test-local-dev.yml

name: Test Local Development Environment

on:
  pull_request:

env:
  # Pin versions for reliable caching
  KUBECTL_VERSION: "v1.31.4"
  KIND_VERSION: "v0.31.0"

jobs:
  # Fast job that doesn't need the cluster - runs in parallel
  validate-manifests:
    runs-on: ubuntu-latest
    timeout-minutes: 5
    steps:
    - name: Checkout code
      uses: actions/checkout@v6

    - name: Validate production manifest safety
      run: |
        echo "Validating production manifests do NOT contain dev mode variables..."

        # Recursively check base and production manifests for DISABLE_AUTH
        while IFS= read -r manifest; do
          if grep -q "DISABLE_AUTH" "$manifest"; then
            echo "CRITICAL: Production manifest contains DISABLE_AUTH: $manifest"
            exit 1
          fi

          if grep -qE "ENVIRONMENT.*[\"']?(local|development)[\"']?" "$manifest"; then
            echo "CRITICAL: Production manifest sets ENVIRONMENT=local/development: $manifest"
            exit 1
          fi
        done < <(find components/manifests/base components/manifests/overlays/production -name "*.yaml" -o -name "*.yml" 2>/dev/null)

        echo "All production manifests are safe"

    - name: Validate Makefile
      run: |
        echo "Validating Makefile quality..."
        make validate-makefile

  test-local-dev-simulation:
    runs-on: ubuntu-latest
    timeout-minutes: 30

    steps:
    - name: Checkout code
      uses: actions/checkout@v6

    - name: Cleanup Diskspace
      id: cleanup
      uses: kubeflow/pipelines/.github/actions/github-disk-cleanup@master
      if: (!cancelled())

    - name: Set up Go
      uses: actions/setup-go@v6
      with:
        go-version-file: 'components/backend/go.mod'
        cache-dependency-path: 'components/backend/go.sum'

    - name: Cache kubectl and kind
      uses: actions/cache@v4
      id: k8s-tools-cache
      with:
        path: |
          ~/k8s-tools/kubectl
          ~/k8s-tools/kind
        key: k8s-tools-${{ runner.os }}-kubectl-${{ env.KUBECTL_VERSION }}-kind-${{ env.KIND_VERSION }}

    - name: Install kind and kubectl
      run: |
        mkdir -p ~/k8s-tools

        # Install kubectl if not cached
        if [[ ! -f ~/k8s-tools/kubectl ]]; then
          echo "Downloading kubectl $KUBECTL_VERSION..."
          curl -sLO "https://dl.k8s.io/release/${KUBECTL_VERSION}/bin/linux/amd64/kubectl"
          chmod +x kubectl
          mv kubectl ~/k8s-tools/
        else
          echo "Using cached kubectl"
        fi
        sudo cp ~/k8s-tools/kubectl /usr/local/bin/kubectl

        # Install kind if not cached
        if [[ ! -f ~/k8s-tools/kind ]]; then
          echo "Downloading kind $KIND_VERSION..."
          curl -sLO "https://kind.sigs.k8s.io/dl/${KIND_VERSION}/kind-linux-amd64"
          chmod +x kind-linux-amd64
          mv kind-linux-amd64 ~/k8s-tools/kind
        else
          echo "Using cached kind"
        fi
        sudo cp ~/k8s-tools/kind /usr/local/bin/kind

        # Verify installations
        kubectl version --client
        kind version

    - name: Deploy using Makefile
      run: |
        echo "Using Makefile to deploy complete stack..."
        make kind-up CONTAINER_ENGINE=docker CI_MODE=true

    - name: Wait for deployments
      run: |
        echo "Waiting for all deployments in parallel..."

        # Wait for all deployments concurrently using background jobs
        kubectl wait --for=condition=available --timeout=180s deployment/backend-api -n ambient-code &
        PID_BACKEND=$!

        kubectl wait --for=condition=available --timeout=180s deployment/frontend -n ambient-code &
        PID_FRONTEND=$!

        kubectl wait --for=condition=available --timeout=180s deployment/agentic-operator -n ambient-code &
        PID_OPERATOR=$!

