is saved in a different directory, you will need to spell out the full path + file name. Update the value of ansible_ssh_private_key_file in the following locations: Line 58 of the TEMPLATE.yml Playbook Line 113 of the SOLUTION.yml Playbook It is necessary to adjust the permissions assigned to the Key Pair file (on your local machine) before attempting to connect via Ansible to AWS. Open a Terminal console on your local machine and navigate ( cd ) to the directory in which the Key Pair file is saved. Execute the following to adjust the permissions assigned to the .pem file: chmod 400 filename.pem Executing the Ansible Playbook At last, you are ready to execute the Playbook and use Ansible automation to deploy both an Amazon EC2 instance and a WordPress web application. With the Terminal console, navigate to the directory ( cd ) where both the Ansible Playbook and your KeyPair.pem file are located. You may use either the TEMPLATE.yml Playbook that you customized or the SOLUTION.yml Playbook for executing the Ansible automation tasks. Regardless of the Playbook being used, you must modify the Playbook with the necessary AWS security and account credentials information. When ready, execute the following instruction with the Terminal console: ansible-playbook TEMPLATE.yml Monitor the console's output as the Ansible Playbook iterates across each of the Tasks customized earlier. A fantastic feature of Ansible's automation is that if any of the Tasks should fail, the steps involved in that Task can be re-tried in subsequent runs; likewise, Tasks which executed successfully won't be repeated, but are validated and skipped instead. If you encounter errors during the litany of automation tasks, return to the Playbook and attempt to debug the issue. Re-issue the instruction in Step 34 to console to attempt the Ansible Playbook execution again. Continue iterating in this manner until all 14 Tasks have been successfully executed. After successfully deploying the Amazon EC2 environment and instantiating a WordPress application, the console will return a URL to the Terminal's output as part of TASK 14 's [debug] instruction. Copy the URL to clipboard and then paste it into your web browser to access the WordPress application. If your web browser loads the quick-start page for WordPress, you have successfully deployed the application and its supporting EC2 environment \u2014 entirely using generative AI and Ansible automation. Well done! Cleaning up and conclusion COSTS & BILLING It is essential that you terminate and deprovision the EC2 instance at the conclusion of the hands on lab. Otherwise, your personal AWS account will continue to be billed for as long as the resources are provisioned. Once you are ready to retire your environment, you will need to return to the AWS Dashboard to terminate the EC2 instance and hosted WordPress application. You can monitor your Amazon EC2 instance from the EC2 dashboard. At the top of the page, locate the Resources table and drill down into the Instances (running) tab. Details about your EC2 environment are summarized within the Instances table. Click the name of your EC2's Instance ID to expose additional details and management options. In the top-right corner of the interface, click the Instance state button and select Terminate instance from the drop-down menu. Confirm the deprovisioning request by clicking the orange Terminate button. All storage associated with the EC2 instance will automatically be deleted. Terminating an instance cannot be undone once underway. It is recommended that you delete all of the Security Groups (3 total) and Key Pairs (1 total) associated with the EC2 instance. To do so, return to the EC2 Dashboard and follow a similar procedure to Steps 38-41 . Search for IAM in the AWS taskbar to return to the Identity Access Management panel. If you wish to further secure your AWS account, delete all unwanted users or consider reducing the user's privilege from AdministratorAccess to a more restrictive class. That concludes the hands-on components to this Level 3 course, but your learning and experimentation doesn't need to end here. Continue to experiment with generating Tasks for Ansible Playbooks. If you have worked extensively with Ansible previously, try recreating Tasks you've written previously from old Playbooks using generative AI \u2014 just be careful not to use confidential or sensitive information as part of those tests. See if you can spot the differences or improvements made from the AI-generated code recommendations. Feel free to reach out to the authors of this coursework if you have suggestions for Tasks or code generation techniques that you'd like to see included in future iterations of this hands-on training.","title":"Advanced"},{"location":"advanced/#_1","text":"","title":""},{"location":"advanced/#use-wca-to-generate-code-for-the-templateyml-playbook-then-compare-against-the-solutionyml-playbook-to-evaluate-the-results","text":"","title":"Use WCA to generate code for the TEMPLATE.yml Playbook, then compare against the SOLUTION.yml Playbook to evaluate the results"},{"location":"advanced/#_2","text":"In the steps ahead, you will use WCA's AI-generated code suggestions to populate the contents of Tasks within the TEMPLATE.yml Playbook. There are 14 Tasks in total and we will examine them each in sequence. For each Task, a tabbed code snippet is provided within this lab documentation to make it easy for you to toggle between the TEMPLATE (the Task prior to AI-generated code suggestions) and the proper SOLUTION (the Task as it should be written for a successful deployment to AWS). Compare the AI-generated code suggestions you receive in VS Code with the expected SOLUTION code. You will find that there are occassional discrepancies and variables that need to be adjusted. Generative AI can save users tremendous amounts of time and effort by automating the creation of powerful blocks of code\u2014 as you will observe in generating these tasks \u2014but it still requires a degree of human supervision and double-checking. However, the time saved with hands on keyboard is enormous and the potential for the future is boundless. Open the TEMPLATE.yml Playbook in VS Code and ensure that the Ansible Lightspeed extension is active. Scroll down MODULE 1 ( Line 4 ) and take note of the comments, as well as the <PLACEHOLDER> values. If you wish to fully deploy this environment to AWS, you will need to register for an AWS account, create an IAM (Identity Access Management) user, and associate an SSH key with the necessary permissions to access the account. These values will then be substituted in your code for the aws_access_key and aws_secret_key variables. Instructions on how to sign up for AWS and create those keys will be included at the end of this module, and you will be reminded at that time to update these variables. However, for now, let's focus on generating the Ansible Tasks. Continue scrolling down until you reach TASK 1 ( Line 18 ) for the creation of a virtual private cloud (VPC) named 'wordpress'. Generate the Task code within VS Code and compare your results with the SOLUTION tab. TEMPLATE SOLUTION # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress amazon.aws.ec2_vpc_net : name : wordpress cidr_block : 10.0.0.0/16 state : present register : wordpress_vpc TAGS One element of AI-generated Task code that you may find yourself frequently needing to update are tags and other naming conventions, such as the register: wordpress_vpc pair in TASK 1 . While the Ansible Lightspeed extension will remember the context of variables that are already named within the Playbook, it will struggle with how to name new variables \u2014 like wordpress_vpc . Take note, however, that as you generate code for the other Tasks in the Playbook that wordpress_vpc will be recalled and referenced. After validating the previous code block, continue scrolling down until you reach TASK 2 ( Line 23 ) for the creation of a security group which allows network traffic over SSH and HTTP/s. Generate, evaluate, and refine the Task code as needed. Notice that the wordpress_vpc attribute which was registered in TASK 2 has been referenced in the creation of the vpc_id in TASK 3 Ignore the additional space between adjacent {``{ characters that are included in SOLUTION code tabs \u2014 these are included only to allow rendering of the code block in GitHub and are not intended as part of the finalized code TEMPLATE SOLUTION # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP amazon.aws.ec2_security_group : name : wordpress description : Allow SSH and HTTP/HTTPS vpc_id : \"{ { wordpress_vpc.vpc.id } }\" # Ignore extra space between { { cbaracters. rules : - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 443 to_port : 443 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 register : wordpress_sg Continue scrolling down until you reach TASK 3 ( Line 28 ) for the creation of an internet gateway for the wordpress VPC. Generate, evaluate, and refine the Task code as needed. TEMPLATE SOLUTION # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ { wordpress_vpc.vpc.id } }\" state : present register : wordpress_igw Continue down to TASK 4 ( Line 32 ) for the creation of an AWS network subnet for the WordPress VPC. TEMPLATE SOLUTION # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc amazon.aws.ec2_vpc_subnet : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" cidr : 10.0.0.0/16 register : wordpress_subnet Continue down to TASK 5 ( Line 36 ) for the creation of a routing table associated with wordpress VPC's subnet and internet gateway. TEMPLATE SOLUTION # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw amazon.aws.ec2_vpc_route_table : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" tags : Name : sg_public Project : phoenix subnets : - \"{ { wordpress_subnet.subnet.id } }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ { wordpress_igw.gateway_id } }\" register : wordpress_route_table WCA can be \"creative\" with its choice of tags and identifiers. You are welcome to adjust these to your liking, or even eliminate them from the Task code altogether. In the majority of cases these tags and identifiers are used for organization and tracking purposes, but not necessarily anything critical to the functioning of the Ansible Playbook or the infrastructure it is automating. Continue further down until you reach TASK 6 ( Line 40 ), which is one of the more consequential steps in the Playbook as it involves the creation of the Amazon EC2 t2.small instance on AWS according to the settings defined in TASKS 1-5 . Note that the image_id and ami values correspond to a specific Amazon Machine Image (ami) for Red Hat Enterprise Linux 9. Hundreds of alternative operating system images are available to select from on AWS, so you may modify this if you so choose; however, for the purposes of deploying WordPress it will make no meaningful difference. The pair key_name: \"bienko-key\" relates to an EC2 instance key (SSH) that you will need to create before attempting to deploy an EC2 instance via Ansible. Instructions for how to create that key will be given at the end of this module for those who wish to deploy a live environment. TEMPLATE SOLUTION # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip amazon.aws.ec2_instance : name : wordpress image_id : \"ami-026ebd4cfe2c043b2\" instance_type : t2.small vpc_subnet_id : \"{ { wordpress_subnet.subnet.id } }\" security_groups : \"{ { wordpress_sg.group_id } }\" network : assign_public_ip : true key_name : \"bienko-key\" state : running register : wordpress_server Continue down to TASK 7 ( Line 48 ). You will notice that in both the TEMPLATE and SOLUTION versions of the Playbook that this Task is fully defined already. For now, leave this as a hard-coded element within your Playbook and do not attempt to recreate the task with code generation. TEMPLATE SOLUTION # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" PLAYBOOK MODIFICATIONS REQUIRED Although TASK 7 does not require WCA to generate any code suggestions, it does require the user to modify the value of the ansible_ssh_private_key_file attribute. Set this attribute to the directory path where the private SSH key ( .pem file) is located on your local machine. Be sure to include the .pem filetype as part of the value and also ensure you have changed the permissions on the file (using Terminal and executing the chmod 400 bienko-key.pem instruction) before attempting to run the Playbook. Scrolling further down the Playbook you will find MODULE B ( Line 60 ), made up Ansible Tasks concerned with installing and configuring WordPress on the newly-provisioned AWS EC2 (VPC) instance. Locate TASK 8 ( Line 67 ) which waits for a connection to be established to the EC2 instance before advancing the sequence of automation tasks. TEMPLATE SOLUTION # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection ansible.builtin.wait_for_connection : delay : 10 timeout : 30 Next, continue further down the Playbook and locate TASK 9 ( Line 71 ) which commences installation of the necessary drivers and services on Amazon EC2 once a connection to the instance has been established. TEMPLATE SOLUTION # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server ansible.builtin.package : name : - httpd - php - php-mysqlnd - mariadb-server state : present Scroll down further until you locate TASK 10 ( Line 75 ) which is responsible for downloading and decompressing (unarchiving) WordPress installation contents onto the EC2 instance. TEMPLATE SOLUTION # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress ansible.builtin.unarchive : src : https://wordpress.org/latest.tar.gz dest : /var/www/html remote_src : true creates : /var/www/html/wordpress Afterwards, locate TASK 11 ( Line 79 ) which sets the ownership attributes for the WordPress environment on Amazon EC2 and prepares it for hosting via Apache webserver. TEMPLATE SOLUTION # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache ansible.builtin.file : path : /var/www/html owner : apache group : apache recurse : true Further down the Playbook, locate TASK 12 ( Line 83 ) which deploys the services installed in TASK 9 . TEMPLATE SOLUTION # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services ansible.builtin.service : name : \"{ { item } }\" state : started enabled : true loop : - httpd - php-fpm - mariadb MODULE C ( Line 87 ) of the Playbook covers the final set of operations. It is responsible for ensuring that WordPress is deployed on the newly-provisioned AWS EC2 (VPC) instance and that the WordPress web page is accessible to the user. As you did previously in Step 2 , you will need to replace the <PLACEHOLDER> values and region details on Lines 96-98 if you want to fully deploy this application into a live AWS environment. Locate TASK 13 ( Line 101 ) which is responsible for gathering facts about AWS EC2 (VPC) instance and the deployed WordPress application that is running atop of it. TEMPLATE SOLUTION # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress amazon.aws.ec2_instance_info : filters : tag:Name : wordpress instance-state-name : running register : ec2_facts Finally, you have arrived at the final step in the Playbook configuration: TASK 14 ( Line 105 ), which is responsible for hosting the WordPress web app and making it accessible to users. TEMPLATE SOLUTION # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name ansible.builtin.debug : msg : \"{ { ec2_facts.instances[0].public_dns_name } }/wordpress/readme.html\"","title":""},{"location":"advanced/#to-deploy-a-live-amazon-ec2-instance-and-wordpress-application-on-aws-follow-along-with-the-steps-below","text":"","title":"To deploy a live Amazon EC2 instance and WordPress application on AWS, follow along with the steps below."},{"location":"advanced/#_3","text":"Congratulations on making it this far! At this stage, your Playbook is ready for execution. However, there are still some preparation that you need to do before this Playbook can be used to automate deployment of the WordPress application into a live Amazon EC2 environment. In order to execute the Ansible Playbook on your local machine, you will need to install Ansible locally. Red Hat provides extensive online documentation for how to go about installing Ansible The authors of your lab guide recommend doing so by executing the following instructions in a Terminal window (macOS), or equivalent steps using PowerShell (Windows) Follow the prompts until you receive confirmation that Ansible is installed brew install ansible HOMEBREW IS MISSING \u2014 CLICK TO EXPAND If you do not have the Homebrew package manager (\"brew\") installed in a macOS environment, execute the following instruction inside a Terminal console: /bin/bash -c \"$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)\" Wait for the installation to conclude and then retry the instructions from Step 18. If you have not registered for an AWS account previously, you need to do so now. You will need to supply a personal credit card for billing expenses If you have an existing Amazon account, your AWS account can be linked to that profile if you so choose AWS Registration : https://aws.amazon.com/resources/create-account/ With a registered AWS account, you next need to navigate from the AWS dashboard into the Identity and Access Management (IAM) tool. You can search for this using the Search bar at the top of the screen and entering IAM . Once inside the IAM tool, look along the left-side of the interface and drill down into Users . Click the Create User orange button located in the top-right corner of the interface. Within the Create User panel, specify your user details and permissions. Set Username to a name of your choosing. Click Next to advance the configuration tool. Under Set Permissions > Permission Options , select the Add user to group tile. Click the Set permissions boundary - optional tile to expand permission options. Toggle the checkmark next to the Use a permissions boundary to control the maximum permissions option. Under Permissions Policies , search the table for the AdministratorAccess policy and enable that policy. You may receive warnings against doing so \u2014 ignore those for now. Continue following the prompts until the User has been created. Once the User has been created, your web browser will reopen the User panel of the IAM tool. Here you can see various details about the User, including their Security Credentials and any keys associated with the User. This section will be empty given that you have only just now created the User. Drill down into the Access Keys category and click the Create Access Key button located below the (empty) table. A secret access key will be created and the web browser will reload to show you details about that key. Record this information to a notepad. Details about the secret access key can only be viewed or downloaded at the time the key is created (now). It cannot be recovered afterwards. However, if you lose your key details, all is not lost \u2014 you can always create a new secret access key at a later time. Record the Access Key , which will be used to replace the <PLACEHOLDER> value assigned to aws_access_key in Step 2 and Step 15 of this lab guide. Record the Secret Access Key , which will replace the <PLACEHOLDER> value assigned to aws_secret_key in Step 2 and Step 15 of this lab guide. Take note at this time as well about the Region in which your AWS account is located. You can easily look up this information by examining the URL in your web browser address bar, as shown below ( us-east-1 ). Record the region's name, as it will replace the region value assigned to the Playbook. PLAYBOOK MODIFICATIONS REQUIRED Replace the <PLACEHOLDER> values for variables aws_access_key and aws_secret_key with the Access Key and Secret Access Key , respectively. Lines 13, 14, 96, and 97 of the TEMPLATE.yml Playbook Lines 13, 14, 179, and 180 of the SOLUTION.yml Playbook Replace the value for variable region with the region assigned to your unique AWS account. Lines 15 and 98 of the TEMPLATE.yml Playbook Lines 15 and 181 of the SOLUTION.yml Playbook Next, you'll need to create an SSH Key Pair with authorization to interact with (and ultimately deploy) Amazon EC2 instances. Return to the search bar at the top of the AWS interface and enter EC2 . Drill down into the Amazon EC2 service page from the results. From the left-hand navigation bar, drill down into the Network & Security > Key Pairs tab. Click the orange Create key pair button in the top-right of the page. A configuration tool will ask for additional information about the Key Pair: Name can be set to a value of your choosing Key pair type set to RSA Private key file format set to .pem When satisfied, click the orange Create key pair button The web browser will redirect to a page where you can view details about the newly-created Key Pair for EC2. Download the .pem file to your local device, preferably to the same directory that your TEMPLATE.yml and SOLUTION.yml Playbooks are located within. PLAYBOOK MODIFICATIONS REQUIRED The value of ansible_ssh_private_key_file must be updated within your Playbooks to account for the full path to the .pem file on your local machine. If you save the Key Pair file to the same directory that your Ansible Playbooks are located within, you do not need to qualify the value with anything more than the name of the file ( bienko-key.pem ). If it is saved in a different directory, you will need to spell out the full path + file name. Update the value of ansible_ssh_private_key_file in the following locations: Line 58 of the TEMPLATE.yml Playbook Line 113 of the SOLUTION.yml Playbook It is necessary to adjust the permissions assigned to the Key Pair file (on your local machine) before attempting to connect via Ansible to AWS. Open a Terminal console on your local machine and navigate ( cd ) to the directory in which the Key Pair file is saved. Execute the following to adjust the permissions assigned to the .pem file: chmod 400 filename.pem","title":""},{"location":"advanced/#executing-the-ansible-playbook","text":"","title":"Executing the Ansible Playbook"},{"location":"advanced/#_4","text":"At last, you are ready to execute the Playbook and use Ansible automation to deploy both an Amazon EC2 instance and a WordPress web application. With the Terminal console, navigate to the directory ( cd ) where both the Ansible Playbook and your KeyPair.pem file are located. You may use either the TEMPLATE.yml Playbook that you customized or the SOLUTION.yml Playbook for executing the Ansible automation tasks. Regardless of the Playbook being used, you must modify the Playbook with the necessary AWS security and account credentials information. When ready, execute the following instruction with the Terminal console: ansible-playbook TEMPLATE.yml Monitor the console's output as the Ansible Playbook iterates across each of the Tasks customized earlier. A fantastic feature of Ansible's automation is that if any of the Tasks should fail, the steps involved in that Task can be re-tried in subsequent runs; likewise, Tasks which executed successfully won't be repeated, but are validated and skipped instead. If you encounter errors during the litany of automation tasks, return to the Playbook and attempt to debug the issue. Re-issue the instruction in Step 34 to console to attempt the Ansible Playbook execution again. Continue iterating in this manner until all 14 Tasks have been successfully executed. After successfully deploying the Amazon EC2 environment and instantiating a WordPress application, the console will return a URL to the Terminal's output as part of TASK 14 's [debug] instruction. Copy the URL to clipboard and then paste it into your web browser to access the WordPress application. If your web browser loads the quick-start page for WordPress, you have successfully deployed the application and its supporting EC2 environment \u2014 entirely using generative AI and Ansible automation. Well done!","title":""},{"location":"advanced/#cleaning-up-and-conclusion","text":"","title":"Cleaning up and conclusion"},{"location":"advanced/#_5","text":"COSTS & BILLING It is essential that you terminate and deprovision the EC2 instance at the conclusion of the hands on lab. Otherwise, your personal AWS account will continue to be billed for as long as the resources are provisioned. Once you are ready to retire your environment, you will need to return to the AWS Dashboard to terminate the EC2 instance and hosted WordPress application. You can monitor your Amazon EC2 instance from the EC2 dashboard. At the top of the page, locate the Resources table and drill down into the Instances (running) tab. Details about your EC2 environment are summarized within the Instances table. Click the name of your EC2's Instance ID to expose additional details and management options. In the top-right corner of the interface, click the Instance state button and select Terminate instance from the drop-down menu. Confirm the deprovisioning request by clicking the orange Terminate button. All storage associated with the EC2 instance will automatically be deleted. Terminating an instance cannot be undone once underway. It is recommended that you delete all of the Security Groups (3 total) and Key Pairs (1 total) associated with the EC2 instance. To do so, return to the EC2 Dashboard and follow a similar procedure to Steps 38-41 . Search for IAM in the AWS taskbar to return to the Identity Access Management panel. If you wish to further secure your AWS account, delete all unwanted users or consider reducing the user's privilege from AdministratorAccess to a more restrictive class. That concludes the hands-on components to this Level 3 course, but your learning and experimentation doesn't need to end here. Continue to experiment with generating Tasks for Ansible Playbooks. If you have worked extensively with Ansible previously, try recreating Tasks you've written previously from old Playbooks using generative AI \u2014 just be careful not to use confidential or sensitive information as part of those tests. See if you can spot the differences or improvements made from the AI-generated code recommendations. Feel free to reach out to the authors of this coursework if you have suggestions for Tasks or code generation techniques that you'd like to see included in future iterations of this hands-on training.","title":""},{"location":"archive-setup-pre-vm/","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. Reserving a lab environment Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Acccount : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.3 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below and select Request a Red Hat Account [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted by IBM Technology Zone after the reservation period has ended Installation of Visual Studio Code and Extensions Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code available for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You might receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] If Extension Settings is not available from the drop-down menu, try right-clicking on the Ansible extension tile instead Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab Click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 5 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 5 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URI? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 12-14 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 12-14 as before; the login process should run smoother (and faster) on the second attempt Preparing the Ansible Playbook materials To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to clone the Ansible Playbook templates into the local (VS Code) environment. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . To perform a clone request with VS Code, perform the following action (depending on your OS): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that you have access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait patiently for the operation to finish (no responses will be prompted to screen if no updates to the code repository were necessary) Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured Troubleshooting and support If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. UNABLE TO INSTALL VS CODE LOCALLY \u2014 REQUIRE A VIRTUAL MACHINE Due to security or regulatory constraints, certain IBM clients or individuals enrolled in this training may be unable to install software (such as VS Code) on their local machines. As an alternative, IBM Technology Zone is able to provision a virtual machine environment that can be reserved at no-charge for completing the hands-on training. Click to expand the instructions embedded below for how to setup the virtual machine environment. Note that the documentation's instructions remain the same for the other modules, regardless of whether you are using a \"local\" or \"virtualized\" VS Code environment \u2014 although the \"local\" option is the preferred option. CLICK TO EXPAND INSTRUCTIONS FOR REQUESTING A VM Follow the link below and select the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.3 [A] (or later) tile: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the ITZ environment is ready to go, you will receive an email to your inbox (similar to the screenshot below). Click the blue View My Reservations at the bottom of the email to open a page with more details about the environment. Alternatively, you can click the My Reservations [A] tab from the ITZ home page to drill down into your environment's details. Near the top of the page, under Published Services , click the noVNC [A] link to access the virtual machine. A new tab will open within your web browser for the VM environment. It is recommended that Firefox be used for optimal performance Along the left-side edge of the page, click the Cog [A] icon From the pop-up menu, click Scaling Mode and then select Remote Resizing The VM window will now be properly scaled to the size of your browser window Click the Connect [B] button to continue with authentication Supply the following password and click Send Credentials : IBMDem0s! Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. Visual Studio (VS) Code is a code editor built on open source technology that is free to use and published by Microsoft. Distributions are available for both Windows and macOS operating systems. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B] If this is your first time logging into the VM, you must authenticate and activate the Ansible Lightspeed extension for VS Code. Click the Ansible [A] extension (denoted by the large A icon) from the left-hand interface open expose details about the Ansible plug-in. In the bottom-right corner of the interface [A] , you may receieve a notification stating that You must be logged in to use the Ansible Lightspeed , which confirms that the Ansible Lightspeed service still needs to be activated for this VM environment. Additionally, along the left-hand side of the interface [B] , you will also receive a welcome notice attached to the Ansible Lightspeed Login panel. Click either the Connect [C] or Login buttons to proceed with authentication and activation of the plugin When prompted with The extension Ansible wants to sign in using Ansible Lightspeed , click Allow [D] Firefox web browser will open within the VM and redirect to a Red Hat authentication page, as shown below. Click Log in with Red Hat [A] Note that the page load may be sluggish or slow \u2014 please be patient and give it a few moments before retrying this step The VM environment already has access credentials saved to Firefox's keychain. Use the stored username and password to authenticate the Ansible Lightspeed plugin with Red Hat. For the Red Hat login or email [A] field, enter ibm-wxca-demo-user For the Password [B] field, enter IBMDem0s! Log in when ready A pop-up notification will ask to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] to proceed. CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 11-14 takes too long (over 60 seconds), activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner. Click the Connect [C] button as shown and repeat Steps 11-14 as before. The login process should run smoother (and faster) on the second attempt. Verify [A] that the environment is logged in as wca-ibm-demo-user and that the User Type: Licensed . Click the Explorer [A] button (as shown) in the top-left corner of the interface. Confirm that the working directory is set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , then hit Return , and wait patiently for the operation to finish Open the console once again, type git pull , and then hit Return (no responses will be prompted to screen if no updates to the code repository were necessary) At this stage, the hands-on environment has been fully configured. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. Next steps The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"Archive setup pre vm"},{"location":"archive-setup-pre-vm/#_1","text":"","title":""},{"location":"archive-setup-pre-vm/#reserving-a-lab-environment","text":"","title":"Reserving a lab environment"},{"location":"archive-setup-pre-vm/#_2","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Acccount : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.3 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below and select Request a Red Hat Account [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted by IBM Technology Zone after the reservation period has ended","title":""},{"location":"archive-setup-pre-vm/#_3","text":"","title":""},{"location":"archive-setup-pre-vm/#installation-of-visual-studio-code-and-extensions","text":"","title":"Installation of Visual Studio Code and Extensions"},{"location":"archive-setup-pre-vm/#_4","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code available for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You might receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] If Extension Settings is not available from the drop-down menu, try right-clicking on the Ansible extension tile instead Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab Click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 5 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 5 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URI? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 12-14 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 12-14 as before; the login process should run smoother (and faster) on the second attempt","title":""},{"location":"archive-setup-pre-vm/#_5","text":"","title":""},{"location":"archive-setup-pre-vm/#preparing-the-ansible-playbook-materials","text":"To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to clone the Ansible Playbook templates into the local (VS Code) environment. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . To perform a clone request with VS Code, perform the following action (depending on your OS): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that you have access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait patiently for the operation to finish (no responses will be prompted to screen if no updates to the code repository were necessary) Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured","title":"Preparing the Ansible Playbook materials"},{"location":"archive-setup-pre-vm/#_6","text":"","title":""},{"location":"archive-setup-pre-vm/#troubleshooting-and-support","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. UNABLE TO INSTALL VS CODE LOCALLY \u2014 REQUIRE A VIRTUAL MACHINE Due to security or regulatory constraints, certain IBM clients or individuals enrolled in this training may be unable to install software (such as VS Code) on their local machines. As an alternative, IBM Technology Zone is able to provision a virtual machine environment that can be reserved at no-charge for completing the hands-on training. Click to expand the instructions embedded below for how to setup the virtual machine environment. Note that the documentation's instructions remain the same for the other modules, regardless of whether you are using a \"local\" or \"virtualized\" VS Code environment \u2014 although the \"local\" option is the preferred option. CLICK TO EXPAND INSTRUCTIONS FOR REQUESTING A VM Follow the link below and select the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.3 [A] (or later) tile: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the ITZ environment is ready to go, you will receive an email to your inbox (similar to the screenshot below). Click the blue View My Reservations at the bottom of the email to open a page with more details about the environment. Alternatively, you can click the My Reservations [A] tab from the ITZ home page to drill down into your environment's details. Near the top of the page, under Published Services , click the noVNC [A] link to access the virtual machine. A new tab will open within your web browser for the VM environment. It is recommended that Firefox be used for optimal performance Along the left-side edge of the page, click the Cog [A] icon From the pop-up menu, click Scaling Mode and then select Remote Resizing The VM window will now be properly scaled to the size of your browser window Click the Connect [B] button to continue with authentication Supply the following password and click Send Credentials : IBMDem0s! Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. Visual Studio (VS) Code is a code editor built on open source technology that is free to use and published by Microsoft. Distributions are available for both Windows and macOS operating systems. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B] If this is your first time logging into the VM, you must authenticate and activate the Ansible Lightspeed extension for VS Code. Click the Ansible [A] extension (denoted by the large A icon) from the left-hand interface open expose details about the Ansible plug-in. In the bottom-right corner of the interface [A] , you may receieve a notification stating that You must be logged in to use the Ansible Lightspeed , which confirms that the Ansible Lightspeed service still needs to be activated for this VM environment. Additionally, along the left-hand side of the interface [B] , you will also receive a welcome notice attached to the Ansible Lightspeed Login panel. Click either the Connect [C] or Login buttons to proceed with authentication and activation of the plugin When prompted with The extension Ansible wants to sign in using Ansible Lightspeed , click Allow [D] Firefox web browser will open within the VM and redirect to a Red Hat authentication page, as shown below. Click Log in with Red Hat [A] Note that the page load may be sluggish or slow \u2014 please be patient and give it a few moments before retrying this step The VM environment already has access credentials saved to Firefox's keychain. Use the stored username and password to authenticate the Ansible Lightspeed plugin with Red Hat. For the Red Hat login or email [A] field, enter ibm-wxca-demo-user For the Password [B] field, enter IBMDem0s! Log in when ready A pop-up notification will ask to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] to proceed. CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 11-14 takes too long (over 60 seconds), activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner. Click the Connect [C] button as shown and repeat Steps 11-14 as before. The login process should run smoother (and faster) on the second attempt. Verify [A] that the environment is logged in as wca-ibm-demo-user and that the User Type: Licensed . Click the Explorer [A] button (as shown) in the top-left corner of the interface. Confirm that the working directory is set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , then hit Return , and wait patiently for the operation to finish Open the console once again, type git pull , and then hit Return (no responses will be prompted to screen if no updates to the code repository were necessary) At this stage, the hands-on environment has been fully configured. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor.","title":"Troubleshooting and support"},{"location":"archive-setup-pre-vm/#_7","text":"","title":""},{"location":"archive-setup-pre-vm/#next-steps","text":"The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"Next steps"},{"location":"archived/","text":"Installation of Visual Studio Code and Extensions Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code availabe for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You may receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab Accessing your Red Hat credentials and authenticating with WCA Red Hat credentials will already have been emailed to you prior to starting this hands-on material, as part of the registration process. The invitation email will have a header similar to Red Hat Login Email Verification , addressed from a no-reply@redhat.com account. Locate this email in your inbox and follow along with the steps below to authenticate the VS Code extension with the IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (WCA) environment that has been pre-provisioned for this training. Click the URL [A] located within the body of the invitation email to finalize your account registration with the WCA environment. An Email Confirmation page will load within your web browser Record the value of Red Hat login to a notepad for reference later Create a new Password and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted after the hands-on training period has ended Return to the VS Code editor and click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 6 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URL? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 7-9 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 7-9 as before; the login process should run smoother (and faster) on the second attempt Preparing the Ansible Playbook materials To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. To perform a clone request with VS Code, perform the following action (depending on your operating system): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait for the operation to finish Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured Troubleshooting and support If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ANSIBLE-LINT IS NOT AVAILABLE ansible-lint checks playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises. RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. Next steps The following section will cover the fundamentals of AI-recommended code generation for Ansible Playbooks using IBM watsonx Code Assistant .","title":"Installation of Visual Studio Code and Extensions"},{"location":"archived/#installation-of-visual-studio-code-and-extensions","text":"","title":"Installation of Visual Studio Code and Extensions"},{"location":"archived/#_1","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code availabe for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You may receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab","title":""},{"location":"archived/#_2","text":"","title":""},{"location":"archived/#accessing-your-red-hat-credentials-and-authenticating-with-wca","text":"Red Hat credentials will already have been emailed to you prior to starting this hands-on material, as part of the registration process. The invitation email will have a header similar to Red Hat Login Email Verification , addressed from a no-reply@redhat.com account. Locate this email in your inbox and follow along with the steps below to authenticate the VS Code extension with the IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (WCA) environment that has been pre-provisioned for this training. Click the URL [A] located within the body of the invitation email to finalize your account registration with the WCA environment. An Email Confirmation page will load within your web browser Record the value of Red Hat login to a notepad for reference later Create a new Password and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted after the hands-on training period has ended Return to the VS Code editor and click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 6 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URL? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 7-9 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 7-9 as before; the login process should run smoother (and faster) on the second attempt","title":"Accessing your Red Hat credentials and authenticating with WCA"},{"location":"archived/#_3","text":"","title":""},{"location":"archived/#preparing-the-ansible-playbook-materials","text":"To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. To perform a clone request with VS Code, perform the following action (depending on your operating system): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait for the operation to finish Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured","title":"Preparing the Ansible Playbook materials"},{"location":"archived/#_4","text":"","title":""},{"location":"archived/#troubleshooting-and-support","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ANSIBLE-LINT IS NOT AVAILABLE ansible-lint checks playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises. RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor.","title":"Troubleshooting and support"},{"location":"archived/#_5","text":"","title":""},{"location":"archived/#next-steps","text":"The following section will cover the fundamentals of AI-recommended code generation for Ansible Playbooks using IBM watsonx Code Assistant .","title":"Next steps"},{"location":"customizing/","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Task Description Tuning and Model Customization module. [15 min] i. Fine-tuning task prompts In this section, you will experiment with customized Ansible Playbooks and test how changes made to an Ansible Task's natural language descriptions can impact the recommended code produced by IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ). UNPREDICTABLE RESULTS FROM GENERATIVE AI & LLMs A consequence of using generative AI and Large Language Models (LLMs) is that the recommended code output of these systems will never be 100% consistent for each and every execution. For this reason, LLMs are referred to as \"non-deterministic\" systems \u2014 as opposed to \"deterministic\" systems. This is both a strength and weakness of LLMs. As you will observe in this module, even a slight modification to the Task description or the smallest change to how a Playbook is structured\u2014 the descriptions used, the variables set, and so on \u2014will influence AI-generated code recommendations. The iterative process that you will go through in this module can be viewed in three different ways: On the one hand, it shows the sensitivity of generative AI models to even the most nuanced change in natural language prompts \u2014 for good or bad. Generative AI can produce tremendous work and that output is further guided along by best practices built-in from Red Hat and IBM. But in the end, the AI can only infer user intent from the natural language descriptions supplied to it. The more clearly a user defines their Task descriptions and intent, the more likely that WCA will correctly generate code which mirrors that intent; conversely, the less precise those descriptions are, the more likely WCA will misinterpret and miss the mark. Precision is key for the disambiguation of natural language prompts. Human feedback and humans-in-the-loop are essential to these formative stages of generative AI. As offerings like WCA mature, the natural language processing capabilities of the service will continue to be refined and improved. Additional packages, functions, and training data from Ansible Galaxy (as well as other sources) are continuously being added to the product's Foundation Models, which will in turn continually improve the AI-generated code recommendations made to users. By using the Model Tuning capabilities built into IBM watsonx Code Assistant , organizations and users are able to customize the recommendations produced by generative AI, tuning a domain-specific LLM with the organization's own Ansible Playbooks. The content and code recommendations that WCA suggests can be tailored to an organization's standards, best practices, and programming styles. These capabilities will be explored later in this module. The take-away here is that your results may vary : they may differ from the SOLUTION code presented in the steps below. Keep this in mind as you work through the examples in this section and understand that it is not a bug, but rather a consequence of working with generative AI in general. The precision with which a Playbook author describes Ansible Tasks in natural language will determine the accuracy and effectiveness of WCA's generated code recommendations. A template CUSTOM PLAYBOOK #1 YAML file has been prepared for you below. Copy and paste the code from the TEMPLATE tab into a New File... within VS Code Name and save the file as you see fit: for example, customplaybook1.yml HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible CUSTOM PLAYBOOK 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 # CUSTOM PLAYBOOK #1 \u2014 Invoke 2 modules to automatically update 2 types of servers. --- # TASK 1 - name : Update web servers hosts : webservers become : true tasks : - name : Ensure apache is at the latest version - name : Write the apache config file # TASK 2 - name : Update db servers hosts : databases become : true tasks : - name : Ensure postgresql is at the latest version - name : Ensure that postgresql is started The template CUSTOM PLAYBOOK #1 contains two sets of tasks: TASK 1 ( Lines 5-12 ) checks if the web server software is up to date and runs an update if necessary TASK 2 ( Lines 15-22 ) checks if the database server software is up to date and runs an update if necessary Examine the instruction on Line 12 , which prompts Ansible to create (\"write\") a configuration file for an Apache webserver: - name : Write the apache config file Place your cursor at the end of Line 12 and press Enter to generate WCA-recommended code for the task. Accept the recommendation by pressing Tab . Two tabs are presented below: AI-GENERATED CODE shows the output from running WCA's generative AI capabilities on Line 12 of the unmodified CUSTOM PLAYBOOK #1 YAML template. SOLUTION CODE shows the expected (correct) code for performing the task that was written by a human programmer. In theory, the AI-generated code should be as good\u2014 or even superior to \u2014the manually-written solution code. