feat: update integrate.yaml env setup, debug logging

.github/workflows/integrate.yaml

@@ -49,77 +49,91 @@ jobs:
     - name: Check out repo
       uses: actions/checkout@v3
 
-    - uses: balchua/[email protected]
+    - name: Setup operator environment
+      uses: charmed-kubernetes/actions-operator@main
       with:
-        # TODO: Pinned to <1.25 until we update to istio 1.16
-        channel: '1.24/stable'
-        addons: '["dns", "storage", "rbac", "metallb:10.64.140.43-10.64.140.49"]'
-
-    - name: Install dependencies
-      run: |
-        set -eux
-        sudo snap install charmcraft --classic --channel=latest/candidate
-        sudo snap install juju --classic
-        sudo snap install juju-bundle --classic
-        sudo snap install juju-wait --classic
-        sudo apt update
-        sudo apt install tox
-
-    - name: Wait before bootstrap
+        provider: microk8s
+        channel: 1.24/stable
+        charmcraft-channel: latest/candidate
+        # TODO: Unpin this when this bug is resolved: https://bugs.launchpad.net/juju/+bug/1992833.
+        #       In particular, these tests failed deploying the prometheus-k8s charm where it gets an error in
+        #       the "metrics-endpoint-relation-changed" hook.
+        bootstrap-options: --agent-version="2.9.34"
+        microk8s-addons: "dns storage rbac metallb:10.64.140.43-10.64.140.49"
+    # TODO: Remove once the actions-operator does this automatically
+    - name: Configure kubectl
       run: |
-        set -eux
-        sg microk8s -c 'microk8s status --wait-ready'
-        kubectl wait --for=condition=available --timeout=5m -nkube-system deployment/coredns deployment/hostpath-provisioner
+        sg microk8s -c "microk8s config > ~/.kube/config"
 
-    - name: Bootstrap
+    - name: Add model
       run: |
-        # Pinning juju agent to 2.9.42 to keep compatibility with pythonlib-juju<3
-        sg microk8s -c 'juju bootstrap microk8s uk8s --agent-version=2.9.42'
-        sg microk8s -c 'juju add-model kubeflow'
+        juju add-model kubeflow
+        juju switch kubeflow
 
+    # DEBUG: added keep-models to ensure our logs are available during dumping
     - run: sg microk8s -c 'KUBECONFIG=/home/runner/.kube/config tox -e integration
-        -- --model kubeflow --destructive-mode'
+        -- --model kubeflow --destructive-mode --keep-models'
       timeout-minutes: 80
 
     - name: Setup Debug Artifact Collection
       run: mkdir tmp
-      if: failure()
+      if: always()
+
+    - name: Collect juju-crashdump from pytest-operator
+      run: |
+        for f in `ls juju-crashdump-*`; do
+          cp $f tmp
+        done
+      if: always()
 
     - name: Collect charmcraft logs
-      if: failure()
+      if: always()
       run: cat /home/runner/snap/charmcraft/common/cache/charmcraft/log/charmcraft-*.log
         | tee tmp/charmcraft.log
 
     - name: Collect Juju status
-      if: failure()
+      if: always()
       run: juju status | tee tmp/juju-status.txt
 
     - name: Collect Juju log
-      if: failure()
-      run: juju debug-log --replay --no-tail | tee tmp/juju-status.txt
+      if: always()
+      run: juju debug-log --replay --no-tail | tee tmp/juju-debug-log.txt
+
+    - name: Collect Juju status log
+      if: always()
+      run: |
+        juju show-status-log istio-pilot/0 | tee tmp/juju-status-log-istio-pilot.txt
+        juju show-status-log istio-ingressgateway/0 | tee tmp/juju-status-log-istio-ingressgateway.txt
 
     - name: Collect Kube status
-      if: failure()
+      if: always()
       run: |
         kubectl get all -A | tee tmp/kube-summary.txt
         kubectl describe virtualservices -A | tee tmp/kube-virtualservices.txt
         kubectl describe gateways -A | tee tmp/kube-gateways.txt
+        kubectl describe envoyfilters -A | tee tmp/kube-envoyfilters.txt
         kubectl describe deployments -A | tee tmp/kube-deployments.txt
         kubectl describe replicasets -A | tee tmp/kubectl-replicasets.txt
         kubectl exec -n kubeflow istio-pilot-0 --container charm -- agents/unit-istio-pilot-0/charm/istioctl analyze -n kubeflow | tee tmp/istioctl-analyze.txt
 
     - name: Collect Kube logs
-      if: failure()
+      if: always()
       run: |
         kubectl logs -n kubeflow --tail 1000 -lapp.kubernetes.io/name=istio-pilot -c charm | tee tmp/istio-pilot.log
         kubectl logs -n kubeflow --tail 1000 -lapp.kubernetes.io/name=istio-ingressgateway-operator -c charm | tee tmp/istio-ingressgateway-operator.log
 
+    # This should get output by pytest-operator, but it is blocked by --keep-models until
+    # https://github.com/charmed-kubernetes/pytest-operator/pull/108 is merged
+    - name: Collect juju-crashdump explicitly
+      run: juju-crashdump -o tmp
+
     - name: Upload debug artifacts
-      if: failure()
+      if: always()
       uses: actions/upload-artifact@v3
       with:
         name: test-run-artifacts
         path: tmp
+        if-no-files-found: error
 
   integration-observability:
     name: Observability Integration Test