ci improvements and adding kots install

Author: diamonwiggins

.github/workflows/mlflow-ci.yml

@@ -32,7 +32,7 @@ jobs:
           version: v3.13.3
 
       - name: Set up Python
-        uses: actions/setup-python@v3
+        uses: actions/setup-python@v4
         with:
           python-version: 3.12
 
@@ -42,21 +42,19 @@ jobs:
       - name: Run chart-testing (lint)
         run: ct lint --config applications/mlflow/ct.yaml --chart-dirs applications/mlflow/charts --charts applications/mlflow/charts/mlflow applications/mlflow/charts/infra
 
-  test:
+  helm-install-test-no-sdk:
     runs-on: ubuntu-22.04
     needs: [lint]
     strategy:
       fail-fast: false
       matrix:
         cluster:
           - distribution: kind
-            version: 1.32.3
-          - distribution: k3s
-            version: 1.32.3
-          - distribution: aks
-            version: 1.31
-          - distribution: gke
             version: 1.32
+          #- distribution: kind
+          #  version: 1.31
+          #- distribution: kind
+          #  version: 1.30
     steps:
       - name: Checkout
         uses: actions/checkout@v4
@@ -82,6 +80,14 @@ jobs:
           ttl: 1h
           export-kubeconfig: true
 
+      # Add required Helm repositories
+      - name: Add Helm repositories
+        run: |
+          helm repo add cnpg https://cloudnative-pg.github.io/charts
+          helm repo add minio-operator https://operator.min.io
+          helm repo update
+          echo "Helm repositories added successfully"
+
       - name: Install Infra Chart
         run: ct install --config applications/mlflow/ct.yaml --chart-dirs applications/mlflow/charts --charts applications/mlflow/charts/infra --skip-clean-up --namespace default
 
@@ -112,7 +118,11 @@ jobs:
 
   create-release:
     runs-on: ubuntu-22.04
-    needs: [test]
+    needs: [lint, helm-install-test-no-sdk]
+    outputs:
+      license-id: ${{ steps.create-customer.outputs.license-id }}
+      customer-id: ${{ steps.create-customer.outputs.customer-id }}
+      channel-slug: ${{ steps.create-release.outputs.channel-slug }}
     if: github.event_name == 'push' || github.event_name == 'release'
     steps:
       - name: Checkout
@@ -125,6 +135,14 @@ jobs:
         with:
           version: v3.13.3
 
+      # Add required Helm repositories
+      - name: Add Helm repositories
+        run: |
+          helm repo add cnpg https://cloudnative-pg.github.io/charts
+          helm repo add minio-operator https://operator.min.io
+          helm repo update
+          echo "Helm repositories added successfully"
+
       - name: Package infra chart
         run: |
           helm package applications/mlflow/charts/infra -d applications/mlflow/kots/ -u
@@ -162,9 +180,116 @@ jobs:
           promote-channel: ci-automation-${{ github.run_id }}
           version: ${{ env.CHART_VERSION }}
 