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 6 7 - name : Write the apache config file ansible.builtin.template : src : templates/apache.conf.j2 dest : /etc/apache2/sites-available/000-default.conf owner : root group : root mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" As you can see, the AI-GENERATED CODE recommendations satisfy some of the requests, but also miss the mark in a few areas. When compared to the SOLUTION CODE tab: Both tabs appropriately utilize the ansible.builtin.template Module, which for this request is absolutely correct. Interesting fact: earlier (circa 2023) versions of WCA produced code recommendations for the same task as this one, which did not align with the expected solution. Instead, the recommendation was for the ansible.builtin.copy: Module to invoke content: \"{{ _content_ }}\" , which is not correct. This speaks to the continuously improving capabilities of the WCA and IBM Granite models. AI-GENERATED CODE recommended including statements that explicitly set owner: root and group: root , which are both appropriate when permissions are set to mode: '0644' . While not strictly necessary, these additional statements arguably are an improvement over the SOLUTION CODE results. It's also worth noting that the request to write the apache config file did not explicitly request mode: '0644' in the natural language description, but WCA nevertheless recommended it (post-processing) as this is a best practice for deploying Apache webservers. The src and dest variables are not in agreement across the AI-GENERATED CODE and SOLUTION CODE tabs. This is an area for improvement. The natural language description wasn't precise about these details; therefore, this is an opportunity where more detailed and verbose instructions could better steer the recommendations WCA returns back with. In general, the more ambiguous the Task description, the greater the likelihood that WCA will misinterpret the author's intent and suggest unwanted Ansible automation jobs. To help disambiguate our intention, Playbook authors should use more precise natural language terms and descriptions. Delete the WCA-suggested lines of code from Step 4 from the Playbook. Rewrite the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations Take note of the much more precise language used to describe the src and dest variables - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config The resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: The src: httpd.j2 recommendation in the AI-GENERATED CODE tab is an exact match to the natural language description set in Step 6 , and only slightly different (in terms of the directory path used) to the src: /srv/httpd.j2 variable in the SOLUTION CODE tab. The dest: /etc/httpd/conf/httpd.conf recommendation in the AI-GENERATER CODE tab, unfortunately, still deviates from the httpd.config (not the same as httpd.conf ) destination that was requested in the natural language description. Close, but still far from exact and not matching our specifications. However, it's a vast improvement over the dest: /etc/apache2/sites-available/000-default.conf destination that was recommended as a result of Step 4 's more ambiguous task description. Perplexingly, the mode: '0644' recommendation that was made previously in Step 4 has been left out of the suggested task code. We can speculate as to why\u2014 perhaps the more precise natural language description made in Step 6 prompted WCA to only generate code for exactly what was specified \u2014but the \"black box\" nature of generative AI means that we cannot know for certain. Perhaps with another, even more precise iteration the AI-GENERATED CODE will match the SOLUTION CODE ? AI-GENERATED CODE SOLUTION CODE 1 2 3 4 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" Iterate on the lessons learned in Step 7 and modify the task description to include more details on the permissions applied to the Apache webserver. Delete the WCA-suggested lines of code from Step 6 from the Playbook. Re-write the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations. Take note of the much more precise language used to describe the mode variable. - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 Once again, the resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: WCA correctly picked up on the mode: '0644' request made in the modified task description. There is now alignment between the two tabs. With only a few iterations and by disambiguating the natural language description of the Ansible Task to be performed, the code recommendations produced by WCA have been markedly improved. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" ii. Model customization Since every organization is different, WCA allows users to customize the AI model output to your organization's unique Ansible Playbooks. This allows for personalized code recommendations that are a better fit to your business' unique needs and more reflective of the programming standards set within your organization. In this scenario, your organization has its own set of Ansible Playbooks that leverage your preferred cloud provider, that uses specific Ansible Modules to manage OpenShift clusters. Open the create-openshift-cluster-ibm-cloud.yml Playbook located within the Model Customization subdirectory of the hands-on lab templates. The full directory address, as well as the Playbook code, are encapsulated in the following code block. ~/ansible-wca-demo-kit/Model Customization/create-openshift-cluster-ibm-cloud.yml 1 2 3 4 5 6 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible Before performing any model tuning tasks, try generating code recommendations for the OpenShift cluster creation task by placing your cursor at the end of Line 6 , pressing Enter , and then accepting the default recommendations by pressing Tab . The resulting code recommendations\u2014 displayed below \u2014represent the standard, unmodified output from WCA's IBM Granite base models. When using the standard IBM Granite model recommendations, it recommends making use of the ansible.builtin.command Ansible Module \u2014 which is a standard, best-practice way to perform these types of deployments. However, your particular organization (and the public cloud provider they use for deploying OpenShift clusters) might require the use of different Ansible Modules. The following steps will explore how to tailor WCA-generated code blocks to the unique needs of a business. STANDARD IBM GRANITE MODEL RECOMMENDATIONS 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" DATA PREPARATION & MODEL TRAINING Given the scope and time available for this hands-on training, participants will not be manually creating the training data or training models themselves. Instead, you will utilize a prepared customized model ( Model ID ) and experiment from Step 12 and onwards with how the tuned Model ID impacts WCA's code recommendations. The process of building Ansible training data and tuning customized models for generative AI is potentially a complex and time-consuming one. For example, the model tuning employed in the steps ahead for the creation of an OpenShift cluster according to an enterprise organization's specific standards requires approximately 4 hours to train and generate a customized AI model. In preparation for model tuning, an organization or user can transform their existing Playbooks into training data for WCA using the open-source Ansible Parser Tool . The tool analyzes Ansible Playbooks and generates a single JSONL ( ftdata.jsonl formatted) file that can be uploaded to WCA's model tuning studio for developing customized AI models. The process of calibrating and running the Ansible Parser Tool is time-consuming and potentially complex, depending on the scope of Playbooks that an organization wishes to analyze and prepare for model tuning. This falls outside the scope of the hands-on material for this training. For the purposes of demonstration, a preconfigured openshift-tune-micro.jsonl was used for the model tuning example. With the Ansible training data prepared, the IBM watsonx Code Assistant on IBM Cloud service was configured to execute a new model tuning experiment. WCA provides a graphical user experience [A] to streamline the model tuning and customization process Tuning experiment name [B] set to Tuning Experiment and confirmed with Create tuning experiment [C] The openshift-tune-micro.jsonl training data from Ansible Parser Tool is uploaded into WCA [D] A summary of metrics\u2014 including parameters for sampling and Ansible modules \u2014are displayed within the model tuning wizard before Start tuning [E] is selected to kick off the model tuning operating After the tuning operation has ended, a training loss graph reveals how accurate the model's predictions are over the training data set, across multiple tuning cycles As the number of tuning cycles increase, the training loss rating will tend to decrease Model tuning for these particular metrics and parameters takes approximately 4 hours to complete CLICK TO EXPAND \u2014 ACCOMPANYING SCREENSHOTS To override the IBM Granite base model recommendations, you must provide a pointer to a customized Model ID (prepared and hosted earlier on WCA) using the VS Code extension. Custom Model ID : b48e3c52-6135-408b-9a1e-8bcc75a47d32<|sepofid|>480ace2b-c58b-4930-b566-674ad6974851 Look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Within the search bar at the top of the panel, add the text Model to filter the results [B] Copy and paste the Model ID value (above) into the empty Ansible > Lightspeed: Model ID Override field [C] , then click the X in the top-left corner of the panel to save and exit Extension Settings With the tuned model, WCA will recommend code using modules, functions, and other details specific to your organization's private IT environment. Return to the tuning-example.yml Playbook from Step 10 and delete the code recommendations generated in Step 11 Regenerate the code recommendations for the task by placing your cursor at the end of Line 6 , hitting Enter , and then Tab Observe how the tuned Model ID has modified WCA's code recommendations by toggling between the TUNED MODEL and STANDARD MODEL tabs below TUNED MODEL STANDARD MODEL 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster register : cluster_create_output when : cluster_output.rc != 0 ibm.cloudcollection.ibm_container_cluster : name : test_cluster datacenter : \"{{ datacenter }}\" machine_type : \"{{ machine_type }}\" hardware : \"{{ hardware }}\" kube_version : 4.14_openshift public_vlan_id : \"{{ public_vlan_id }}\" private_vlan_id : \"{{ private_vlan_id }}\" default_pool_size : \"{{ default_worker_pool_size }}\" entitlement : \"{{ entitlement }}\" 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" Using the tuned Model ID , WCA has returned a recommendation to use the ibm.cloudcollection.ibm_container_cluster Module for the IBM Cloud, instead of the standard ansible.builtin.command Module from Step 11 . This more accurately conforms to the organization's Ansible automation standards and requirements. REMOVE THE CUSTOMIZED MODEL ID WHEN FINISHED Remember to clear the Ansible > Lightspeed: Model ID Override in the Extension Settings after you have completed Step 13 , otherwise all subsequent code generation requests made to WCA will be produced using the customized AI model. iii. Conclusion This concludes the hands-on components of the Level 3 course, but your learning and experimentation doesn't need to end here. Participants are encouraged to follow the Level 3 accreditation steps (depending on your role).","title":"Task Description Tuning and Model Customization"},{"location":"customizing/#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Task Description Tuning and Model Customization module. [15 min]","title":""},{"location":"customizing/#i-fine-tuning-task-prompts","text":"In this section, you will experiment with customized Ansible Playbooks and test how changes made to an Ansible Task's natural language descriptions can impact the recommended code produced by IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ). UNPREDICTABLE RESULTS FROM GENERATIVE AI & LLMs A consequence of using generative AI and Large Language Models (LLMs) is that the recommended code output of these systems will never be 100% consistent for each and every execution. For this reason, LLMs are referred to as \"non-deterministic\" systems \u2014 as opposed to \"deterministic\" systems. This is both a strength and weakness of LLMs. As you will observe in this module, even a slight modification to the Task description or the smallest change to how a Playbook is structured\u2014 the descriptions used, the variables set, and so on \u2014will influence AI-generated code recommendations. The iterative process that you will go through in this module can be viewed in three different ways: On the one hand, it shows the sensitivity of generative AI models to even the most nuanced change in natural language prompts \u2014 for good or bad. Generative AI can produce tremendous work and that output is further guided along by best practices built-in from Red Hat and IBM. But in the end, the AI can only infer user intent from the natural language descriptions supplied to it. The more clearly a user defines their Task descriptions and intent, the more likely that WCA will correctly generate code which mirrors that intent; conversely, the less precise those descriptions are, the more likely WCA will misinterpret and miss the mark. Precision is key for the disambiguation of natural language prompts. Human feedback and humans-in-the-loop are essential to these formative stages of generative AI. As offerings like WCA mature, the natural language processing capabilities of the service will continue to be refined and improved. Additional packages, functions, and training data from Ansible Galaxy (as well as other sources) are continuously being added to the product's Foundation Models, which will in turn continually improve the AI-generated code recommendations made to users. By using the Model Tuning capabilities built into IBM watsonx Code Assistant , organizations and users are able to customize the recommendations produced by generative AI, tuning a domain-specific LLM with the organization's own Ansible Playbooks. The content and code recommendations that WCA suggests can be tailored to an organization's standards, best practices, and programming styles. These capabilities will be explored later in this module. The take-away here is that your results may vary : they may differ from the SOLUTION code presented in the steps below. Keep this in mind as you work through the examples in this section and understand that it is not a bug, but rather a consequence of working with generative AI in general. The precision with which a Playbook author describes Ansible Tasks in natural language will determine the accuracy and effectiveness of WCA's generated code recommendations. A template CUSTOM PLAYBOOK #1 YAML file has been prepared for you below. Copy and paste the code from the TEMPLATE tab into a New File... within VS Code Name and save the file as you see fit: for example, customplaybook1.yml HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible CUSTOM PLAYBOOK 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 # CUSTOM PLAYBOOK #1 \u2014 Invoke 2 modules to automatically update 2 types of servers. --- # TASK 1 - name : Update web servers hosts : webservers become : true tasks : - name : Ensure apache is at the latest version - name : Write the apache config file # TASK 2 - name : Update db servers hosts : databases become : true tasks : - name : Ensure postgresql is at the latest version - name : Ensure that postgresql is started The template CUSTOM PLAYBOOK #1 contains two sets of tasks: TASK 1 ( Lines 5-12 ) checks if the web server software is up to date and runs an update if necessary TASK 2 ( Lines 15-22 ) checks if the database server software is up to date and runs an update if necessary Examine the instruction on Line 12 , which prompts Ansible to create (\"write\") a configuration file for an Apache webserver: - name : Write the apache config file Place your cursor at the end of Line 12 and press Enter to generate WCA-recommended code for the task. Accept the recommendation by pressing Tab . Two tabs are presented below: AI-GENERATED CODE shows the output from running WCA's generative AI capabilities on Line 12 of the unmodified CUSTOM PLAYBOOK #1 YAML template. SOLUTION CODE shows the expected (correct) code for performing the task that was written by a human programmer. In theory, the AI-generated code should be as good\u2014 or even superior to \u2014the manually-written solution code. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 6 7 - name : Write the apache config file ansible.builtin.template : src : templates/apache.conf.j2 dest : /etc/apache2/sites-available/000-default.conf owner : root group : root mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" As you can see, the AI-GENERATED CODE recommendations satisfy some of the requests, but also miss the mark in a few areas. When compared to the SOLUTION CODE tab: Both tabs appropriately utilize the ansible.builtin.template Module, which for this request is absolutely correct. Interesting fact: earlier (circa 2023) versions of WCA produced code recommendations for the same task as this one, which did not align with the expected solution. Instead, the recommendation was for the ansible.builtin.copy: Module to invoke content: \"{{ _content_ }}\" , which is not correct. This speaks to the continuously improving capabilities of the WCA and IBM Granite models. AI-GENERATED CODE recommended including statements that explicitly set owner: root and group: root , which are both appropriate when permissions are set to mode: '0644' . While not strictly necessary, these additional statements arguably are an improvement over the SOLUTION CODE results. It's also worth noting that the request to write the apache config file did not explicitly request mode: '0644' in the natural language description, but WCA nevertheless recommended it (post-processing) as this is a best practice for deploying Apache webservers. The src and dest variables are not in agreement across the AI-GENERATED CODE and SOLUTION CODE tabs. This is an area for improvement. The natural language description wasn't precise about these details; therefore, this is an opportunity where more detailed and verbose instructions could better steer the recommendations WCA returns back with. In general, the more ambiguous the Task description, the greater the likelihood that WCA will misinterpret the author's intent and suggest unwanted Ansible automation jobs. To help disambiguate our intention, Playbook authors should use more precise natural language terms and descriptions. Delete the WCA-suggested lines of code from Step 4 from the Playbook. Rewrite the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations Take note of the much more precise language used to describe the src and dest variables - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config The resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: The src: httpd.j2 recommendation in the AI-GENERATED CODE tab is an exact match to the natural language description set in Step 6 , and only slightly different (in terms of the directory path used) to the src: /srv/httpd.j2 variable in the SOLUTION CODE tab. The dest: /etc/httpd/conf/httpd.conf recommendation in the AI-GENERATER CODE tab, unfortunately, still deviates from the httpd.config (not the same as httpd.conf ) destination that was requested in the natural language description. Close, but still far from exact and not matching our specifications. However, it's a vast improvement over the dest: /etc/apache2/sites-available/000-default.conf destination that was recommended as a result of Step 4 's more ambiguous task description. Perplexingly, the mode: '0644' recommendation that was made previously in Step 4 has been left out of the suggested task code. We can speculate as to why\u2014 perhaps the more precise natural language description made in Step 6 prompted WCA to only generate code for exactly what was specified \u2014but the \"black box\" nature of generative AI means that we cannot know for certain. Perhaps with another, even more precise iteration the AI-GENERATED CODE will match the SOLUTION CODE ? AI-GENERATED CODE SOLUTION CODE 1 2 3 4 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" Iterate on the lessons learned in Step 7 and modify the task description to include more details on the permissions applied to the Apache webserver. Delete the WCA-suggested lines of code from Step 6 from the Playbook. Re-write the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations. Take note of the much more precise language used to describe the mode variable. - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 Once again, the resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: WCA correctly picked up on the mode: '0644' request made in the modified task description. There is now alignment between the two tabs. With only a few iterations and by disambiguating the natural language description of the Ansible Task to be performed, the code recommendations produced by WCA have been markedly improved. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\"","title":"i. Fine-tuning task prompts"},{"location":"customizing/#ii-model-customization","text":"Since every organization is different, WCA allows users to customize the AI model output to your organization's unique Ansible Playbooks. This allows for personalized code recommendations that are a better fit to your business' unique needs and more reflective of the programming standards set within your organization. In this scenario, your organization has its own set of Ansible Playbooks that leverage your preferred cloud provider, that uses specific Ansible Modules to manage OpenShift clusters. Open the create-openshift-cluster-ibm-cloud.yml Playbook located within the Model Customization subdirectory of the hands-on lab templates. The full directory address, as well as the Playbook code, are encapsulated in the following code block. ~/ansible-wca-demo-kit/Model Customization/create-openshift-cluster-ibm-cloud.yml 1 2 3 4 5 6 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible Before performing any model tuning tasks, try generating code recommendations for the OpenShift cluster creation task by placing your cursor at the end of Line 6 , pressing Enter , and then accepting the default recommendations by pressing Tab . The resulting code recommendations\u2014 displayed below \u2014represent the standard, unmodified output from WCA's IBM Granite base models. When using the standard IBM Granite model recommendations, it recommends making use of the ansible.builtin.command Ansible Module \u2014 which is a standard, best-practice way to perform these types of deployments. However, your particular organization (and the public cloud provider they use for deploying OpenShift clusters) might require the use of different Ansible Modules. The following steps will explore how to tailor WCA-generated code blocks to the unique needs of a business. STANDARD IBM GRANITE MODEL RECOMMENDATIONS 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" DATA PREPARATION & MODEL TRAINING Given the scope and time available for this hands-on training, participants will not be manually creating the training data or training models themselves. Instead, you will utilize a prepared customized model ( Model ID ) and experiment from Step 12 and onwards with how the tuned Model ID impacts WCA's code recommendations. The process of building Ansible training data and tuning customized models for generative AI is potentially a complex and time-consuming one. For example, the model tuning employed in the steps ahead for the creation of an OpenShift cluster according to an enterprise organization's specific standards requires approximately 4 hours to train and generate a customized AI model. In preparation for model tuning, an organization or user can transform their existing Playbooks into training data for WCA using the open-source Ansible Parser Tool . The tool analyzes Ansible Playbooks and generates a single JSONL ( ftdata.jsonl formatted) file that can be uploaded to WCA's model tuning studio for developing customized AI models. The process of calibrating and running the Ansible Parser Tool is time-consuming and potentially complex, depending on the scope of Playbooks that an organization wishes to analyze and prepare for model tuning. This falls outside the scope of the hands-on material for this training. For the purposes of demonstration, a preconfigured openshift-tune-micro.jsonl was used for the model tuning example. With the Ansible training data prepared, the IBM watsonx Code Assistant on IBM Cloud service was configured to execute a new model tuning experiment. WCA provides a graphical user experience [A] to streamline the model tuning and customization process Tuning experiment name [B] set to Tuning Experiment and confirmed with Create tuning experiment [C] The openshift-tune-micro.jsonl training data from Ansible Parser Tool is uploaded into WCA [D] A summary of metrics\u2014 including parameters for sampling and Ansible modules \u2014are displayed within the model tuning wizard before Start tuning [E] is selected to kick off the model tuning operating After the tuning operation has ended, a training loss graph reveals how accurate the model's predictions are over the training data set, across multiple tuning cycles As the number of tuning cycles increase, the training loss rating will tend to decrease Model tuning for these particular metrics and parameters takes approximately 4 hours to complete CLICK TO EXPAND \u2014 ACCOMPANYING SCREENSHOTS To override the IBM Granite base model recommendations, you must provide a pointer to a customized Model ID (prepared and hosted earlier on WCA) using the VS Code extension. Custom Model ID : b48e3c52-6135-408b-9a1e-8bcc75a47d32<|sepofid|>480ace2b-c58b-4930-b566-674ad6974851 Look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Within the search bar at the top of the panel, add the text Model to filter the results [B] Copy and paste the Model ID value (above) into the empty Ansible > Lightspeed: Model ID Override field [C] , then click the X in the top-left corner of the panel to save and exit Extension Settings With the tuned model, WCA will recommend code using modules, functions, and other details specific to your organization's private IT environment. Return to the tuning-example.yml Playbook from Step 10 and delete the code recommendations generated in Step 11 Regenerate the code recommendations for the task by placing your cursor at the end of Line 6 , hitting Enter , and then Tab Observe how the tuned Model ID has modified WCA's code recommendations by toggling between the TUNED MODEL and STANDARD MODEL tabs below TUNED MODEL STANDARD MODEL 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster register : cluster_create_output when : cluster_output.rc != 0 ibm.cloudcollection.ibm_container_cluster : name : test_cluster datacenter : \"{{ datacenter }}\" machine_type : \"{{ machine_type }}\" hardware : \"{{ hardware }}\" kube_version : 4.14_openshift public_vlan_id : \"{{ public_vlan_id }}\" private_vlan_id : \"{{ private_vlan_id }}\" default_pool_size : \"{{ default_worker_pool_size }}\" entitlement : \"{{ entitlement }}\" 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" Using the tuned Model ID , WCA has returned a recommendation to use the ibm.cloudcollection.ibm_container_cluster Module for the IBM Cloud, instead of the standard ansible.builtin.command Module from Step 11 . This more accurately conforms to the organization's Ansible automation standards and requirements. REMOVE THE CUSTOMIZED MODEL ID WHEN FINISHED Remember to clear the Ansible > Lightspeed: Model ID Override in the Extension Settings after you have completed Step 13 , otherwise all subsequent code generation requests made to WCA will be produced using the customized AI model.","title":"ii. Model customization"},{"location":"customizing/#iii-conclusion","text":"This concludes the hands-on components of the Level 3 course, but your learning and experimentation doesn't need to end here. Participants are encouraged to follow the Level 3 accreditation steps (depending on your role).","title":"iii. Conclusion"},{"location":"evaluation/","text":"i. Evaluation Criteria for IBM Technical Sellers and Business Partners To receive the Level 3 badge ( Generative AI for Code with watsonx Code Assistant Technical Sales Intermediate ), IBMers and Business Partners must demonstrate mastery of the skills learned throughout the various modules of these hands-on labs and coursework. Level 3 skills requirements\u2014 and the way participants will be evaluated \u2014differs depending on job role. IBM TECHNICAL SELLERS IBM Sales and Tech Sales must develop and record a Stand & Deliver presentation, which will be uploaded to the IBM Stand & Deliver platform for evaluation. This video is intended to simulate your delivery of a \u201clive\u201d demo in front of a client \u2014 on camera. IBMers will have flexibility in defining a hypothetical client, the pain points that customer has, and the goals they aspire to. The recording will then cover the seller\u2019s hands-on demonstration and pitch to the client of the value of the IBM solution using the environments and techniques of this lab. BUSINESS PARTNERS Business partners must pass a skills evaluation quiz after completing the hands-on portion of the course. The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. Participants must pass the quiz with a grade of 80% or higher. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Answers to quiz questions can only be determined by carefully following the instructions of the hands-on lab. ii. IBMer Stand & Deliver Assessment IBMers \u2014 SUBMIT RECORDINGS HERE Submit your Stand & Deliver recording online using IBM YourLearning. Hands-on demonstrations of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed are available on Seismic, as well as embedded within the various modules of this documentation. The authors highly recommend tailoring the Stand & Deliver to your own presentation style and expertise. Captivating and delighting clients with your unique delivery is paramount! Introduction [6 min] Generating Code [10 min] Content Source Matching and Post-Processing [10 min] Task Description Tuning and Model Customization [15 min] Stand & Deliver with Paul Zikopoulos [8 min] The evaluation criteria described below only applies to IBMers , who must record a Stand & Deliver to receive credit for this Level 3 course. By default, IBMers will be evaluated by their First Line Manager (FLM) \u2014 although they may request another manager to evaluate their Stand & Deliver, if appropriate. Instructions on how to submit a Stand & Deliver are included within the activity on YourLearning. IBM Technical Sellers need to include all six of the following elements in their Stand & Deliver recording to receive a Level 3 badge: Seller articulated their client's pain point(s) and the value proposition of using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller highlighted use cases for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller demonstrated and discussed several of the key differentiated capabilities of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed that deliver on the value proposition on point one. Seller highlighted benefits to their client (this is the why the client can\u2019t live without these benefits section). Seller highlighted benefits to their client's customers (what will the client be able to deliver to their customers that they could not without this product). Seller closed the demonstration with a call to action for their client that could include: a workshop, a deeper dive into the product meeting, or a Proof of Experience (PoX). iii. Business Partner Quiz Assessment PARTNERS \u2014 COMPLETE ASSESSMENT HERE Complete your assessment online using IBM Training: https://learn.ibm.com/mod/subcourse/view.php?id=284786 The skills evaluation quiz only applies to business partners \u2013 IBMers must record a Stand & Deliver . The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Participants must pass the quiz with a grade of 80% or higher. Answers to the quiz can only be determined by carefully following the instructions of the hands-on lab. iv. Next steps In the following section, you will prepare your hands-on lab environment with the necessary services and configurations.","title":"Evaluation"},{"location":"evaluation/#_1","text":"","title":""},{"location":"evaluation/#i-evaluation-criteria-for-ibm-technical-sellers-and-business-partners","text":"To receive the Level 3 badge ( Generative AI for Code with watsonx Code Assistant Technical Sales Intermediate ), IBMers and Business Partners must demonstrate mastery of the skills learned throughout the various modules of these hands-on labs and coursework. Level 3 skills requirements\u2014 and the way participants will be evaluated \u2014differs depending on job role. IBM TECHNICAL SELLERS IBM Sales and Tech Sales must develop and record a Stand & Deliver presentation, which will be uploaded to the IBM Stand & Deliver platform for evaluation. This video is intended to simulate your delivery of a \u201clive\u201d demo in front of a client \u2014 on camera. IBMers will have flexibility in defining a hypothetical client, the pain points that customer has, and the goals they aspire to. The recording will then cover the seller\u2019s hands-on demonstration and pitch to the client of the value of the IBM solution using the environments and techniques of this lab. BUSINESS PARTNERS Business partners must pass a skills evaluation quiz after completing the hands-on portion of the course. The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. Participants must pass the quiz with a grade of 80% or higher. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Answers to quiz questions can only be determined by carefully following the instructions of the hands-on lab.","title":"i. Evaluation Criteria for IBM Technical Sellers and Business Partners"},{"location":"evaluation/#ii-ibmer-stand-deliver-assessment","text":"IBMers \u2014 SUBMIT RECORDINGS HERE Submit your Stand & Deliver recording online using IBM YourLearning. Hands-on demonstrations of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed are available on Seismic, as well as embedded within the various modules of this documentation. The authors highly recommend tailoring the Stand & Deliver to your own presentation style and expertise. Captivating and delighting clients with your unique delivery is paramount! Introduction [6 min] Generating Code [10 min] Content Source Matching and Post-Processing [10 min] Task Description Tuning and Model Customization [15 min] Stand & Deliver with Paul Zikopoulos [8 min] The evaluation criteria described below only applies to IBMers , who must record a Stand & Deliver to receive credit for this Level 3 course. By default, IBMers will be evaluated by their First Line Manager (FLM) \u2014 although they may request another manager to evaluate their Stand & Deliver, if appropriate. Instructions on how to submit a Stand & Deliver are included within the activity on YourLearning. IBM Technical Sellers need to include all six of the following elements in their Stand & Deliver recording to receive a Level 3 badge: Seller articulated their client's pain point(s) and the value proposition of using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller highlighted use cases for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller demonstrated and discussed several of the key differentiated capabilities of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed that deliver on the value proposition on point one. Seller highlighted benefits to their client (this is the why the client can\u2019t live without these benefits section). Seller highlighted benefits to their client's customers (what will the client be able to deliver to their customers that they could not without this product). Seller closed the demonstration with a call to action for their client that could include: a workshop, a deeper dive into the product meeting, or a Proof of Experience (PoX).","title":"ii. IBMer Stand &amp; Deliver Assessment"},{"location":"evaluation/#iii-business-partner-quiz-assessment","text":"PARTNERS \u2014 COMPLETE ASSESSMENT HERE Complete your assessment online using IBM Training: https://learn.ibm.com/mod/subcourse/view.php?id=284786 The skills evaluation quiz only applies to business partners \u2013 IBMers must record a Stand & Deliver . The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Participants must pass the quiz with a grade of 80% or higher. Answers to the quiz can only be determined by carefully following the instructions of the hands-on lab.","title":"iii. Business Partner Quiz Assessment"},{"location":"evaluation/#iv-next-steps","text":"In the following section, you will prepare your hands-on lab environment with the necessary services and configurations.","title":"iv. Next steps"},{"location":"generating/","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements and hands-on components of the Generating Code module. [10 min] i. Generating Code with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed An Ansible Task is a statement in Ansible's automation script (the YAML-based Playbooks you will be working with) that declares a single action to be executed. This might be installing a package, copying a file, or shutting down a service on a remote machine. Each Task represents an idempotent operation (an action that can be repeated multiple times and deliver the same result every time) that aligns the remote managed node to the specified state. Idempotent operations also ensure consistency across multiple executions, guaranteeing the same steps are taken on each execution of the task. After you have learned the fundamentals of generating Ansible Task code blocks using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you'll be ready to shape and tailor the AI-generated code recommendations using WCA's model tuning capabilities. ii. Single task Ansible operations The process of creating AI-generated code recommendations is as simple as modifying the natural language (plain English) Task descriptions of an action that is to be executed, which always start with - name: and are followed by some description of the task to be performed. Ansible Tasks are often preceded with the prefix # , indicating developer comments or documentation. After the natural language description of the automation Task has been set by the user, WCA handles the rest. WCA is also capable of generating multiple Ansible Tasks from more complex natural language descriptions\u2014 what is referred to as multi-task code generation \u2014which you will experiment with later in this module. However, to get started, let's begin with the basics of generating code for single task use cases. Begin by opening the install_cockpit_single-task.yml Playbook from the list of assets in the Explorer browser. Click the Explorer tab from the left-hand interface [A] Drill down into the Install and configure Cockpit using Ansible subdirectory [B] Double-click the install_cockpit_single-task.yml Playbook A replica of the Playbook code is also included below in the documentation The red highlighting within the editor reminds users that the tasks: section contains no valid -name: task definitions. This is part of WCA's code validation process which runs automatically and alerts users to syntax errors in their code. You can safely ignore these warnings for now, as you will be un-commenting and generating valid -name: task definitions in the following steps. ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_single-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 --- - name : Install and configure Cockpit hosts : rhel become : true # module_defaults: # ansible.builtin.service: # enabled: true # state: started tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - The suggestion included Ansible best practices by using Fully Qualified Collection name. # - name: Install cockpit package # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used \"ansible.builtin.template\" module based on the \".j2\" file extension. # # Note - The suggestion set the file permissions (\"0644\"), owner, and group based on Ansible best practices. # - name: Copy cockpit.conf.j2 to /etc/cockpit # TASK 3 # # 3a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used the generic \"Start and enable service\" prompt # # and full Playbook context to infer the recommendation should start the \"cockpit\" service. # # 3b. Uncomment the \"module_defaults\" section at the top of the Playbook. # # 3c. Clear current task suggestion and request updated suggestion. # # Note - Ansible Lightspeed used the full Playbook context and evaluated the \"module_defaults\" # # when generating a suggestion. # # The updated suggestion no longer includes \"enabled:\" and \"state:\" arguments. # - name: Start and enable service # TASK 4 # # 4a. Uncomment task description below and generate an Ansible Playbook task suggestion. # # 4b. Save the Playbook. # - name: Wait 15 seconds port 9090 The install_cockpit_single-task.yml Playbook code above warrants some explanation before we move on with making AI-generated modifications to it: Line 2 essentially marks the beginning of the Playbook instructions, the purpose of which is to automate the process of installing and configuring Cockpit for Red Hat Ansible. Lines 3-4 define variables that will remain static throughout the remainder of the Playbook. These variables will be referenced by the AI-generated code suggestions at a later stage. This is a key capability of the offering and one which you will explore in much finer details later on in this module. Lines 6-9 are variables which have been commented out and therefore are invisible to the execution of the Ansible script and not examined by WCA for context when generating code recommendations. You will experiment with how removing the # comment blocks impacts the recommendations of task block code. \"Uncommenting\" these lines of code will make them viable for execution and these lines will afterwards be considered as valid Playbook \"context\" for AI code generation. Locate TASK 1 on Line 15 of the YAML file, which handles installation of Cockpit for Ansible. Cockpit is an interactive server administration interface that provides a graphical overview of statistics and configurations for a system or systems within a network. # - name: Install cockpit package Pay attention to the indentation and characters used on Line 15 , which in sequence from left to right are as follows: begins with Tab (or Space whitespaces) for indentation a # character to \"comment out\" the line's contents a whitespace Space character - name: which signifies the start of a Task definition and finally the natural language description of the Task INDENTATION LEVELS AND WHITESPACE Similar to Python, Ansible and YAML-based Playbooks are very sensitive to whitespacing and indentation. Indentations (such as the Tab in this example) denote different hierarchies and code nesting levels within the YAML structure. You may use Space instead of Tab if you prefer, but be sure to use indentations consistently : choose to use either Tab or Space for indenting lines of code, and do not interchange between the two. To generate code for TASK 1 , first uncomment the line of code (remove the # character from the start of a line). Highlight the line(s) of code you wish to uncomment and then press Cmd + ? for macOS or Ctrl + ? for Windows You can repeat those keystrokes with the line(s) selected to toggle between commenting or uncommenting lines of code Tip: commented out lines of code in VS Code will appear as green text Afterwards, Line 15 should look like the following \u2014 beginning with a single Tab - name : Install cockpit package Now you are ready to begin generating code. Place your cursor on Line 15 and hit Enter Wait for WCA to engage and generate the suggested (in grey, italicized text ) code block for executing the task This temporary code suggestion is entirely generated by AI As a user, you have the option to either: Accept the code recommendation as-given by pressing Tab Modify the recommended code by highlighting and replacing the italicized text FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint Hit Tab to accept the suggested code and then compare with the SOLUTION tab below. TEMPLATE SOLUTION # TASK 1 - name : Install cockpit package # TASK 1 - name : Install cockpit package ansible.builtin.package : name : cockpit state : present As part of the plain-text description of the Task, WCA was asked to include the cockpit Role, part of the Red Hat Enterprise Linux System Roles Certified Content Collection. The AI-generated code suggestion invoked a Fully Qualified Collection Name ( FQCN ) - ansible.builtin.package Making use of FQCNs where possible is a recommended best practice and is a prime example of the many ways in which the offering infuses post-processing capabilities within the AI-generated code produced by WCA. Additional examples of infusing best-practices into AI-generated code recommendations can be found in TASK 2 ( Line 21 of the unmodified template or Line 25 after Step 5 ): Uncomment - name: Copy cockpit.conf.j2 to /etc/cockpit Hit Enter to generate the task code recommendation and accept the AI-suggested code (without modifications) by pressing Tab Compare your results with the SOLUTION tab below TEMPLATE SOLUTION # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' The AI-generated code recommendation will copy cockpit.conf to the target host. Take note of the fact that the recommendation included the mode: argument and set the Linux file permissions to 0644 , neither of which were things explicitly requested in the Task -name description, but are both additions which adhere to best practices around defining Ansible automaton tasks. Setting a file permission to 0644 specifies read and write permissions for User and Group levels within the Linux OS, and provides only read permissions to all others. iii. Multi-task Ansible operations Up to this point, we've kept a narrow aperture on AI-generated recommendations for single tasks \u2014 examining and experimenting with generating Ansible code task by task, one at a time. However, a powerful WCA feature is the ability to combine multiple task descriptions into a single natural language prompt; in turn, WCA is able to parse that instruction, decompose the instruction into discrete Ansible Task parts, and return a complete code recommendation for achieving the author's intended goal. Syntactically, multiple tasks are combined into a single natural language expression through the use of ampersand ( & ) characters. Simply write out all the automation task descriptions on a single line, separating each description with a & character. The line must also begin with a # character for reasons that will be explained shortly. To illustrate, let's look at a multi-task Ansible Playbook: install_cockpit_multi-task.yml The contents of this Playbook should look familiar to you already: it is essentially the same Playbook examined in Steps 1-6 ( install_cockpit_single-task.yml ), re-written in an equivalent multi-task expression Each of the Task descriptions from the previous Playbook have been consolidated into a single description on Line 12 , separated by & characters ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_multi-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : # Install cockpit package & Copy cockpit.conf.j2 to /etc/cockpit & Start and enable service & Wait 15 seconds port 9090 There are two crucial distinctions between single task and multi-task code generation: formatting and execution. Formatting : Notice that Line 12 does not begin with -name: , as was the case with single task descriptions Execution : In order to generate AI code recommendations for multi-task descriptions, Line 12 must stay commented out (the # must remain at the start of the line) What is the rationale behind this? When WCA's generative AI capabilities parse Line 12 , its output will include multiple -name: tasks, each containing potentially multiple lines of instructions, based on how many & -delineated task descriptions are included on the line. Therefore, the way in which the code generation step is executed on Line 12 is a consequence of the formatting decision. Execution of a code generation step on a commented-out ( # ) line containing & delineators is recognized by WCA as a unique case that will be acted upon as a multi-task statement. Place your cursor at the end of Line 12 , and without removing the # character, press Enter to execute the code generation step. Be aware that generating code for multi-task descriptions will take longer compared to a single task. Compare the MULTI-TASK solution tab with the SINGLE TASK solution (copied over from Step 6 ). How did the multi-task code generation fare compared to the single task approach? MULTI-TASK SINGLE TASK 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit.socket - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit state : started enabled : true - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 Comparing the two results, the only notable difference between the two approaches are Lines 27-29 from the SINGLE TASK generative AI approach. Both the SINGLE TASK and MULTI-TASK suggestions for that particular task satisfy the request made by the user. However, whether the single task or multi-task approach resulted in a better code suggestion is up to the judgement of the programmer. Nearly 90% of the remaining code was identical between the two approaches and was achieved in far fewer lines of code (and less typing) using the multi-task approach. The variability of generative AI suggestions is a fascinating topic and one that we will dive more deeply into with the module ahead. Before moving on to other product features, experiment by creating a new Ansible Playbook in your workspace using the code template below. Suggestions will be given on how to perform the same automation task using single and multi-task generation approaches. Save the YAML file as create_ec2_single_multi.yml (if you forget to save the file, WCA will not generate recommendations) Copy the following code block to your clipboard using the + icon in the top-right corner of the panel and paste into the newly created YAML file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ~/Documents/ansible-wca-demo-kit/create_ec2_single_multi.yml 1 2 3 4 --- - name : Provision an EC2 instance hosts : all tasks : First, configure the Playbook for single task generation by adding the following code snippet into Line 5 (remembering to properly indent with Tab or Space characters): - name : create vpc named demo Your workspace Playbook should look identical to the TEMPLATE tab below. Place your cursor at the end of the newly-created Line 5 and hit Enter to execute single task code generation. Compare your results to the SOLUTION tab. Record your results to a notepad so that you can compare the results later. TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc Now it's time to re-write the Playbook for multi-task code generation. The beauty of natural language statements is that your approach can be as terse or verbose as you want. The more verbose and descriptive, the more prescriptive you can be in terms of influencing the AI-generated code recommendations from WCA. The topic of \"prompt tuning\" will be explored in much greater detail in the Task Description Tuning and Model Customization module. But for now: Replace all of the code inside create_ec2_single_multi.yml with the TEMPLATE code block (below) Place your cursor at the end of Line 5 (remember not to remove the # character) and hit Enter , then accept the recommendation with Tab Compare your results to the SOLUTION tab and to the result from Step 13 TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet - name : Create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc - name : Create security group for demo vpc and allow ssh and http amazon.aws.ec2_security_group : name : demo description : demo security group vpc_id : \"{ vpc.vpc.id }\" rules : - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 register : sg - name : Create internet gateway named demogateway for demo vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demogateway register : igw - name : Create subnet named demosubnet in demo vpc amazon.aws.ec2_vpc_subnet : cidr : 10.0.0.80/24 vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet register : subnet - name : Create route table for subnet demosubnet and gateway demogateway amazon.aws.ec2_vpc_route_table : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet subnets : - \"{ subnet.subnet.id }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ igw.gateway_id }\" register : route_table - name : Create t2.small ec2 instance in subnet demosubnet amazon.aws.ec2_instance : key_name : \"{ _key_name_ }\" instance_type : t2.small image : \"{ _image_ }\" wait : true vpc_subnet_id : \"{ subnet.subnet.id }\" security_group : demo register : ec2 TIMEOUT WARNING It may take several moments for WCA to process and return code recommendations for a multi-task description as complex as this one. If you receive a time-out warning, try executing the code generation step by pressing Enter a second time. iv. Next steps The following module will examine in detail WCA's post-processing and content source attribution capabilities.","title":"Generating Code"},{"location":"generating/#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements and hands-on components of the Generating Code module. [10 min]","title":""},{"location":"generating/#i-generating-code-with-ibm-watsonx-code-assistant-for-red-hat-ansible-lightspeed","text":"An Ansible Task is a statement in Ansible's automation script (the YAML-based Playbooks you will be working with) that declares a single action to be executed. This might be installing a package, copying a file, or shutting down a service on a remote machine. Each Task represents an idempotent operation (an action that can be repeated multiple times and deliver the same result every time) that aligns the remote managed node to the specified state. Idempotent operations also ensure consistency across multiple executions, guaranteeing the same steps are taken on each execution of the task. After you have learned the fundamentals of generating Ansible Task code blocks using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you'll be ready to shape and tailor the AI-generated code recommendations using WCA's model tuning capabilities.","title":"i. Generating Code with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed"},{"location":"generating/#ii-single-task-ansible-operations","text":"The process of creating AI-generated code recommendations is as simple as modifying the natural language (plain English) Task descriptions of an action that is to be executed, which always start with - name: and are followed by some description of the task to be performed. Ansible Tasks are often preceded with the prefix # , indicating developer comments or documentation. After the natural language description of the automation Task has been set by the user, WCA handles the rest. WCA is also capable of generating multiple Ansible Tasks from more complex natural language descriptions\u2014 what is referred to as multi-task code generation \u2014which you will experiment with later in this module. However, to get started, let's begin with the basics of generating code for single task use cases. Begin by opening the install_cockpit_single-task.yml Playbook from the list of assets in the Explorer browser. Click the Explorer tab from the left-hand interface [A] Drill down into the Install and configure Cockpit using Ansible subdirectory [B] Double-click the install_cockpit_single-task.yml Playbook A replica of the Playbook code is also included below in the documentation The red highlighting within the editor reminds users that the tasks: section contains no valid -name: task definitions. This is part of WCA's code validation process which runs automatically and alerts users to syntax errors in their code. You can safely ignore these warnings for now, as you will be un-commenting and generating valid -name: task definitions in the following steps. ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_single-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 --- - name : Install and configure Cockpit hosts : rhel become : true # module_defaults: # ansible.builtin.service: # enabled: true # state: started tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - The suggestion included Ansible best practices by using Fully Qualified Collection name. # - name: Install cockpit package # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used \"ansible.builtin.template\" module based on the \".j2\" file extension. # # Note - The suggestion set the file permissions (\"0644\"), owner, and group based on Ansible best practices. # - name: Copy cockpit.conf.j2 to /etc/cockpit # TASK 3 # # 3a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used the generic \"Start and enable service\" prompt # # and full Playbook context to infer the recommendation should start the \"cockpit\" service. # # 3b. Uncomment the \"module_defaults\" section at the top of the Playbook. # # 3c. Clear current task suggestion and request updated suggestion. # # Note - Ansible Lightspeed used the full Playbook context and evaluated the \"module_defaults\" # # when generating a suggestion. # # The updated suggestion no longer includes \"enabled:\" and \"state:\" arguments. # - name: Start and enable service # TASK 4 # # 4a. Uncomment task description below and generate an Ansible Playbook task suggestion. # # 4b. Save the Playbook. # - name: Wait 15 seconds port 9090 The install_cockpit_single-task.yml Playbook code above warrants some explanation before we move on with making AI-generated modifications to it: Line 2 essentially marks the beginning of the Playbook instructions, the purpose of which is to automate the process of installing and configuring Cockpit for Red Hat Ansible. Lines 3-4 define variables that will remain static throughout the remainder of the Playbook. These variables will be referenced by the AI-generated code suggestions at a later stage. This is a key capability of the offering and one which you will explore in much finer details later on in this module. Lines 6-9 are variables which have been commented out and therefore are invisible to the execution of the Ansible script and not examined by WCA for context when generating code recommendations. You will experiment with how removing the # comment blocks impacts the recommendations of task block code. \"Uncommenting\" these lines of code will make them viable for execution and these lines will afterwards be considered as valid Playbook \"context\" for AI code generation. Locate TASK 1 on Line 15 of the YAML file, which handles installation of Cockpit for Ansible. Cockpit is an interactive server administration interface that provides a graphical overview of statistics and configurations for a system or systems within a network. # - name: Install cockpit package Pay attention to the indentation and characters used on Line 15 , which in sequence from left to right are as follows: begins with Tab (or Space whitespaces) for indentation a # character to \"comment out\" the line's contents a whitespace Space character - name: which signifies the start of a Task definition and finally the natural language description of the Task INDENTATION LEVELS AND WHITESPACE Similar to Python, Ansible and YAML-based Playbooks are very sensitive to whitespacing and indentation. Indentations (such as the Tab in this example) denote different hierarchies and code nesting levels within the YAML structure. You may use Space instead of Tab if you prefer, but be sure to use indentations consistently : choose to use either Tab or Space for indenting lines of code, and do not interchange between the two. To generate code for TASK 1 , first uncomment the line of code (remove the # character from the start of a line). Highlight the line(s) of code you wish to uncomment and then press Cmd + ? for macOS or Ctrl + ? for Windows You can repeat those keystrokes with the line(s) selected to toggle between commenting or uncommenting lines of code Tip: commented out lines of code in VS Code will appear as green text Afterwards, Line 15 should look like the following \u2014 beginning with a single Tab - name : Install cockpit package Now you are ready to begin generating code. Place your cursor on Line 15 and hit Enter Wait for WCA to engage and generate the suggested (in grey, italicized text ) code block for executing the task This temporary code suggestion is entirely generated by AI As a user, you have the option to either: Accept the code recommendation as-given by pressing Tab Modify the recommended code by highlighting and replacing the italicized text FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint Hit Tab to accept the suggested code and then compare with the SOLUTION tab below. TEMPLATE SOLUTION # TASK 1 - name : Install cockpit package # TASK 1 - name : Install cockpit package ansible.builtin.package : name : cockpit state : present As part of the plain-text description of the Task, WCA was asked to include the cockpit Role, part of the Red Hat Enterprise Linux System Roles Certified Content Collection. The AI-generated code suggestion invoked a Fully Qualified Collection Name ( FQCN ) - ansible.builtin.package Making use of FQCNs where possible is a recommended best practice and is a prime example of the many ways in which the offering infuses post-processing capabilities within the AI-generated code produced by WCA. Additional examples of infusing best-practices into AI-generated code recommendations can be found in TASK 2 ( Line 21 of the unmodified template or Line 25 after Step 5 ): Uncomment - name: Copy cockpit.conf.j2 to /etc/cockpit Hit Enter to generate the task code recommendation and accept the AI-suggested code (without modifications) by pressing Tab Compare your results with the SOLUTION tab below TEMPLATE SOLUTION # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' The AI-generated code recommendation will copy cockpit.conf to the target host. Take note of the fact that the recommendation included the mode: argument and set the Linux file permissions to 0644 , neither of which were things explicitly requested in the Task -name description, but are both additions which adhere to best practices around defining Ansible automaton tasks. Setting a file permission to 0644 specifies read and write permissions for User and Group levels within the Linux OS, and provides only read permissions to all others.","title":"ii. Single task Ansible operations"},{"location":"generating/#iii-multi-task-ansible-operations","text":"Up to this point, we've kept a narrow aperture on AI-generated recommendations for single tasks \u2014 examining and experimenting with generating Ansible code task by task, one at a time. However, a powerful WCA feature is the ability to combine multiple task descriptions into a single natural language prompt; in turn, WCA is able to parse that instruction, decompose the instruction into discrete Ansible Task parts, and return a complete code recommendation for achieving the author's intended goal. Syntactically, multiple tasks are combined into a single natural language expression through the use of ampersand ( & ) characters. Simply write out all the automation task descriptions on a single line, separating each description with a & character. The line must also begin with a # character for reasons that will be explained shortly. To illustrate, let's look at a multi-task Ansible Playbook: install_cockpit_multi-task.yml The contents of this Playbook should look familiar to you already: it is essentially the same Playbook examined in Steps 1-6 ( install_cockpit_single-task.yml ), re-written in an equivalent multi-task expression Each of the Task descriptions from the previous Playbook have been consolidated into a single description on Line 12 , separated by & characters ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_multi-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : # Install cockpit package & Copy cockpit.conf.j2 to /etc/cockpit & Start and enable service & Wait 15 seconds port 9090 There are two crucial distinctions between single task and multi-task code generation: formatting and execution. Formatting : Notice that Line 12 does not begin with -name: , as was the case with single task descriptions Execution : In order to generate AI code recommendations for multi-task descriptions, Line 12 must stay commented out (the # must remain at the start of the line) What is the rationale behind this? When WCA's generative AI capabilities parse Line 12 , its output will include multiple -name: tasks, each containing potentially multiple lines of instructions, based on how many & -delineated task descriptions are included on the line. Therefore, the way in which the code generation step is executed on Line 12 is a consequence of the formatting decision. Execution of a code generation step on a commented-out ( # ) line containing & delineators is recognized by WCA as a unique case that will be acted upon as a multi-task statement. Place your cursor at the end of Line 12 , and without removing the # character, press Enter to execute the code generation step. Be aware that generating code for multi-task descriptions will take longer compared to a single task. Compare the MULTI-TASK solution tab with the SINGLE TASK solution (copied over from Step 6 ). How did the multi-task code generation fare compared to the single task approach? MULTI-TASK SINGLE TASK 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit.socket - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit state : started enabled : true - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 Comparing the two results, the only notable difference between the two approaches are Lines 27-29 from the SINGLE TASK generative AI approach. Both the SINGLE TASK and MULTI-TASK suggestions for that particular task satisfy the request made by the user. However, whether the single task or multi-task approach resulted in a better code suggestion is up to the judgement of the programmer. Nearly 90% of the remaining code was identical between the two approaches and was achieved in far fewer lines of code (and less typing) using the multi-task approach. The variability of generative AI suggestions is a fascinating topic and one that we will dive more deeply into with the module ahead. Before moving on to other product features, experiment by creating a new Ansible Playbook in your workspace using the code template below. Suggestions will be given on how to perform the same automation task using single and multi-task generation approaches. Save the YAML file as create_ec2_single_multi.yml (if you forget to save the file, WCA will not generate recommendations) Copy the following code block to your clipboard using the + icon in the top-right corner of the panel and paste into the newly created YAML file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ~/Documents/ansible-wca-demo-kit/create_ec2_single_multi.yml 1 2 3 4 --- - name : Provision an EC2 instance hosts : all tasks : First, configure the Playbook for single task generation by adding the following code snippet into Line 5 (remembering to properly indent with Tab or Space characters): - name : create vpc named demo Your workspace Playbook should look identical to the TEMPLATE tab below. Place your cursor at the end of the newly-created Line 5 and hit Enter to execute single task code generation. Compare your results to the SOLUTION tab. Record your results to a notepad so that you can compare the results later. TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc Now it's time to re-write the Playbook for multi-task code generation. The beauty of natural language statements is that your approach can be as terse or verbose as you want. The more verbose and descriptive, the more prescriptive you can be in terms of influencing the AI-generated code recommendations from WCA. The topic of \"prompt tuning\" will be explored in much greater detail in the Task Description Tuning and Model Customization module. But for now: Replace all of the code inside create_ec2_single_multi.yml with the TEMPLATE code block (below) Place your cursor at the end of Line 5 (remember not to remove the # character) and hit Enter , then accept the recommendation with Tab Compare your results to the SOLUTION tab and to the result from Step 13 TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet - name : Create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc - name : Create security group for demo vpc and allow ssh and http amazon.aws.ec2_security_group : name : demo description : demo security group vpc_id : \"{ vpc.vpc.id }\" rules : - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 register : sg - name : Create internet gateway named demogateway for demo vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demogateway register : igw - name : Create subnet named demosubnet in demo vpc amazon.aws.ec2_vpc_subnet : cidr : 10.0.0.80/24 vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet register : subnet - name : Create route table for subnet demosubnet and gateway demogateway amazon.aws.ec2_vpc_route_table : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet subnets : - \"{ subnet.subnet.id }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ igw.gateway_id }\" register : route_table - name : Create t2.small ec2 instance in subnet demosubnet amazon.aws.ec2_instance : key_name : \"{ _key_name_ }\" instance_type : t2.small image : \"{ _image_ }\" wait : true vpc_subnet_id : \"{ subnet.subnet.id }\" security_group : demo register : ec2 TIMEOUT WARNING It may take several moments for WCA to process and return code recommendations for a multi-task description as complex as this one. If you receive a time-out warning, try executing the code generation step by pressing Enter a second time.","title":"iii. Multi-task Ansible operations"},{"location":"generating/#iv-next-steps","text":"The following module will examine in detail WCA's post-processing and content source attribution capabilities.","title":"iv. Next steps"},{"location":"setup/","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. Frequently asked questions and troubleshooting steps are documented below . i. Reserving the lab environments Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Account : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.4 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below. Locate the Request a Red Hat Account tile, hover over the IBM Cloud environment button with your cursor, and then click Reserve it [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. End Date and Time Select a time and date for when the reservation will expire. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 5-10 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will automatically be de-authorized and deleted by IBM Technology Zone after the reservation period has ended Now you must request your second ITZ environment, this time for the virtualized machine (VM) environment. Open the IBM Technology Zone reservation link below: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments Locate the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.5 tile Hover over the IBM Cloud environment button with your cursor Click Reserve it [A] to continue From the Single environment reservation options , select Reserve now . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. Customer Data Select No, I will not be using customer data if using for education purposes. End Date and Time Select a time and date for when the reservation will expire. VPN Access Set to Disabled . When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. ii. Accessing the VM Once the ITZ reservation has been marked as \"Ready\" , access connection details for the environment by either clicking the shortcut in the ITZ email or by drilling down into the My Reservations tab [A] on the ITZ web portal. Scroll down to the bottom of the reservation page and click the blue VM Remote Console button [A] to launch the VM interface. Select the techzone [A] login when prompted. Supply the following for the password: IBMDem0s! Hit Enter to complete the VM login Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B] iii. Authorizing VS Code with WCA Now you must authorize the VS Code environment for use with WCA, using the Red Hat account details that were generated for you in Steps 5-6 . Within your VS Code environment click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button Provide the Username and Password recorded in Step 6 of this module Click Submit to continue When prompted for We need a little more information , set Job Role to Student Click the Submit button to finalize your account activation COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not possible to \"paste\" lab instructions or information from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Allow this site to open the vscode link with Visual Studio Code - URL Handler? : click Open URL At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 15-17 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 15-17 as before; the login process should run smoother (and faster) on the second attempt iv. [OPTIONAL] Download demo assets to local machine If you wish to complete the hands-on lab using a local installation of VS Code (instead of using the provided Virtual Machine), you may do so \u2014 but you'll need to clone (download) the accompanying demo assets first. To do so, first install the GitHub CLI library. Execute the following command within a Terminal console to clone (via git ) the supporting demo assets repository ( ansible-wca-demo-kit.git ) to your local machine: git clone https://github.com/chetan-hireholi/ansible-wca-demo-kit Open the local ansible-wca-demo-kit folder within VS Code to access the demo assets for the remaining sections of the lab v. Troubleshooting and support If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. vi. Next steps The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"Setup and Troubleshooting"},{"location":"setup/#_1","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. Frequently asked questions and troubleshooting steps are documented below .","title":""},{"location":"setup/#i-reserving-the-lab-environments","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Account : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.4 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below. Locate the Request a Red Hat Account tile, hover over the IBM Cloud environment button with your cursor, and then click Reserve it [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. End Date and Time Select a time and date for when the reservation will expire. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 5-10 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will automatically be de-authorized and deleted by IBM Technology Zone after the reservation period has ended Now you must request your second ITZ environment, this time for the virtualized machine (VM) environment. Open the IBM Technology Zone reservation link below: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments Locate the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.5 tile Hover over the IBM Cloud environment button with your cursor Click Reserve it [A] to continue From the Single environment reservation options , select Reserve now . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. Customer Data Select No, I will not be using customer data if using for education purposes. End Date and Time Select a time and date for when the reservation will expire. VPN Access Set to Disabled . When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab.","title":"i. Reserving the lab environments"},{"location":"setup/#ii-accessing-the-vm","text":"Once the ITZ reservation has been marked as \"Ready\" , access connection details for the environment by either clicking the shortcut in the ITZ email or by drilling down into the My Reservations tab [A] on the ITZ web portal. Scroll down to the bottom of the reservation page and click the blue VM Remote Console button [A] to launch the VM interface. Select the techzone [A] login when prompted. Supply the following for the password: IBMDem0s! Hit Enter to complete the VM login Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B]","title":"ii. Accessing the VM"},{"location":"setup/#iii-authorizing-vs-code-with-wca","text":"Now you must authorize the VS Code environment for use with WCA, using the Red Hat account details that were generated for you in Steps 5-6 . Within your VS Code environment click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button Provide the Username and Password recorded in Step 6 of this module Click Submit to continue When prompted for We need a little more information , set Job Role to Student Click the Submit button to finalize your account activation COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not possible to \"paste\" lab instructions or information from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Allow this site to open the vscode link with Visual Studio Code - URL Handler? : click Open URL At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 15-17 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 15-17 as before; the login process should run smoother (and faster) on the second attempt","title":"iii. Authorizing VS Code with WCA"},{"location":"setup/#iv-optional-download-demo-assets-to-local-machine","text":"If you wish to complete the hands-on lab using a local installation of VS Code (instead of using the provided Virtual Machine), you may do so \u2014 but you'll need to clone (download) the accompanying demo assets first. To do so, first install the GitHub CLI library. Execute the following command within a Terminal console to clone (via git ) the supporting demo assets repository ( ansible-wca-demo-kit.git ) to your local machine: git clone https://github.com/chetan-hireholi/ansible-wca-demo-kit Open the local ansible-wca-demo-kit folder within VS Code to access the demo assets for the remaining sections of the lab","title":"iv. [OPTIONAL] Download demo assets to local machine"},{"location":"setup/#v-troubleshooting-and-support","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor.","title":"v. Troubleshooting and support"},{"location":"setup/#vi-next-steps","text":"The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"vi. Next steps"},{"location":"source-matching-post-processing/","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Content Source Matching and Post-Processing module. [10 min] i. Why code attribution and post-processing matters for generative AI At this point, you should now be well acquainted with Ansible Tasks and Playbooks . Ansible Playbooks are comprised of multiple Tasks \u2014 which you have been using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed's ( WCA ) generative AI capabilities to define and expand upon. If a Playbook contains multiple \"tasks\" that together achieve some singular goal or purpose, you might want to group these tasks into a single reusable unit. For example, a string of \"tasks\" might provision infrastructure for separate development, test, and production clusters from a major cloud vendor. Patterns for automating endpoints from these vendors can be codified into Ansible Modules , which are supported by Red Hat and the Ansible community writ large. And finally, collections of Playbooks and Modules can be organized into \"blueprints\" for automation tasks that are frequently used together to achieve some goal; a goal which you might want to make repeatable or shareable with other teams within your business. These blueprints are referred to as Ansible Roles . They provide a structured way to organize tasks, templates, files, and variables, and are available to drop directly into Ansible Playbooks. Roles make it possible to easily manage and set up complex automation tasks, essentially providing a rubric to streamline automation projects. Collectively, these Playbooks, Modules, and Roles form a comprehensive ecosystem of business and community-driven support for patterning automation to work with the broadest range of vendors, technologies, and clouds. Ansible Galaxy is a Red Hat-curated, community-driven repository for Ansible Roles . Via communities such as Galaxy, thousands of Roles are available for Ansible users to leverage within their own Playbooks. Ansible Galaxy is also key to how generative AI code recommendations from WCA are attributed back to original content sources and authors. ii. Content source matching A powerful capability within WCA is Content Source Matching (often referred to as \"code explainability\"), which attempts to match AI-generated code suggestions to the training data and sources that were utilized in generating the suggested Task code. These code attribution suggestions are created using a k-NN ( K-Nearest Neighbors ) algorithm that examines Ansible Galaxy and training data repositories in search of the nearest related content to the AI-generated code suggestions. To enable Content Source Matching capabilities within WCA, navigate to the main menu bar for VS Code and drill down into View > Open View... [A] The console [A] along the top of the VS Code interface is now activated and awaiting a prompt. Type Lightspeed Training Matches and hit Enter to confirm the selection. After making this configuration change to VS Code, all requests for WCA to generate code recommendations will now open a panel at the bottom of the VS Code interface It is under the Ansible tab that content source matching results will be displayed Other tabs will also displayed for variety of options: Output , Debug Console , Terminal , Ports , and Comments Open the install_pgsql-single-task.yml Ansible Playbook, which is included within the ansible-wca-demo-kit directory. The full location of the Playbook, as well as the contents, are available in the code block below. === \"~/Documents/ansible-wca-demo-kit/install and configure PostgreSQL and PGAdmin container/install_pgsql-single-task.yml\" ``` yaml linenums=\"1\" --- - name: Configure Database servers hosts: databases become: true tasks: # TASK 1 # - name: Install postgresql-server # TASK 2 # Ansible Lightspeed used an easy-to-understand natural language prompt and suggested the correct, more complex PostgreSQL CLI command to initiate the database. # Ansible Lightspeed used best practices and kept the task idempotent by including creates: /var/lib/pgsql/data/postgresql.conf in the suggestion. # - name: Run postgresql setup command # TASK 3 # Ansible Lightspeed used natural language prompt and added state: started and enabled: true module arguments based on Start and enable... in the Ansible task description. # - name: Start and enable postgresql service ``` Generate a code recommendation for the task on Line 8 by placing your cursor at the end of the line and hitting Enter . Pay attention to the code attribution details associated with this recommendation, which will be appearing under the Ansible tab at the bottom of the VS Code interface once the code recommendation is accepted with Tab You need to accept the AI-generated code suggestions (using the Tab key) before the Ansible content source matching tab will provide details about the code's origins - name : Install postgresql-server ansible.builtin.package : name : postgresql-server state : present The three most likely content sources used in training the WCA model\u2014 which produced the AI-generated code recommendations \u2014are listed within the Ansible tab [A] . Recall from earlier that these code attribution suggestions are created using a k-NN algorithm that searches Ansible Galaxy repositories for the nearest related content to the AI-generated code suggestions. Clicking the arrow icon to the left of each attribution [B] will expand further details about the source. Information about the URL , Path , Data Source , License , and Score are displayed (where available) under each listing. Red Hat-certified and maintained collections, as well as contributors to open source projects on Ansible Galaxy, are the primary sources for Ansible Lightspeed model training and are the content sources you are most likely to see matched to AI-generated code recommendations. Clicking on the URL field will redirect your web browser back to the precise collections and sources on Ansible Galaxy from which the code recommendations were derived. Here you can learn much richer details about the status of the project, any associated open source repositories involved (such as GitHub), contributions and activities ongoing with the code base, the author(s) involved, and many more intricacies. iii. Post-processing capabilities Another element of code generation that WCA excels at is understanding context within the Playbook it is executing against. If a variable or attribute is defined earlier within that Playbook, it will be recalled and referenced\u2014 where it makes sense to do so \u2014in the generation of subsequent lines of code. You may have already noticed these post-processing capabilities in your experimentations with WCA-generated code suggestions. Post-processing of Task descriptions and YAML file contents helps generate contextually aware, accurate Ansible content suggestions. However, one way to make this feature quite obvious is to take a previously-generated block of Ansible Task code, update the value assigned to a named variable earlier in the Playbook, and then regenerate the Task code block. In theory, the newly-generated Task block will use the updated variable name (and differ from how the code block was originally generated). Using the Explorer tab, create a New File... YAML Playbook named demo_provision_ec2_instance.yml anywhere within the /ansible-wca-demo-kit/ directory. You may save the Playbook in the root of the folder or any of the subdirectories Be sure to reference the expandable tooltips below for instructions on how to perform these operations and how to configure the Playbook language for use with WCA Copy and paste the following code block into the demo_provision_ec2_instance.yml Playbook and then save the file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible demo_provision_ec2_instance.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 --- - name : EC2 Cloud Operations hosts : localhost connection : local gather_facts : false module_defaults : group/aws : region : us-east-1 # vars: # ec2_instance: # name: lightspeed-instance-01 # key_name: lightspeed-keypair # image_id: ami-016eb5d644c333ccb # RHEL 9 us-east-1 # tags: # function: lightspeed-demo # security_group: secgroup-lightspeed tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - Best practices: The suggestion used the Fully Qualified Collection name. # # Note - Context: Ansible Lightspeed used the Playbook name \"EC2 Cloud Operations\" to use the correct \"amazon.aws.ec2_vpc_subnet_info\" module. - name : Gather subnet info tag:Name subnet-lightspeed # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Context: The suggestion included the previous task's registered variable in the suggestion. # # Note - Accuracy: The suggestion provides the correct key value from the previously task's registered variable. - name : Create vpc_subnet_id var # TASK 3 # # 3a. Uncomment task description \"Provision t3.micro instance\" below and generate a task suggestion. # # Note - Efficiency: The suggestion provides practical variable examples to improve efficiency. - name : Provision t3.micro instance # # 3b. Remove the above task and suggestion. # # Uncomment 2nd task description \"Provision t3.micro instance using ec2_instance var\". # # Generate an updated suggestion. # # Note - Context: The updated suggestion includes the \"ec2_instance variable fields in the suggestion\" - name : Provision t3.micro instance using ec2_instance var TASK 1 ( Line 26 ) and TASK 2 ( Line 33 ) are responsible for gathering information about a network subnet that is to be provisioned and then creating a virtual private cloud (VPC) definition based on those details. Generate suggested code for TASK 1 ( Line 26 ) by placing the cursor at the end of the line, hitting Enter , and then confirming with Tab . Afterwards, perform the same steps to generate code for TASK 2 as well. The first round of WCA-generated code produces a code block [A] with a register: subnet_info ( Line 30 ), the result of which is to assign this VPC definition to a variable subnet_info . Nothing terribly complicated or surprising about that. The WCA-generated code that follows for the task on Line 38 (previously Line 33 in the raw template) recommends a code block [B] with vpc_subnet_id: \"{ { subnet_info.subnets[0].subnet_id } }\" as the value associated with the VPC's subnet ID. Critically, the variable subnet_info that was generated in the previous Task [A] is also referenced in the second Task. This demonstrates the contextual awareness of WCA in action. Change the name of variable subnet_info to subnet_name on Line 30 . Then delete the code block recommendations under Line 38 and regenerate the task. Notice that the new block [A] of TASK 2 now references the variable subnet_name that was modified just a moment ago in TASK 1 WCA has generated code for TASK 2 that takes into account the modified context and variables from TASK 1 of the Playbook Continue experimenting with WCA's contextual awareness and post-processing capabilities. Try adjusting other variables within the Playbook and study how these modifications impact the generation of later blocks of Task code within the Playbook. iv. Next steps In the next section, you will begin experimenting with customized Ansible Playbooks and testing how changes to Ansible Task natural language descriptions impacts the recommended code produced by WCA.","title":"Content Source Matching and Post-Processing"},{"location":"source-matching-post-processing/#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Content Source Matching and Post-Processing module. [10 min]","title":""},{"location":"source-matching-post-processing/#i-why-code-attribution-and-post-processing-matters-for-generative-ai","text":"At this point, you should now be well acquainted with Ansible Tasks and Playbooks . Ansible Playbooks are comprised of multiple Tasks \u2014 which you have been using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed's ( WCA ) generative AI capabilities to define and expand upon. If a Playbook contains multiple \"tasks\" that together achieve some singular goal or purpose, you might want to group these tasks into a single reusable unit. For example, a string of \"tasks\" might provision infrastructure for separate development, test, and production clusters from a major cloud vendor. Patterns for automating endpoints from these vendors can be codified into Ansible Modules , which are supported by Red Hat and the Ansible community writ large. And finally, collections of Playbooks and Modules can be organized into \"blueprints\" for automation tasks that are frequently used together to achieve some goal; a goal which you might want to make repeatable or shareable with other teams within your business. These blueprints are referred to as Ansible Roles . They provide a structured way to organize tasks, templates, files, and variables, and are available to drop directly into Ansible Playbooks. Roles make it possible to easily manage and set up complex automation tasks, essentially providing a rubric to streamline automation projects. Collectively, these Playbooks, Modules, and Roles form a comprehensive ecosystem of business and community-driven support for patterning automation to work with the broadest range of vendors, technologies, and clouds. Ansible Galaxy is a Red Hat-curated, community-driven repository for Ansible Roles . Via communities such as Galaxy, thousands of Roles are available for Ansible users to leverage within their own Playbooks. Ansible Galaxy is also key to how generative AI code recommendations from WCA are attributed back to original content sources and authors.","title":"i. Why code attribution and post-processing matters for generative AI"},{"location":"source-matching-post-processing/#ii-content-source-matching","text":"A powerful capability within WCA is Content Source Matching (often referred to as \"code explainability\"), which attempts to match AI-generated code suggestions to the training data and sources that were utilized in generating the suggested Task code. These code attribution suggestions are created using a k-NN ( K-Nearest Neighbors ) algorithm that examines Ansible Galaxy and training data repositories in search of the nearest related content to the AI-generated code suggestions. To enable Content Source Matching capabilities within WCA, navigate to the main menu bar for VS Code and drill down into View > Open View... [A] The console [A] along the top of the VS Code interface is now activated and awaiting a prompt. Type Lightspeed Training Matches and hit Enter to confirm the selection. After making this configuration change to VS Code, all requests for WCA to generate code recommendations will now open a panel at the bottom of the VS Code interface It is under the Ansible tab that content source matching results will be displayed Other tabs will also displayed for variety of options: Output , Debug Console , Terminal , Ports , and Comments Open the install_pgsql-single-task.yml Ansible Playbook, which is included within the ansible-wca-demo-kit directory. The full location of the Playbook, as well as the contents, are available in the code block below. === \"~/Documents/ansible-wca-demo-kit/install and configure PostgreSQL and PGAdmin container/install_pgsql-single-task.yml\" ``` yaml linenums=\"1\" --- - name: Configure Database servers hosts: databases become: true tasks: # TASK 1 # - name: Install postgresql-server # TASK 2 # Ansible Lightspeed used an easy-to-understand natural language prompt and suggested the correct, more complex PostgreSQL CLI command to initiate the database. # Ansible Lightspeed used best practices and kept the task idempotent by including creates: /var/lib/pgsql/data/postgresql.conf in the suggestion. # - name: Run postgresql setup command # TASK 3 # Ansible Lightspeed used natural language prompt and added state: started and enabled: true module arguments based on Start and enable... in the Ansible task description. # - name: Start and enable postgresql service ``` Generate a code recommendation for the task on Line 8 by placing your cursor at the end of the line and hitting Enter . Pay attention to the code attribution details associated with this recommendation, which will be appearing under the Ansible tab at the bottom of the VS Code interface once the code recommendation is accepted with Tab You need to accept the AI-generated code suggestions (using the Tab key) before the Ansible content source matching tab will provide details about the code's origins - name : Install postgresql-server ansible.builtin.package : name : postgresql-server state : present The three most likely content sources used in training the WCA model\u2014 which produced the AI-generated code recommendations \u2014are listed within the Ansible tab [A] . Recall from earlier that these code attribution suggestions are created using a k-NN algorithm that searches Ansible Galaxy repositories for the nearest related content to the AI-generated code suggestions. Clicking the arrow icon to the left of each attribution [B] will expand further details about the source. Information about the URL , Path , Data Source , License , and Score are displayed (where available) under each listing. Red Hat-certified and maintained collections, as well as contributors to open source projects on Ansible Galaxy, are the primary sources for Ansible Lightspeed model training and are the content sources you are most likely to see matched to AI-generated code recommendations. Clicking on the URL field will redirect your web browser back to the precise collections and sources on Ansible Galaxy from which the code recommendations were derived. Here you can learn much richer details about the status of the project, any associated open source repositories involved (such as GitHub), contributions and activities ongoing with the code base, the author(s) involved, and many more intricacies.","title":"ii. Content source matching"},{"location":"source-matching-post-processing/#iii-post-processing-capabilities","text":"Another element of code generation that WCA excels at is understanding context within the Playbook it is executing against. If a variable or attribute is defined earlier within that Playbook, it will be recalled and referenced\u2014 where it makes sense to do so \u2014in the generation of subsequent lines of code. You may have already noticed these post-processing capabilities in your experimentations with WCA-generated code suggestions. Post-processing of Task descriptions and YAML file contents helps generate contextually aware, accurate Ansible content suggestions. However, one way to make this feature quite obvious is to take a previously-generated block of Ansible Task code, update the value assigned to a named variable earlier in the Playbook, and then regenerate the Task code block. In theory, the newly-generated Task block will use the updated variable name (and differ from how the code block was originally generated). Using the Explorer tab, create a New File... YAML Playbook named demo_provision_ec2_instance.yml anywhere within the /ansible-wca-demo-kit/ directory. You may save the Playbook in the root of the folder or any of the subdirectories Be sure to reference the expandable tooltips below for instructions on how to perform these operations and how to configure the Playbook language for use with WCA Copy and paste the following code block into the demo_provision_ec2_instance.yml Playbook and then save the file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible demo_provision_ec2_instance.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 --- - name : EC2 Cloud Operations hosts : localhost connection : local gather_facts : false module_defaults : group/aws : region : us-east-1 # vars: # ec2_instance: # name: lightspeed-instance-01 # key_name: lightspeed-keypair # image_id: ami-016eb5d644c333ccb # RHEL 9 us-east-1 # tags: # function: lightspeed-demo # security_group: secgroup-lightspeed tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - Best practices: The suggestion used the Fully Qualified Collection name. # # Note - Context: Ansible Lightspeed used the Playbook name \"EC2 Cloud Operations\" to use the correct \"amazon.aws.ec2_vpc_subnet_info\" module. - name : Gather subnet info tag:Name subnet-lightspeed # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Context: The suggestion included the previous task's registered variable in the suggestion. # # Note - Accuracy: The suggestion provides the correct key value from the previously task's registered variable. - name : Create vpc_subnet_id var # TASK 3 # # 3a. Uncomment task description \"Provision t3.micro instance\" below and generate a task suggestion. # # Note - Efficiency: The suggestion provides practical variable examples to improve efficiency. - name : Provision t3.micro instance # # 3b. Remove the above task and suggestion. # # Uncomment 2nd task description \"Provision t3.micro instance using ec2_instance var\". # # Generate an updated suggestion. # # Note - Context: The updated suggestion includes the \"ec2_instance variable fields in the suggestion\" - name : Provision t3.micro instance using ec2_instance var TASK 1 ( Line 26 ) and TASK 2 ( Line 33 ) are responsible for gathering information about a network subnet that is to be provisioned and then creating a virtual private cloud (VPC) definition based on those details. Generate suggested code for TASK 1 ( Line 26 ) by placing the cursor at the end of the line, hitting Enter , and then confirming with Tab . Afterwards, perform the same steps to generate code for TASK 2 as well. The first round of WCA-generated code produces a code block [A] with a register: subnet_info ( Line 30 ), the result of which is to assign this VPC definition to a variable subnet_info . Nothing terribly complicated or surprising about that. The WCA-generated code that follows for the task on Line 38 (previously Line 33 in the raw template) recommends a code block [B] with vpc_subnet_id: \"{ { subnet_info.subnets[0].subnet_id } }\" as the value associated with the VPC's subnet ID. Critically, the variable subnet_info that was generated in the previous Task [A] is also referenced in the second Task. This demonstrates the contextual awareness of WCA in action. Change the name of variable subnet_info to subnet_name on Line 30 . Then delete the code block recommendations under Line 38 and regenerate the task. Notice that the new block [A] of TASK 2 now references the variable subnet_name that was modified just a moment ago in TASK 1 WCA has generated code for TASK 2 that takes into account the modified context and variables from TASK 1 of the Playbook Continue experimenting with WCA's contextual awareness and post-processing capabilities. Try adjusting other variables within the Playbook and study how these modifications impact the generation of later blocks of Task code within the Playbook.","title":"iii. Post-processing capabilities"},{"location":"source-matching-post-processing/#iv-next-steps","text":"In the next section, you will begin experimenting with customized Ansible Playbooks and testing how changes to Ansible Task natural language descriptions impacts the recommended code produced by WCA.","title":"iv. Next steps"}]}
