Run PR check for guided notebooks

.github/workflows/guided_notebook_tests.yaml

@@ -0,0 +1,159 @@
+name: Guided notebooks tests
+
+on:
+  pull_request:
+    branches:
+      - main
+      - 'release-*'
+    paths-ignore:
+      - 'docs/**'
+      - '**.adoc'
+      - '**.md'
+      - 'LICENSE'
+  push:
+    branches:
+      - main
+      - 'release-*'
+    paths-ignore:
+      - 'docs/**'
+      - '**.adoc'
+      - '**.md'
+      - 'LICENSE'
+
+concurrency:
+  group: ${{ github.head_ref }}-${{ github.workflow }}
+  cancel-in-progress: true
+
+env:
+  CODEFLARE_OPERATOR_IMG: "quay.io/project-codeflare/codeflare-operator:dev"
+
+jobs:
+  kubernetes:
+
+    runs-on: ubuntu-20.04-4core
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Checkout common repo code
+        uses: actions/checkout@v4
+        with:
+          repository: 'project-codeflare/codeflare-common'
+          ref: 'main'
+          path: 'common'
+
+      - name: Checkout CodeFlare operator repository
+        uses: actions/checkout@v4
+        with:
+          repository: project-codeflare/codeflare-operator
+          path: codeflare-operator
+
+      - name: Set Go
+        uses: actions/setup-go@v5
+        with:
+          go-version-file: './codeflare-operator/go.mod'
+          cache-dependency-path: "./codeflare-operator/go.sum"
+
+      - name: Set up gotestfmt
+        uses: gotesttools/gotestfmt-action@v2
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Set up specific Python version
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.9'
+          cache: 'pip' # caching pip dependencies
+
+      - name: Setup and start KinD cluster
+        uses: ./common/github-actions/kind
+
+      - name: Deploy CodeFlare stack
+        id: deploy
+        run: |
+          cd codeflare-operator
+          echo Setting up CodeFlare stack
+          make setup-e2e
+          echo Deploying CodeFlare operator
+          make deploy -e IMG="${CODEFLARE_OPERATOR_IMG}" -e ENV="e2e"
+          kubectl wait --timeout=120s --for=condition=Available=true deployment -n openshift-operators codeflare-operator-manager
+          cd ..
+
+      - name: Add user to KinD
+        uses: ./common/github-actions/kind-add-user
+        with:
+          user-name: sdk-user
+
+      - name: Configure RBAC for sdk user with limited permissions
+        run: |
+          kubectl create clusterrole list-ingresses --verb=get,list --resource=ingresses
+          kubectl create clusterrolebinding sdk-user-list-ingresses --clusterrole=list-ingresses --user=sdk-user
+          kubectl create clusterrole namespace-creator --verb=get,list,create,delete,patch --resource=namespaces
+          kubectl create clusterrolebinding sdk-user-namespace-creator --clusterrole=namespace-creator --user=sdk-user
+          kubectl create clusterrole raycluster-creator --verb=get,list,create,delete,patch --resource=rayclusters
+          kubectl create clusterrolebinding sdk-user-raycluster-creator --clusterrole=raycluster-creator --user=sdk-user
+          kubectl create clusterrole appwrapper-creator --verb=get,list,create,delete,patch --resource=appwrappers
+          kubectl create clusterrolebinding sdk-user-appwrapper-creator --clusterrole=appwrapper-creator --user=sdk-user
+          kubectl create clusterrole resourceflavor-creator --verb=get,list,create,delete --resource=resourceflavors
+          kubectl create clusterrolebinding sdk-user-resourceflavor-creator --clusterrole=resourceflavor-creator --user=sdk-user
+          kubectl create clusterrole clusterqueue-creator --verb=get,list,create,delete,patch --resource=clusterqueues
+          kubectl create clusterrolebinding sdk-user-clusterqueue-creator --clusterrole=clusterqueue-creator --user=sdk-user
+          kubectl create clusterrole localqueue-creator --verb=get,list,create,delete,patch --resource=localqueues
+          kubectl create clusterrolebinding sdk-user-localqueue-creator --clusterrole=localqueue-creator --user=sdk-user
+          kubectl create clusterrole list-secrets --verb=get,list --resource=secrets
+          kubectl create clusterrolebinding sdk-user-list-secrets --clusterrole=list-secrets --user=sdk-user
+          kubectl config use-context sdk-user
+
+      - name: Run e2e tests
+        run: |
+          set -euo pipefail
+          pip install poetry papermill
+          poetry install --with test,docs
+
+          echo "Running Guided notebooks..."
+          cd demo-notebooks/guided-demos
+          jq 'del(.cells[] | select(.source[] | contains("Create authentication object for user permissions")))' 0_basic_ray.ipynb > 0_basic_ray.ipynb.tmp
+          poetry run papermill 0_basic_ray.ipynb.tmp 0_basic_ray_out.ipynb --log-output --execution-timeout 600
+        env:
+          GRPC_DNS_RESOLVER: "native"
+
+      - name: Switch to kind-cluster context to print logs
+        if: always() && steps.deploy.outcome == 'success'
+        run: kubectl config use-context kind-cluster
+
+      - name: Print CodeFlare operator logs
+        if: always() && steps.deploy.outcome == 'success'
+        run: |
+          echo "Printing CodeFlare operator logs"
+          kubectl logs -n openshift-operators --tail -1 -l app.kubernetes.io/name=codeflare-operator | tee ${TEMP_DIR}/codeflare-operator.log
+
+      - name: Print Kueue operator logs
+        if: always() && steps.deploy.outcome == 'success'
+        run: |
+          echo "Printing Kueue operator logs"
+          KUEUE_CONTROLLER_POD=$(kubectl get pods -n kueue-system | grep kueue-controller | awk '{print $1}')
+          kubectl logs -n kueue-system --tail -1 ${KUEUE_CONTROLLER_POD} | tee ${TEMP_DIR}/kueue.log
+
+      - name: Print KubeRay operator logs
+        if: always() && steps.deploy.outcome == 'success'
+        run: |
+          echo "Printing KubeRay operator logs"
+          kubectl logs -n ray-system --tail -1 -l app.kubernetes.io/name=kuberay | tee ${TEMP_DIR}/kuberay.log
+
+      - name: Export all KinD pod logs
+        uses: ./common/github-actions/kind-export-logs
+        if: always() && steps.deploy.outcome == 'success'
+        with:
+          output-directory: ${TEMP_DIR}
+
+      - name: Upload logs
+        uses: actions/upload-artifact@v4
+        if: always() && steps.deploy.outcome == 'success'
+        with:
+          name: logs
+          retention-days: 10
+          path: |
+            ${{ env.TEMP_DIR }}/**/*.log