Publish API-only Docker image workflow (aigc-apps/PAI-RAG)

The Publish API-only Docker image workflow from aigc-apps/PAI-RAG, explained and optimized by Latchkey.

CI health: C - fair

Point runs-on at Latchkey and get run de-duplication, job timeouts, SHA-pinned actions, self-healing for flaky steps, and up to 58% lower cost, applied automatically.

Grade your own workflow free or run it on Latchkey →

Source: aigc-apps/PAI-RAG.github/workflows/docker-api.ymlLicense MITView source

What it does

This is the Publish API-only Docker image workflow from the aigc-apps/PAI-RAG repository, a real project running GitHub Actions. It is shown here with attribution under its MIT license.

Below, Latchkey shows a faster, safer version produced by its optimization engine.

The workflow

workflow (.yml)

name: Publish API-only Docker image

# Configures this workflow to run on manual dispatch or push to specific branches
on:
  workflow_dispatch:
    inputs:
      push_to_prod:
        description: 'Push to production registry'
        required: false
        default: false
        type: boolean
  push:
    branches: ["feature"]
    paths:
      - 'backend/**'
      - 'integrations/**'
      - 'resources/**'
      - 'alembic/**'
      - 'alembic.ini'
      - 'scripts/**'
      - 'Dockerfile.api'
      - '**/pyproject.toml'
      - '**/poetry.lock'
      - '.github/workflows/docker-api.yml'

# Environment variables for Container registry
env:
  API_REGISTRY: mybigpai-public-registry.cn-beijing.cr.aliyuncs.com

jobs:
  build-and-push-api-image:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout repository
        uses: actions/checkout@v4

      - name: Check disk space
        run: df . -h

      - name: Free disk space
        run: |
          sudo docker rmi $(docker image ls -aq) >/dev/null 2>&1 || true
          sudo rm -rf \
            /usr/share/dotnet /usr/local/lib/android /opt/ghc \
            /usr/local/share/powershell /usr/share/swift /usr/local/.ghcup \
            /usr/lib/jvm || true
          df . -h

      - name: Extract version
        run: |
          pip install poetry
          CURRENT_TIME="`date +%Y%m%d%H%M%S`"
          echo "API_VERSION_TAG=$CURRENT_TIME" >> $GITHUB_ENV

      - name: Login to ACR Hangzhou region
        uses: docker/login-action@v1
        with:
          registry: ${{ env.API_REGISTRY }}
          username: ${{ secrets.ACR_USER }}
          password: ${{ secrets.ACR_PUBLIC_PASSWORD }}

      - name: Build and push API-only image
        env:
          API_IMAGE_TAG: ${{ env.API_VERSION_TAG }}
        run: |
          # Build using Dockerfile.api
          docker build -f Dockerfile.api -t ${{ env.API_REGISTRY }}/mybigpai/pairag-api:$API_IMAGE_TAG .
          
          # Always push to test registry
          docker push ${{ env.API_REGISTRY }}/mybigpai/pairag-api:$API_IMAGE_TAG
          
      - name: Free disk space
        run: |
          docker image prune --filter "until=168h" --force -a
          docker builder prune --filter "until=24h" -f

The same workflow, on Latchkey

Removes redundant runs and caps runaway jobs. Added and changed lines are highlighted.

name: Publish API-only Docker image
 
# Configures this workflow to run on manual dispatch or push to specific branches
on:
  workflow_dispatch:
    inputs:
      push_to_prod:
        description: 'Push to production registry'
        required: false
        default: false
        type: boolean
  push:
    branches: ["feature"]
    paths:
      - 'backend/**'
      - 'integrations/**'
      - 'resources/**'
      - 'alembic/**'
      - 'alembic.ini'
      - 'scripts/**'
      - 'Dockerfile.api'
      - '**/pyproject.toml'
      - '**/poetry.lock'
      - '.github/workflows/docker-api.yml'
 
# Environment variables for Container registry
env:
  API_REGISTRY: mybigpai-public-registry.cn-beijing.cr.aliyuncs.com
 
concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true
 
jobs:
  build-and-push-api-image:
    timeout-minutes: 30
    runs-on: latchkey-small
    steps:
      - name: Checkout repository
        uses: actions/checkout@v4
 
      - name: Check disk space
        run: df . -h
 
      - name: Free disk space
        run: |
          sudo docker rmi $(docker image ls -aq) >/dev/null 2>&1 || true
          sudo rm -rf \
            /usr/share/dotnet /usr/local/lib/android /opt/ghc \
            /usr/local/share/powershell /usr/share/swift /usr/local/.ghcup \
            /usr/lib/jvm || true
          df . -h
 
      - name: Extract version
        run: |
          pip install poetry
          CURRENT_TIME="`date +%Y%m%d%H%M%S`"
          echo "API_VERSION_TAG=$CURRENT_TIME" >> $GITHUB_ENV
 
      - name: Login to ACR Hangzhou region
        uses: docker/login-action@v1
        with:
          registry: ${{ env.API_REGISTRY }}
          username: ${{ secrets.ACR_USER }}
          password: ${{ secrets.ACR_PUBLIC_PASSWORD }}
 
      - name: Build and push API-only image
        env:
          API_IMAGE_TAG: ${{ env.API_VERSION_TAG }}
        run: |
          # Build using Dockerfile.api
          docker build -f Dockerfile.api -t ${{ env.API_REGISTRY }}/mybigpai/pairag-api:$API_IMAGE_TAG .
          
          # Always push to test registry
          docker push ${{ env.API_REGISTRY }}/mybigpai/pairag-api:$API_IMAGE_TAG
          
      - name: Free disk space
        run: |
          docker image prune --filter "until=168h" --force -a
          docker builder prune --filter "until=24h" -f

What changed

Run on Latchkey managed runners with one line (runs-on), which apply the fixes below automatically and self-heal transient failures. This example uses latchkey-small; pick the runner size that fits the job.
Cancel superseded runs when a branch or PR gets a newer push.
Add a job timeout so a hung step cannot burn hours of runner time.

1 third-party action is referenced by a movable tag. Pin it to the commit SHA (Latchkey resolves and applies this automatically) so a repointed tag cannot change what runs.

What Latchkey heals here

This workflow has steps that commonly fail on transient issues (network, registries, flaky browsers). On Latchkey managed runners they are detected, retried, and self-healed instead of failing your build:

Dependency installs
Container pulls and builds

This workflow runs 1 job per trigger. On Latchkey the same minutes cost up to 58% less than GitHub-hosted, with zero queue time.

Actions used in this workflow

actions/checkout docker/login-action