\ No newline at end of file
+{"config":{"indexing":"full","lang":["en"],"min_search_length":3,"prebuild_index":false,"separator":"[\\s\\-]+"},"docs":[{"location":"","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) introduces IBM watsonx Code Assistant's generative AI capabilities and lays the groundwork for the hands-on training that will follow. [6 min] i. Automation is indispensable to modern IT strategy Despite the innovations and advancements made in the domain of automation, IBM sellers and partners know first-hand from discussions with clients that many businesses are still struggling to keep up with their IT operations. The rapid pace of technological innovation\u2014 in particular, areas such as AI and machine learning \u2014are obviously challenging for any organization to strategize and plan around. But smaller, more practical challenges also stand in the way of these businesses. The fact remains that IT operations, and wrangling those operations in an efficient and streamlined manner, remains a difficult problem to solve. Three primary pain points that IBM consistently hears from the marketplace include: an ever-increasing skills gap in IT management; that Day 2 operations continue to be labor-intensive, mostly manual endeavors; and that the complexity of the systems needing to be managed are out-pacing many organization\u2019s ability to adapt. All of these pain points are potential automation challenges to be solved. Each of them impedes a company's ability to move quickly and adapt for the future. And as such, for many IBM clients, solving these automation challenges have become an indispensable element in their strategy to modernize IT. UNPRECEDENTED RATE OF GENERATIVE AI ADOPTION Even though generative AI is relatively new, the widespread popularity of ChatGPT has created significant interest in the notion of large language models (LLMs) and foundation models (FMs) \u2014 and what they can do for business. It took quite some time for enterprises to start moving toward traditional AI. In contrast, generative AI has experienced massive early adoption: 80% of enterprises are already working with, or planning to leverage FMs, and plan to adopt generative AI in their use cases and workflow. Moreover, the following data points to an ever-growing adoption trend for generative AI: Scale Zeitgeist 2023 AI Readiness Report notes that with the companies they reviewed, 21% have generative AI models in production; 29% are experimenting with generative AI and another 31% are planning to work with generative AI models; a total of 81% are either working with or planning to work with generative AI models Goldman Sachs has estimated that generative AI will have a very deep economic impact \u2013 raising global Gross Domestic Product (GDP) by 7% within 10 years, reflecting the technology\u2019s huge potential. Boston Consulting Group (BCG) noted that generative AI is expected to represent 30% of the overall market by 2025 ii. Generative AI-assisted code lifecycle management achieves what LLMs alone cannot Following the debut of OpenAI's ChatGPT, the marketplace has been awash with competing large language model (LLM) and generative AI-based assistants. It's one thing to train and deploy an LLM; it's another thing entirely to make it applicable and tangibly beneficial for business. What separates IBM watsonx Code Assistant (WCA) offerings from competing vendors in the marketplace? The design and implementation of WCA is purposely built to assist, using generative AI, software and code lifecycle management. In short, generative AI-assisted code lifecycle management helps to achieve what large language models cannot achieve on their own. It is what distinguished WCA from other code assistants in the marketplace today. Code lifecycle management begins with understanding client code, through training across a myriad of programming languages and specializations in paradigms such as Ansible Automation Platform, and applies that understanding across a client\u2019s application and runtime environments. Users are able to plan next steps based on generative AI analysis of their existing application code. Operations teams can rapidly transform their codebases with optimized design and architecture that is recommended according to IBM Granite's best-practice models. Administrators can validate the outcomes with automatically generated unit tests. Afterwards, they can deploy those services and applications using automated processes like Ansible's automation engine. Over the course of that application or code's lifecycle, generative AI can maintain healthy operations with runtime insights. iii. Introducing the IBM watsonx Code Assistant product family IBM watsonx Code Assistant is the flagship offering in a suite of generative AI code assistant products, which also include offerings for Ansible Automation Platform (IBM watsonx Code Assistant for Red Hat Ansible Lightspeed) and IBM Systems modernization (IBM watsonx Code Assistant for Z). These solutions accelerate software development tasks with AI-powered capabilities including context-aware code generation, explanation, documentation, translation, and unit test generation. It does so while maintaining the principles of trust, security, and compliance with regards to IBM client's data and intellectual property. Developers and IT Operators can utilize WCA to speed up application modernization efforts and generate Ansible-based automation jobs to rapidly scale out (or scale up) IT environments. IBM watsonx Code Assistant products are powered by IBM Granite foundation models that include state-of-the-art large language models designed for code. For offerings such as WCA for Ansible Lightspeed and WCA for Z, bespoke code models\u2014 tailored to working with Ansible Automation Platform and COBOL-to-Z use cases, respectively \u2014are invoked. Universally true for all of the watsonx Code Assistant offerings is that they are geared towards helping IT teams create high-quality code using AI-generated recommendations, based on natural language requests or existing source code. These AI models, and the recommendations they generate, are seamlessly integrated via extensions with the world's most popular developer integrated development environments (IDE), including Visual Studio Code and Eclipse. Granite is IBM\u2019s flagship brand of open and proprietary LLMs, spanning multiple modalities. Granite models exist for code, languages, time series, and GeoSpatial \u2014 with additional modalities expected in future. IBM Granite code models are a series of decoder-only models for code generative tasks, trained with code written in 116 different programming languages. The Granite code models family consists of models ranging in size from 3 to 34 billion parameters, in both a base model and instruction-following model variants. These models have a range of uses, from complex application modernization tasks to on-device memory-constrained use cases. The larger the block size for a particular language on this chart, the larger percentage of training corpus data of that language was used to train the Granite code model. Languages and formats such as Java, C, JSON, JavaScript, HTML, and PHP are subjects in which the model \u201cMajors\u201d and excels. Other languages such as Ruby, SQL, and Swift could be considered \u201cMinors\u201d where the generalized code model can work with the language, but has less training data to base those recommendations on. These percentages and training data volumes will continue to evolve as the Granite code models mature. WATSONX CODE ASSISTANT vs. WCA FOR ANSIBLE LIGHTSPEED? For those familiar with other IBM watsonx Code Assistant offerings\u2014 such as WCA for Red Hat Ansible Lightspeed and WCA for Z \u2014the generalized code model approach, as seen here, differs from the specialized code model approach of those two aforementioned offerings. The WCA for Ansible Lightspeed flavor of IBM Granite code models specializes (Majors) only in Ansible Playbooks and YAML Similarly, the IBM Granite code model used by WCA for Z specializes in transforming COBOL mainframe code into modernized Java code Ansible Playbooks (YAML) and mainframe (COBOL) code are both supported (Minor) languages for the generalized IBM Granite code models\u2014 and therefore are supported by IBM watsonx Code Assistant \u2014but if a client wishes to specialize in those particular languages and frameworks, they would be well advised to utilize the bespoke WCA for Ansible Lightspeed and WCA for Z offerings, respectively, to do so. iv. Solution architecture of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed IBM watsonx Code Assistant for Red Hat Ansible Lightspeed meets developers where they are: with a rich plugin via VS Code extensions, where developers input their prompts directly in the code editor. Prompts are sent to the Ansible Lightspeed service, and the service sends a suggestion back (a completion ) that\u2019s powered by IBM Granite LLMs for code. It is important to note that all data in transit is encrypted and ephemeral so users can be confident and have trust in the security of the service during this exchange. In terms of data security, client Ansible playbooks and customized models that they may potentially have are stored in client-owned Cloud Object Storage and are not shared with IBM, Red Hat, or any other clients. In order to utilize IBM watsonx Code Assistant for Red Hat Ansible Lightspeed, a client must have an existing license for Red Hat Ansible Automation Platform (the \u201dred tile\u201d component in the center of the diagram), as well as a license for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (the \u201cblue tile\u201d on the right of the diagram). Generative AI has recently demonstrated proficiency in creating syntactically correct and contextually relevant application code in a variety of programming languages. For example, if trained on a large dataset of Ansible Playbooks, generative AI models can be fine-tuned to understand the nuances of Playbook syntax and structure. An enterprise organization with dozens or hundreds of Playbooks within their IT estate today would have a rich corpus of training data on-hand that could be used to fine-tune AI models that are tailored to the automation needs and programming style or standards of that particular company. WHAT ARE PLAYBOOKS? Ansible Playbooks instruct Ansible\u2019s automation engine on how to execute tasks in a step-by-step manner. Playbooks defines roles, tasks, handlers, and other configurations; in turn, these attributes allow developers and users to codify complex orchestration scenarios. Conceptually, think of a Playbook as a recipe book for system administration: each recipe (or Playbook) spells out the steps required to achieve a particular system state or to complete a given operation. One of the standout features of Ansible Playbooks is that they are idempotent : executing Playbooks multiple times on the same system won't create additional \"side effects\" (unintended operations or creation of unwanted artifacts) after the first successful run. This ensures consistency and reliability across deployments of the Red Hat Ansible Automation Platform ( AAP ). As you will see throughout the hands-on training material, generative AI models provide a natural language prompt to users which in turn is understood and translated by the AI models into the necessary Ansible Task code. For example, a user might describe a desired system state in plain language ( \"I want a Playbook to install and start an Apache web server\" ) and the model will generate the appropriate Ansible Tasks for a Playbook. All of this is achieved without physically writing code or requiring much programming expertise. Not only does this speed up the automation process by cutting the time needed to author Playbooks, but it also democratizes access to automation in general. Even those within a company with limited Ansible or programming expertise will be able to produce effective Playbooks. There are plenty of caveats of course, and thorough validation and testing of AI-generated code will be needed before being put into production. However, the productivity gains and broadening of skillsets within an organization can be tremendous. And as a whole, generative AI brings the original goals of Red Hat Ansible Automation Platform (the democratization of automation for everything) that much closer to a reality. v. Lab objectives The material covered for this hands-on training is intended to prepare IBM sellers and business partners with the skills necessary to create Ansible automation tasks using the generative AI capabilities of WCA. The curriculum will leverage WCA's generative AI code recommendations for automating cloud-based and infrastructure-based automation tasks. In-depth explanations accompanying Ansible Playbook templates will also explain: How WCA uses natural language prompts , as well as Ansible Playbook contents, to generate contextually-aware Task code recommendations Post-processing capabilities that refine the generative AI suggestions into syntactically correct code (adherent to best practices) How WCA provides content source matching attribution and \"explainability\" for all AI-generated content Leveraging WCA's model tuning capabilities to tailor content and code recommendations to an organization's standards, best practices, and programming styles vi. Next steps The module ahead will outline the evaluation criteria for IBM sellers and business partners. Afterwards, you will setup your local environment with the necessary pre-requisites for getting started with the hands-on material.","title":"Introduction"},{"location":"#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) introduces IBM watsonx Code Assistant's generative AI capabilities and lays the groundwork for the hands-on training that will follow. [6 min]","title":""},{"location":"#i-automation-is-indispensable-to-modern-it-strategy","text":"Despite the innovations and advancements made in the domain of automation, IBM sellers and partners know first-hand from discussions with clients that many businesses are still struggling to keep up with their IT operations. The rapid pace of technological innovation\u2014 in particular, areas such as AI and machine learning \u2014are obviously challenging for any organization to strategize and plan around. But smaller, more practical challenges also stand in the way of these businesses. The fact remains that IT operations, and wrangling those operations in an efficient and streamlined manner, remains a difficult problem to solve. Three primary pain points that IBM consistently hears from the marketplace include: an ever-increasing skills gap in IT management; that Day 2 operations continue to be labor-intensive, mostly manual endeavors; and that the complexity of the systems needing to be managed are out-pacing many organization\u2019s ability to adapt. All of these pain points are potential automation challenges to be solved. Each of them impedes a company's ability to move quickly and adapt for the future. And as such, for many IBM clients, solving these automation challenges have become an indispensable element in their strategy to modernize IT. UNPRECEDENTED RATE OF GENERATIVE AI ADOPTION Even though generative AI is relatively new, the widespread popularity of ChatGPT has created significant interest in the notion of large language models (LLMs) and foundation models (FMs) \u2014 and what they can do for business. It took quite some time for enterprises to start moving toward traditional AI. In contrast, generative AI has experienced massive early adoption: 80% of enterprises are already working with, or planning to leverage FMs, and plan to adopt generative AI in their use cases and workflow. Moreover, the following data points to an ever-growing adoption trend for generative AI: Scale Zeitgeist 2023 AI Readiness Report notes that with the companies they reviewed, 21% have generative AI models in production; 29% are experimenting with generative AI and another 31% are planning to work with generative AI models; a total of 81% are either working with or planning to work with generative AI models Goldman Sachs has estimated that generative AI will have a very deep economic impact \u2013 raising global Gross Domestic Product (GDP) by 7% within 10 years, reflecting the technology\u2019s huge potential. Boston Consulting Group (BCG) noted that generative AI is expected to represent 30% of the overall market by 2025","title":"i. Automation is indispensable to modern IT strategy"},{"location":"#ii-generative-ai-assisted-code-lifecycle-management-achieves-what-llms-alone-cannot","text":"Following the debut of OpenAI's ChatGPT, the marketplace has been awash with competing large language model (LLM) and generative AI-based assistants. It's one thing to train and deploy an LLM; it's another thing entirely to make it applicable and tangibly beneficial for business. What separates IBM watsonx Code Assistant (WCA) offerings from competing vendors in the marketplace? The design and implementation of WCA is purposely built to assist, using generative AI, software and code lifecycle management. In short, generative AI-assisted code lifecycle management helps to achieve what large language models cannot achieve on their own. It is what distinguished WCA from other code assistants in the marketplace today. Code lifecycle management begins with understanding client code, through training across a myriad of programming languages and specializations in paradigms such as Ansible Automation Platform, and applies that understanding across a client\u2019s application and runtime environments. Users are able to plan next steps based on generative AI analysis of their existing application code. Operations teams can rapidly transform their codebases with optimized design and architecture that is recommended according to IBM Granite's best-practice models. Administrators can validate the outcomes with automatically generated unit tests. Afterwards, they can deploy those services and applications using automated processes like Ansible's automation engine. Over the course of that application or code's lifecycle, generative AI can maintain healthy operations with runtime insights.","title":"ii. Generative AI-assisted code lifecycle management achieves what LLMs alone cannot"},{"location":"#iii-introducing-the-ibm-watsonx-code-assistant-product-family","text":"IBM watsonx Code Assistant is the flagship offering in a suite of generative AI code assistant products, which also include offerings for Ansible Automation Platform (IBM watsonx Code Assistant for Red Hat Ansible Lightspeed) and IBM Systems modernization (IBM watsonx Code Assistant for Z). These solutions accelerate software development tasks with AI-powered capabilities including context-aware code generation, explanation, documentation, translation, and unit test generation. It does so while maintaining the principles of trust, security, and compliance with regards to IBM client's data and intellectual property. Developers and IT Operators can utilize WCA to speed up application modernization efforts and generate Ansible-based automation jobs to rapidly scale out (or scale up) IT environments. IBM watsonx Code Assistant products are powered by IBM Granite foundation models that include state-of-the-art large language models designed for code. For offerings such as WCA for Ansible Lightspeed and WCA for Z, bespoke code models\u2014 tailored to working with Ansible Automation Platform and COBOL-to-Z use cases, respectively \u2014are invoked. Universally true for all of the watsonx Code Assistant offerings is that they are geared towards helping IT teams create high-quality code using AI-generated recommendations, based on natural language requests or existing source code. These AI models, and the recommendations they generate, are seamlessly integrated via extensions with the world's most popular developer integrated development environments (IDE), including Visual Studio Code and Eclipse. Granite is IBM\u2019s flagship brand of open and proprietary LLMs, spanning multiple modalities. Granite models exist for code, languages, time series, and GeoSpatial \u2014 with additional modalities expected in future. IBM Granite code models are a series of decoder-only models for code generative tasks, trained with code written in 116 different programming languages. The Granite code models family consists of models ranging in size from 3 to 34 billion parameters, in both a base model and instruction-following model variants. These models have a range of uses, from complex application modernization tasks to on-device memory-constrained use cases. The larger the block size for a particular language on this chart, the larger percentage of training corpus data of that language was used to train the Granite code model. Languages and formats such as Java, C, JSON, JavaScript, HTML, and PHP are subjects in which the model \u201cMajors\u201d and excels. Other languages such as Ruby, SQL, and Swift could be considered \u201cMinors\u201d where the generalized code model can work with the language, but has less training data to base those recommendations on. These percentages and training data volumes will continue to evolve as the Granite code models mature. WATSONX CODE ASSISTANT vs. WCA FOR ANSIBLE LIGHTSPEED? For those familiar with other IBM watsonx Code Assistant offerings\u2014 such as WCA for Red Hat Ansible Lightspeed and WCA for Z \u2014the generalized code model approach, as seen here, differs from the specialized code model approach of those two aforementioned offerings. The WCA for Ansible Lightspeed flavor of IBM Granite code models specializes (Majors) only in Ansible Playbooks and YAML Similarly, the IBM Granite code model used by WCA for Z specializes in transforming COBOL mainframe code into modernized Java code Ansible Playbooks (YAML) and mainframe (COBOL) code are both supported (Minor) languages for the generalized IBM Granite code models\u2014 and therefore are supported by IBM watsonx Code Assistant \u2014but if a client wishes to specialize in those particular languages and frameworks, they would be well advised to utilize the bespoke WCA for Ansible Lightspeed and WCA for Z offerings, respectively, to do so.","title":"iii. Introducing the IBM watsonx Code Assistant product family"},{"location":"#iv-solution-architecture-of-ibm-watsonx-code-assistant-for-red-hat-ansible-lightspeed","text":"IBM watsonx Code Assistant for Red Hat Ansible Lightspeed meets developers where they are: with a rich plugin via VS Code extensions, where developers input their prompts directly in the code editor. Prompts are sent to the Ansible Lightspeed service, and the service sends a suggestion back (a completion ) that\u2019s powered by IBM Granite LLMs for code. It is important to note that all data in transit is encrypted and ephemeral so users can be confident and have trust in the security of the service during this exchange. In terms of data security, client Ansible playbooks and customized models that they may potentially have are stored in client-owned Cloud Object Storage and are not shared with IBM, Red Hat, or any other clients. In order to utilize IBM watsonx Code Assistant for Red Hat Ansible Lightspeed, a client must have an existing license for Red Hat Ansible Automation Platform (the \u201dred tile\u201d component in the center of the diagram), as well as a license for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (the \u201cblue tile\u201d on the right of the diagram). Generative AI has recently demonstrated proficiency in creating syntactically correct and contextually relevant application code in a variety of programming languages. For example, if trained on a large dataset of Ansible Playbooks, generative AI models can be fine-tuned to understand the nuances of Playbook syntax and structure. An enterprise organization with dozens or hundreds of Playbooks within their IT estate today would have a rich corpus of training data on-hand that could be used to fine-tune AI models that are tailored to the automation needs and programming style or standards of that particular company. WHAT ARE PLAYBOOKS? Ansible Playbooks instruct Ansible\u2019s automation engine on how to execute tasks in a step-by-step manner. Playbooks defines roles, tasks, handlers, and other configurations; in turn, these attributes allow developers and users to codify complex orchestration scenarios. Conceptually, think of a Playbook as a recipe book for system administration: each recipe (or Playbook) spells out the steps required to achieve a particular system state or to complete a given operation. One of the standout features of Ansible Playbooks is that they are idempotent : executing Playbooks multiple times on the same system won't create additional \"side effects\" (unintended operations or creation of unwanted artifacts) after the first successful run. This ensures consistency and reliability across deployments of the Red Hat Ansible Automation Platform ( AAP ). As you will see throughout the hands-on training material, generative AI models provide a natural language prompt to users which in turn is understood and translated by the AI models into the necessary Ansible Task code. For example, a user might describe a desired system state in plain language ( \"I want a Playbook to install and start an Apache web server\" ) and the model will generate the appropriate Ansible Tasks for a Playbook. All of this is achieved without physically writing code or requiring much programming expertise. Not only does this speed up the automation process by cutting the time needed to author Playbooks, but it also democratizes access to automation in general. Even those within a company with limited Ansible or programming expertise will be able to produce effective Playbooks. There are plenty of caveats of course, and thorough validation and testing of AI-generated code will be needed before being put into production. However, the productivity gains and broadening of skillsets within an organization can be tremendous. And as a whole, generative AI brings the original goals of Red Hat Ansible Automation Platform (the democratization of automation for everything) that much closer to a reality.","title":"iv. Solution architecture of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed"},{"location":"#v-lab-objectives","text":"The material covered for this hands-on training is intended to prepare IBM sellers and business partners with the skills necessary to create Ansible automation tasks using the generative AI capabilities of WCA. The curriculum will leverage WCA's generative AI code recommendations for automating cloud-based and infrastructure-based automation tasks. In-depth explanations accompanying Ansible Playbook templates will also explain: How WCA uses natural language prompts , as well as Ansible Playbook contents, to generate contextually-aware Task code recommendations Post-processing capabilities that refine the generative AI suggestions into syntactically correct code (adherent to best practices) How WCA provides content source matching attribution and \"explainability\" for all AI-generated content Leveraging WCA's model tuning capabilities to tailor content and code recommendations to an organization's standards, best practices, and programming styles","title":"v. Lab objectives"},{"location":"#vi-next-steps","text":"The module ahead will outline the evaluation criteria for IBM sellers and business partners. Afterwards, you will setup your local environment with the necessary pre-requisites for getting started with the hands-on material.","title":"vi. Next steps"},{"location":"advanced/","text":"OPTIONAL SECTION \u2014 NOT REQUIRED FOR LEVEL 3 CREDIT IBM technical sellers and business partners are NOT required to complete this section to receive Level 3 credit. However, it is highly recommended by the authors that you attempt the material, given the tremendous demonstration value it provides for clients. COSTS & BILLING Be aware that deploying a live Amazon EC2 instance on AWS does incur real-world charges which are the responsibility of the individual to pay for. If you wish to go ahead with deploying the live environment\u2014 which you are encouraged, but NOT REQUIRED, to complete \u2014you will need to do so on your personal AWS account, with any potential billings and charges incurred to your personal credit card. The billing associated with an Amazon EC2 t2.small instance and other services needed for the demo is minimal : an on-demand hourly rate of $0.023 USD at the time of publication. Running the fully-deployed WordPress and EC2 environments for 24 hours cost USD $2.03 in billing when testing for publication. Additional AWS pricing details and plans are available online. To conclude the Level 3 Technical Sales education on IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will be provided with a complete set of instructions for deploying a live Amazon Web Services (AWS) environment with Ansible automation, consisting of a WordPress web application running atop an Amazon EC2 VPC . The deployment of the environments will be conducted entirely using Red Hat Ansible automation, executed on your local machine, and deployed to AWS cloud. The generation of the Ansible Tasks and code required to automate these deployments will be created using the generative AI capabilities of the WCA extension for VS Code. Ansible Playbooks to support this demo were prepared ahead of time in a public GitHub repository and are available for your use at the links below. TEMPLATE ( .yml ) : https://github.com/bienko/WCA-Lightspeed-L3/blob/main/TEMPLATE.yml The TEMPLATE.yml Playbook contains the skeleton of the application you will ultimately build. Playbook structure, key variables, Task descriptions, and comments are provided. Use the generative AI capabilities of WCA to fill out the details of the various Ansible Tasks, then compare the results with the SOLUTION.yml file below. SOLUTION ( .yml ) : https://github.com/bienko/WCA-Lightspeed-L3/blob/main/SOLUTION.yml The SOLUTION.yml Playbook contains the complete set of Ansible Playbook instructions needed to fully deploy a live EC2 VPC instance on AWS and host a WordPress application within that environment. You will need to replace certain variables (indicated by comments in the code) with details specific to your personal AWS account and environment. Afterwards, you can\u2014 and are encouraged to \u2014execute the Playbook with Ansible and deploy a live , at-cost instance to your personal AWS account. Once the two YAML files have been downloaded to your local machine, open them with VS Code and follow along with the instructions below. DOWNLOAD .yml FILES TO DESKTOP You can easily downlod the TEMPLATE and SOLUTION .yml files to your local machine. Navigate to the GitHub links provided (above) and click the Download raw file button. HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible Use WCA to generate code for the TEMPLATE.yml Playbook, then compare against the SOLUTION.yml Playbook to evaluate the results In the steps ahead, you will use WCA's AI-generated code suggestions to populate the contents of Tasks within the TEMPLATE.yml Playbook. There are 14 Tasks in total and we will examine them each in sequence. For each Task, a tabbed code snippet is provided within this lab documentation to make it easy for you to toggle between the TEMPLATE (the Task prior to AI-generated code suggestions) and the proper SOLUTION (the Task as it should be written for a successful deployment to AWS). Compare the AI-generated code suggestions you receive in VS Code with the expected SOLUTION code. You will find that there are occassional discrepancies and variables that need to be adjusted. Generative AI can save users tremendous amounts of time and effort by automating the creation of powerful blocks of code\u2014 as you will observe in generating these tasks \u2014but it still requires a degree of human supervision and double-checking. However, the time saved with hands on keyboard is enormous and the potential for the future is boundless. Open the TEMPLATE.yml Playbook in VS Code and ensure that the Ansible Lightspeed extension is active. Scroll down MODULE 1 ( Line 4 ) and take note of the comments, as well as the <PLACEHOLDER> values. If you wish to fully deploy this environment to AWS, you will need to register for an AWS account, create an IAM (Identity Access Management) user, and associate an SSH key with the necessary permissions to access the account. These values will then be substituted in your code for the aws_access_key and aws_secret_key variables. Instructions on how to sign up for AWS and create those keys will be included at the end of this module, and you will be reminded at that time to update these variables. However, for now, let's focus on generating the Ansible Tasks. Continue scrolling down until you reach TASK 1 ( Line 18 ) for the creation of a virtual private cloud (VPC) named 'wordpress'. Generate the Task code within VS Code and compare your results with the SOLUTION tab. TEMPLATE SOLUTION # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress amazon.aws.ec2_vpc_net : name : wordpress cidr_block : 10.0.0.0/16 state : present register : wordpress_vpc TAGS One element of AI-generated Task code that you may find yourself frequently needing to update are tags and other naming conventions, such as the register: wordpress_vpc pair in TASK 1 . While the Ansible Lightspeed extension will remember the context of variables that are already named within the Playbook, it will struggle with how to name new variables \u2014 like wordpress_vpc . Take note, however, that as you generate code for the other Tasks in the Playbook that wordpress_vpc will be recalled and referenced. After validating the previous code block, continue scrolling down until you reach TASK 2 ( Line 23 ) for the creation of a security group which allows network traffic over SSH and HTTP/s. Generate, evaluate, and refine the Task code as needed. Notice that the wordpress_vpc attribute which was registered in TASK 2 has been referenced in the creation of the vpc_id in TASK 3 Ignore the additional space between adjacent {``{ characters that are included in SOLUTION code tabs \u2014 these are included only to allow rendering of the code block in GitHub and are not intended as part of the finalized code TEMPLATE SOLUTION # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP amazon.aws.ec2_security_group : name : wordpress description : Allow SSH and HTTP/HTTPS vpc_id : \"{ { wordpress_vpc.vpc.id } }\" # Ignore extra space between { { cbaracters. rules : - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 443 to_port : 443 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 register : wordpress_sg Continue scrolling down until you reach TASK 3 ( Line 28 ) for the creation of an internet gateway for the wordpress VPC. Generate, evaluate, and refine the Task code as needed. TEMPLATE SOLUTION # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ { wordpress_vpc.vpc.id } }\" state : present register : wordpress_igw Continue down to TASK 4 ( Line 32 ) for the creation of an AWS network subnet for the WordPress VPC. TEMPLATE SOLUTION # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc amazon.aws.ec2_vpc_subnet : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" cidr : 10.0.0.0/16 register : wordpress_subnet Continue down to TASK 5 ( Line 36 ) for the creation of a routing table associated with wordpress VPC's subnet and internet gateway. TEMPLATE SOLUTION # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw amazon.aws.ec2_vpc_route_table : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" tags : Name : sg_public Project : phoenix subnets : - \"{ { wordpress_subnet.subnet.id } }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ { wordpress_igw.gateway_id } }\" register : wordpress_route_table WCA can be \"creative\" with its choice of tags and identifiers. You are welcome to adjust these to your liking, or even eliminate them from the Task code altogether. In the majority of cases these tags and identifiers are used for organization and tracking purposes, but not necessarily anything critical to the functioning of the Ansible Playbook or the infrastructure it is automating. Continue further down until you reach TASK 6 ( Line 40 ), which is one of the more consequential steps in the Playbook as it involves the creation of the Amazon EC2 t2.small instance on AWS according to the settings defined in TASKS 1-5 . Note that the image_id and ami values correspond to a specific Amazon Machine Image (ami) for Red Hat Enterprise Linux 9. Hundreds of alternative operating system images are available to select from on AWS, so you may modify this if you so choose; however, for the purposes of deploying WordPress it will make no meaningful difference. The pair key_name: \"bienko-key\" relates to an EC2 instance key (SSH) that you will need to create before attempting to deploy an EC2 instance via Ansible. Instructions for how to create that key will be given at the end of this module for those who wish to deploy a live environment. TEMPLATE SOLUTION # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip amazon.aws.ec2_instance : name : wordpress image_id : \"ami-026ebd4cfe2c043b2\" instance_type : t2.small vpc_subnet_id : \"{ { wordpress_subnet.subnet.id } }\" security_groups : \"{ { wordpress_sg.group_id } }\" network : assign_public_ip : true key_name : \"bienko-key\" state : running register : wordpress_server Continue down to TASK 7 ( Line 48 ). You will notice that in both the TEMPLATE and SOLUTION versions of the Playbook that this Task is fully defined already. For now, leave this as a hard-coded element within your Playbook and do not attempt to recreate the task with code generation. TEMPLATE SOLUTION # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" PLAYBOOK MODIFICATIONS REQUIRED Although TASK 7 does not require WCA to generate any code suggestions, it does require the user to modify the value of the ansible_ssh_private_key_file attribute. Set this attribute to the directory path where the private SSH key ( .pem file) is located on your local machine. Be sure to include the .pem filetype as part of the value and also ensure you have changed the permissions on the file (using Terminal and executing the chmod 400 bienko-key.pem instruction) before attempting to run the Playbook. Scrolling further down the Playbook you will find MODULE B ( Line 60 ), made up Ansible Tasks concerned with installing and configuring WordPress on the newly-provisioned AWS EC2 (VPC) instance. Locate TASK 8 ( Line 67 ) which waits for a connection to be established to the EC2 instance before advancing the sequence of automation tasks. TEMPLATE SOLUTION # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection ansible.builtin.wait_for_connection : delay : 10 timeout : 30 Next, continue further down the Playbook and locate TASK 9 ( Line 71 ) which commences installation of the necessary drivers and services on Amazon EC2 once a connection to the instance has been established. TEMPLATE SOLUTION # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server ansible.builtin.package : name : - httpd - php - php-mysqlnd - mariadb-server state : present Scroll down further until you locate TASK 10 ( Line 75 ) which is responsible for downloading and decompressing (unarchiving) WordPress installation contents onto the EC2 instance. TEMPLATE SOLUTION # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress ansible.builtin.unarchive : src : https://wordpress.org/latest.tar.gz dest : /var/www/html remote_src : true creates : /var/www/html/wordpress Afterwards, locate TASK 11 ( Line 79 ) which sets the ownership attributes for the WordPress environment on Amazon EC2 and prepares it for hosting via Apache webserver. TEMPLATE SOLUTION # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache ansible.builtin.file : path : /var/www/html owner : apache group : apache recurse : true Further down the Playbook, locate TASK 12 ( Line 83 ) which deploys the services installed in TASK 9 . TEMPLATE SOLUTION # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services ansible.builtin.service : name : \"{ { item } }\" state : started enabled : true loop : - httpd - php-fpm - mariadb MODULE C ( Line 87 ) of the Playbook covers the final set of operations. It is responsible for ensuring that WordPress is deployed on the newly-provisioned AWS EC2 (VPC) instance and that the WordPress web page is accessible to the user. As you did previously in Step 2 , you will need to replace the <PLACEHOLDER> values and region details on Lines 96-98 if you want to fully deploy this application into a live AWS environment. Locate TASK 13 ( Line 101 ) which is responsible for gathering facts about AWS EC2 (VPC) instance and the deployed WordPress application that is running atop of it. TEMPLATE SOLUTION # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress amazon.aws.ec2_instance_info : filters : tag:Name : wordpress instance-state-name : running register : ec2_facts Finally, you have arrived at the final step in the Playbook configuration: TASK 14 ( Line 105 ), which is responsible for hosting the WordPress web app and making it accessible to users. TEMPLATE SOLUTION # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name ansible.builtin.debug : msg : \"{ { ec2_facts.instances[0].public_dns_name } }/wordpress/readme.html\" To deploy a live Amazon EC2 instance and WordPress application on AWS, follow along with the steps below. Congratulations on making it this far! At this stage, your Playbook is ready for execution. However, there are still some preparation that you need to do before this Playbook can be used to automate deployment of the WordPress application into a live Amazon EC2 environment. In order to execute the Ansible Playbook on your local machine, you will need to install Ansible locally. Red Hat provides extensive online documentation for how to go about installing Ansible The authors of your lab guide recommend doing so by executing the following instructions in a Terminal window (macOS), or equivalent steps using PowerShell (Windows) Follow the prompts until you receive confirmation that Ansible is installed brew install ansible HOMEBREW IS MISSING \u2014 CLICK TO EXPAND If you do not have the Homebrew package manager (\"brew\") installed in a macOS environment, execute the following instruction inside a Terminal console: /bin/bash -c \"$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)\" Wait for the installation to conclude and then retry the instructions from Step 18. If you have not registered for an AWS account previously, you need to do so now. You will need to supply a personal credit card for billing expenses If you have an existing Amazon account, your AWS account can be linked to that profile if you so