Skip to content

Add GitHub workflow to deploy InferenceSystem configmaps on push#434

Merged
dthaler merged 3 commits intomainfrom
copilot/create-github-publish-workflow
Mar 10, 2026
Merged

Add GitHub workflow to deploy InferenceSystem configmaps on push#434
dthaler merged 3 commits intomainfrom
copilot/create-github-publish-workflow

Conversation

Copy link
Copy Markdown
Contributor

Copilot AI commented Mar 9, 2026

Merging a PR that modifies InferenceSystem/deploy/*-configmap.yaml files had no automated deployment path — changes required manual kubectl apply runs.

Changes

  • New workflow .github/workflows/InferenceSystem-deploy-configmaps.yaml:
    • Triggers on push to main when any InferenceSystem/deploy/*-configmap.yaml file changes
    • Supports workflow_dispatch for manual runs (applies all configmaps)
    • Uses fetch-depth: 2 to correctly diff HEAD~1 on push events
    • Authenticates to the cluster via KUBE_CONFIG secret using azure/k8s-set-context
    • Applies only the changed files on push; all configmap files on manual dispatch
    • After applying each configmap, restarts the inference-system deployment in the matching namespace so pods pick up the updated configuration (namespace is derived from the filename, e.g. bush-point-configmap.yaml → namespace bush-point)
- name: Apply configmaps and restart deployments
  run: |
    for FILE in $CHANGED; do
      echo "Applying $FILE"
      kubectl apply -f "$FILE"
      NAMESPACE=$(basename "$FILE" -configmap.yaml)
      kubectl rollout restart deployment/inference-system -n "$NAMESPACE"
    done

Required secret

KUBE_CONFIG must be set in repository secrets with a kubeconfig granting access to the target cluster.

Original prompt

This section details on the original issue you should resolve

<issue_title>Create a github publish workflow for publishing config maps</issue_title>
<issue_description>When any ./InferenceSystem/deploy/*-configmap.yaml file is changed, then on push (i.e., when a PR is merged), a github workflow should deploy it to Azure by doing kubectl apply -f deploy/_filename_.yaml

For example, when #432 is merged, this should happen for the relevant files changed.

Create a new github workflow to do this, e.g., with filename .github/workflows/InferenceSystem-deploy-configmaps.yaml</issue_description>

Comments on the Issue (you are @copilot in this section)


✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

Co-authored-by: dthaler <6547784+dthaler@users.noreply.github.com>
Copilot AI changed the title [WIP] Create GitHub publish workflow for config maps Add GitHub workflow to deploy InferenceSystem configmaps on push Mar 9, 2026
@dthaler dthaler marked this pull request as ready for review March 9, 2026 17:02
@dthaler dthaler requested review from micya and scottveirs as code owners March 9, 2026 17:02
@dthaler dthaler requested a review from akashmjn March 9, 2026 17:02
@dthaler dthaler added the inference system Code to perform inference with the trained model(s) label Mar 10, 2026
Copy link
Copy Markdown
Collaborator

@akashmjn akashmjn left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approving assuming pod restart is handled

Comment thread .github/workflows/InferenceSystem-deploy-configmaps.yaml
Co-authored-by: dthaler <6547784+dthaler@users.noreply.github.com>
@dthaler dthaler merged commit bd3568d into main Mar 10, 2026
25 checks passed
@dthaler dthaler deleted the copilot/create-github-publish-workflow branch March 10, 2026 20:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

inference system Code to perform inference with the trained model(s)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Create a github publish workflow for publishing config maps

3 participants