Skip to content

Latest commit

 

History

History
154 lines (107 loc) · 6.36 KB

README.md

File metadata and controls

154 lines (107 loc) · 6.36 KB

ComfyUI Helper Scripts

This directory contains helper scripts to simplify the deployment and management of ComfyStream:

  • Ansible Playbook (ansible/plays/setup_comfystream.yml) – Deploys ComfyStream on any cloud provider.
  • spinup_comfystream_tensordock.py – Fully automates VM creation and ComfyStream setup on a TensorDock server.
  • monitor_pid_resources.py: Monitors and profiles the resource usage of a running ComfyStream server.

Usage Instructions

Ansible Playbook (Cloud-Agnostic Deployment)

This repository includes an Ansible playbook to deploy ComfyStream on any cloud provider. Follow the steps below:

  1. Create a VM:
    Deploy a VM on a cloud provider such as TensorDock, AWS, Google Cloud, or Azure with the following minimum specifications:

    • GPU: 20GB VRAM
    • RAM: 16GB
    • CPU: 4 vCPUs
    • Storage: 100GB

    [!TIP] You can use the spinup_comfystream_tensordock.py --bare-vm script to create a compatible VM on TensorDock

  2. Open Required Ports:
    Ensure the following ports are open inbound and outbound on the VM's firewall/security group:

    • SSH (Port 22) – Remote access
    • HTTPS (Port 8189) – ComfyStream access

  3. Install Ansible:
    Follow the official Ansible installation guide.

  4. Configure the Inventory File: Add the VM’s public IP to ansible/inventory.yml.

  5. Change the ComfyUI Password:
    Open the ansible/plays/setup_comfystream.yaml file and replace the comfyui_password value with your own secure password.

  6. Run the Playbook:
    Execute:

    ansible-playbook -i ansible/inventory.yaml ansible/plays/setup_comfystream.yaml

    [!IMPORTANT]
    When using a non-sudo user, add --ask-become-pass to provide the sudo password or use an Ansible vault for secure storage.

  7. Access the Server:
    After the playbook completes, ComfyStream will start, and you can access ComfyUI at https://<VM_IP>:<PORT_FOR_8189>. Credentials are shown in the output and regenerated each time. To persist the password, set the comfyui_password variable when running the playbook:

    ansible-playbook -i ansible/inventory.yaml ansible/plays/setup_comfystream.yaml -e "comfyui_password=YourSecurePasswordHere"

Important

If you encounter a toomanyrequests error while pulling the Docker image, either wait a few minutes or provide your Docker credentials when running the playbook:

ansible-playbook -i ansible/inventory.yaml ansible/plays/setup_comfystream.yaml -e "docker_hub_username=your_dockerhub_username docker_hub_password=your_dockerhub_pat"

Tip

The ComfyStream Docker image is ~20GB. To check download progress, SSH into the VM and run:

docker pull livepeer/comfystream:latest

TensorDock Spinup Script (Fully Automated)

The spinup_comfystream_tensordock.py script automates VM provisioning, setup, and server launch on TensorDock. Follow the steps below:

  1. Create a TensorDock Account: Sign up at Tensordock, add a payment method, and generate API credentials.

  2. Set Up a Python Virtual Environment: To prevent dependency conflicts, create and activate a virtual environment with Conda and install the required dependencies:

    conda create -n comfystream python=3.8
conda activate comfystream
pip install -r requirements.txt

  3. View Available Script Options (Optional):
    To see all available options, run:

    python spinup_comfystream_tensordock.py --help

  4. Run the Script:
    Execute the following command to provision a VM and set up ComfyStream automatically:

    python spinup_comfystream_tensordock.py --api-key <API_KEY> --api-token <API_TOKEN>

  5. Access the Server:
    Once the setup is complete, the script will display the URLs to access ComfyStream.

  6. Stop & Delete the VM (When No Longer Needed): To stop and remove the instance, run:

    python spinup_comfystream_tensordock.py --delete <VM_ID>

    Replace <VM_ID> with the VM ID found in the script logs or the TensorDock dashboard.

Warning

If you encounter max retries exceeded with URL errors, the VM might have been created but is inaccessible.
Check the TensorDock dashboard, delete the VM manually, wait 2-3 minutes, then rerun the script.

Profiling a Running ComfyStream Server

To monitor the resource consumption of a running ComfyStream server, use the monitor_pid_resources.py script:

  1. Start the ComfyStream server and execute a streaming workflow.

  2. Run the profiling script:

    python monitor_pid_resources.py --name app.py

    The script will automatically try to find the process ID (PID) of the server. If you prefer to specify the PID manually, you can retrieve it using:

    pgrep -f app.py | xargs ps -o pid,cmd --pid

    Then run the profiling script with the retrieved PID:

    python monitor_pid_resources.py --pid <PID>

  3. Running Inside a Container: If you are running the script inside a container, use the --host-pid option to provide the host PID for accurate GPU monitoring:

    python monitor_pid_resources.py --name app.py --host-pid <HOST_PID>

    Find <HOST_PID> with nvidia-smi on the host.

The script will continuously track CPU and memory usage at specified intervals. If the --spy flag is used, it will also generate a detailed Py-Spy profiler report for deeper performance insights.

Additional Options

For a complete list of available options, run:

python monitor_pid_resources.py --help