Skip to content

Latest commit

 

History

History
69 lines (41 loc) · 3.83 KB

README.md

File metadata and controls

69 lines (41 loc) · 3.83 KB

microservices-synchronizer-config

Configuration repository for microservices Synchronizer. Synchronizer is a service to deal with synchronization issues in microservices-based CD pipelines backed by a service registry component.

This repository keeps Jenkins pipeline examples and YAML files for configuring a simple continuous deployment pipeline. It also keeps links to support repositories.

Synchronizer could be also implemented with other service registries and could also have other adapter implementations beside a Jenkins Shared library.

In future, we are also thinking of implementing Synchronizer by exploiting the Sidecar design pattern.

jenkins-synchronizer-sharedlibrary

Jenkins-etcdwatcher-sharedlibrary https://github.com/woland7/jenkins-synchronizer-sharedlibrary keeps the code for the implementation of the adatper as a Jenkins shared library.

Synchronizer

Synchronizer https://github.com/woland7/synchronizer keeps the code for the implementation of the synchronizer for etcd for the OpenShift platform, which has been called etcdwatcher.

jmeter4docker

Jmeter4docker https://github.com/woland7/jmeter4docker keeps the code for a JMeter container image along with scripts to test Synchronizer.

TestSynchronizer

TestSynchronizer https://github.com/woland7/TestSynchronizer is a web application to set up a case study in order to evaluate if Synchronizer works as expected for a continous deployment pipeline use case. It could be changed as one sees fit.

Microservices Demo

Microservices Demo https://github.com/microservices-demo is a benchmark microservices application; we have used it to test the application bootstrapping use case. You should edit the catalogue microservice to disable the circuit-breaker, which they have implemented.

How To-Continuos Deployment Pipeline

This section briefly explains how to set up a CD pipeline on OpenShift/Kubernetes.

What you need

An active OpenShift cluster

Steps

  • Spin up a Jenkins microservice from the OpenShift catalogue.

  • Spin up a Wildfly microservice from the OpenShift catalogue and use TestEtcdWatcher repositor to create the application

  • Use the configurations file provided in this repository to create the imagestream, the buildconfig and the deployconfig to create a JMeter application which will be used to perform the tests.

    • As there is no JMeter default template in the OpenShift catalogue, one should create the application manually by providing the configuration files.
    • Instead, when there are such templates, one could simply use the OpenShift catalogue and then all these operations would be performed by the platform under the hood.

  • Once the previous steps have been accomplished one should create the Jenkins pipeline. A Jenkins pipeline is also provided in this repository.

    • Of course, IPs should be edited accordingly.

  • To use Synchronizer, one should set the shared library in Jenkins. Then, as it was done for testing purposes, etcdwatcher could be preloaded on the platform and just downloaded on the pod from there or could instead be downloaded from the repository etcdwatcher-Openshift.

How To-Application Bootstrapping Use case

This section briefly explains how to set up an application bootstrapping use case.

What you need

An active OpenShift cluster

Steps

  • Retrieve the YAML file, containing the deployment of the application from Microservices Demo.

Now, there are two possibilities:

  • Modify the file by adding an init container to each microservices, whose startup wanna be delayed.

    • Multiple dependencies can be specified as a coma separated list

  • Modifiy the file by adding an init container to each microservice according to the dependencies relationships described in Microservices Demo; in this way the startup order is correctly constrained

The same workflow can be applied to any microservices application.