        # Wait for all and collect exit codes
        FAILED=0

        wait $PID_BACKEND || {
          echo "Backend deployment timeout - showing status"
          kubectl get pods -n ambient-code -o wide
          kubectl describe deployment backend-api -n ambient-code | tail -50
          kubectl get events -n ambient-code --field-selector involvedObject.kind=Pod --sort-by='.lastTimestamp' | grep backend-api || true
          FAILED=1
        }

        wait $PID_FRONTEND || {
          echo "Frontend deployment timeout - showing status"
          kubectl get pods -n ambient-code -o wide
          kubectl describe deployment frontend -n ambient-code | tail -50
          FAILED=1
        }

        wait $PID_OPERATOR || {
          echo "Operator deployment timeout - showing status"
          kubectl get pods -n ambient-code -o wide
          kubectl describe deployment agentic-operator -n ambient-code | tail -50
          FAILED=1
        }

        if [[ $FAILED -eq 1 ]]; then
          exit 1
        fi

        echo "All deployments ready"

    - name: Run backend integration tests (real k8s auth path)
      run: |
        set -euo pipefail

        echo "Setting up ServiceAccount + RBAC for backend integration tests..."
        kubectl -n ambient-code create serviceaccount backend-integration-test 2>/dev/null || true
        kubectl -n ambient-code create role backend-integration-test \
          --verb=get,list \
          --resource=configmaps 2>/dev/null || true
        kubectl -n ambient-code create rolebinding backend-integration-test \
          --role=backend-integration-test \
          --serviceaccount=ambient-code:backend-integration-test 2>/dev/null || true

        TEST_TOKEN="$(kubectl -n ambient-code create token backend-integration-test)"
        echo "::add-mask::$TEST_TOKEN"

        echo "Running Go integration tests (skips any external-provider tests without env)..."
        cd components/backend
        INTEGRATION_TESTS=true \
        K8S_TEST_TOKEN="$TEST_TOKEN" \
        K8S_TEST_NAMESPACE="ambient-code" \
        go test ./tests/integration/... -count=1 -timeout=10m

    - name: Run Makefile smoke tests
      run: |
        echo "Running Makefile smoke tests..."
        make local-test-quick CONTAINER_ENGINE=docker || {
          echo "Smoke tests failed - showing debugging information..."
          make local-troubleshoot
          exit 1
        }

    - name: Run comprehensive test suite
      run: |
        echo "Running local development test suite..."
        chmod +x tests/local-dev-test.sh

        # Run tests in CI mode (known failures tracked separately)
        ./tests/local-dev-test.sh --skip-setup --ci || {
          echo "Test suite failed - showing debugging information..."
          make local-troubleshoot
          exit 1
        }

    - name: Show deployment status
      if: always()
      run: |
        echo "=== Namespace ==="
        kubectl get namespace ambient-code 2>&1 || echo "(Namespace not found - may not have been created)"

        echo ""
        echo "=== Deployments ==="
        kubectl get deployments -n ambient-code -o wide 2>&1 || echo "(No deployments found or namespace does not exist)"

        echo ""
        echo "=== ReplicaSets ==="
        kubectl get replicasets -n ambient-code -o wide 2>&1 || echo "(No replicasets found or namespace does not exist)"

        echo ""
        echo "=== Pods ==="
        kubectl get pods -n ambient-code -o wide 2>&1 || echo "(No pods found or namespace does not exist)"

        echo ""
        echo "=== Services ==="
        kubectl get svc -n ambient-code 2>&1 || echo "(No services found or namespace does not exist)"

        echo ""
        echo "=== Ingress ==="
        kubectl get ingress -n ambient-code 2>&1 || echo "(No ingress found or namespace does not exist)"

        echo ""
        echo "=== CRDs ==="
        kubectl get crd 2>&1 | grep vteam || echo "(No vteam CRDs found)"

        echo ""
        echo "=== Events (last 30) ==="
        kubectl get events -n ambient-code --sort-by='.lastTimestamp' 2>&1 | tail -30 || echo "(No events found or namespace does not exist)"

        echo ""
        echo "=== Deployment describe (if no pods) ==="
        if ! kubectl get pods -n ambient-code 2>/dev/null | grep -q "backend-api\|frontend\|agentic-operator"; then
          echo "No pods found - describing deployments for details:"
          kubectl describe deployment backend-api -n ambient-code 2>&1 | tail -50 || echo "(backend-api deployment not found)"
          kubectl describe deployment frontend -n ambient-code 2>&1 | tail -50 || echo "(frontend deployment not found)"
          kubectl describe deployment agentic-operator -n ambient-code 2>&1 | tail -50 || echo "(agentic-operator deployment not found)"
        fi

    - name: Cleanup
      if: always()
      run: |
        echo "Cleaning up using Makefile..."
        make kind-down 2>&1 || echo "(Cleanup failed or cluster already removed)"