Skip to content

Latest commit

 

History

History
85 lines (69 loc) · 4.52 KB

README.md

File metadata and controls

85 lines (69 loc) · 4.52 KB
Raw

GitOps

Prerequisite

We need a Ubuntu VM, which is configured as a self-hosted runner. This VM, should have network connectivity to the Prism-Central and the clusters in consideration. Click here to read how to configure self-hosted runner to your repository.

Required parameters

Read about the required parameters to run the framework here.

Github pipelines

We can use Github workflows/ pipelines to manage GitOps i.e using this repository to manage Infrastructure as a Code (IaaC), that gets triggered from Github.

A Github workflow is a configurable automated process that will run one or more jobs. Workflows are defined by a YAML file checked in to your repository and will run when triggered by an event in your repository, or they can be triggered manually, or at a defined schedule.

How to auto-trigger pre-configured workflows

There are several pre-configured Github workflows/ pipelines. All the below mentioned workflows are triggered by the approval of a PR containing the configuration file changes. Here is the step-by-step guide.

That's it. The corresponding workflow/s kick in. Below is the details of workflows that are triggered by the configuration files changes.

  • Imaging workflow - This Github workflow gets triggered when the file config/new-site.json is part of the PR. This Github workflow will setup the necessary Python environment and calls the framework as below 
    > cd framework && python main.py --workflow imaging -f config/new-site.json
  • Pod config workflow - This Github workflow gets triggered when the file config/pod-config.yml is part of the PR. This Github workflow will setup the necessary Python environment and calls the framework as below 
    > cd framework && python main.py --workflow pod-config -f config/pod-config.yml
  • VM workloads workflow - This Github workflow gets triggered when the file config/create_workloads.json is part of the PR. This Github workflow will setup the necessary Python environment and calls the framework as below 
    > cd framework && python main.py --workflow calm-vm-workloads -f config/create-vm-workloads.json
  • Edge AI workload workflow - This Github workflow gets triggered when the file config/edge-ai.json is part of the PR. This Github workflow will setup the necessary Python environment and calls the framework as below 
    > cd framework && python main.py --workflow calm-edgeai-vm-workload -f config/edge-ai.json

Note: The PR can contain configuration changes in multiple files config/new-site.json , config/pod-config.yml, config/create_workloads.json , config/edge-ai.json. In this scenario, the workflows are triggered one after the other in the below order. Imaging workflow -> Pod config workflow -> VM workloads workflow -> Edge AI workload workflow.

Snapshots of auto-trigger

  • Modify any of the files configured for auto-trigger (config/new-site.json/ config/pod-config.yml/ config/create_workloads.json) and commit to a new branch.
  • Create a PR.
  • Get approval for the PR. This triggers the workflow.
  • Then monitor the workflow from Actions tab under Trigger Python Workflows section, click on latest workflow run and expand Run python script for streaming logs.
  • The logs and input config file, used for the run will be pushed back to the branch that triggered the run.

Example of triggering workflows manually from the Github UI

  • This workflow can be manually triggered from the Actions Tab in Github.