+      - name: Create customer
+        id: create-customer
+        uses: replicatedhq/replicated-actions/create-customer@main
+        with:
+          app-slug: ${{ env.APP_SLUG }}
+          api-token: ${{ secrets.REPLICATED_PLATFORM_EXAMPLES_TOKEN }}
+          customer-name: automated-${{ github.run_id }}
+          customer-email: [email protected]
+          license-type: dev
+          channel-slug: ${{ steps.create-release.outputs.channel-slug }}
+          is-kots-install-enabled: "true"
+
+  kots-install-test:
+    runs-on: ubuntu-22.04
+    needs: [create-release]
+    env:
+      LICENSE_ID: ${{ needs.create-release.outputs.license-id }}
+      CHANNEL_SLUG: ${{ needs.create-release.outputs.channel-slug }}
+    strategy:
+      fail-fast: false
+      matrix:
+        cluster:
+          - distribution: kind
+            version: 1.32
+          #- distribution: kind
+          #  version: 1.31
+          #- distribution: kind
+          #  version: 1.30
+          #- distribution: aks
+          #  version: 1.31
+          #- distribution: aks
+          #  version: 1.30
+          #- distribution: gke
+          #  version: 1.32
+          #- distribution: gke
+          #  version: 1.31
+          #- distribution: gke
+          #  version: 1.30
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Create Cluster
+        id: create-cluster
+        uses: replicatedhq/replicated-actions/[email protected]
+        with:
+          api-token: ${{ secrets.REPLICATED_PLATFORM_EXAMPLES_TOKEN }}
+          kubernetes-distribution: ${{ matrix.cluster.distribution }}
+          kubernetes-version: ${{ matrix.cluster.version }}
+          cluster-name: mlflow-kots-${{ github.run_id }}-${{ matrix.cluster.distribution }}-${{ matrix.cluster.version }}
+          ttl: 1h
+          export-kubeconfig: true
+
+      # The following step extracts the chart version to use it for the KOTS installation
+      - name: Extract MLflow chart version
+        id: chart-version
+        run: |
+          CHART_VERSION=$(grep 'chartVersion:' applications/mlflow/kots/mlflow-chart.yaml | awk '{print $2}')
+          echo "CHART_VERSION=$CHART_VERSION" >> $GITHUB_ENV
+          echo "Using MLflow chart version: $CHART_VERSION"
+
+      # Download license using Replicated vendor-cli Docker container
+      - name: Download license
+        uses: docker://replicated/vendor-cli:latest
+        with:
+          args: customer download-license --customer ${{ needs.create-release.outputs.customer-id }} --output /tmp/license.yaml
+        env:
+          REPLICATED_API_TOKEN: ${{ secrets.REPLICATED_PLATFORM_EXAMPLES_TOKEN }}
+          REPLICATED_APP: ${{ env.APP_SLUG }}
+
+      # Install using KOTS
+      - name: KOTS Install
+        uses: replicatedhq/replicated-actions/[email protected]
+        with:
+          kubeconfig: ${{ steps.create-cluster.outputs.cluster-kubeconfig }}
+          kots-version: latest
+          app-slug: ${{ env.APP_SLUG }}
+          app-version-label: ${{ env.CHART_VERSION }}
+          license-file: /tmp/license.yaml
+          namespace: default
+          wait-duration: 10m
+          shared-password: ${{ secrets.KOTS_SHARED_PASSWORD || 'replicatedmlflow' }}
+
+      - name: Install troubleshoot
+        run: curl -L https://github.com/replicatedhq/troubleshoot/releases/latest/download/support-bundle_linux_amd64.tar.gz | tar xzvf -
+        if: failure()
+
+      - name: Collect bundle
+        run: ./support-bundle --kubeconfig=${{ steps.create-cluster.outputs.cluster-kubeconfig }} --interactive=false -o kots-ci-bundle-${{ matrix.cluster.distribution }}-${{ matrix.cluster.version }} https://raw.githubusercontent.com/replicatedhq/troubleshoot-specs/main/in-cluster/default.yaml
+        if: failure()
+
+      - name: Upload support bundle artifact
+        uses: actions/upload-artifact@v4
+        if: failure()
+        with:
+          name: mlflow-kots-bundle-${{ matrix.cluster.distribution }}-${{ matrix.cluster.version }}
+          path: 'kots-ci-bundle-${{ matrix.cluster.distribution }}-${{ matrix.cluster.version }}.tar.gz'
+
+      - name: Remove Cluster
+        uses: replicatedhq/replicated-actions/[email protected]
+        if: ${{ always() && steps.create-cluster.outputs.cluster-id != '' }}
+        with:
+          api-token: ${{ secrets.REPLICATED_PLATFORM_EXAMPLES_TOKEN }}
+          cluster-id: ${{ steps.create-cluster.outputs.cluster-id }}
+
   cleanup-test-release:
     runs-on: ubuntu-22.04
-    needs: [test, create-release]
+    needs: [helm-install-test-no-sdk, create-release, kots-install-test]
     steps:
       - name: Archive Customer
         if: ${{ needs.create-release.outputs.customer-id != '' }}