choose AWS Registration : https://aws.amazon.com/resources/create-account/ With a registered AWS account, you next need to navigate from the AWS dashboard into the Identity and Access Management (IAM) tool. You can search for this using the Search bar at the top of the screen and entering IAM . Once inside the IAM tool, look along the left-side of the interface and drill down into Users . Click the Create User orange button located in the top-right corner of the interface. Within the Create User panel, specify your user details and permissions. Set Username to a name of your choosing. Click Next to advance the configuration tool. Under Set Permissions > Permission Options , select the Add user to group tile. Click the Set permissions boundary - optional tile to expand permission options. Toggle the checkmark next to the Use a permissions boundary to control the maximum permissions option. Under Permissions Policies , search the table for the AdministratorAccess policy and enable that policy. You may receive warnings against doing so \u2014 ignore those for now. Continue following the prompts until the User has been created. Once the User has been created, your web browser will reopen the User panel of the IAM tool. Here you can see various details about the User, including their Security Credentials and any keys associated with the User. This section will be empty given that you have only just now created the User. Drill down into the Access Keys category and click the Create Access Key button located below the (empty) table. A secret access key will be created and the web browser will reload to show you details about that key. Record this information to a notepad. Details about the secret access key can only be viewed or downloaded at the time the key is created (now). It cannot be recovered afterwards. However, if you lose your key details, all is not lost \u2014 you can always create a new secret access key at a later time. Record the Access Key , which will be used to replace the <PLACEHOLDER> value assigned to aws_access_key in Step 2 and Step 15 of this lab guide. Record the Secret Access Key , which will replace the <PLACEHOLDER> value assigned to aws_secret_key in Step 2 and Step 15 of this lab guide. Take note at this time as well about the Region in which your AWS account is located. You can easily look up this information by examining the URL in your web browser address bar, as shown below ( us-east-1 ). Record the region's name, as it will replace the region value assigned to the Playbook. PLAYBOOK MODIFICATIONS REQUIRED Replace the <PLACEHOLDER> values for variables aws_access_key and aws_secret_key with the Access Key and Secret Access Key , respectively. Lines 13, 14, 96, and 97 of the TEMPLATE.yml Playbook Lines 13, 14, 179, and 180 of the SOLUTION.yml Playbook Replace the value for variable region with the region assigned to your unique AWS account. Lines 15 and 98 of the TEMPLATE.yml Playbook Lines 15 and 181 of the SOLUTION.yml Playbook Next, you'll need to create an SSH Key Pair with authorization to interact with (and ultimately deploy) Amazon EC2 instances. Return to the search bar at the top of the AWS interface and enter EC2 . Drill down into the Amazon EC2 service page from the results. From the left-hand navigation bar, drill down into the Network & Security > Key Pairs tab. Click the orange Create key pair button in the top-right of the page. A configuration tool will ask for additional information about the Key Pair: Name can be set to a value of your choosing Key pair type set to RSA Private key file format set to .pem When satisfied, click the orange Create key pair button The web browser will redirect to a page where you can view details about the newly-created Key Pair for EC2. Download the .pem file to your local device, preferably to the same directory that your TEMPLATE.yml and SOLUTION.yml Playbooks are located within. PLAYBOOK MODIFICATIONS REQUIRED The value of ansible_ssh_private_key_file must be updated within your Playbooks to account for the full path to the .pem file on your local machine. If you save the Key Pair file to the same directory that your Ansible Playbooks are located within, you do not need to qualify the value with anything more than the name of the file ( bienko-key.pem ). If it is saved in a different directory, you will need to spell out the full path + file name. Update the value of ansible_ssh_private_key_file in the following locations: Line 58 of the TEMPLATE.yml Playbook Line 113 of the SOLUTION.yml Playbook It is necessary to adjust the permissions assigned to the Key Pair file (on your local machine) before attempting to connect via Ansible to AWS. Open a Terminal console on your local machine and navigate ( cd ) to the directory in which the Key Pair file is saved. Execute the following to adjust the permissions assigned to the .pem file: chmod 400 filename.pem Executing the Ansible Playbook At last, you are ready to execute the Playbook and use Ansible automation to deploy both an Amazon EC2 instance and a WordPress web application. With the Terminal console, navigate to the directory ( cd ) where both the Ansible Playbook and your KeyPair.pem file are located. You may use either the TEMPLATE.yml Playbook that you customized or the SOLUTION.yml Playbook for executing the Ansible automation tasks. Regardless of the Playbook being used, you must modify the Playbook with the necessary AWS security and account credentials information. When ready, execute the following instruction with the Terminal console: ansible-playbook TEMPLATE.yml Monitor the console's output as the Ansible Playbook iterates across each of the Tasks customized earlier. A fantastic feature of Ansible's automation is that if any of the Tasks should fail, the steps involved in that Task can be re-tried in subsequent runs; likewise, Tasks which executed successfully won't be repeated, but are validated and skipped instead. If you encounter errors during the litany of automation tasks, return to the Playbook and attempt to debug the issue. Re-issue the instruction in Step 34 to console to attempt the Ansible Playbook execution again. Continue iterating in this manner until all 14 Tasks have been successfully executed. After successfully deploying the Amazon EC2 environment and instantiating a WordPress application, the console will return a URL to the Terminal's output as part of TASK 14 's [debug] instruction. Copy the URL to clipboard and then paste it into your web browser to access the WordPress application. If your web browser loads the quick-start page for WordPress, you have successfully deployed the application and its supporting EC2 environment \u2014 entirely using generative AI and Ansible automation. Well done! Cleaning up and conclusion COSTS & BILLING It is essential that you terminate and deprovision the EC2 instance at the conclusion of the hands on lab. Otherwise, your personal AWS account will continue to be billed for as long as the resources are provisioned. Once you are ready to retire your environment, you will need to return to the AWS Dashboard to terminate the EC2 instance and hosted WordPress application. You can monitor your Amazon EC2 instance from the EC2 dashboard. At the top of the page, locate the Resources table and drill down into the Instances (running) tab. Details about your EC2 environment are summarized within the Instances table. Click the name of your EC2's Instance ID to expose additional details and management options. In the top-right corner of the interface, click the Instance state button and select Terminate instance from the drop-down menu. Confirm the deprovisioning request by clicking the orange Terminate button. All storage associated with the EC2 instance will automatically be deleted. Terminating an instance cannot be undone once underway. It is recommended that you delete all of the Security Groups (3 total) and Key Pairs (1 total) associated with the EC2 instance. To do so, return to the EC2 Dashboard and follow a similar procedure to Steps 38-41 . Search for IAM in the AWS taskbar to return to the Identity Access Management panel. If you wish to further secure your AWS account, delete all unwanted users or consider reducing the user's privilege from AdministratorAccess to a more restrictive class. That concludes the hands-on components to this Level 3 course, but your learning and experimentation doesn't need to end here. Continue to experiment with generating Tasks for Ansible Playbooks. If you have worked extensively with Ansible previously, try recreating Tasks you've written previously from old Playbooks using generative AI \u2014 just be careful not to use confidential or sensitive information as part of those tests. See if you can spot the differences or improvements made from the AI-generated code recommendations. Feel free to reach out to the authors of this coursework if you have suggestions for Tasks or code generation techniques that you'd like to see included in future iterations of this hands-on training.","title":"Advanced"},{"location":"advanced/#_1","text":"","title":""},{"location":"advanced/#use-wca-to-generate-code-for-the-templateyml-playbook-then-compare-against-the-solutionyml-playbook-to-evaluate-the-results","text":"","title":"Use WCA to generate code for the TEMPLATE.yml Playbook, then compare against the SOLUTION.yml Playbook to evaluate the results"},{"location":"advanced/#_2","text":"In the steps ahead, you will use WCA's AI-generated code suggestions to populate the contents of Tasks within the TEMPLATE.yml Playbook. There are 14 Tasks in total and we will examine them each in sequence. For each Task, a tabbed code snippet is provided within this lab documentation to make it easy for you to toggle between the TEMPLATE (the Task prior to AI-generated code suggestions) and the proper SOLUTION (the Task as it should be written for a successful deployment to AWS). Compare the AI-generated code suggestions you receive in VS Code with the expected SOLUTION code. You will find that there are occassional discrepancies and variables that need to be adjusted. Generative AI can save users tremendous amounts of time and effort by automating the creation of powerful blocks of code\u2014 as you will observe in generating these tasks \u2014but it still requires a degree of human supervision and double-checking. However, the time saved with hands on keyboard is enormous and the potential for the future is boundless. Open the TEMPLATE.yml Playbook in VS Code and ensure that the Ansible Lightspeed extension is active. Scroll down MODULE 1 ( Line 4 ) and take note of the comments, as well as the <PLACEHOLDER> values. If you wish to fully deploy this environment to AWS, you will need to register for an AWS account, create an IAM (Identity Access Management) user, and associate an SSH key with the necessary permissions to access the account. These values will then be substituted in your code for the aws_access_key and aws_secret_key variables. Instructions on how to sign up for AWS and create those keys will be included at the end of this module, and you will be reminded at that time to update these variables. However, for now, let's focus on generating the Ansible Tasks. Continue scrolling down until you reach TASK 1 ( Line 18 ) for the creation of a virtual private cloud (VPC) named 'wordpress'. Generate the Task code within VS Code and compare your results with the SOLUTION tab. TEMPLATE SOLUTION # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress # TASK 1 # Creation of a virtual private cloud (VPC) named 'wordpress'. # Should have value 10.0.0.0/16 associated with cidr_block. - name : Create VPC named wordpress amazon.aws.ec2_vpc_net : name : wordpress cidr_block : 10.0.0.0/16 state : present register : wordpress_vpc TAGS One element of AI-generated Task code that you may find yourself frequently needing to update are tags and other naming conventions, such as the register: wordpress_vpc pair in TASK 1 . While the Ansible Lightspeed extension will remember the context of variables that are already named within the Playbook, it will struggle with how to name new variables \u2014 like wordpress_vpc . Take note, however, that as you generate code for the other Tasks in the Playbook that wordpress_vpc will be recalled and referenced. After validating the previous code block, continue scrolling down until you reach TASK 2 ( Line 23 ) for the creation of a security group which allows network traffic over SSH and HTTP/s. Generate, evaluate, and refine the Task code as needed. Notice that the wordpress_vpc attribute which was registered in TASK 2 has been referenced in the creation of the vpc_id in TASK 3 Ignore the additional space between adjacent {``{ characters that are included in SOLUTION code tabs \u2014 these are included only to allow rendering of the code block in GitHub and are not intended as part of the finalized code TEMPLATE SOLUTION # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP # TASK 2 # Creation of the security group which allows traffic over SSH and HTTP/s # TCP ports 80-80, 443-443, 22-22 - name : Create and register wordpress_vpc VPC security group allow SSH and HTTP amazon.aws.ec2_security_group : name : wordpress description : Allow SSH and HTTP/HTTPS vpc_id : \"{ { wordpress_vpc.vpc.id } }\" # Ignore extra space between { { cbaracters. rules : - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 443 to_port : 443 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 register : wordpress_sg Continue scrolling down until you reach TASK 3 ( Line 28 ) for the creation of an internet gateway for the wordpress VPC. Generate, evaluate, and refine the Task code as needed. TEMPLATE SOLUTION # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc # TASK 3 # Creation of an internet gateway for the wordpress VPC - name : Create internet gateway for VPC wordpress_vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ { wordpress_vpc.vpc.id } }\" state : present register : wordpress_igw Continue down to TASK 4 ( Line 32 ) for the creation of an AWS network subnet for the WordPress VPC. TEMPLATE SOLUTION # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc # TASK 4 # Creation of a network subnet for the wordpress VPC - name : Create subnet in wordpress_vpc amazon.aws.ec2_vpc_subnet : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" cidr : 10.0.0.0/16 register : wordpress_subnet Continue down to TASK 5 ( Line 36 ) for the creation of a routing table associated with wordpress VPC's subnet and internet gateway. TEMPLATE SOLUTION # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw # TASK 5 # Creation of a routing table associated with wordpress VPC's subnet and internet gateway - name : Create route table for subnet and gateway wordpress_igw amazon.aws.ec2_vpc_route_table : state : present vpc_id : \"{ { wordpress_vpc.vpc.id } }\" tags : Name : sg_public Project : phoenix subnets : - \"{ { wordpress_subnet.subnet.id } }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ { wordpress_igw.gateway_id } }\" register : wordpress_route_table WCA can be \"creative\" with its choice of tags and identifiers. You are welcome to adjust these to your liking, or even eliminate them from the Task code altogether. In the majority of cases these tags and identifiers are used for organization and tracking purposes, but not necessarily anything critical to the functioning of the Ansible Playbook or the infrastructure it is automating. Continue further down until you reach TASK 6 ( Line 40 ), which is one of the more consequential steps in the Playbook as it involves the creation of the Amazon EC2 t2.small instance on AWS according to the settings defined in TASKS 1-5 . Note that the image_id and ami values correspond to a specific Amazon Machine Image (ami) for Red Hat Enterprise Linux 9. Hundreds of alternative operating system images are available to select from on AWS, so you may modify this if you so choose; however, for the purposes of deploying WordPress it will make no meaningful difference. The pair key_name: \"bienko-key\" relates to an EC2 instance key (SSH) that you will need to create before attempting to deploy an EC2 instance via Ansible. Instructions for how to create that key will be given at the end of this module for those who wish to deploy a live environment. TEMPLATE SOLUTION # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip # TASK 6 # Creation of an EC2 t2.small instance with attributes defined in Tasks 1-5 # Amazon Machine Image (ami) image_id: ami-026ebd4cfe2c043b2 # Registered to Red Hat Enterprise Linux 9 (HVM, 64-bit x86) image # Replace `key_name:` value with your EC2 .pem keypair - DO NOT include .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Create t2.small instance named wordpress in wordpress_subnet assign public ip amazon.aws.ec2_instance : name : wordpress image_id : \"ami-026ebd4cfe2c043b2\" instance_type : t2.small vpc_subnet_id : \"{ { wordpress_subnet.subnet.id } }\" security_groups : \"{ { wordpress_sg.group_id } }\" network : assign_public_ip : true key_name : \"bienko-key\" state : running register : wordpress_server Continue down to TASK 7 ( Line 48 ). You will notice that in both the TEMPLATE and SOLUTION versions of the Playbook that this Task is fully defined already. For now, leave this as a hard-coded element within your Playbook and do not attempt to recreate the task with code generation. TEMPLATE SOLUTION # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" # TASK 7 # This section has been hard-coded ahead of time. # DO NOT regenerate using AI-suggested code. # Replace `ansible_ssh_private_key_file` value with path on local machine to SSH key \u2014 INCLUDE .pem as part of filename # Use 'chmod 400' to change file permissions of .pem file before executing Playbook - name : Add host to inventory using tunnel using wordpress_instance public ip and ansible user ec2-user ansible.builtin.add_host : name : wordpress ansible_host : \"{ { wordpress_server.instances[0].public_ip_address } }\" ansible_user : ec2-user ansible_ssh_private_key_file : \"bienko-key.pem\" PLAYBOOK MODIFICATIONS REQUIRED Although TASK 7 does not require WCA to generate any code suggestions, it does require the user to modify the value of the ansible_ssh_private_key_file attribute. Set this attribute to the directory path where the private SSH key ( .pem file) is located on your local machine. Be sure to include the .pem filetype as part of the value and also ensure you have changed the permissions on the file (using Terminal and executing the chmod 400 bienko-key.pem instruction) before attempting to run the Playbook. Scrolling further down the Playbook you will find MODULE B ( Line 60 ), made up Ansible Tasks concerned with installing and configuring WordPress on the newly-provisioned AWS EC2 (VPC) instance. Locate TASK 8 ( Line 67 ) which waits for a connection to be established to the EC2 instance before advancing the sequence of automation tasks. TEMPLATE SOLUTION # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection # TASK 8 # Wait for a connection to be established to the EC2 instance. - name : Wait for connection ansible.builtin.wait_for_connection : delay : 10 timeout : 30 Next, continue further down the Playbook and locate TASK 9 ( Line 71 ) which commences installation of the necessary drivers and services on Amazon EC2 once a connection to the instance has been established. TEMPLATE SOLUTION # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server # TASK 9 # After connecting, begin installation of necessary drivers and services. - name : Install httpd, php, php-mysqli, and mariadb-server ansible.builtin.package : name : - httpd - php - php-mysqlnd - mariadb-server state : present Scroll down further until you locate TASK 10 ( Line 75 ) which is responsible for downloading and decompressing (unarchiving) WordPress installation contents onto the EC2 instance. TEMPLATE SOLUTION # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress # TASK 10 # Download and decompress (unarchive) WordPress contents on EC2 instance. - name : Download and unarchive wordpress ansible.builtin.unarchive : src : https://wordpress.org/latest.tar.gz dest : /var/www/html remote_src : true creates : /var/www/html/wordpress Afterwards, locate TASK 11 ( Line 79 ) which sets the ownership attributes for the WordPress environment on Amazon EC2 and prepares it for hosting via Apache webserver. TEMPLATE SOLUTION # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache # TASK 11 # Set owner attributes for WordPress environment. - name : Change owner of /var/www/html/wordpress to apache:apache ansible.builtin.file : path : /var/www/html owner : apache group : apache recurse : true Further down the Playbook, locate TASK 12 ( Line 83 ) which deploys the services installed in TASK 9 . TEMPLATE SOLUTION # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services # TASK 12 # Deploy services installed in Task 9. - name : Start and enable httpd, php-fpm and mariadb services ansible.builtin.service : name : \"{ { item } }\" state : started enabled : true loop : - httpd - php-fpm - mariadb MODULE C ( Line 87 ) of the Playbook covers the final set of operations. It is responsible for ensuring that WordPress is deployed on the newly-provisioned AWS EC2 (VPC) instance and that the WordPress web page is accessible to the user. As you did previously in Step 2 , you will need to replace the <PLACEHOLDER> values and region details on Lines 96-98 if you want to fully deploy this application into a live AWS environment. Locate TASK 13 ( Line 101 ) which is responsible for gathering facts about AWS EC2 (VPC) instance and the deployed WordPress application that is running atop of it. TEMPLATE SOLUTION # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress # TASK 13 # Gather facts about AWS EC2 (VPC) instance and deployed WordPress application. - name : Gather ec2 instance info for tag name wordpress amazon.aws.ec2_instance_info : filters : tag:Name : wordpress instance-state-name : running register : ec2_facts Finally, you have arrived at the final step in the Playbook configuration: TASK 14 ( Line 105 ), which is responsible for hosting the WordPress web app and making it accessible to users. TEMPLATE SOLUTION # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name # TASK 14 # Host the WordPress web application and make accessible to user. - name : Debug ec2_facts public dns name ansible.builtin.debug : msg : \"{ { ec2_facts.instances[0].public_dns_name } }/wordpress/readme.html\"","title":""},{"location":"advanced/#to-deploy-a-live-amazon-ec2-instance-and-wordpress-application-on-aws-follow-along-with-the-steps-below","text":"","title":"To deploy a live Amazon EC2 instance and WordPress application on AWS, follow along with the steps below."},{"location":"advanced/#_3","text":"Congratulations on making it this far! At this stage, your Playbook is ready for execution. However, there are still some preparation that you need to do before this Playbook can be used to automate deployment of the WordPress application into a live Amazon EC2 environment. In order to execute the Ansible Playbook on your local machine, you will need to install Ansible locally. Red Hat provides extensive online documentation for how to go about installing Ansible The authors of your lab guide recommend doing so by executing the following instructions in a Terminal window (macOS), or equivalent steps using PowerShell (Windows) Follow the prompts until you receive confirmation that Ansible is installed brew install ansible HOMEBREW IS MISSING \u2014 CLICK TO EXPAND If you do not have the Homebrew package manager (\"brew\") installed in a macOS environment, execute the following instruction inside a Terminal console: /bin/bash -c \"$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)\" Wait for the installation to conclude and then retry the instructions from Step 18. If you have not registered for an AWS account previously, you need to do so now. You will need to supply a personal credit card for billing expenses If you have an existing Amazon account, your AWS account can be linked to that profile if you so choose AWS Registration : https://aws.amazon.com/resources/create-account/ With a registered AWS account, you next need to navigate from the AWS dashboard into the Identity and Access Management (IAM) tool. You can search for this using the Search bar at the top of the screen and entering IAM . Once inside the IAM tool, look along the left-side of the interface and drill down into Users . Click the Create User orange button located in the top-right corner of the interface. Within the Create User panel, specify your user details and permissions. Set Username to a name of your choosing. Click Next to advance the configuration tool. Under Set Permissions > Permission Options , select the Add user to group tile. Click the Set permissions boundary - optional tile to expand permission options. Toggle the checkmark next to the Use a permissions boundary to control the maximum permissions option. Under Permissions Policies , search the table for the AdministratorAccess policy and enable that policy. You may receive warnings against doing so \u2014 ignore those for now. Continue following the prompts until the User has been created. Once the User has been created, your web browser will reopen the User panel of the IAM tool. Here you can see various details about the User, including their Security Credentials and any keys associated with the User. This section will be empty given that you have only just now created the User. Drill down into the Access Keys category and click the Create Access Key button located below the (empty) table. A secret access key will be created and the web browser will reload to show you details about that key. Record this information to a notepad. Details about the secret access key can only be viewed or downloaded at the time the key is created (now). It cannot be recovered afterwards. However, if you lose your key details, all is not lost \u2014 you can always create a new secret access key at a later time. Record the Access Key , which will be used to replace the <PLACEHOLDER> value assigned to aws_access_key in Step 2 and Step 15 of this lab guide. Record the Secret Access Key , which will replace the <PLACEHOLDER> value assigned to aws_secret_key in Step 2 and Step 15 of this lab guide. Take note at this time as well about the Region in which your AWS account is located. You can easily look up this information by examining the URL in your web browser address bar, as shown below ( us-east-1 ). Record the region's name, as it will replace the region value assigned to the Playbook. PLAYBOOK MODIFICATIONS REQUIRED Replace the <PLACEHOLDER> values for variables aws_access_key and aws_secret_key with the Access Key and Secret Access Key , respectively. Lines 13, 14, 96, and 97 of the TEMPLATE.yml Playbook Lines 13, 14, 179, and 180 of the SOLUTION.yml Playbook Replace the value for variable region with the region assigned to your unique AWS account. Lines 15 and 98 of the TEMPLATE.yml Playbook Lines 15 and 181 of the SOLUTION.yml Playbook Next, you'll need to create an SSH Key Pair with authorization to interact with (and ultimately deploy) Amazon EC2 instances. Return to the search bar at the top of the AWS interface and enter EC2 . Drill down into the Amazon EC2 service page from the results. From the left-hand navigation bar, drill down into the Network & Security > Key Pairs tab. Click the orange Create key pair button in the top-right of the page. A configuration tool will ask for additional information about the Key Pair: Name can be set to a value of your choosing Key pair type set to RSA Private key file format set to .pem When satisfied, click the orange Create key pair button The web browser will redirect to a page where you can view details about the newly-created Key Pair for EC2. Download the .pem file to your local device, preferably to the same directory that your TEMPLATE.yml and SOLUTION.yml Playbooks are located within. PLAYBOOK MODIFICATIONS REQUIRED The value of ansible_ssh_private_key_file must be updated within your Playbooks to account for the full path to the .pem file on your local machine. If you save the Key Pair file to the same directory that your Ansible Playbooks are located within, you do not need to qualify the value with anything more than the name of the file ( bienko-key.pem ). If it is saved in a different directory, you will need to spell out the full path + file name. Update the value of ansible_ssh_private_key_file in the following locations: Line 58 of the TEMPLATE.yml Playbook Line 113 of the SOLUTION.yml Playbook It is necessary to adjust the permissions assigned to the Key Pair file (on your local machine) before attempting to connect via Ansible to AWS. Open a Terminal console on your local machine and navigate ( cd ) to the directory in which the Key Pair file is saved. Execute the following to adjust the permissions assigned to the .pem file: chmod 400 filename.pem","title":""},{"location":"advanced/#executing-the-ansible-playbook","text":"","title":"Executing the Ansible Playbook"},{"location":"advanced/#_4","text":"At last, you are ready to execute the Playbook and use Ansible automation to deploy both an Amazon EC2 instance and a WordPress web application. With the Terminal console, navigate to the directory ( cd ) where both the Ansible Playbook and your KeyPair.pem file are located. You may use either the TEMPLATE.yml Playbook that you customized or the SOLUTION.yml Playbook for executing the Ansible automation tasks. Regardless of the Playbook being used, you must modify the Playbook with the necessary AWS security and account credentials information. When ready, execute the following instruction with the Terminal console: ansible-playbook TEMPLATE.yml Monitor the console's output as the Ansible Playbook iterates across each of the Tasks customized earlier. A fantastic feature of Ansible's automation is that if any of the Tasks should fail, the steps involved in that Task can be re-tried in subsequent runs; likewise, Tasks which executed successfully won't be repeated, but are validated and skipped instead. If you encounter errors during the litany of automation tasks, return to the Playbook and attempt to debug the issue. Re-issue the instruction in Step 34 to console to attempt the Ansible Playbook execution again. Continue iterating in this manner until all 14 Tasks have been successfully executed. After successfully deploying the Amazon EC2 environment and instantiating a WordPress application, the console will return a URL to the Terminal's output as part of TASK 14 's [debug] instruction. Copy the URL to clipboard and then paste it into your web browser to access the WordPress application. If your web browser loads the quick-start page for WordPress, you have successfully deployed the application and its supporting EC2 environment \u2014 entirely using generative AI and Ansible automation. Well done!","title":""},{"location":"advanced/#cleaning-up-and-conclusion","text":"","title":"Cleaning up and conclusion"},{"location":"advanced/#_5","text":"COSTS & BILLING It is essential that you terminate and deprovision the EC2 instance at the conclusion of the hands on lab. Otherwise, your personal AWS account will continue to be billed for as long as the resources are provisioned. Once you are ready to retire your environment, you will need to return to the AWS Dashboard to terminate the EC2 instance and hosted WordPress application. You can monitor your Amazon EC2 instance from the EC2 dashboard. At the top of the page, locate the Resources table and drill down into the Instances (running) tab. Details about your EC2 environment are summarized within the Instances table. Click the name of your EC2's Instance ID to expose additional details and management options. In the top-right corner of the interface, click the Instance state button and select Terminate instance from the drop-down menu. Confirm the deprovisioning request by clicking the orange Terminate button. All storage associated with the EC2 instance will automatically be deleted. Terminating an instance cannot be undone once underway. It is recommended that you delete all of the Security Groups (3 total) and Key Pairs (1 total) associated with the EC2 instance. To do so, return to the EC2 Dashboard and follow a similar procedure to Steps 38-41 . Search for IAM in the AWS taskbar to return to the Identity Access Management panel. If you wish to further secure your AWS account, delete all unwanted users or consider reducing the user's privilege from AdministratorAccess to a more restrictive class. That concludes the hands-on components to this Level 3 course, but your learning and experimentation doesn't need to end here. Continue to experiment with generating Tasks for Ansible Playbooks. If you have worked extensively with Ansible previously, try recreating Tasks you've written previously from old Playbooks using generative AI \u2014 just be careful not to use confidential or sensitive information as part of those tests. See if you can spot the differences or improvements made from the AI-generated code recommendations. Feel free to reach out to the authors of this coursework if you have suggestions for Tasks or code generation techniques that you'd like to see included in future iterations of this hands-on training.","title":""},{"location":"archive-setup-pre-vm/","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. Reserving a lab environment Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Acccount : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.3 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below and select Request a Red Hat Account [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted by IBM Technology Zone after the reservation period has ended Installation of Visual Studio Code and Extensions Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code available for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You might receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] If Extension Settings is not available from the drop-down menu, try right-clicking on the Ansible extension tile instead Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab Click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 5 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 5 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URI? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 12-14 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 12-14 as before; the login process should run smoother (and faster) on the second attempt Preparing the Ansible Playbook materials To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to clone the Ansible Playbook templates into the local (VS Code) environment. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . To perform a clone request with VS Code, perform the following action (depending on your OS): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that you have access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait patiently for the operation to finish (no responses will be prompted to screen if no updates to the code repository were necessary) Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured Troubleshooting and support If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. UNABLE TO INSTALL VS CODE LOCALLY \u2014 REQUIRE A VIRTUAL MACHINE Due to security or regulatory constraints, certain IBM clients or individuals enrolled in this training may be unable to install software (such as VS Code) on their local machines. As an alternative, IBM Technology Zone is able to provision a virtual machine environment that can be reserved at no-charge for completing the hands-on training. Click to expand the instructions embedded below for how to setup the virtual machine environment. Note that the documentation's instructions remain the same for the other modules, regardless of whether you are using a \"local\" or \"virtualized\" VS Code environment \u2014 although the \"local\" option is the preferred option. CLICK TO EXPAND INSTRUCTIONS FOR REQUESTING A VM Follow the link below and select the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.3 [A] (or later) tile: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the ITZ environment is ready to go, you will receive an email to your inbox (similar to the screenshot below). Click the blue View My Reservations at the bottom of the email to open a page with more details about the environment. Alternatively, you can click the My Reservations [A] tab from the ITZ home page to drill down into your environment's details. Near the top of the page, under Published Services , click the noVNC [A] link to access the virtual machine. A new tab will open within your web browser for the VM environment. It is recommended that Firefox be used for optimal performance Along the left-side edge of the page, click the Cog [A] icon From the pop-up menu, click Scaling Mode and then select Remote Resizing The VM window will now be properly scaled to the size of your browser window Click the Connect [B] button to continue with authentication Supply the following password and click Send Credentials : IBMDem0s! Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. Visual Studio (VS) Code is a code editor built on open source technology that is free to use and published by Microsoft. Distributions are available for both Windows and macOS operating systems. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B] If this is your first time logging into the VM, you must authenticate and activate the Ansible Lightspeed extension for VS Code. Click the Ansible [A] extension (denoted by the large A icon) from the left-hand interface open expose details about the Ansible plug-in. In the bottom-right corner of the interface [A] , you may receieve a notification stating that You must be logged in to use the Ansible Lightspeed , which confirms that the Ansible Lightspeed service still needs to be activated for this VM environment. Additionally, along the left-hand side of the interface [B] , you will also receive a welcome notice attached to the Ansible Lightspeed Login panel. Click either the Connect [C] or Login buttons to proceed with authentication and activation of the plugin When prompted with The extension Ansible wants to sign in using Ansible Lightspeed , click Allow [D] Firefox web browser will open within the VM and redirect to a Red Hat authentication page, as shown below. Click Log in with Red Hat [A] Note that the page load may be sluggish or slow \u2014 please be patient and give it a few moments before retrying this step The VM environment already has access credentials saved to Firefox's keychain. Use the stored username and password to authenticate the Ansible Lightspeed plugin with Red Hat. For the Red Hat login or email [A] field, enter ibm-wxca-demo-user For the Password [B] field, enter IBMDem0s! Log in when ready A pop-up notification will ask to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] to proceed. CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 11-14 takes too long (over 60 seconds), activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner. Click the Connect [C] button as shown and repeat Steps 11-14 as before. The login process should run smoother (and faster) on the second attempt. Verify [A] that the environment is logged in as wca-ibm-demo-user and that the User Type: Licensed . Click the Explorer [A] button (as shown) in the top-left corner of the interface. Confirm that the working directory is set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , then hit Return , and wait patiently for the operation to finish Open the console once again, type git pull , and then hit Return (no responses will be prompted to screen if no updates to the code repository were necessary) At this stage, the hands-on environment has been fully configured. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. Next steps The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"Archive setup pre vm"},{"location":"archive-setup-pre-vm/#_1","text":"","title":""},{"location":"archive-setup-pre-vm/#reserving-a-lab-environment","text":"","title":"Reserving a lab environment"},{"location":"archive-setup-pre-vm/#_2","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Acccount : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.3 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below and select Request a Red Hat Account [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted by IBM Technology Zone after the reservation period has ended","title":""},{"location":"archive-setup-pre-vm/#_3","text":"","title":""},{"location":"archive-setup-pre-vm/#installation-of-visual-studio-code-and-extensions","text":"","title":"Installation of Visual Studio Code and Extensions"},{"location":"archive-setup-pre-vm/#_4","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code available for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You might receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] If Extension Settings is not available from the drop-down menu, try right-clicking on the Ansible extension tile instead Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab Click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 5 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 5 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URI? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 12-14 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 12-14 as before; the login process should run smoother (and faster) on the second attempt","title":""},{"location":"archive-setup-pre-vm/#_5","text":"","title":""},{"location":"archive-setup-pre-vm/#preparing-the-ansible-playbook-materials","text":"To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to clone the Ansible Playbook templates into the local (VS Code) environment. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A clone request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. GIT NOT INSTALLED \u2014 CLICK TO EXPAND If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . To perform a clone request with VS Code, perform the following action (depending on your OS): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that you have access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait patiently for the operation to finish (no responses will be prompted to screen if no updates to the code repository were necessary) Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured","title":"Preparing the Ansible Playbook materials"},{"location":"archive-setup-pre-vm/#_6","text":"","title":""},{"location":"archive-setup-pre-vm/#troubleshooting-and-support","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. UNABLE TO INSTALL VS CODE LOCALLY \u2014 REQUIRE A VIRTUAL MACHINE Due to security or regulatory constraints, certain IBM clients or individuals enrolled in this training may be unable to install software (such as VS Code) on their local machines. As an alternative, IBM Technology Zone is able to provision a virtual machine environment that can be reserved at no-charge for completing the hands-on training. Click to expand the instructions embedded below for how to setup the virtual machine environment. Note that the documentation's instructions remain the same for the other modules, regardless of whether you are using a \"local\" or \"virtualized\" VS Code environment \u2014 although the \"local\" option is the preferred option. CLICK TO EXPAND INSTRUCTIONS FOR REQUESTING A VM Follow the link below and select the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.3 [A] (or later) tile: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Name [A] : Give your reservation a unique name. Purpose [B] : Set to Practice / Self-Education and affirm that customer data will not be used with the environment. If you are replicating this hands-on demonstration with a client, you must select Customer Demo and supply a sales opportunity number. Purpose Description [C] : Provide a brief summary of how the environment will be used. Preferred Geography [D] : Select the data center region that is closest to your location. End Date & Time [E] : Select a time and date for when the reservation will expire. The recommended amount is 2 days, although it is possible to finish the hands-on demonstration within a few hours. Additional time extensions are available. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit [F] . PROVISIONING TIMES Reservations take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the ITZ environment is ready to go, you will receive an email to your inbox (similar to the screenshot below). Click the blue View My Reservations at the bottom of the email to open a page with more details about the environment. Alternatively, you can click the My Reservations [A] tab from the ITZ home page to drill down into your environment's details. Near the top of the page, under Published Services , click the noVNC [A] link to access the virtual machine. A new tab will open within your web browser for the VM environment. It is recommended that Firefox be used for optimal performance Along the left-side edge of the page, click the Cog [A] icon From the pop-up menu, click Scaling Mode and then select Remote Resizing The VM window will now be properly scaled to the size of your browser window Click the Connect [B] button to continue with authentication Supply the following password and click Send Credentials : IBMDem0s! Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. Visual Studio (VS) Code is a code editor built on open source technology that is free to use and published by Microsoft. Distributions are available for both Windows and macOS operating systems. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B] If this is your first time logging into the VM, you must authenticate and activate the Ansible Lightspeed extension for VS Code. Click the Ansible [A] extension (denoted by the large A icon) from the left-hand interface open expose details about the Ansible plug-in. In the bottom-right corner of the interface [A] , you may receieve a notification stating that You must be logged in to use the Ansible Lightspeed , which confirms that the Ansible Lightspeed service still needs to be activated for this VM environment. Additionally, along the left-hand side of the interface [B] , you will also receive a welcome notice attached to the Ansible Lightspeed Login panel. Click either the Connect [C] or Login buttons to proceed with authentication and activation of the plugin When prompted with The extension Ansible wants to sign in using Ansible Lightspeed , click Allow [D] Firefox web browser will open within the VM and redirect to a Red Hat authentication page, as shown below. Click Log in with Red Hat [A] Note that the page load may be sluggish or slow \u2014 please be patient and give it a few moments before retrying this step The VM environment already has access credentials saved to Firefox's keychain. Use the stored username and password to authenticate the Ansible Lightspeed plugin with Red Hat. For the Red Hat login or email [A] field, enter ibm-wxca-demo-user For the Password [B] field, enter IBMDem0s! Log in when ready A pop-up notification will ask to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] to proceed. CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 11-14 takes too long (over 60 seconds), activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner. Click the Connect [C] button as shown and repeat Steps 11-14 as before. The login process should run smoother (and faster) on the second attempt. Verify [A] that the environment is logged in as wca-ibm-demo-user and that the User Type: Licensed . Click the Explorer [A] button (as shown) in the top-left corner of the interface. Confirm that the working directory is set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , then hit Return , and wait patiently for the operation to finish Open the console once again, type git pull , and then hit Return (no responses will be prompted to screen if no updates to the code repository were necessary) At this stage, the hands-on environment has been fully configured. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor.","title":"Troubleshooting and support"},{"location":"archive-setup-pre-vm/#_7","text":"","title":""},{"location":"archive-setup-pre-vm/#next-steps","text":"The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"Next steps"},{"location":"archived/","text":"Installation of Visual Studio Code and Extensions Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code availabe for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You may receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab Accessing your Red Hat credentials and authenticating with WCA Red Hat credentials will already have been emailed to you prior to starting this hands-on material, as part of the registration process. The invitation email will have a header similar to Red Hat Login Email Verification , addressed from a no-reply@redhat.com account. Locate this email in your inbox and follow along with the steps below to authenticate the VS Code extension with the IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (WCA) environment that has been pre-provisioned for this training. Click the URL [A] located within the body of the invitation email to finalize your account registration with the WCA environment. An Email Confirmation page will load within your web browser Record the value of Red Hat login to a notepad for reference later Create a new Password and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted after the hands-on training period has ended Return to the VS Code editor and click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 6 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URL? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 7-9 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 7-9 as before; the login process should run smoother (and faster) on the second attempt Preparing the Ansible Playbook materials To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. To perform a clone request with VS Code, perform the following action (depending on your operating system): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait for the operation to finish Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured Troubleshooting and support If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ANSIBLE-LINT IS NOT AVAILABLE ansible-lint checks playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises. RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. Next steps The following section will cover the fundamentals of AI-recommended code generation for Ansible Playbooks using IBM watsonx Code Assistant .","title":"Installation of Visual Studio Code and Extensions"},{"location":"archived/#installation-of-visual-studio-code-and-extensions","text":"","title":"Installation of Visual Studio Code and Extensions"},{"location":"archived/#_1","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you will first setup Visual Studio Code (commonly referred to as VS Code ) on a local machine. This will provide an integrated development environment for experimenting with WCA's generative AI capabilities. Download the latest Stable Build [A] release of VS Code availabe for your machine's operating system using the link below. Download : https://code.visualstudio.com Follow along with the installer wizard steps and continue with the hands-on lab instructions once VS Code is running on your local machine. FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. Launch the VS Code application and take note of the sidebar along the left-side. Click the Extensions icon [A] to open the marketplace of services and open source technologies that can be integrated with VS Code If you have used VS Code previously, extensions that have already been integrated with the environment will be listed along the left side At the top of the Extensions panel is a search bar: Type Ansible into the search bar [A] and then hit Enter Click the blue Install button [B] for the official Ansible extension for VS Code, published by Red Hat (blue checkmark) INSTALLATION PROMPTS You may receive two different prompts during the installation process: Do you trust the authors of the files in this workspace? : select Trust Workspace & Install Do you want to allow untrusted files in this window? : select Open Installation of the Ansible extension for VS Code should only take a moment \u2014 an Extension:Ansible welcome panel will open when it is finished Once the Ansible extension has been integrated with VS Code, close any Welcome tabs that open and look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Settings for the Ansible extension will be displayed within a new panel. Ensure that User [A] is selected at the top of the panel \u2014 do not edit Workspace Using the search bar [B] at the top of the panel, add the text Lightspeed to filter the available options Check the option for Ansible > Lightspeed [C] Check the option for Ansible > Lightspeed: Suggestions [D] Check the option for Ansible > Lightspeed: Disable Content Suggestion Header [E] Changes to Settings are automatically saved and applied \u2013 click the X button in the top-left corner of the panel's tab","title":""},{"location":"archived/#_2","text":"","title":""},{"location":"archived/#accessing-your-red-hat-credentials-and-authenticating-with-wca","text":"Red Hat credentials will already have been emailed to you prior to starting this hands-on material, as part of the registration process. The invitation email will have a header similar to Red Hat Login Email Verification , addressed from a no-reply@redhat.com account. Locate this email in your inbox and follow along with the steps below to authenticate the VS Code extension with the IBM watsonx Code Assistant for Red Hat Ansible Lightspeed (WCA) environment that has been pre-provisioned for this training. Click the URL [A] located within the body of the invitation email to finalize your account registration with the WCA environment. An Email Confirmation page will load within your web browser Record the value of Red Hat login to a notepad for reference later Create a new Password and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will be de-authorized and deleted after the hands-on training period has ended Return to the VS Code editor and click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button [A] If you had previously logged in to Red Hat with your browser, you might not be asked again for those credentials If you are asked to provide a Username and Password , supply the values recorded in Step 6 of this module After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Do you want to allow this website to open Visual Studio Code? : click Allow [B] Allow Ansible extension to open this URL? : click Open [C] At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 7-9 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 7-9 as before; the login process should run smoother (and faster) on the second attempt","title":"Accessing your Red Hat credentials and authenticating with WCA"},{"location":"archived/#_3","text":"","title":""},{"location":"archived/#preparing-the-ansible-playbook-materials","text":"To begin experimenting with WCA's generative AI capabilities, you will first need access to some Ansible Playbooks to generate Tasks with. Playbook templates have already been prepared ahead of time for this training so that you can get straight to work. Click the Explorer button at the top of the left-hand VS Code interface. Depending on your VS Code environment, the Explorer tab will look one of two ways. Click to expand whichever one of the two options best describes your situation and follow the instructions. I AM NEW TO VS CODE If you are working within a new installation of VS Code, the Explorer tab [A] will display NO FOLDER OPENED and give options to either Open Folder or Clone Repository . You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. Click the Clone Repository button [B] , which will open an executable console [C] at the top of VS Code You must specify the public GitHub repository from which to \"clone\" the Ansible Playbook templates into the local (VS Code) environment Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit If prompted Would you like to open the cloned repository? : click Open [D] If prompted Do you trust the authors of the files in this folder? : click Yes, I trust the authors [E] I HAVE USED VS CODE BEFORE If you have worked with VS Code before and have added projects or folders to the environment previously, those folders (and their contents) will be displayed within the Explorer tab. However, you still need to clone (replicate) the Ansible Playbook templates from GitHub to a folder on your local machine. A \"clone\" request in GitHub is essentially a request to replicate code from the cloud-hosted repository into the local environment. To perform a clone request with VS Code, perform the following action (depending on your operating system): Windows : Press Ctrl + Shift + P to open an executable console at the top of VS Code macOS : Press Cmd + Shift + P to open an executable console at the top of VS Code Enter following command into the console and hit Enter to confirm: git:clone Next, specify the public repository from which to clone the Ansible Playbook templates. Enter following GitHub repository address into the console and hit Enter to confirm: https://github.com/chetan-hireholi/ansible-wca-demo-kit Regardless of whether you are new to VS Code or have used VS Code previously, the remainder of the steps in the Setup & Troubleshooting module are the same. Confirm that the working directory [A] is now set to ansible-wca-demo-kit and that it contains files similar to those listed in the screenshot below. As a final step, verify that the VM has access to the latest demo code by performing the following git fetch and git pull commands: From the top of the VM interface, drill down into View and then Command Palette [A] Into the pop-up console [B] , type git fetch , hit Enter , and wait for the operation to finish Open the console once again, type git pull , and then hit Enter At this stage, the hands-on environment has been fully configured","title":"Preparing the Ansible Playbook materials"},{"location":"archived/#_4","text":"","title":""},{"location":"archived/#troubleshooting-and-support","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ANSIBLE-LINT IS NOT AVAILABLE ansible-lint checks playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises. RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE If you are running the lab environment inside a virtual machine (VM), it might not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor.","title":"Troubleshooting and support"},{"location":"archived/#_5","text":"","title":""},{"location":"archived/#next-steps","text":"The following section will cover the fundamentals of AI-recommended code generation for Ansible Playbooks using IBM watsonx Code Assistant .","title":"Next steps"},{"location":"customizing/","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Task Description Tuning and Model Customization module. [15 min] i. Fine-tuning task prompts In this section, you will experiment with customized Ansible Playbooks and test how changes made to an Ansible Task's natural language descriptions can impact the recommended code produced by IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ). UNPREDICTABLE RESULTS FROM GENERATIVE AI & LLMs A consequence of using generative AI and Large Language Models (LLMs) is that the recommended code output of these systems will never be 100% consistent for each and every execution. For this reason, LLMs are referred to as \"non-deterministic\" systems \u2014 as opposed to \"deterministic\" systems. This is both a strength and weakness of LLMs. As you will observe in this module, even a slight modification to the Task description or the smallest change to how a Playbook is structured\u2014 the descriptions used, the variables set, and so on \u2014will influence AI-generated code recommendations. The iterative process that you will go through in this module can be viewed in three different ways: On the one hand, it shows the sensitivity of generative AI models to even the most nuanced change in natural language prompts \u2014 for good or bad. Generative AI can produce tremendous work and that output is further guided along by best practices built-in from Red Hat and IBM. But in the end, the AI can only infer user intent from the natural language descriptions supplied to it. The more clearly a user defines their Task descriptions and intent, the more likely that WCA will correctly generate code which mirrors that intent; conversely, the less precise those descriptions are, the more likely WCA will misinterpret and miss the mark. Precision is key for the disambiguation of natural language prompts. Human feedback and humans-in-the-loop are essential to these formative stages of generative AI. As offerings like WCA mature, the natural language processing capabilities of the service will continue to be refined and improved. Additional packages, functions, and training data from Ansible Galaxy (as well as other sources) are continuously being added to the product's Foundation Models, which will in turn continually improve the AI-generated code recommendations made to users. By using the Model Tuning capabilities built into IBM watsonx Code Assistant , organizations and users are able to customize the recommendations produced by generative AI, tuning a domain-specific LLM with the organization's own Ansible Playbooks. The content and code recommendations that WCA suggests can be tailored to an organization's standards, best practices, and programming styles. These capabilities will be explored later in this module. The take-away here is that your results may vary : they may differ from the SOLUTION code presented in the steps below. Keep this in mind as you work through the examples in this section and understand that it is not a bug, but rather a consequence of working with generative AI in general. The precision with which a Playbook author describes Ansible Tasks in natural language will determine the accuracy and effectiveness of WCA's generated code recommendations. A template CUSTOM PLAYBOOK #1 YAML file has been prepared for you below. Copy and paste the code from the TEMPLATE tab into a New File... within VS Code Name and save the file as you see fit: for example, customplaybook1.yml HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible CUSTOM PLAYBOOK 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 # CUSTOM PLAYBOOK #1 \u2014 Invoke 2 modules to automatically update 2 types of servers. --- # TASK 1 - name : Update web servers hosts : webservers become : true tasks : - name : Ensure apache is at the latest version - name : Write the apache config file # TASK 2 - name : Update db servers hosts : databases become : true tasks : - name : Ensure postgresql is at the latest version - name : Ensure that postgresql is started The template CUSTOM PLAYBOOK #1 contains two sets of tasks: TASK 1 ( Lines 5-12 ) checks if the web server software is up to date and runs an update if necessary TASK 2 ( Lines 15-22 ) checks if the database server software is up to date and runs an update if necessary Examine the instruction on Line 12 , which prompts Ansible to create (\"write\") a configuration file for an Apache webserver: - name : Write the apache config file Place your cursor at the end of Line 12 and press Enter to generate WCA-recommended code for the task. Accept the recommendation by pressing Tab . Two tabs are presented below: AI-GENERATED CODE shows the output from running WCA's generative AI capabilities on Line 12 of the unmodified CUSTOM PLAYBOOK #1 YAML template. SOLUTION CODE shows the expected (correct) code for performing the task that was written by a human programmer. In theory, the AI-generated code should be as good\u2014 or even superior to \u2014the manually-written solution code. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 6 7 - name : Write the apache config file ansible.builtin.template : src : templates/apache.conf.j2 dest : /etc/apache2/sites-available/000-default.conf owner : root group : root mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" As you can see, the AI-GENERATED CODE recommendations satisfy some of the requests, but also miss the mark in a few areas. When compared to the SOLUTION CODE tab: Both tabs appropriately utilize the ansible.builtin.template Module, which for this request is absolutely correct. Interesting fact: earlier (circa 2023) versions of WCA produced code recommendations for the same task as this one, which did not align with the expected solution. Instead, the recommendation was for the ansible.builtin.copy: Module to invoke content: \"{{ _content_ }}\" , which is not correct. This speaks to the continuously improving capabilities of the WCA and IBM Granite models. AI-GENERATED CODE recommended including statements that explicitly set owner: root and group: root , which are both appropriate when permissions are set to mode: '0644' . While not strictly necessary, these additional statements arguably are an improvement over the SOLUTION CODE results. It's also worth noting that the request to write the apache config file did not explicitly request mode: '0644' in the natural language description, but WCA nevertheless recommended it (post-processing) as this is a best practice for deploying Apache webservers. The src and dest variables are not in agreement across the AI-GENERATED CODE and SOLUTION CODE tabs. This is an area for improvement. The natural language description wasn't precise about these details; therefore, this is an opportunity where more detailed and verbose instructions could better steer the recommendations WCA returns back with. In general, the more ambiguous the Task description, the greater the likelihood that WCA will misinterpret the author's intent and suggest unwanted Ansible automation jobs. To help disambiguate our intention, Playbook authors should use more precise natural language terms and descriptions. Delete the WCA-suggested lines of code from Step 4 from the Playbook. Rewrite the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations Take note of the much more precise language used to describe the src and dest variables - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config The resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: The src: httpd.j2 recommendation in the AI-GENERATED CODE tab is an exact match to the natural language description set in Step 6 , and only slightly different (in terms of the directory path used) to the src: /srv/httpd.j2 variable in the SOLUTION CODE tab. The dest: /etc/httpd/conf/httpd.conf recommendation in the AI-GENERATER CODE tab, unfortunately, still deviates from the httpd.config (not the same as httpd.conf ) destination that was requested in the natural language description. Close, but still far from exact and not matching our specifications. However, it's a vast improvement over the dest: /etc/apache2/sites-available/000-default.conf destination that was recommended as a result of Step 4 's more ambiguous task description. Perplexingly, the mode: '0644' recommendation that was made previously in Step 4 has been left out of the suggested task code. We can speculate as to why\u2014 perhaps the more precise natural language description made in Step 6 prompted WCA to only generate code for exactly what was specified \u2014but the \"black box\" nature of generative AI means that we cannot know for certain. Perhaps with another, even more precise iteration the AI-GENERATED CODE will match the SOLUTION CODE ? AI-GENERATED CODE SOLUTION CODE 1 2 3 4 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" Iterate on the lessons learned in Step 7 and modify the task description to include more details on the permissions applied to the Apache webserver. Delete the WCA-suggested lines of code from Step 6 from the Playbook. Re-write the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations. Take note of the much more precise language used to describe the mode variable. - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 Once again, the resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: WCA correctly picked up on the mode: '0644' request made in the modified task description. There is now alignment between the two tabs. With only a few iterations and by disambiguating the natural language description of the Ansible Task to be performed, the code recommendations produced by WCA have been markedly improved. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" ii. Model customization Since every organization is different, WCA allows users to customize the AI model output to your organization's unique Ansible Playbooks. This allows for personalized code recommendations that are a better fit to your business' unique needs and more reflective of the programming standards set within your organization. In this scenario, your organization has its own set of Ansible Playbooks that leverage your preferred cloud provider, that uses specific Ansible Modules to manage OpenShift clusters. Open the create-openshift-cluster-ibm-cloud.yml Playbook located within the Model Customization subdirectory of the hands-on lab templates. The full directory address, as well as the Playbook code, are encapsulated in the following code block. ~/ansible-wca-demo-kit/Model Customization/create-openshift-cluster-ibm-cloud.yml 1 2 3 4 5 6 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible Before performing any model tuning tasks, try generating code recommendations for the OpenShift cluster creation task by placing your cursor at the end of Line 6 , pressing Enter , and then accepting the default recommendations by pressing Tab . The resulting code recommendations\u2014 displayed below \u2014represent the standard, unmodified output from WCA's IBM Granite base models. When using the standard IBM Granite model recommendations, it recommends making use of the ansible.builtin.command Ansible Module \u2014 which is a standard, best-practice way to perform these types of deployments. However, your particular organization (and the public cloud provider they use for deploying OpenShift clusters) might require the use of different Ansible Modules. The following steps will explore how to tailor WCA-generated code blocks to the unique needs of a business. STANDARD IBM GRANITE MODEL RECOMMENDATIONS 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" DATA PREPARATION & MODEL TRAINING Given the scope and time available for this hands-on training, participants will not be manually creating the training data or training models themselves. Instead, you will utilize a prepared customized model ( Model ID ) and experiment from Step 12 and onwards with how the tuned Model ID impacts WCA's code recommendations. The process of building Ansible training data and tuning customized models for generative AI is potentially a complex and time-consuming one. For example, the model tuning employed in the steps ahead for the creation of an OpenShift cluster according to an enterprise organization's specific standards requires approximately 4 hours to train and generate a customized AI model. In preparation for model tuning, an organization or user can transform their existing Playbooks into training data for WCA using the open-source Ansible Parser Tool . The tool analyzes Ansible Playbooks and generates a single JSONL ( ftdata.jsonl formatted) file that can be uploaded to WCA's model tuning studio for developing customized AI models. The process of calibrating and running the Ansible Parser Tool is time-consuming and potentially complex, depending on the scope of Playbooks that an organization wishes to analyze and prepare for model tuning. This falls outside the scope of the hands-on material for this training. For the purposes of demonstration, a preconfigured openshift-tune-micro.jsonl was used for the model tuning example. With the Ansible training data prepared, the IBM watsonx Code Assistant on IBM Cloud service was configured to execute a new model tuning experiment. WCA provides a graphical user experience [A] to streamline the model tuning and customization process Tuning experiment name [B] set to Tuning Experiment and confirmed with Create tuning experiment [C] The openshift-tune-micro.jsonl training data from Ansible Parser Tool is uploaded into WCA [D] A summary of metrics\u2014 including parameters for sampling and Ansible modules \u2014are displayed within the model tuning wizard before Start tuning [E] is selected to kick off the model tuning operating After the tuning operation has ended, a training loss graph reveals how accurate the model's predictions are over the training data set, across multiple tuning cycles As the number of tuning cycles increase, the training loss rating will tend to decrease Model tuning for these particular metrics and parameters takes approximately 4 hours to complete CLICK TO EXPAND \u2014 ACCOMPANYING SCREENSHOTS To override the IBM Granite base model recommendations, you must provide a pointer to a customized Model ID (prepared and hosted earlier on WCA) using the VS Code extension. Custom Model ID : b48e3c52-6135-408b-9a1e-8bcc75a47d32<|sepofid|>480ace2b-c58b-4930-b566-674ad6974851 Look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Within the search bar at the top of the panel, add the text Model to filter the results [B] Copy and paste the Model ID value (above) into the empty Ansible > Lightspeed: Model ID Override field [C] , then click the X in the top-left corner of the panel to save and exit Extension Settings With the tuned model, WCA will recommend code using modules, functions, and other details specific to your organization's private IT environment. Return to the tuning-example.yml Playbook from Step 10 and delete the code recommendations generated in Step 11 Regenerate the code recommendations for the task by placing your cursor at the end of Line 6 , hitting Enter , and then Tab Observe how the tuned Model ID has modified WCA's code recommendations by toggling between the TUNED MODEL and STANDARD MODEL tabs below TUNED MODEL STANDARD MODEL 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster register : cluster_create_output when : cluster_output.rc != 0 ibm.cloudcollection.ibm_container_cluster : name : test_cluster datacenter : \"{{ datacenter }}\" machine_type : \"{{ machine_type }}\" hardware : \"{{ hardware }}\" kube_version : 4.14_openshift public_vlan_id : \"{{ public_vlan_id }}\" private_vlan_id : \"{{ private_vlan_id }}\" default_pool_size : \"{{ default_worker_pool_size }}\" entitlement : \"{{ entitlement }}\" 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" Using the tuned Model ID , WCA has returned a recommendation to use the ibm.cloudcollection.ibm_container_cluster Module for the IBM Cloud, instead of the standard ansible.builtin.command Module from Step 11 . This more accurately conforms to the organization's Ansible automation standards and requirements. REMOVE THE CUSTOMIZED MODEL ID WHEN FINISHED Remember to clear the Ansible > Lightspeed: Model ID Override in the Extension Settings after you have completed Step 13 , otherwise all subsequent code generation requests made to WCA will be produced using the customized AI model. iii. Conclusion This concludes the hands-on components of the Level 3 course, but your learning and experimentation doesn't need to end here. Participants are encouraged to follow the Level 3 accreditation steps (depending on your role).","title":"Task Description Tuning and Model Customization"},{"location":"customizing/#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Task Description Tuning and Model Customization module. [15 min]","title":""},{"location":"customizing/#i-fine-tuning-task-prompts","text":"In this section, you will experiment with customized Ansible Playbooks and test how changes made to an Ansible Task's natural language descriptions can impact the recommended code produced by IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ). UNPREDICTABLE RESULTS FROM GENERATIVE AI & LLMs A consequence of using generative AI and Large Language Models (LLMs) is that the recommended code output of these systems will never be 100% consistent for each and every execution. For this reason, LLMs are referred to as \"non-deterministic\" systems \u2014 as opposed to \"deterministic\" systems. This is both a strength and weakness of LLMs. As you will observe in this module, even a slight modification to the Task description or the smallest change to how a Playbook is structured\u2014 the descriptions used, the variables set, and so on \u2014will influence AI-generated code recommendations. The iterative process that you will go through in this module can be viewed in three different ways: On the one hand, it shows the sensitivity of generative AI models to even the most nuanced change in natural language prompts \u2014 for good or bad. Generative AI can produce tremendous work and that output is further guided along by best practices built-in from Red Hat and IBM. But in the end, the AI can only infer user intent from the natural language descriptions supplied to it. The more clearly a user defines their Task descriptions and intent, the more likely that WCA will correctly generate code which mirrors that intent; conversely, the less precise those descriptions are, the more likely WCA will misinterpret and miss the mark. Precision is key for the disambiguation of natural language prompts. Human feedback and humans-in-the-loop are essential to these formative stages of generative AI. As offerings like WCA mature, the natural language processing capabilities of the service will continue to be refined and improved. Additional packages, functions, and training data from Ansible Galaxy (as well as other sources) are continuously being added to the product's Foundation Models, which will in turn continually improve the AI-generated code recommendations made to users. By using the Model Tuning capabilities built into IBM watsonx Code Assistant , organizations and users are able to customize the recommendations produced by generative AI, tuning a domain-specific LLM with the organization's own Ansible Playbooks. The content and code recommendations that WCA suggests can be tailored to an organization's standards, best practices, and programming styles. These capabilities will be explored later in this module. The take-away here is that your results may vary : they may differ from the SOLUTION code presented in the steps below. Keep this in mind as you work through the examples in this section and understand that it is not a bug, but rather a consequence of working with generative AI in general. The precision with which a Playbook author describes Ansible Tasks in natural language will determine the accuracy and effectiveness of WCA's generated code recommendations. A template CUSTOM PLAYBOOK #1 YAML file has been prepared for you below. Copy and paste the code from the TEMPLATE tab into a New File... within VS Code Name and save the file as you see fit: for example, customplaybook1.yml HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible CUSTOM PLAYBOOK 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 # CUSTOM PLAYBOOK #1 \u2014 Invoke 2 modules to automatically update 2 types of servers. --- # TASK 1 - name : Update web servers hosts : webservers become : true tasks : - name : Ensure apache is at the latest version - name : Write the apache config file # TASK 2 - name : Update db servers hosts : databases become : true tasks : - name : Ensure postgresql is at the latest version - name : Ensure that postgresql is started The template CUSTOM PLAYBOOK #1 contains two sets of tasks: TASK 1 ( Lines 5-12 ) checks if the web server software is up to date and runs an update if necessary TASK 2 ( Lines 15-22 ) checks if the database server software is up to date and runs an update if necessary Examine the instruction on Line 12 , which prompts Ansible to create (\"write\") a configuration file for an Apache webserver: - name : Write the apache config file Place your cursor at the end of Line 12 and press Enter to generate WCA-recommended code for the task. Accept the recommendation by pressing Tab . Two tabs are presented below: AI-GENERATED CODE shows the output from running WCA's generative AI capabilities on Line 12 of the unmodified CUSTOM PLAYBOOK #1 YAML template. SOLUTION CODE shows the expected (correct) code for performing the task that was written by a human programmer. In theory, the AI-generated code should be as good\u2014 or even superior to \u2014the manually-written solution code. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 6 7 - name : Write the apache config file ansible.builtin.template : src : templates/apache.conf.j2 dest : /etc/apache2/sites-available/000-default.conf owner : root group : root mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" As you can see, the AI-GENERATED CODE recommendations satisfy some of the requests, but also miss the mark in a few areas. When compared to the SOLUTION CODE tab: Both tabs appropriately utilize the ansible.builtin.template Module, which for this request is absolutely correct. Interesting fact: earlier (circa 2023) versions of WCA produced code recommendations for the same task as this one, which did not align with the expected solution. Instead, the recommendation was for the ansible.builtin.copy: Module to invoke content: \"{{ _content_ }}\" , which is not correct. This speaks to the continuously improving capabilities of the WCA and IBM Granite models. AI-GENERATED CODE recommended including statements that explicitly set owner: root and group: root , which are both appropriate when permissions are set to mode: '0644' . While not strictly necessary, these additional statements arguably are an improvement over the SOLUTION CODE results. It's also worth noting that the request to write the apache config file did not explicitly request mode: '0644' in the natural language description, but WCA nevertheless recommended it (post-processing) as this is a best practice for deploying Apache webservers. The src and dest variables are not in agreement across the AI-GENERATED CODE and SOLUTION CODE tabs. This is an area for improvement. The natural language description wasn't precise about these details; therefore, this is an opportunity where more detailed and verbose instructions could better steer the recommendations WCA returns back with. In general, the more ambiguous the Task description, the greater the likelihood that WCA will misinterpret the author's intent and suggest unwanted Ansible automation jobs. To help disambiguate our intention, Playbook authors should use more precise natural language terms and descriptions. Delete the WCA-suggested lines of code from Step 4 from the Playbook. Rewrite the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations Take note of the much more precise language used to describe the src and dest variables - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config The resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: The src: httpd.j2 recommendation in the AI-GENERATED CODE tab is an exact match to the natural language description set in Step 6 , and only slightly different (in terms of the directory path used) to the src: /srv/httpd.j2 variable in the SOLUTION CODE tab. The dest: /etc/httpd/conf/httpd.conf recommendation in the AI-GENERATER CODE tab, unfortunately, still deviates from the httpd.config (not the same as httpd.conf ) destination that was requested in the natural language description. Close, but still far from exact and not matching our specifications. However, it's a vast improvement over the dest: /etc/apache2/sites-available/000-default.conf destination that was recommended as a result of Step 4 's more ambiguous task description. Perplexingly, the mode: '0644' recommendation that was made previously in Step 4 has been left out of the suggested task code. We can speculate as to why\u2014 perhaps the more precise natural language description made in Step 6 prompted WCA to only generate code for exactly what was specified \u2014but the \"black box\" nature of generative AI means that we cannot know for certain. Perhaps with another, even more precise iteration the AI-GENERATED CODE will match the SOLUTION CODE ? AI-GENERATED CODE SOLUTION CODE 1 2 3 4 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\" Iterate on the lessons learned in Step 7 and modify the task description to include more details on the permissions applied to the Apache webserver. Delete the WCA-suggested lines of code from Step 6 from the Playbook. Re-write the description on Line 12 to the following code block, then press Enter and Tab to accept the new WCA code recommendations. Take note of the much more precise language used to describe the mode variable. - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 Once again, the resulting WCA-recommended code should be similar to the AI-GENERATED CODE tab below. When compared to the SOLUTION CODE tab: WCA correctly picked up on the mode: '0644' request made in the modified task description. There is now alignment between the two tabs. With only a few iterations and by disambiguating the natural language description of the Ansible Task to be performed, the code recommendations produced by WCA have been markedly improved. AI-GENERATED CODE SOLUTION CODE 1 2 3 4 5 - name : Write the apache config file where src equals httpd.j2 and dest equals httpd.config and mode equals 0644 ansible.builtin.template : src : httpd.j2 dest : /etc/httpd/conf/httpd.conf mode : '0644' 1 2 3 4 5 - name : Write the apache config file ansible.builtin.template : src : /srv/httpd.j2 dest : /etc/httpd.config mode : \"0644\"","title":"i. Fine-tuning task prompts"},{"location":"customizing/#ii-model-customization","text":"Since every organization is different, WCA allows users to customize the AI model output to your organization's unique Ansible Playbooks. This allows for personalized code recommendations that are a better fit to your business' unique needs and more reflective of the programming standards set within your organization. In this scenario, your organization has its own set of Ansible Playbooks that leverage your preferred cloud provider, that uses specific Ansible Modules to manage OpenShift clusters. Open the create-openshift-cluster-ibm-cloud.yml Playbook located within the Model Customization subdirectory of the hands-on lab templates. The full directory address, as well as the Playbook code, are encapsulated in the following code block. ~/ansible-wca-demo-kit/Model Customization/create-openshift-cluster-ibm-cloud.yml 1 2 3 4 5 6 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible Before performing any model tuning tasks, try generating code recommendations for the OpenShift cluster creation task by placing your cursor at the end of Line 6 , pressing Enter , and then accepting the default recommendations by pressing Tab . The resulting code recommendations\u2014 displayed below \u2014represent the standard, unmodified output from WCA's IBM Granite base models. When using the standard IBM Granite model recommendations, it recommends making use of the ansible.builtin.command Ansible Module \u2014 which is a standard, best-practice way to perform these types of deployments. However, your particular organization (and the public cloud provider they use for deploying OpenShift clusters) might require the use of different Ansible Modules. The following steps will explore how to tailor WCA-generated code blocks to the unique needs of a business. STANDARD IBM GRANITE MODEL RECOMMENDATIONS 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" DATA PREPARATION & MODEL TRAINING Given the scope and time available for this hands-on training, participants will not be manually creating the training data or training models themselves. Instead, you will utilize a prepared customized model ( Model ID ) and experiment from Step 12 and onwards with how the tuned Model ID impacts WCA's code recommendations. The process of building Ansible training data and tuning customized models for generative AI is potentially a complex and time-consuming one. For example, the model tuning employed in the steps ahead for the creation of an OpenShift cluster according to an enterprise organization's specific standards requires approximately 4 hours to train and generate a customized AI model. In preparation for model tuning, an organization or user can transform their existing Playbooks into training data for WCA using the open-source Ansible Parser Tool . The tool analyzes Ansible Playbooks and generates a single JSONL ( ftdata.jsonl formatted) file that can be uploaded to WCA's model tuning studio for developing customized AI models. The process of calibrating and running the Ansible Parser Tool is time-consuming and potentially complex, depending on the scope of Playbooks that an organization wishes to analyze and prepare for model tuning. This falls outside the scope of the hands-on material for this training. For the purposes of demonstration, a preconfigured openshift-tune-micro.jsonl was used for the model tuning example. With the Ansible training data prepared, the IBM watsonx Code Assistant on IBM Cloud service was configured to execute a new model tuning experiment. WCA provides a graphical user experience [A] to streamline the model tuning and customization process Tuning experiment name [B] set to Tuning Experiment and confirmed with Create tuning experiment [C] The openshift-tune-micro.jsonl training data from Ansible Parser Tool is uploaded into WCA [D] A summary of metrics\u2014 including parameters for sampling and Ansible modules \u2014are displayed within the model tuning wizard before Start tuning [E] is selected to kick off the model tuning operating After the tuning operation has ended, a training loss graph reveals how accurate the model's predictions are over the training data set, across multiple tuning cycles As the number of tuning cycles increase, the training loss rating will tend to decrease Model tuning for these particular metrics and parameters takes approximately 4 hours to complete CLICK TO EXPAND \u2014 ACCOMPANYING SCREENSHOTS To override the IBM Granite base model recommendations, you must provide a pointer to a customized Model ID (prepared and hosted earlier on WCA) using the VS Code extension. Custom Model ID : b48e3c52-6135-408b-9a1e-8bcc75a47d32<|sepofid|>480ace2b-c58b-4930-b566-674ad6974851 Look for Ansible under the Installed services in the Extensions panel. Click the Manage (\"cogwheel\") icon located just to the right of the Ansible tile From the drop-down options, click Extension Settings [A] Within the search bar at the top of the panel, add the text Model to filter the results [B] Copy and paste the Model ID value (above) into the empty Ansible > Lightspeed: Model ID Override field [C] , then click the X in the top-left corner of the panel to save and exit Extension Settings With the tuned model, WCA will recommend code using modules, functions, and other details specific to your organization's private IT environment. Return to the tuning-example.yml Playbook from Step 10 and delete the code recommendations generated in Step 11 Regenerate the code recommendations for the task by placing your cursor at the end of Line 6 , hitting Enter , and then Tab Observe how the tuned Model ID has modified WCA's code recommendations by toggling between the TUNED MODEL and STANDARD MODEL tabs below TUNED MODEL STANDARD MODEL 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster register : cluster_create_output when : cluster_output.rc != 0 ibm.cloudcollection.ibm_container_cluster : name : test_cluster datacenter : \"{{ datacenter }}\" machine_type : \"{{ machine_type }}\" hardware : \"{{ hardware }}\" kube_version : 4.14_openshift public_vlan_id : \"{{ public_vlan_id }}\" private_vlan_id : \"{{ private_vlan_id }}\" default_pool_size : \"{{ default_worker_pool_size }}\" entitlement : \"{{ entitlement }}\" 1 2 3 4 5 6 7 8 9 10 --- - name : Deploy infrastructure hosts : all tasks : - name : Create an OpenShift cluster ansible.builtin.command : oadm create-cluster --wait=false --name=openshift-cluster register : oadm_create_cluster changed_when : \"'created' in oadm_create_cluster.stdout\" failed_when : \"'already exists' not in oadm_create_cluster.stdout\" Using the tuned Model ID , WCA has returned a recommendation to use the ibm.cloudcollection.ibm_container_cluster Module for the IBM Cloud, instead of the standard ansible.builtin.command Module from Step 11 . This more accurately conforms to the organization's Ansible automation standards and requirements. REMOVE THE CUSTOMIZED MODEL ID WHEN FINISHED Remember to clear the Ansible > Lightspeed: Model ID Override in the Extension Settings after you have completed Step 13 , otherwise all subsequent code generation requests made to WCA will be produced using the customized AI model.","title":"ii. Model customization"},{"location":"customizing/#iii-conclusion","text":"This concludes the hands-on components of the Level 3 course, but your learning and experimentation doesn't need to end here. Participants are encouraged to follow the Level 3 accreditation steps (depending on your role).","title":"iii. Conclusion"},{"location":"evaluation/","text":"i. Evaluation Criteria for IBM Technical Sellers and Business Partners To receive the Level 3 badge ( Generative AI for Code with watsonx Code Assistant Technical Sales Intermediate ), IBMers and Business Partners must demonstrate mastery of the skills learned throughout the various modules of these hands-on labs and coursework. Level 3 skills requirements\u2014 and the way participants will be evaluated \u2014differs depending on job role. IBM TECHNICAL SELLERS IBM Sales and Tech Sales must develop and record a Stand & Deliver presentation, which will be uploaded to the IBM Stand & Deliver platform for evaluation. This video is intended to simulate your delivery of a \u201clive\u201d demo in front of a client \u2014 on camera. IBMers will have flexibility in defining a hypothetical client, the pain points that customer has, and the goals they aspire to. The recording will then cover the seller\u2019s hands-on demonstration and pitch to the client of the value of the IBM solution using the environments and techniques of this lab. BUSINESS PARTNERS Business partners must pass a skills evaluation quiz after completing the hands-on portion of the course. The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. Participants must pass the quiz with a grade of 80% or higher. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Answers to quiz questions can only be determined by carefully following the instructions of the hands-on lab. ii. IBMer Stand & Deliver Assessment IBMers \u2014 SUBMIT RECORDINGS HERE Submit your Stand & Deliver recording online using IBM YourLearning. Hands-on demonstrations of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed are available on Seismic, as well as embedded within the various modules of this documentation. The authors highly recommend tailoring the Stand & Deliver to your own presentation style and expertise. Captivating and delighting clients with your unique delivery is paramount! Introduction [6 min] Generating Code [10 min] Content Source Matching and Post-Processing [10 min] Task Description Tuning and Model Customization [15 min] Stand & Deliver with Paul Zikopoulos [8 min] The evaluation criteria described below only applies to IBMers , who must record a Stand & Deliver to receive credit for this Level 3 course. By default, IBMers will be evaluated by their First Line Manager (FLM) \u2014 although they may request another manager to evaluate their Stand & Deliver, if appropriate. Instructions on how to submit a Stand & Deliver are included within the activity on YourLearning. IBM Technical Sellers need to include all six of the following elements in their Stand & Deliver recording to receive a Level 3 badge: Seller articulated their client's pain point(s) and the value proposition of using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller highlighted use cases for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller demonstrated and discussed several of the key differentiated capabilities of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed that deliver on the value proposition on point one. Seller highlighted benefits to their client (this is the why the client can\u2019t live without these benefits section). Seller highlighted benefits to their client's customers (what will the client be able to deliver to their customers that they could not without this product). Seller closed the demonstration with a call to action for their client that could include: a workshop, a deeper dive into the product meeting, or a Proof of Experience (PoX). iii. Business Partner Quiz Assessment PARTNERS \u2014 COMPLETE ASSESSMENT HERE Complete your assessment online using IBM Training: https://learn.ibm.com/mod/subcourse/view.php?id=284786 The skills evaluation quiz only applies to business partners \u2013 IBMers must record a Stand & Deliver . The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Participants must pass the quiz with a grade of 80% or higher. Answers to the quiz can only be determined by carefully following the instructions of the hands-on lab. iv. Next steps In the following section, you will prepare your hands-on lab environment with the necessary services and configurations.","title":"Evaluation"},{"location":"evaluation/#_1","text":"","title":""},{"location":"evaluation/#i-evaluation-criteria-for-ibm-technical-sellers-and-business-partners","text":"To receive the Level 3 badge ( Generative AI for Code with watsonx Code Assistant Technical Sales Intermediate ), IBMers and Business Partners must demonstrate mastery of the skills learned throughout the various modules of these hands-on labs and coursework. Level 3 skills requirements\u2014 and the way participants will be evaluated \u2014differs depending on job role. IBM TECHNICAL SELLERS IBM Sales and Tech Sales must develop and record a Stand & Deliver presentation, which will be uploaded to the IBM Stand & Deliver platform for evaluation. This video is intended to simulate your delivery of a \u201clive\u201d demo in front of a client \u2014 on camera. IBMers will have flexibility in defining a hypothetical client, the pain points that customer has, and the goals they aspire to. The recording will then cover the seller\u2019s hands-on demonstration and pitch to the client of the value of the IBM solution using the environments and techniques of this lab. BUSINESS PARTNERS Business partners must pass a skills evaluation quiz after completing the hands-on portion of the course. The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. Participants must pass the quiz with a grade of 80% or higher. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Answers to quiz questions can only be determined by carefully following the instructions of the hands-on lab.","title":"i. Evaluation Criteria for IBM Technical Sellers and Business Partners"},{"location":"evaluation/#ii-ibmer-stand-deliver-assessment","text":"IBMers \u2014 SUBMIT RECORDINGS HERE Submit your Stand & Deliver recording online using IBM YourLearning. Hands-on demonstrations of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed are available on Seismic, as well as embedded within the various modules of this documentation. The authors highly recommend tailoring the Stand & Deliver to your own presentation style and expertise. Captivating and delighting clients with your unique delivery is paramount! Introduction [6 min] Generating Code [10 min] Content Source Matching and Post-Processing [10 min] Task Description Tuning and Model Customization [15 min] Stand & Deliver with Paul Zikopoulos [8 min] The evaluation criteria described below only applies to IBMers , who must record a Stand & Deliver to receive credit for this Level 3 course. By default, IBMers will be evaluated by their First Line Manager (FLM) \u2014 although they may request another manager to evaluate their Stand & Deliver, if appropriate. Instructions on how to submit a Stand & Deliver are included within the activity on YourLearning. IBM Technical Sellers need to include all six of the following elements in their Stand & Deliver recording to receive a Level 3 badge: Seller articulated their client's pain point(s) and the value proposition of using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller highlighted use cases for IBM watsonx Code Assistant for Red Hat Ansible Lightspeed . Seller demonstrated and discussed several of the key differentiated capabilities of IBM watsonx Code Assistant for Red Hat Ansible Lightspeed that deliver on the value proposition on point one. Seller highlighted benefits to their client (this is the why the client can\u2019t live without these benefits section). Seller highlighted benefits to their client's customers (what will the client be able to deliver to their customers that they could not without this product). Seller closed the demonstration with a call to action for their client that could include: a workshop, a deeper dive into the product meeting, or a Proof of Experience (PoX).","title":"ii. IBMer Stand &amp; Deliver Assessment"},{"location":"evaluation/#iii-business-partner-quiz-assessment","text":"PARTNERS \u2014 COMPLETE ASSESSMENT HERE Complete your assessment online using IBM Training: https://learn.ibm.com/mod/subcourse/view.php?id=284786 The skills evaluation quiz only applies to business partners \u2013 IBMers must record a Stand & Deliver . The quiz consists of multiple choice questions, with four possible responses (and only one correct answer) for each question. The quiz questions will ask you about on-screen text or descriptions that come up as you work through the lab guide. Participants must pass the quiz with a grade of 80% or higher. Answers to the quiz can only be determined by carefully following the instructions of the hands-on lab.","title":"iii. Business Partner Quiz Assessment"},{"location":"evaluation/#iv-next-steps","text":"In the following section, you will prepare your hands-on lab environment with the necessary services and configurations.","title":"iv. Next steps"},{"location":"generating/","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements and hands-on components of the Generating Code module. [10 min] i. Generating Code with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed An Ansible Task is a statement in Ansible's automation script (the YAML-based Playbooks you will be working with) that declares a single action to be executed. This might be installing a package, copying a file, or shutting down a service on a remote machine. Each Task represents an idempotent operation (an action that can be repeated multiple times and deliver the same result every time) that aligns the remote managed node to the specified state. Idempotent operations also ensure consistency across multiple executions, guaranteeing the same steps are taken on each execution of the task. After you have learned the fundamentals of generating Ansible Task code blocks using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you'll be ready to shape and tailor the AI-generated code recommendations using WCA's model tuning capabilities. ii. Single task Ansible operations The process of creating AI-generated code recommendations is as simple as modifying the natural language (plain English) Task descriptions of an action that is to be executed, which always start with - name: and are followed by some description of the task to be performed. Ansible Tasks are often preceded with the prefix # , indicating developer comments or documentation. After the natural language description of the automation Task has been set by the user, WCA handles the rest. WCA is also capable of generating multiple Ansible Tasks from more complex natural language descriptions\u2014 what is referred to as multi-task code generation \u2014which you will experiment with later in this module. However, to get started, let's begin with the basics of generating code for single task use cases. Begin by opening the install_cockpit_single-task.yml Playbook from the list of assets in the Explorer browser. Click the Explorer tab from the left-hand interface [A] Drill down into the Install and configure Cockpit using Ansible subdirectory [B] Double-click the install_cockpit_single-task.yml Playbook A replica of the Playbook code is also included below in the documentation The red highlighting within the editor reminds users that the tasks: section contains no valid -name: task definitions. This is part of WCA's code validation process which runs automatically and alerts users to syntax errors in their code. You can safely ignore these warnings for now, as you will be un-commenting and generating valid -name: task definitions in the following steps. ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_single-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 --- - name : Install and configure Cockpit hosts : rhel become : true # module_defaults: # ansible.builtin.service: # enabled: true # state: started tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - The suggestion included Ansible best practices by using Fully Qualified Collection name. # - name: Install cockpit package # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used \"ansible.builtin.template\" module based on the \".j2\" file extension. # # Note - The suggestion set the file permissions (\"0644\"), owner, and group based on Ansible best practices. # - name: Copy cockpit.conf.j2 to /etc/cockpit # TASK 3 # # 3a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used the generic \"Start and enable service\" prompt # # and full Playbook context to infer the recommendation should start the \"cockpit\" service. # # 3b. Uncomment the \"module_defaults\" section at the top of the Playbook. # # 3c. Clear current task suggestion and request updated suggestion. # # Note - Ansible Lightspeed used the full Playbook context and evaluated the \"module_defaults\" # # when generating a suggestion. # # The updated suggestion no longer includes \"enabled:\" and \"state:\" arguments. # - name: Start and enable service # TASK 4 # # 4a. Uncomment task description below and generate an Ansible Playbook task suggestion. # # 4b. Save the Playbook. # - name: Wait 15 seconds port 9090 The install_cockpit_single-task.yml Playbook code above warrants some explanation before we move on with making AI-generated modifications to it: Line 2 essentially marks the beginning of the Playbook instructions, the purpose of which is to automate the process of installing and configuring Cockpit for Red Hat Ansible. Lines 3-4 define variables that will remain static throughout the remainder of the Playbook. These variables will be referenced by the AI-generated code suggestions at a later stage. This is a key capability of the offering and one which you will explore in much finer details later on in this module. Lines 6-9 are variables which have been commented out and therefore are invisible to the execution of the Ansible script and not examined by WCA for context when generating code recommendations. You will experiment with how removing the # comment blocks impacts the recommendations of task block code. \"Uncommenting\" these lines of code will make them viable for execution and these lines will afterwards be considered as valid Playbook \"context\" for AI code generation. Locate TASK 1 on Line 15 of the YAML file, which handles installation of Cockpit for Ansible. Cockpit is an interactive server administration interface that provides a graphical overview of statistics and configurations for a system or systems within a network. # - name: Install cockpit package Pay attention to the indentation and characters used on Line 15 , which in sequence from left to right are as follows: begins with Tab (or Space whitespaces) for indentation a # character to \"comment out\" the line's contents a whitespace Space character - name: which signifies the start of a Task definition and finally the natural language description of the Task INDENTATION LEVELS AND WHITESPACE Similar to Python, Ansible and YAML-based Playbooks are very sensitive to whitespacing and indentation. Indentations (such as the Tab in this example) denote different hierarchies and code nesting levels within the YAML structure. You may use Space instead of Tab if you prefer, but be sure to use indentations consistently : choose to use either Tab or Space for indenting lines of code, and do not interchange between the two. To generate code for TASK 1 , first uncomment the line of code (remove the # character from the start of a line). Highlight the line(s) of code you wish to uncomment and then press Cmd + ? for macOS or Ctrl + ? for Windows You can repeat those keystrokes with the line(s) selected to toggle between commenting or uncommenting lines of code Tip: commented out lines of code in VS Code will appear as green text Afterwards, Line 15 should look like the following \u2014 beginning with a single Tab - name : Install cockpit package Now you are ready to begin generating code. Place your cursor on Line 15 and hit Enter Wait for WCA to engage and generate the suggested (in grey, italicized text ) code block for executing the task This temporary code suggestion is entirely generated by AI As a user, you have the option to either: Accept the code recommendation as-given by pressing Tab Modify the recommended code by highlighting and replacing the italicized text FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint Hit Tab to accept the suggested code and then compare with the SOLUTION tab below. TEMPLATE SOLUTION # TASK 1 - name : Install cockpit package # TASK 1 - name : Install cockpit package ansible.builtin.package : name : cockpit state : present As part of the plain-text description of the Task, WCA was asked to include the cockpit Role, part of the Red Hat Enterprise Linux System Roles Certified Content Collection. The AI-generated code suggestion invoked a Fully Qualified Collection Name ( FQCN ) - ansible.builtin.package Making use of FQCNs where possible is a recommended best practice and is a prime example of the many ways in which the offering infuses post-processing capabilities within the AI-generated code produced by WCA. Additional examples of infusing best-practices into AI-generated code recommendations can be found in TASK 2 ( Line 21 of the unmodified template or Line 25 after Step 5 ): Uncomment - name: Copy cockpit.conf.j2 to /etc/cockpit Hit Enter to generate the task code recommendation and accept the AI-suggested code (without modifications) by pressing Tab Compare your results with the SOLUTION tab below TEMPLATE SOLUTION # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' The AI-generated code recommendation will copy cockpit.conf to the target host. Take note of the fact that the recommendation included the mode: argument and set the Linux file permissions to 0644 , neither of which were things explicitly requested in the Task -name description, but are both additions which adhere to best practices around defining Ansible automaton tasks. Setting a file permission to 0644 specifies read and write permissions for User and Group levels within the Linux OS, and provides only read permissions to all others. iii. Multi-task Ansible operations Up to this point, we've kept a narrow aperture on AI-generated recommendations for single tasks \u2014 examining and experimenting with generating Ansible code task by task, one at a time. However, a powerful WCA feature is the ability to combine multiple task descriptions into a single natural language prompt; in turn, WCA is able to parse that instruction, decompose the instruction into discrete Ansible Task parts, and return a complete code recommendation for achieving the author's intended goal. Syntactically, multiple tasks are combined into a single natural language expression through the use of ampersand ( & ) characters. Simply write out all the automation task descriptions on a single line, separating each description with a & character. The line must also begin with a # character for reasons that will be explained shortly. To illustrate, let's look at a multi-task Ansible Playbook: install_cockpit_multi-task.yml The contents of this Playbook should look familiar to you already: it is essentially the same Playbook examined in Steps 1-6 ( install_cockpit_single-task.yml ), re-written in an equivalent multi-task expression Each of the Task descriptions from the previous Playbook have been consolidated into a single description on Line 12 , separated by & characters ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_multi-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : # Install cockpit package & Copy cockpit.conf.j2 to /etc/cockpit & Start and enable service & Wait 15 seconds port 9090 There are two crucial distinctions between single task and multi-task code generation: formatting and execution. Formatting : Notice that Line 12 does not begin with -name: , as was the case with single task descriptions Execution : In order to generate AI code recommendations for multi-task descriptions, Line 12 must stay commented out (the # must remain at the start of the line) What is the rationale behind this? When WCA's generative AI capabilities parse Line 12 , its output will include multiple -name: tasks, each containing potentially multiple lines of instructions, based on how many & -delineated task descriptions are included on the line. Therefore, the way in which the code generation step is executed on Line 12 is a consequence of the formatting decision. Execution of a code generation step on a commented-out ( # ) line containing & delineators is recognized by WCA as a unique case that will be acted upon as a multi-task statement. Place your cursor at the end of Line 12 , and without removing the # character, press Enter to execute the code generation step. Be aware that generating code for multi-task descriptions will take longer compared to a single task. Compare the MULTI-TASK solution tab with the SINGLE TASK solution (copied over from Step 6 ). How did the multi-task code generation fare compared to the single task approach? MULTI-TASK SINGLE TASK 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit.socket - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit state : started enabled : true - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 Comparing the two results, the only notable difference between the two approaches are Lines 27-29 from the SINGLE TASK generative AI approach. Both the SINGLE TASK and MULTI-TASK suggestions for that particular task satisfy the request made by the user. However, whether the single task or multi-task approach resulted in a better code suggestion is up to the judgement of the programmer. Nearly 90% of the remaining code was identical between the two approaches and was achieved in far fewer lines of code (and less typing) using the multi-task approach. The variability of generative AI suggestions is a fascinating topic and one that we will dive more deeply into with the module ahead. Before moving on to other product features, experiment by creating a new Ansible Playbook in your workspace using the code template below. Suggestions will be given on how to perform the same automation task using single and multi-task generation approaches. Save the YAML file as create_ec2_single_multi.yml (if you forget to save the file, WCA will not generate recommendations) Copy the following code block to your clipboard using the + icon in the top-right corner of the panel and paste into the newly created YAML file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ~/Documents/ansible-wca-demo-kit/create_ec2_single_multi.yml 1 2 3 4 --- - name : Provision an EC2 instance hosts : all tasks : First, configure the Playbook for single task generation by adding the following code snippet into Line 5 (remembering to properly indent with Tab or Space characters): - name : create vpc named demo Your workspace Playbook should look identical to the TEMPLATE tab below. Place your cursor at the end of the newly-created Line 5 and hit Enter to execute single task code generation. Compare your results to the SOLUTION tab. Record your results to a notepad so that you can compare the results later. TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc Now it's time to re-write the Playbook for multi-task code generation. The beauty of natural language statements is that your approach can be as terse or verbose as you want. The more verbose and descriptive, the more prescriptive you can be in terms of influencing the AI-generated code recommendations from WCA. The topic of \"prompt tuning\" will be explored in much greater detail in the Task Description Tuning and Model Customization module. But for now: Replace all of the code inside create_ec2_single_multi.yml with the TEMPLATE code block (below) Place your cursor at the end of Line 5 (remember not to remove the # character) and hit Enter , then accept the recommendation with Tab Compare your results to the SOLUTION tab and to the result from Step 13 TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet - name : Create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc - name : Create security group for demo vpc and allow ssh and http amazon.aws.ec2_security_group : name : demo description : demo security group vpc_id : \"{ vpc.vpc.id }\" rules : - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 register : sg - name : Create internet gateway named demogateway for demo vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demogateway register : igw - name : Create subnet named demosubnet in demo vpc amazon.aws.ec2_vpc_subnet : cidr : 10.0.0.80/24 vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet register : subnet - name : Create route table for subnet demosubnet and gateway demogateway amazon.aws.ec2_vpc_route_table : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet subnets : - \"{ subnet.subnet.id }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ igw.gateway_id }\" register : route_table - name : Create t2.small ec2 instance in subnet demosubnet amazon.aws.ec2_instance : key_name : \"{ _key_name_ }\" instance_type : t2.small image : \"{ _image_ }\" wait : true vpc_subnet_id : \"{ subnet.subnet.id }\" security_group : demo register : ec2 TIMEOUT WARNING It may take several moments for WCA to process and return code recommendations for a multi-task description as complex as this one. If you receive a time-out warning, try executing the code generation step by pressing Enter a second time. iv. Next steps The following module will examine in detail WCA's post-processing and content source attribution capabilities.","title":"Generating Code"},{"location":"generating/#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements and hands-on components of the Generating Code module. [10 min]","title":""},{"location":"generating/#i-generating-code-with-ibm-watsonx-code-assistant-for-red-hat-ansible-lightspeed","text":"An Ansible Task is a statement in Ansible's automation script (the YAML-based Playbooks you will be working with) that declares a single action to be executed. This might be installing a package, copying a file, or shutting down a service on a remote machine. Each Task represents an idempotent operation (an action that can be repeated multiple times and deliver the same result every time) that aligns the remote managed node to the specified state. Idempotent operations also ensure consistency across multiple executions, guaranteeing the same steps are taken on each execution of the task. After you have learned the fundamentals of generating Ansible Task code blocks using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), you'll be ready to shape and tailor the AI-generated code recommendations using WCA's model tuning capabilities.","title":"i. Generating Code with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed"},{"location":"generating/#ii-single-task-ansible-operations","text":"The process of creating AI-generated code recommendations is as simple as modifying the natural language (plain English) Task descriptions of an action that is to be executed, which always start with - name: and are followed by some description of the task to be performed. Ansible Tasks are often preceded with the prefix # , indicating developer comments or documentation. After the natural language description of the automation Task has been set by the user, WCA handles the rest. WCA is also capable of generating multiple Ansible Tasks from more complex natural language descriptions\u2014 what is referred to as multi-task code generation \u2014which you will experiment with later in this module. However, to get started, let's begin with the basics of generating code for single task use cases. Begin by opening the install_cockpit_single-task.yml Playbook from the list of assets in the Explorer browser. Click the Explorer tab from the left-hand interface [A] Drill down into the Install and configure Cockpit using Ansible subdirectory [B] Double-click the install_cockpit_single-task.yml Playbook A replica of the Playbook code is also included below in the documentation The red highlighting within the editor reminds users that the tasks: section contains no valid -name: task definitions. This is part of WCA's code validation process which runs automatically and alerts users to syntax errors in their code. You can safely ignore these warnings for now, as you will be un-commenting and generating valid -name: task definitions in the following steps. ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_single-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 --- - name : Install and configure Cockpit hosts : rhel become : true # module_defaults: # ansible.builtin.service: # enabled: true # state: started tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - The suggestion included Ansible best practices by using Fully Qualified Collection name. # - name: Install cockpit package # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used \"ansible.builtin.template\" module based on the \".j2\" file extension. # # Note - The suggestion set the file permissions (\"0644\"), owner, and group based on Ansible best practices. # - name: Copy cockpit.conf.j2 to /etc/cockpit # TASK 3 # # 3a. Uncomment task description below and generate a task suggestion. # # Note - Ansible Lightspeed used the generic \"Start and enable service\" prompt # # and full Playbook context to infer the recommendation should start the \"cockpit\" service. # # 3b. Uncomment the \"module_defaults\" section at the top of the Playbook. # # 3c. Clear current task suggestion and request updated suggestion. # # Note - Ansible Lightspeed used the full Playbook context and evaluated the \"module_defaults\" # # when generating a suggestion. # # The updated suggestion no longer includes \"enabled:\" and \"state:\" arguments. # - name: Start and enable service # TASK 4 # # 4a. Uncomment task description below and generate an Ansible Playbook task suggestion. # # 4b. Save the Playbook. # - name: Wait 15 seconds port 9090 The install_cockpit_single-task.yml Playbook code above warrants some explanation before we move on with making AI-generated modifications to it: Line 2 essentially marks the beginning of the Playbook instructions, the purpose of which is to automate the process of installing and configuring Cockpit for Red Hat Ansible. Lines 3-4 define variables that will remain static throughout the remainder of the Playbook. These variables will be referenced by the AI-generated code suggestions at a later stage. This is a key capability of the offering and one which you will explore in much finer details later on in this module. Lines 6-9 are variables which have been commented out and therefore are invisible to the execution of the Ansible script and not examined by WCA for context when generating code recommendations. You will experiment with how removing the # comment blocks impacts the recommendations of task block code. \"Uncommenting\" these lines of code will make them viable for execution and these lines will afterwards be considered as valid Playbook \"context\" for AI code generation. Locate TASK 1 on Line 15 of the YAML file, which handles installation of Cockpit for Ansible. Cockpit is an interactive server administration interface that provides a graphical overview of statistics and configurations for a system or systems within a network. # - name: Install cockpit package Pay attention to the indentation and characters used on Line 15 , which in sequence from left to right are as follows: begins with Tab (or Space whitespaces) for indentation a # character to \"comment out\" the line's contents a whitespace Space character - name: which signifies the start of a Task definition and finally the natural language description of the Task INDENTATION LEVELS AND WHITESPACE Similar to Python, Ansible and YAML-based Playbooks are very sensitive to whitespacing and indentation. Indentations (such as the Tab in this example) denote different hierarchies and code nesting levels within the YAML structure. You may use Space instead of Tab if you prefer, but be sure to use indentations consistently : choose to use either Tab or Space for indenting lines of code, and do not interchange between the two. To generate code for TASK 1 , first uncomment the line of code (remove the # character from the start of a line). Highlight the line(s) of code you wish to uncomment and then press Cmd + ? for macOS or Ctrl + ? for Windows You can repeat those keystrokes with the line(s) selected to toggle between commenting or uncommenting lines of code Tip: commented out lines of code in VS Code will appear as green text Afterwards, Line 15 should look like the following \u2014 beginning with a single Tab - name : Install cockpit package Now you are ready to begin generating code. Place your cursor on Line 15 and hit Enter Wait for WCA to engage and generate the suggested (in grey, italicized text ) code block for executing the task This temporary code suggestion is entirely generated by AI As a user, you have the option to either: Accept the code recommendation as-given by pressing Tab Modify the recommended code by highlighting and replacing the italicized text FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint Hit Tab to accept the suggested code and then compare with the SOLUTION tab below. TEMPLATE SOLUTION # TASK 1 - name : Install cockpit package # TASK 1 - name : Install cockpit package ansible.builtin.package : name : cockpit state : present As part of the plain-text description of the Task, WCA was asked to include the cockpit Role, part of the Red Hat Enterprise Linux System Roles Certified Content Collection. The AI-generated code suggestion invoked a Fully Qualified Collection Name ( FQCN ) - ansible.builtin.package Making use of FQCNs where possible is a recommended best practice and is a prime example of the many ways in which the offering infuses post-processing capabilities within the AI-generated code produced by WCA. Additional examples of infusing best-practices into AI-generated code recommendations can be found in TASK 2 ( Line 21 of the unmodified template or Line 25 after Step 5 ): Uncomment - name: Copy cockpit.conf.j2 to /etc/cockpit Hit Enter to generate the task code recommendation and accept the AI-suggested code (without modifications) by pressing Tab Compare your results with the SOLUTION tab below TEMPLATE SOLUTION # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit # TASK 2 - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' The AI-generated code recommendation will copy cockpit.conf to the target host. Take note of the fact that the recommendation included the mode: argument and set the Linux file permissions to 0644 , neither of which were things explicitly requested in the Task -name description, but are both additions which adhere to best practices around defining Ansible automaton tasks. Setting a file permission to 0644 specifies read and write permissions for User and Group levels within the Linux OS, and provides only read permissions to all others.","title":"ii. Single task Ansible operations"},{"location":"generating/#iii-multi-task-ansible-operations","text":"Up to this point, we've kept a narrow aperture on AI-generated recommendations for single tasks \u2014 examining and experimenting with generating Ansible code task by task, one at a time. However, a powerful WCA feature is the ability to combine multiple task descriptions into a single natural language prompt; in turn, WCA is able to parse that instruction, decompose the instruction into discrete Ansible Task parts, and return a complete code recommendation for achieving the author's intended goal. Syntactically, multiple tasks are combined into a single natural language expression through the use of ampersand ( & ) characters. Simply write out all the automation task descriptions on a single line, separating each description with a & character. The line must also begin with a # character for reasons that will be explained shortly. To illustrate, let's look at a multi-task Ansible Playbook: install_cockpit_multi-task.yml The contents of this Playbook should look familiar to you already: it is essentially the same Playbook examined in Steps 1-6 ( install_cockpit_single-task.yml ), re-written in an equivalent multi-task expression Each of the Task descriptions from the previous Playbook have been consolidated into a single description on Line 12 , separated by & characters ~/Documents/ansible-wca-demo-kit/install and configure Cockpit using Ansible/install_cockpit_multi-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : # Install cockpit package & Copy cockpit.conf.j2 to /etc/cockpit & Start and enable service & Wait 15 seconds port 9090 There are two crucial distinctions between single task and multi-task code generation: formatting and execution. Formatting : Notice that Line 12 does not begin with -name: , as was the case with single task descriptions Execution : In order to generate AI code recommendations for multi-task descriptions, Line 12 must stay commented out (the # must remain at the start of the line) What is the rationale behind this? When WCA's generative AI capabilities parse Line 12 , its output will include multiple -name: tasks, each containing potentially multiple lines of instructions, based on how many & -delineated task descriptions are included on the line. Therefore, the way in which the code generation step is executed on Line 12 is a consequence of the formatting decision. Execution of a code generation step on a commented-out ( # ) line containing & delineators is recognized by WCA as a unique case that will be acted upon as a multi-task statement. Place your cursor at the end of Line 12 , and without removing the # character, press Enter to execute the code generation step. Be aware that generating code for multi-task descriptions will take longer compared to a single task. Compare the MULTI-TASK solution tab with the SINGLE TASK solution (copied over from Step 6 ). How did the multi-task code generation fare compared to the single task approach? MULTI-TASK SINGLE TASK 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit.socket - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 --- - name : Install and configure Cockpit hosts : rhel become : true module_defaults : ansible.builtin.service : enabled : true state : started tasks : - name : Install cockpit package ansible.builtin.package : name : cockpit state : present - name : Copy cockpit.conf.j2 to /etc/cockpit ansible.builtin.template : src : cockpit.conf.j2 dest : /etc/cockpit/cockpit.conf owner : root group : root mode : '0644' - name : Start and enable service ansible.builtin.service : name : cockpit state : started enabled : true - name : Wait 15 seconds port 9090 ansible.builtin.wait_for : port : 9090 delay : 15 Comparing the two results, the only notable difference between the two approaches are Lines 27-29 from the SINGLE TASK generative AI approach. Both the SINGLE TASK and MULTI-TASK suggestions for that particular task satisfy the request made by the user. However, whether the single task or multi-task approach resulted in a better code suggestion is up to the judgement of the programmer. Nearly 90% of the remaining code was identical between the two approaches and was achieved in far fewer lines of code (and less typing) using the multi-task approach. The variability of generative AI suggestions is a fascinating topic and one that we will dive more deeply into with the module ahead. Before moving on to other product features, experiment by creating a new Ansible Playbook in your workspace using the code template below. Suggestions will be given on how to perform the same automation task using single and multi-task generation approaches. Save the YAML file as create_ec2_single_multi.yml (if you forget to save the file, WCA will not generate recommendations) Copy the following code block to your clipboard using the + icon in the top-right corner of the panel and paste into the newly created YAML file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible ~/Documents/ansible-wca-demo-kit/create_ec2_single_multi.yml 1 2 3 4 --- - name : Provision an EC2 instance hosts : all tasks : First, configure the Playbook for single task generation by adding the following code snippet into Line 5 (remembering to properly indent with Tab or Space characters): - name : create vpc named demo Your workspace Playbook should look identical to the TEMPLATE tab below. Place your cursor at the end of the newly-created Line 5 and hit Enter to execute single task code generation. Compare your results to the SOLUTION tab. Record your results to a notepad so that you can compare the results later. TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo 1 2 3 4 5 6 7 8 9 10 11 12 --- - name : Provision an EC2 instance hosts : all tasks : - name : create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc Now it's time to re-write the Playbook for multi-task code generation. The beauty of natural language statements is that your approach can be as terse or verbose as you want. The more verbose and descriptive, the more prescriptive you can be in terms of influencing the AI-generated code recommendations from WCA. The topic of \"prompt tuning\" will be explored in much greater detail in the Task Description Tuning and Model Customization module. But for now: Replace all of the code inside create_ec2_single_multi.yml with the TEMPLATE code block (below) Place your cursor at the end of Line 5 (remember not to remove the # character) and hit Enter , then accept the recommendation with Tab Compare your results to the SOLUTION tab and to the result from Step 13 TEMPLATE SOLUTION 1 2 3 4 5 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 --- - name : Provision an EC2 instance hosts : all tasks : # create vpc named demo & create security group for demo vpc and allow ssh and http & create internet gateway named demogateway for demo vpc & create subnet named demosubnet in demo vpc & create route table for subnet demosubnet and gateway demogateway & create t2.small ec2 instance in subnet demosubnet - name : Create vpc named demo amazon.aws.ec2_vpc_net : name : demo cidr_block : 10.0.0.80/16 tags : Name : demo tenancy : default register : vpc - name : Create security group for demo vpc and allow ssh and http amazon.aws.ec2_security_group : name : demo description : demo security group vpc_id : \"{ vpc.vpc.id }\" rules : - proto : tcp from_port : 22 to_port : 22 cidr_ip : 0.0.0.0/0 - proto : tcp from_port : 80 to_port : 80 cidr_ip : 0.0.0.0/0 register : sg - name : Create internet gateway named demogateway for demo vpc amazon.aws.ec2_vpc_igw : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demogateway register : igw - name : Create subnet named demosubnet in demo vpc amazon.aws.ec2_vpc_subnet : cidr : 10.0.0.80/24 vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet register : subnet - name : Create route table for subnet demosubnet and gateway demogateway amazon.aws.ec2_vpc_route_table : vpc_id : \"{ vpc.vpc.id }\" tags : Name : demosubnet subnets : - \"{ subnet.subnet.id }\" routes : - dest : 0.0.0.0/0 gateway_id : \"{ igw.gateway_id }\" register : route_table - name : Create t2.small ec2 instance in subnet demosubnet amazon.aws.ec2_instance : key_name : \"{ _key_name_ }\" instance_type : t2.small image : \"{ _image_ }\" wait : true vpc_subnet_id : \"{ subnet.subnet.id }\" security_group : demo register : ec2 TIMEOUT WARNING It may take several moments for WCA to process and return code recommendations for a multi-task description as complex as this one. If you receive a time-out warning, try executing the code generation step by pressing Enter a second time.","title":"iii. Multi-task Ansible operations"},{"location":"generating/#iv-next-steps","text":"The following module will examine in detail WCA's post-processing and content source attribution capabilities.","title":"iv. Next steps"},{"location":"setup/","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. Frequently asked questions and troubleshooting steps are documented below . i. Reserving the lab environments Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Account : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.4 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below. Locate the Request a Red Hat Account tile, hover over the IBM Cloud environment button with your cursor, and then click Reserve it [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. End Date and Time Select a time and date for when the reservation will expire. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 5-10 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will automatically be de-authorized and deleted by IBM Technology Zone after the reservation period has ended Now you must request your second ITZ environment, this time for the virtualized machine (VM) environment. Open the IBM Technology Zone reservation link below: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments Locate the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.5 tile Hover over the IBM Cloud environment button with your cursor Click Reserve it [A] to continue From the Single environment reservation options , select Reserve now . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. Customer Data Select No, I will not be using customer data if using for education purposes. End Date and Time Select a time and date for when the reservation will expire. VPN Access Set to Disabled . When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. ii. Accessing the VM Once the ITZ reservation has been marked as \"Ready\" , access connection details for the environment by either clicking the shortcut in the ITZ email or by drilling down into the My Reservations tab [A] on the ITZ web portal. Scroll down to the bottom of the reservation page and click the blue VM Remote Console button [A] to launch the VM interface. Select the techzone [A] login when prompted. Supply the following for the password: IBMDem0s! Hit Enter to complete the VM login Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B] iii. Authorizing VS Code with WCA Now you must authorize the VS Code environment for use with WCA, using the Red Hat account details that were generated for you in Steps 5-6 . Within your VS Code environment click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button Provide the Username and Password recorded in Step 6 of this module Click Submit to continue When prompted for We need a little more information , set Job Role to Student Click the Submit button to finalize your account activation COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not possible to \"paste\" lab instructions or information from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Allow this site to open the vscode link with Visual Studio Code - URL Handler? : click Open URL At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 15-17 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 15-17 as before; the login process should run smoother (and faster) on the second attempt iv. [OPTIONAL] Download demo assets to local machine If you wish to complete the hands-on lab using a local installation of VS Code (instead of using the provided Virtual Machine), you may do so \u2014 but you'll need to clone (download) the accompanying demo assets first. To do so, first install the GitHub CLI library. Execute the following command within a Terminal console to clone (via git ) the supporting demo assets repository ( ansible-wca-demo-kit.git ) to your local machine: git clone https://github.com/chetan-hireholi/ansible-wca-demo-kit Open the local ansible-wca-demo-kit folder within VS Code to access the demo assets for the remaining sections of the lab v. Troubleshooting and support If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. vi. Next steps The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"Setup and Troubleshooting"},{"location":"setup/#_1","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. Frequently asked questions and troubleshooting steps are documented below .","title":""},{"location":"setup/#i-reserving-the-lab-environments","text":"Before getting started with IBM watsonx Code Assistant for Red Hat Ansible Lightspeed ( WCA ), an environment must be reserved and deployed via the IBM Technology Zone ( ITZ ). You will require access to the ITZ in order to reserve your environment and complete the lab. If you do not yet have access or an account with the ITZ, you will need to register for one . There are TWO environments that you must reserve from ITZ: Request a Red Hat Account : responsible for generating unique access credentials for IBM watsonx Code Assistant and Red Hat Ansible Lightspeed authorizations WCA for Ansible Essentials Plan - Visual Studio Desktop 1.4 : virtualized machine prepared with Visual Studio Code and lab demonstration scripts pre-installed; you will authenticate within this environment using the Red Hat account requested from ITZ Follow along with the instructions below to request and configure these environments. Click the IBM Technology Zone link below. Locate the Request a Red Hat Account tile, hover over the IBM Cloud environment button with your cursor, and then click Reserve it [A] : URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments FULLSCREEN IMAGES Click on any of the screenshots within this documentation to enlarge the image. From the Single environment reservation options , select Reserve now [A] . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. End Date and Time Select a time and date for when the reservation will expire. When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 5-10 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab. When the Red Hat account request has been processed by IBM Technology Zone, you will receive a pair of emails: one from ITZ and a second from Red Hat. Reservation Ready on IBM Technology Zone : You can ignore the contents of this email, as the relevant account and licensing information are contained in the Red Hat email. Confirm that the ITZ email states that Status Update: Ready [A] . Red Hat Login Email Verification : This email, addressed from a no-reply@redhat.com account [B] , contains the resources necessary for accessing your uniquely-generated Red Hat credentials. The lab guide steps that follow will instruct you on how to set those up and how to use them for accessing your IBM watsonx Code Assistant for Red Hat Ansible Lightspeed entitlements inside VS Code. With the Red Hat Login Email Verification email open, take note of two essential pieces of information: Locate the Account Information [A] details at the bottom of the email. Your login: is the unique Red Hat account assigned for you by ITZ. Your email address: will be the address associated with your ITZ account. Record both to a notepad for reference later. Click the URL [B] located within the body of the invitation email to finalize your account registration with Red Hat. An Email Confirmation page will load within your web browser. Note that the value of Red Hat login is the same as the one recorded in Step 5 Create a new Password [A] and record this to a notepad for reference later When ready, click Save [B] to finalize registration REGISTRATION IS REQUIRED If you already have a personal account with Red Hat, you must still register for a new account using the invitation URL provided Do not attempt to use a personal Red Hat account in the later steps of the Setup & Troubleshooting guide, as that account will not have access to the WCA services needed to perform the training Red Hat accounts created for this training will automatically be de-authorized and deleted by IBM Technology Zone after the reservation period has ended Now you must request your second ITZ environment, this time for the virtualized machine (VM) environment. Open the IBM Technology Zone reservation link below: URL: https://techzone.ibm.com/collection/ibm-watson-x-code-assistant-for-ansible-lightspeed/environments Locate the WCA for Ansible Essentials Plan: Visual Studio Desktop 1.5 tile Hover over the IBM Cloud environment button with your cursor Click Reserve it [A] to continue From the Single environment reservation options , select Reserve now . Supply additional details about your ITZ reservation request: Field Value Name Give your reservation a unique name. Purpose If reserving for L3 training, select Education . If delivering a PoC, select Pilot and provide a Sales Opportunity number. Describe If reserving for L3, enter WCA for Ansible Lightspeed training . If delivering a PoC, enter the PoC and client details. Preferred Geography Select the region and data center geographically closest to your location. Customer Data Select No, I will not be using customer data if using for education purposes. End Date and Time Select a time and date for when the reservation will expire. VPN Access Set to Disabled . When satisfied, verify that you agree to the Terms and Conditions for the environment and finalize your reservation request by clicking Submit . PROVISIONING TIMES Red Hat account creation take approximately 15-30 minutes to complete from the time that you click submit. If you navigate to the My Reservations tab of the ITZ, you can monitor the progress of your reservation. Wait for the ITZ reservation to be marked as \"Ready\" before attempting to start the lab.","title":"i. Reserving the lab environments"},{"location":"setup/#ii-accessing-the-vm","text":"Once the ITZ reservation has been marked as \"Ready\" , access connection details for the environment by either clicking the shortcut in the ITZ email or by drilling down into the My Reservations tab [A] on the ITZ web portal. Scroll down to the bottom of the reservation page and click the blue VM Remote Console button [A] to launch the VM interface. Select the techzone [A] login when prompted. Supply the following for the password: IBMDem0s! Hit Enter to complete the VM login Once you have successfully authenticated, click the Activities [A] button in the top-left corner of the interface to pull open the list of available applications from the bottom of the screen. Click the Visual Studio Code [B] application shortcut at the bottom of the desktop to start up the service. If you wish to copy and paste instructions directly from the lab documentation into the virtual machine (VM), it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. You may access the web browser at any time by clicking the Activities [A] button and then launching Firefox [C] , as shown. VS Code will load and then present you with an Authentication Required splash screen. Enter the same password [A] used to log into the VM: IBMDem0s! Click Unlock [B]","title":"ii. Accessing the VM"},{"location":"setup/#iii-authorizing-vs-code-with-wca","text":"Now you must authorize the VS Code environment for use with WCA, using the Red Hat account details that were generated for you in Steps 5-6 . Within your VS Code environment click the Ansible plugin [A] (denoted by the A logo) on the left-hand side of the interface. Two panels will open along the left side of the interface Within the Ansible Lightspeed Login panel, click the blue Connect button [B] The extension Ansible wants to sign in using Ansible Lightspeed : click Allow [C] Do you want Code to open the external website? : click Open [D] A web browser will load with the header Log in to Ansible Lightspeed with IBM watsonx Code Assistant \u2014 this is where you will supply your registration details recorded in Step 6 in order to authenticate the VS Code plugin with WCA. Click the Log in with Red Hat button Provide the Username and Password recorded in Step 6 of this module Click Submit to continue When prompted for We need a little more information , set Job Role to Student Click the Submit button to finalize your account activation COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not possible to \"paste\" lab instructions or information from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. After logging in with Red Hat, the web browser will display the prompt to Authorize Ansible Lightspeed for VS Code . Click Authorize [A] Allow this site to open the vscode link with Visual Studio Code - URL Handler? : click Open URL At this stage, the Ansible extension for VS Code is now authenticated and connected to IBM watsonx Code Assistant . Verify [A] that the environment is logged in as your unique Username and that the User Type: Licensed A notification pop-up will also appear in the bottom-right corner of the VS Code interface confirming the successful login CLICK TO EXPAND \u2014 FAILURE TO LOGIN OR TIMED OUT If the authentication procedure in Steps 15-17 takes too long, activation of the plugin will be \"timed out\" and the VS Code environment will display an error message [B] in the bottom-right corner Click the Connect [C] button as shown and repeat Steps 15-17 as before; the login process should run smoother (and faster) on the second attempt","title":"iii. Authorizing VS Code with WCA"},{"location":"setup/#iv-optional-download-demo-assets-to-local-machine","text":"If you wish to complete the hands-on lab using a local installation of VS Code (instead of using the provided Virtual Machine), you may do so \u2014 but you'll need to clone (download) the accompanying demo assets first. To do so, first install the GitHub CLI library. Execute the following command within a Terminal console to clone (via git ) the supporting demo assets repository ( ansible-wca-demo-kit.git ) to your local machine: git clone https://github.com/chetan-hireholi/ansible-wca-demo-kit Open the local ansible-wca-demo-kit folder within VS Code to access the demo assets for the remaining sections of the lab","title":"iv. [OPTIONAL] Download demo assets to local machine"},{"location":"setup/#v-troubleshooting-and-support","text":"If you require assistance or run into issues with the hands-on lab, help is available. Environment issues: The lab environment is managed by IBM Technology Zone. Opening a support case ticket is recommended for issues related to the hands-on environment (provisioning, running, and so on.) Documentation issues: If there is an error in the lab documentation, or if you require additional support in completing the material, open a thread on the #wca-ansible-techzone-support Slack channel. Product questions: For questions related to IBM watsonx Code Assistant capabilities, sales opportunities, roadmap, and other such matters, open a thread on the #watsonx-code-assistant Slack channel. As you settle in to the environment and begin your training, you may encounter unexpected warnings or errors. Many of these can be safely ignored or can be easily rectified. This section will serve as a running list of frequently asked questions and troubleshooting techniques. Click on any of the following topics for additional details. FAILED TO CONNECT TO THE SERVER / \"YOU DON'T HAVE ACCESS TO IBM WATSONX...\" This warning will occur when the Ansible plugin for VS Code needs to be re-authenticated with WCA. It can occur after an extended period of inactivity or a system restart. For example, if your lab environment is running inside a VM, pausing or restarting the VM may produce this error. To re-authenticate: Sign out from the VS Code application by clicking the User icon [A] in the bottom-left corner of the interface, hover over your username, and then click Sign Out [B] If you are running this environment inside a virtual machine (VM) , closing and restarting the VM will not resolve the issue \u2014 you must sign out from the VS Code application, not the VM Once logged out, follow from Step 7 of the Setup & Troubleshooting to re-authenticate with WCA CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible \"ANSIBLE-LINT IS NOT AVAILABLE.\" ansible-lint checks Playbooks for practices and behavior that could potentially be improved and can fix some of the most common ones for you. It will constantly check your Ansible syntax as you type and provide recommendations for how to improve it. You can safely ignore this error if it occurs during the lab exercises If you wish to install ansible-lint on your local machine, execute the following instruction within a Terminal console: python3 -m pip install --upgrade --user ansible-lint RED ANSIBLE ICON ALONG BOTTOM-RIGHT INTERFACE The Ansible extension for VS Code will check your local machine to determine if Red Hat Ansible has been installed locally. If you have not set up Ansible (the standalone version) on your local machine previously, this tile will display as red. You can safely ignore this error if it occurs during the lab exercises. CANNOT SET PROPERTIES OF UNDEFINED (SETTING 'currentModelValue') Make sure that the Model ID Override field is set to empty in your Ansible for VS Code extension settings. To verify this: Click the Extensions tab [A] along the left-hand interface Click the Manage icon [B] on the right side of the Ansible extension tile, then drill down into Extension Settings [C] Add the text Model to the search filter [D] at the top of the Extension Settings panel Clear the input field [E] of any model IDs and leave it blank Close the Extension Settings panel by clicking X and return to the Ansible Playbook SPAWN C:\\Windows\\system32\\cmd.exe ENOENT This warning is not related to Ansible or WCA. You can safely ignore this error if it occurs during the lab exercises. PYTHON DRIVERS ARE MISSING The WCA extension for VS Code requires that Python drivers are included within the workspace. These are usually configured within VS Code by default, but can be easily set if necessary. Look for a Python tile adjacent to the Ansible tile along the bottom-right corner of the VS Code interface. If it is not set, click the tile and select the Python 3.11.5 64-bit drivers. Click the gold-colored Select python environment button at the bottom-right of the interface From the console at the top of the VS Code environment, select the recommended Python 3.11.5 64-bit option and hit Enter to confirm GIT NOT INSTALLED If your machine has not used Git previously, you may be prompted by VS Code to install it before attempting a clone request. This is more commonplace on Windows operating systems, but some MacOS users may need to install Git as well. The following instructions will guide you through the process: With the VS Code application open, click the Source Control [A] tab from the left-hand interface and then click Download Git [B] . A web browser will open to the git-scm.com Downloads page. Download [C] the version recommended for your machine's particular operating system. The recommended version will be displayed first at the top of the list. Execute the installer on your machine and follow along with the prompts to finish installing Git. Accept the license agreement and accept the default values on each page. Return to VS Code and once again click the Source Control [D] tab, then click the blue Initialize Repository [E] button. From the top of the Source Control tab, click the three dots ... [F] icon to expand a drop-down menu of options. Click on the Clone [G] option. Follow the instructions for the remainder of this section to clone the ansible-wca-demo-kit repository to your local machine [H] . COPY AND PASTE INSTRUCTIONS INTO A VIRTUAL MACHINE As you are running the lab environment inside a virtual machine (VM), it is not be possible to \"paste\" lab instructions from your local machine's clipboard directly into the VM. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor.","title":"v. Troubleshooting and support"},{"location":"setup/#vi-next-steps","text":"The following section will cover the fundamentals of AI-recommended code generation for Ansible Tasks.","title":"vi. Next steps"},{"location":"source-matching-post-processing/","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Content Source Matching and Post-Processing module. [10 min] i. Why code attribution and post-processing matters for generative AI At this point, you should now be well acquainted with Ansible Tasks and Playbooks . Ansible Playbooks are comprised of multiple Tasks \u2014 which you have been using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed's ( WCA ) generative AI capabilities to define and expand upon. If a Playbook contains multiple \"tasks\" that together achieve some singular goal or purpose, you might want to group these tasks into a single reusable unit. For example, a string of \"tasks\" might provision infrastructure for separate development, test, and production clusters from a major cloud vendor. Patterns for automating endpoints from these vendors can be codified into Ansible Modules , which are supported by Red Hat and the Ansible community writ large. And finally, collections of Playbooks and Modules can be organized into \"blueprints\" for automation tasks that are frequently used together to achieve some goal; a goal which you might want to make repeatable or shareable with other teams within your business. These blueprints are referred to as Ansible Roles . They provide a structured way to organize tasks, templates, files, and variables, and are available to drop directly into Ansible Playbooks. Roles make it possible to easily manage and set up complex automation tasks, essentially providing a rubric to streamline automation projects. Collectively, these Playbooks, Modules, and Roles form a comprehensive ecosystem of business and community-driven support for patterning automation to work with the broadest range of vendors, technologies, and clouds. Ansible Galaxy is a Red Hat-curated, community-driven repository for Ansible Roles . Via communities such as Galaxy, thousands of Roles are available for Ansible users to leverage within their own Playbooks. Ansible Galaxy is also key to how generative AI code recommendations from WCA are attributed back to original content sources and authors. ii. Content source matching A powerful capability within WCA is Content Source Matching (often referred to as \"code explainability\"), which attempts to match AI-generated code suggestions to the training data and sources that were utilized in generating the suggested Task code. These code attribution suggestions are created using a k-NN ( K-Nearest Neighbors ) algorithm that examines Ansible Galaxy and training data repositories in search of the nearest related content to the AI-generated code suggestions. To enable Content Source Matching capabilities within WCA, navigate to the main menu bar for VS Code and drill down into View > Open View... [A] The console [A] along the top of the VS Code interface is now activated and awaiting a prompt. Type Lightspeed Training Matches and hit Enter to confirm the selection. After making this configuration change to VS Code, all requests for WCA to generate code recommendations will now open a panel at the bottom of the VS Code interface It is under the Ansible tab that content source matching results will be displayed Other tabs will also displayed for variety of options: Output , Debug Console , Terminal , Ports , and Comments Open the install_pgsql-single-task.yml Ansible Playbook, which is included within the ansible-wca-demo-kit directory. The full location of the Playbook, as well as the contents, are available in the code block below. ~/Documents/ansible-wca-demo-kit/install and configure PostgreSQL and PGAdmin container/install_pgsql-single-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 --- - name : Configure Database servers hosts : databases become : true tasks : # TASK 1 # - name: Install postgresql-server # TASK 2 # Ansible Lightspeed used an easy-to-understand natural language prompt and suggested the correct, more complex PostgreSQL CLI command to initiate the database. # Ansible Lightspeed used best practices and kept the task idempotent by including creates: /var/lib/pgsql/data/postgresql.conf in the suggestion. # - name: Run postgresql setup command # TASK 3 # Ansible Lightspeed used natural language prompt and added state: started and enabled: true module arguments based on Start and enable... in the Ansible task description. # - name: Start and enable postgresql service Generate a code recommendation for the task on Line 8 by placing your cursor at the end of the line and hitting Enter . Pay attention to the code attribution details associated with this recommendation, which will be appearing under the Ansible tab at the bottom of the VS Code interface once the code recommendation is accepted with Tab You need to accept the AI-generated code suggestions (using the Tab key) before the Ansible content source matching tab will provide details about the code's origins - name : Install postgresql-server ansible.builtin.package : name : postgresql-server state : present The three most likely content sources used in training the WCA model\u2014 which produced the AI-generated code recommendations \u2014are listed within the Ansible tab [A] . Recall from earlier that these code attribution suggestions are created using a k-NN algorithm that searches Ansible Galaxy repositories for the nearest related content to the AI-generated code suggestions. Clicking the arrow icon to the left of each attribution [B] will expand further details about the source. Information about the URL , Path , Data Source , License , and Score are displayed (where available) under each listing. Red Hat-certified and maintained collections, as well as contributors to open source projects on Ansible Galaxy, are the primary sources for Ansible Lightspeed model training and are the content sources you are most likely to see matched to AI-generated code recommendations. Clicking on the URL field will redirect your web browser back to the precise collections and sources on Ansible Galaxy from which the code recommendations were derived. Here you can learn much richer details about the status of the project, any associated open source repositories involved (such as GitHub), contributions and activities ongoing with the code base, the author(s) involved, and many more intricacies. iii. Post-processing capabilities Another element of code generation that WCA excels at is understanding context within the Playbook it is executing against. If a variable or attribute is defined earlier within that Playbook, it will be recalled and referenced\u2014 where it makes sense to do so \u2014in the generation of subsequent lines of code. You may have already noticed these post-processing capabilities in your experimentations with WCA-generated code suggestions. Post-processing of Task descriptions and YAML file contents helps generate contextually aware, accurate Ansible content suggestions. However, one way to make this feature quite obvious is to take a previously-generated block of Ansible Task code, update the value assigned to a named variable earlier in the Playbook, and then regenerate the Task code block. In theory, the newly-generated Task block will use the updated variable name (and differ from how the code block was originally generated). Using the Explorer tab, create a New File... YAML Playbook named demo_provision_ec2_instance.yml anywhere within the /ansible-wca-demo-kit/ directory. You may save the Playbook in the root of the folder or any of the subdirectories Be sure to reference the expandable tooltips below for instructions on how to perform these operations and how to configure the Playbook language for use with WCA Copy and paste the following code block into the demo_provision_ec2_instance.yml Playbook and then save the file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible demo_provision_ec2_instance.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 --- - name : EC2 Cloud Operations hosts : localhost connection : local gather_facts : false module_defaults : group/aws : region : us-east-1 # vars: # ec2_instance: # name: lightspeed-instance-01 # key_name: lightspeed-keypair # image_id: ami-016eb5d644c333ccb # RHEL 9 us-east-1 # tags: # function: lightspeed-demo # security_group: secgroup-lightspeed tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - Best practices: The suggestion used the Fully Qualified Collection name. # # Note - Context: Ansible Lightspeed used the Playbook name \"EC2 Cloud Operations\" to use the correct \"amazon.aws.ec2_vpc_subnet_info\" module. - name : Gather subnet info tag:Name subnet-lightspeed # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Context: The suggestion included the previous task's registered variable in the suggestion. # # Note - Accuracy: The suggestion provides the correct key value from the previously task's registered variable. - name : Create vpc_subnet_id var # TASK 3 # # 3a. Uncomment task description \"Provision t3.micro instance\" below and generate a task suggestion. # # Note - Efficiency: The suggestion provides practical variable examples to improve efficiency. - name : Provision t3.micro instance # # 3b. Remove the above task and suggestion. # # Uncomment 2nd task description \"Provision t3.micro instance using ec2_instance var\". # # Generate an updated suggestion. # # Note - Context: The updated suggestion includes the \"ec2_instance variable fields in the suggestion\" - name : Provision t3.micro instance using ec2_instance var TASK 1 ( Line 26 ) and TASK 2 ( Line 33 ) are responsible for gathering information about a network subnet that is to be provisioned and then creating a virtual private cloud (VPC) definition based on those details. Generate suggested code for TASK 1 ( Line 26 ) by placing the cursor at the end of the line, hitting Enter , and then confirming with Tab . Afterwards, perform the same steps to generate code for TASK 2 as well. The first round of WCA-generated code produces a code block [A] with a register: subnet_info ( Line 30 ), the result of which is to assign this VPC definition to a variable subnet_info . Nothing terribly complicated or surprising about that. The WCA-generated code that follows for the task on Line 38 (previously Line 33 in the raw template) recommends a code block [B] with vpc_subnet_id: \"{ { subnet_info.subnets[0].subnet_id } }\" as the value associated with the VPC's subnet ID. Critically, the variable subnet_info that was generated in the previous Task [A] is also referenced in the second Task. This demonstrates the contextual awareness of WCA in action. Change the name of variable subnet_info to subnet_name on Line 30 . Then delete the code block recommendations under Line 38 and regenerate the task. Notice that the new block [A] of TASK 2 now references the variable subnet_name that was modified just a moment ago in TASK 1 WCA has generated code for TASK 2 that takes into account the modified context and variables from TASK 1 of the Playbook Continue experimenting with WCA's contextual awareness and post-processing capabilities. Try adjusting other variables within the Playbook and study how these modifications impact the generation of later blocks of Task code within the Playbook. iv. Next steps In the next section, you will begin experimenting with customized Ansible Playbooks and testing how changes to Ansible Task natural language descriptions impacts the recommended code produced by WCA.","title":"Content Source Matching and Post-Processing"},{"location":"source-matching-post-processing/#_1","text":"Christopher Bienko (Principal, IBM Global Sales Enablement) demonstrates key elements of the Content Source Matching and Post-Processing module. [10 min]","title":""},{"location":"source-matching-post-processing/#i-why-code-attribution-and-post-processing-matters-for-generative-ai","text":"At this point, you should now be well acquainted with Ansible Tasks and Playbooks . Ansible Playbooks are comprised of multiple Tasks \u2014 which you have been using IBM watsonx Code Assistant for Red Hat Ansible Lightspeed's ( WCA ) generative AI capabilities to define and expand upon. If a Playbook contains multiple \"tasks\" that together achieve some singular goal or purpose, you might want to group these tasks into a single reusable unit. For example, a string of \"tasks\" might provision infrastructure for separate development, test, and production clusters from a major cloud vendor. Patterns for automating endpoints from these vendors can be codified into Ansible Modules , which are supported by Red Hat and the Ansible community writ large. And finally, collections of Playbooks and Modules can be organized into \"blueprints\" for automation tasks that are frequently used together to achieve some goal; a goal which you might want to make repeatable or shareable with other teams within your business. These blueprints are referred to as Ansible Roles . They provide a structured way to organize tasks, templates, files, and variables, and are available to drop directly into Ansible Playbooks. Roles make it possible to easily manage and set up complex automation tasks, essentially providing a rubric to streamline automation projects. Collectively, these Playbooks, Modules, and Roles form a comprehensive ecosystem of business and community-driven support for patterning automation to work with the broadest range of vendors, technologies, and clouds. Ansible Galaxy is a Red Hat-curated, community-driven repository for Ansible Roles . Via communities such as Galaxy, thousands of Roles are available for Ansible users to leverage within their own Playbooks. Ansible Galaxy is also key to how generative AI code recommendations from WCA are attributed back to original content sources and authors.","title":"i. Why code attribution and post-processing matters for generative AI"},{"location":"source-matching-post-processing/#ii-content-source-matching","text":"A powerful capability within WCA is Content Source Matching (often referred to as \"code explainability\"), which attempts to match AI-generated code suggestions to the training data and sources that were utilized in generating the suggested Task code. These code attribution suggestions are created using a k-NN ( K-Nearest Neighbors ) algorithm that examines Ansible Galaxy and training data repositories in search of the nearest related content to the AI-generated code suggestions. To enable Content Source Matching capabilities within WCA, navigate to the main menu bar for VS Code and drill down into View > Open View... [A] The console [A] along the top of the VS Code interface is now activated and awaiting a prompt. Type Lightspeed Training Matches and hit Enter to confirm the selection. After making this configuration change to VS Code, all requests for WCA to generate code recommendations will now open a panel at the bottom of the VS Code interface It is under the Ansible tab that content source matching results will be displayed Other tabs will also displayed for variety of options: Output , Debug Console , Terminal , Ports , and Comments Open the install_pgsql-single-task.yml Ansible Playbook, which is included within the ansible-wca-demo-kit directory. The full location of the Playbook, as well as the contents, are available in the code block below. ~/Documents/ansible-wca-demo-kit/install and configure PostgreSQL and PGAdmin container/install_pgsql-single-task.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 --- - name : Configure Database servers hosts : databases become : true tasks : # TASK 1 # - name: Install postgresql-server # TASK 2 # Ansible Lightspeed used an easy-to-understand natural language prompt and suggested the correct, more complex PostgreSQL CLI command to initiate the database. # Ansible Lightspeed used best practices and kept the task idempotent by including creates: /var/lib/pgsql/data/postgresql.conf in the suggestion. # - name: Run postgresql setup command # TASK 3 # Ansible Lightspeed used natural language prompt and added state: started and enabled: true module arguments based on Start and enable... in the Ansible task description. # - name: Start and enable postgresql service Generate a code recommendation for the task on Line 8 by placing your cursor at the end of the line and hitting Enter . Pay attention to the code attribution details associated with this recommendation, which will be appearing under the Ansible tab at the bottom of the VS Code interface once the code recommendation is accepted with Tab You need to accept the AI-generated code suggestions (using the Tab key) before the Ansible content source matching tab will provide details about the code's origins - name : Install postgresql-server ansible.builtin.package : name : postgresql-server state : present The three most likely content sources used in training the WCA model\u2014 which produced the AI-generated code recommendations \u2014are listed within the Ansible tab [A] . Recall from earlier that these code attribution suggestions are created using a k-NN algorithm that searches Ansible Galaxy repositories for the nearest related content to the AI-generated code suggestions. Clicking the arrow icon to the left of each attribution [B] will expand further details about the source. Information about the URL , Path , Data Source , License , and Score are displayed (where available) under each listing. Red Hat-certified and maintained collections, as well as contributors to open source projects on Ansible Galaxy, are the primary sources for Ansible Lightspeed model training and are the content sources you are most likely to see matched to AI-generated code recommendations. Clicking on the URL field will redirect your web browser back to the precise collections and sources on Ansible Galaxy from which the code recommendations were derived. Here you can learn much richer details about the status of the project, any associated open source repositories involved (such as GitHub), contributions and activities ongoing with the code base, the author(s) involved, and many more intricacies.","title":"ii. Content source matching"},{"location":"source-matching-post-processing/#iii-post-processing-capabilities","text":"Another element of code generation that WCA excels at is understanding context within the Playbook it is executing against. If a variable or attribute is defined earlier within that Playbook, it will be recalled and referenced\u2014 where it makes sense to do so \u2014in the generation of subsequent lines of code. You may have already noticed these post-processing capabilities in your experimentations with WCA-generated code suggestions. Post-processing of Task descriptions and YAML file contents helps generate contextually aware, accurate Ansible content suggestions. However, one way to make this feature quite obvious is to take a previously-generated block of Ansible Task code, update the value assigned to a named variable earlier in the Playbook, and then regenerate the Task code block. In theory, the newly-generated Task block will use the updated variable name (and differ from how the code block was originally generated). Using the Explorer tab, create a New File... YAML Playbook named demo_provision_ec2_instance.yml anywhere within the /ansible-wca-demo-kit/ directory. You may save the Playbook in the root of the folder or any of the subdirectories Be sure to reference the expandable tooltips below for instructions on how to perform these operations and how to configure the Playbook language for use with WCA Copy and paste the following code block into the demo_provision_ec2_instance.yml Playbook and then save the file HOW TO CREATE NEW YAML PLAYBOOKS Note: You need to copy and paste the contents of the Playbook into a New File... within the same Lightspeed project directory that was used for the previous lab modules in order for the VS Code extension to engage. To create a new YAML Playbook within a VS Code environment: a. Copy the contents of the Playbook to clipboard using the button in the top-right corner of the lab guide code block. b. Return to your VS Code environment. In the top-left corner of the interface, with your Ansible Lightspeed folder selected, click the New File... button. c. Name the file to a description of your choosing, ending with .yml as the filetype. Set it to CustomPlaybook.yml , for example. Save it to one of the directories in the ansible-wca-demo-kit folder. d. Paste the clipboard contents into the YAML file and follow along with the suggestions below. COPY AND PASTE CODE WITHIN THE VM Information \"copied\" to your local machine's clipboard cannot be \"pasted\" directly into the virtual machine (VM) environment or VS Code. If you wish to copy and paste instructions directly from the lab documentation, it is recommended that you open the GitHub instructions inside the VM's web browser (Firefox). This will allow you to copy instructions to the VM's clipboard and paste instructions inside the VS Code editor. ANSIBLE LIGHTSPEED IS MISSING OR CODE RECOMMENDATIONS ARE NOT GENERATING Ansible Lightspeed and WCA will only generate code recommendations for Ansible Playbooks and YAML files. VS Code will typically auto-detect the programming language of the document you're working with, but on occassion you may need to manually specify the language. Even if working with a YAML file, you'll still need to specify the language mode as Ansible for the Lightspeed plugin to engage. To set the language mode correctly: In the bottom-right corner of the VS Code interface, hover over the Select Language Mode toggle [A] A console will appear at the top of VS Code with a drop-down list of options [B] Click Ansible from the suggested languages, or enter the text yourself and hit Enter Confirm that the Select Language Mode toggle in the bottom-right corner displays Ansible demo_provision_ec2_instance.yml 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 --- - name : EC2 Cloud Operations hosts : localhost connection : local gather_facts : false module_defaults : group/aws : region : us-east-1 # vars: # ec2_instance: # name: lightspeed-instance-01 # key_name: lightspeed-keypair # image_id: ami-016eb5d644c333ccb # RHEL 9 us-east-1 # tags: # function: lightspeed-demo # security_group: secgroup-lightspeed tasks : # TASK 1 # # 1a. Uncomment task description below and generate a task suggestion. # # Note - Best practices: The suggestion used the Fully Qualified Collection name. # # Note - Context: Ansible Lightspeed used the Playbook name \"EC2 Cloud Operations\" to use the correct \"amazon.aws.ec2_vpc_subnet_info\" module. - name : Gather subnet info tag:Name subnet-lightspeed # TASK 2 # # 2a. Uncomment task description below and generate a task suggestion. # # Note - Context: The suggestion included the previous task's registered variable in the suggestion. # # Note - Accuracy: The suggestion provides the correct key value from the previously task's registered variable. - name : Create vpc_subnet_id var # TASK 3 # # 3a. Uncomment task description \"Provision t3.micro instance\" below and generate a task suggestion. # # Note - Efficiency: The suggestion provides practical variable examples to improve efficiency. - name : Provision t3.micro instance # # 3b. Remove the above task and suggestion. # # Uncomment 2nd task description \"Provision t3.micro instance using ec2_instance var\". # # Generate an updated suggestion. # # Note - Context: The updated suggestion includes the \"ec2_instance variable fields in the suggestion\" - name : Provision t3.micro instance using ec2_instance var TASK 1 ( Line 26 ) and TASK 2 ( Line 33 ) are responsible for gathering information about a network subnet that is to be provisioned and then creating a virtual private cloud (VPC) definition based on those details. Generate suggested code for TASK 1 ( Line 26 ) by placing the cursor at the end of the line, hitting Enter , and then confirming with Tab . Afterwards, perform the same steps to generate code for TASK 2 as well. The first round of WCA-generated code produces a code block [A] with a register: subnet_info ( Line 30 ), the result of which is to assign this VPC definition to a variable subnet_info . Nothing terribly complicated or surprising about that. The WCA-generated code that follows for the task on Line 38 (previously Line 33 in the raw template) recommends a code block [B] with vpc_subnet_id: \"{ { subnet_info.subnets[0].subnet_id } }\" as the value associated with the VPC's subnet ID. Critically, the variable subnet_info that was generated in the previous Task [A] is also referenced in the second Task. This demonstrates the contextual awareness of WCA in action. Change the name of variable subnet_info to subnet_name on Line 30 . Then delete the code block recommendations under Line 38 and regenerate the task. Notice that the new block [A] of TASK 2 now references the variable subnet_name that was modified just a moment ago in TASK 1 WCA has generated code for TASK 2 that takes into account the modified context and variables from TASK 1 of the Playbook Continue experimenting with WCA's contextual awareness and post-processing capabilities. Try adjusting other variables within the Playbook and study how these modifications impact the generation of later blocks of Task code within the Playbook.","title":"iii. Post-processing capabilities"},{"location":"source-matching-post-processing/#iv-next-steps","text":"In the next section, you will begin experimenting with customized Ansible Playbooks and testing how changes to Ansible Task natural language descriptions impacts the recommended code produced by WCA.","title":"iv. Next steps"}]}
\ No newline at end of file
diff --git a/sitemap.xml.gz b/sitemap.xml.gz
index e559283..e2e9426 100644
Binary files a/sitemap.xml.gz and b/sitemap.xml.gz differ
diff --git a/source-matching-post-processing/index.html b/source-matching-post-processing/index.html
index 2971e8a..d906081 100644
--- a/source-matching-post-processing/index.html
+++ b/source-matching-post-processing/index.html
@@ -451,28 +451,47 @@ <h2 id="ii-content-source-matching"><strong>ii. Content source matching</strong>
 </ol>
 <hr />
 <ol start="3">
-<li>Open the <code>install_pgsql-single-task.yml</code> Ansible Playbook, which is included within the <code>ansible-wca-demo-kit</code> directory. The full location of the Playbook, as well as the contents, are available in the code block below.<div class="highlight"><pre><span></span><code>=== &quot;~/Documents/ansible-wca-demo-kit/install and configure PostgreSQL and PGAdmin container/install_pgsql-single-task.yml&quot;
-
-``` yaml linenums=&quot;1&quot;
----
-- name: Configure Database servers
-hosts: databases
-become: true
-
-tasks:
-    # TASK 1
-    # - name: Install postgresql-server
-
-    # TASK 2
-    # Ansible Lightspeed used an easy-to-understand natural language prompt and suggested the correct, more complex PostgreSQL CLI command to initiate the database.
-    # Ansible Lightspeed used best practices and kept the task idempotent by including creates: /var/lib/pgsql/data/postgresql.conf in the suggestion.
-    # - name: Run postgresql setup command
-
-    # TASK 3
-    # Ansible Lightspeed used natural language prompt and added state: started and enabled: true module arguments based on Start and enable... in the Ansible task description.
-    # - name: Start and enable postgresql service
-```
-</code></pre></div>
+<li>
+<p>Open the <code>install_pgsql-single-task.yml</code> Ansible Playbook, which is included within the <code>ansible-wca-demo-kit</code> directory. The full location of the Playbook, as well as the contents, are available in the code block below.</p>
+<div class="tabbed-set tabbed-alternate" data-tabs="1:1"><input checked="checked" id="__tabbed_1_1" name="__tabbed_1" type="radio" /><div class="tabbed-labels"><label for="__tabbed_1_1">~/Documents/ansible-wca-demo-kit/install and configure PostgreSQL and PGAdmin container/install_pgsql-single-task.yml</label></div>
+<div class="tabbed-content">
+<div class="tabbed-block"></div>
+</div>
+</div>
+<div class="highlight"><table class="highlighttable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span><span class="normal"><a href="#__codelineno-0-1"> 1</a></span>
+<span class="normal"><a href="#__codelineno-0-2"> 2</a></span>
+<span class="normal"><a href="#__codelineno-0-3"> 3</a></span>
+<span class="normal"><a href="#__codelineno-0-4"> 4</a></span>
+<span class="normal"><a href="#__codelineno-0-5"> 5</a></span>
+<span class="normal"><a href="#__codelineno-0-6"> 6</a></span>
+<span class="normal"><a href="#__codelineno-0-7"> 7</a></span>
+<span class="normal"><a href="#__codelineno-0-8"> 8</a></span>
+<span class="normal"><a href="#__codelineno-0-9"> 9</a></span>
+<span class="normal"><a href="#__codelineno-0-10">10</a></span>
+<span class="normal"><a href="#__codelineno-0-11">11</a></span>
+<span class="normal"><a href="#__codelineno-0-12">12</a></span>
+<span class="normal"><a href="#__codelineno-0-13">13</a></span>
+<span class="normal"><a href="#__codelineno-0-14">14</a></span>
+<span class="normal"><a href="#__codelineno-0-15">15</a></span>
+<span class="normal"><a href="#__codelineno-0-16">16</a></span>
+<span class="normal"><a href="#__codelineno-0-17">17</a></span></pre></div></td><td class="code"><div><pre><span></span><code><a id="__codelineno-0-1" name="__codelineno-0-1"></a><span class="nn">---</span><span class="w"></span>
+<a id="__codelineno-0-2" name="__codelineno-0-2"></a><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">Configure Database servers</span><span class="w"></span>
+<a id="__codelineno-0-3" name="__codelineno-0-3"></a><span class="nt">hosts</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">databases</span><span class="w"></span>
+<a id="__codelineno-0-4" name="__codelineno-0-4"></a><span class="nt">become</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">true</span><span class="w"></span>
+<a id="__codelineno-0-5" name="__codelineno-0-5"></a>
+<a id="__codelineno-0-6" name="__codelineno-0-6"></a><span class="nt">tasks</span><span class="p">:</span><span class="w"></span>
+<a id="__codelineno-0-7" name="__codelineno-0-7"></a><span class="w">    </span><span class="c1"># TASK 1</span><span class="w"></span>
+<a id="__codelineno-0-8" name="__codelineno-0-8"></a><span class="w">    </span><span class="c1"># - name: Install postgresql-server</span><span class="w"></span>
+<a id="__codelineno-0-9" name="__codelineno-0-9"></a>
+<a id="__codelineno-0-10" name="__codelineno-0-10"></a><span class="w">    </span><span class="c1"># TASK 2</span><span class="w"></span>
+<a id="__codelineno-0-11" name="__codelineno-0-11"></a><span class="w">    </span><span class="c1"># Ansible Lightspeed used an easy-to-understand natural language prompt and suggested the correct, more complex PostgreSQL CLI command to initiate the database.</span><span class="w"></span>
+<a id="__codelineno-0-12" name="__codelineno-0-12"></a><span class="w">    </span><span class="c1"># Ansible Lightspeed used best practices and kept the task idempotent by including creates: /var/lib/pgsql/data/postgresql.conf in the suggestion.</span><span class="w"></span>
+<a id="__codelineno-0-13" name="__codelineno-0-13"></a><span class="w">    </span><span class="c1"># - name: Run postgresql setup command</span><span class="w"></span>
+<a id="__codelineno-0-14" name="__codelineno-0-14"></a>
+<a id="__codelineno-0-15" name="__codelineno-0-15"></a><span class="w">    </span><span class="c1"># TASK 3</span><span class="w"></span>
+<a id="__codelineno-0-16" name="__codelineno-0-16"></a><span class="w">    </span><span class="c1"># Ansible Lightspeed used natural language prompt and added state: started and enabled: true module arguments based on Start and enable... in the Ansible task description.</span><span class="w"></span>
+<a id="__codelineno-0-17" name="__codelineno-0-17"></a><span class="w">    </span><span class="c1"># - name: Start and enable postgresql service</span><span class="w"></span>
+</code></pre></div></td></tr></table></div>
 </li>
 </ol>
 <hr />
@@ -498,10 +517,9 @@ <h2 id="ii-content-source-matching"><strong>ii. Content source matching</strong>
 <ol start="5">
 <li>
 <p>The three most likely content sources used in training the WCA model— which produced the AI-generated code recommendations —are listed within the <code>Ansible</code> <strong>tab</strong><sup>[A]</sup>. Recall from earlier that these code attribution suggestions are created using a <strong>k-NN</strong>  algorithm that searches <strong>Ansible Galaxy</strong> repositories for the nearest related content to the AI-generated code suggestions.</p>
+<p>Clicking the arrow icon to the left of each <strong>attribution</strong><sup>[B]</sup> will expand further details about the source. Information about the <code>URL</code>, <code>Path</code>, <code>Data Source</code>, <code>License</code>, and <code>Score</code> are displayed (where available) under each listing. Red Hat-certified and maintained collections, as well as contributors to open source projects on Ansible Galaxy, are the primary sources for Ansible Lightspeed model training and are the content sources you are most likely to see matched to AI-generated code recommendations.</p>
 <p><br/>
 <a class="glightbox" href="../_attachments/contentmatching5.png" data-type="image" data-width="auto" data-height="auto" data-desc-position="bottom"><img alt="" loading="lazy" src="../_attachments/contentmatching5.png" width="600" /></a></p>
-<p><br/>
-Clicking the arrow icon to the left of each <strong>attribution</strong><sup>[B]</sup> will expand further details about the source. Information about the <code>URL</code>, <code>Path</code>, <code>Data Source</code>, <code>License</code>, and <code>Score</code> are displayed (where available) under each listing. Red Hat-certified and maintained collections, as well as contributors to open source projects on Ansible Galaxy, are the primary sources for Ansible Lightspeed model training and are the content sources you are most likely to see matched to AI-generated code recommendations.</p>
 </li>
 </ol>
 <hr />
@@ -566,7 +584,7 @@ <h2 id="iii-post-processing-capabilities"><strong>iii. Post-processing capabilit
 <a class="glightbox" href="../_attachments/troubleshooting-2.png" data-type="image" data-width="auto" data-height="auto" data-desc-position="bottom"><img alt="" loading="lazy" src="../_attachments/troubleshooting-2.png" width="600" /></a></p>
 <p><a class="glightbox" href="../_attachments/troubleshooting-3.png" data-type="image" data-width="auto" data-height="auto" data-desc-position="bottom"><img alt="" loading="lazy" src="../_attachments/troubleshooting-3.png" width="600" /></a></p>
 </details>
-<div class="tabbed-set tabbed-alternate" data-tabs="1:1"><input checked="checked" id="__tabbed_1_1" name="__tabbed_1" type="radio" /><div class="tabbed-labels"><label for="__tabbed_1_1">demo_provision_ec2_instance.yml</label></div>
+<div class="tabbed-set tabbed-alternate" data-tabs="2:1"><input checked="checked" id="__tabbed_2_1" name="__tabbed_2" type="radio" /><div class="tabbed-labels"><label for="__tabbed_2_1">demo_provision_ec2_instance.yml</label></div>
 <div class="tabbed-content">
 <div class="tabbed-block"></div>
 </div>
@@ -618,11 +636,11 @@ <h2 id="iii-post-processing-capabilities"><strong>iii. Post-processing capabilit
 <span class="normal"><a href="#__codelineno-2-45">45</a></span>
 <span class="normal"><a href="#__codelineno-2-46">46</a></span></pre></div></td><td class="code"><div><pre><span></span><code><a id="__codelineno-2-1" name="__codelineno-2-1"></a><span class="nn">---</span><span class="w"></span>
 <a id="__codelineno-2-2" name="__codelineno-2-2"></a><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">name</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">EC2 Cloud Operations</span><span class="w"></span>
-<a id="__codelineno-2-3" name="__codelineno-2-3"></a><span class="nt">hosts</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">localhost</span><span class="w"></span>
-<a id="__codelineno-2-4" name="__codelineno-2-4"></a><span class="nt">connection</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">local</span><span class="w"></span>
-<a id="__codelineno-2-5" name="__codelineno-2-5"></a><span class="nt">gather_facts</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">false</span><span class="w"></span>
+<a id="__codelineno-2-3" name="__codelineno-2-3"></a><span class="w">  </span><span class="nt">hosts</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">localhost</span><span class="w"></span>
+<a id="__codelineno-2-4" name="__codelineno-2-4"></a><span class="w">  </span><span class="nt">connection</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">local</span><span class="w"></span>
+<a id="__codelineno-2-5" name="__codelineno-2-5"></a><span class="w">  </span><span class="nt">gather_facts</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">false</span><span class="w"></span>
 <a id="__codelineno-2-6" name="__codelineno-2-6"></a>
-<a id="__codelineno-2-7" name="__codelineno-2-7"></a><span class="nt">module_defaults</span><span class="p">:</span><span class="w"></span>
+<a id="__codelineno-2-7" name="__codelineno-2-7"></a><span class="w">  </span><span class="nt">module_defaults</span><span class="p">:</span><span class="w"></span>
 <a id="__codelineno-2-8" name="__codelineno-2-8"></a><span class="w">    </span><span class="nt">group/aws</span><span class="p">:</span><span class="w"></span>
 <a id="__codelineno-2-9" name="__codelineno-2-9"></a><span class="w">    </span><span class="nt">region</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">us-east-1</span><span class="w"></span>
 <a id="__codelineno-2-10" name="__codelineno-2-10"></a>
@@ -635,7 +653,7 @@ <h2 id="iii-post-processing-capabilities"><strong>iii. Post-processing capabilit
 <a id="__codelineno-2-17" name="__codelineno-2-17"></a><span class="c1">#       function: lightspeed-demo</span><span class="w"></span>
 <a id="__codelineno-2-18" name="__codelineno-2-18"></a><span class="c1">#     security_group: secgroup-lightspeed</span><span class="w"></span>
 <a id="__codelineno-2-19" name="__codelineno-2-19"></a>
-<a id="__codelineno-2-20" name="__codelineno-2-20"></a><span class="nt">tasks</span><span class="p">:</span><span class="w"></span>
+<a id="__codelineno-2-20" name="__codelineno-2-20"></a><span class="w">  </span><span class="nt">tasks</span><span class="p">:</span><span class="w"></span>
 <a id="__codelineno-2-21" name="__codelineno-2-21"></a><span class="w">        </span><span class="c1"># TASK 1</span><span class="w"></span>
 <a id="__codelineno-2-22" name="__codelineno-2-22"></a><span class="w">        </span><span class="c1"># # 1a. Uncomment task description below and generate a task suggestion.</span><span class="w"></span>
 <a id="__codelineno-2-23" name="__codelineno-2-23"></a><span class="w">        </span><span class="c1"># #     Note - Best practices: The suggestion used the Fully Qualified Collection name.</span><span class="w"></span>