chore: soak interruptions for checkresync test

.github/workflows/soak-interrupts.yaml

@@ -0,0 +1,212 @@
+name: Soak Interrupts Test
+
+permissions: read-all
+on:
+  workflow_dispatch:
+    inputs:
+      branch:
+        description: 'Branch to run the workflow on'
+        required: true
+        default: 'main'
+      http2:
+        description: 'use http2 watcher'
+        required: false
+        default: 'false'
+  schedule:
+    - cron: '0 4 * * *' # 12AM EST/9PM PST
+
+jobs:
+  pepr-build:
+    name: controller image
+    runs-on: ubuntu-latest
+    steps:
+      - name: Harden Runner
+        uses: step-security/harden-runner@91182cccc01eb5e619899d80e4e971d6181294a7 # v2.10.1
+        with:
+          egress-policy: audit
+
+      - name: clone pepr
+        uses: actions/checkout@d632683dd7b4114ad314bca15554477dd762a938 # v4.2.0
+        with:
+          repository: defenseunicorns/pepr
+          path: pepr
+
+      - name: "set env: PEPR"
+        run: echo "PEPR=${GITHUB_WORKSPACE}/pepr" >> "$GITHUB_ENV"
+
+      - name: setup node
+        uses: actions/setup-node@0a44ba7841725637a19e28fa30b79a866c81b0a6 # v4.0.4
+        with:
+          node-version: 20
+          cache: "npm"
+          cache-dependency-path: pepr
+
+      - name: install pepr deps
+        run: |
+          cd "$PEPR"
+          npm ci
+
+      - name: build pepr image
+        run: |
+          cd "$PEPR"
+          npm run build:image
+
+      - name: tar pepr image
+        run: |
+          PEPR_TAR="${GITHUB_WORKSPACE}/pepr-img.tar"
+          echo "PEPR_TAR=${PEPR_TAR}" >> "$GITHUB_ENV"
+          docker image save --output "$PEPR_TAR" pepr:dev
+
+      - name: upload image tar artifact
+        uses: actions/upload-artifact@50769540e7f4bd5e21e526ee35c689e35e0d6874 # v4.4.0
+        with:
+          name: pepr-img.tar
+          path: pepr-img.tar
+          retention-days: 1
+
+  soak-test:
+    name: soak-test
+    runs-on: ubuntu-latest
+    needs:
+      - pepr-build
+
+    steps:
+      - name: Harden Runner
+        uses: step-security/harden-runner@91182cccc01eb5e619899d80e4e971d6181294a7 # v2.10.1
+        with:
+          egress-policy: audit
+
+      - uses: actions/checkout@d632683dd7b4114ad314bca15554477dd762a938 # v4.2.0
+
+      - name: "install k3d"
+        run: "curl -s https://raw.githubusercontent.com/k3d-io/k3d/main/install.sh | bash"
+        shell: bash
+
+      - name: dowload image tar artifact
+        uses: actions/download-artifact@fa0a91b85d4f404e444e00e005971372dc801d16 # v4.1.8
+        with:
+          name: pepr-img.tar
+          path: ${{ github.workspace }}
+
+      - name: import pepr image from tar
+        run: |
+          PEPR_TAR="${GITHUB_WORKSPACE}/pepr-img.tar"
+          echo "PEPR_TAR=${PEPR_TAR}" >> "$GITHUB_ENV"
+          docker image load --input "$PEPR_TAR"
+
+      - name: Create k3d Cluster
+        run: "k3d cluster create"
+        shell: bash
+
+      - name: Import pepr image into k3d
+        run: "k3d image import pepr:dev -c k3s-default"
+        shell: bash
+
+      - name: Install istioctl
+        run: |
+          curl -L https://istio.io/downloadIstio | sh -
+        shell: bash
+
+      - name: Install default profile
+        run: |
+          cd istio*/bin
+          ./istioctl install --set profile=demo -y
+
+      - name: Set up Kubernetes
+        uses: azure/setup-kubectl@3e0aec4d80787158d308d7b364cb1b702e7feb7f # v4.0.0
+        with:
+          version: 'latest'
+
+      - name: Create logs directory
+        run: mkdir -p logs
+
+      - name: Deploy Pepr 
+        if: ${{ (github.event.inputs.http2 || 'none') != 'true' }}
+        run: |
+          kubectl apply -f hack/soak.ci.yaml
+
+      - name: Deploy http2 Pepr
+        if: ${{ (github.event.inputs.http2 || 'none') == 'true' }}
+        run: |
+          kubectl apply -f hack/soak-http2.ci.yaml
+
+      - name: Deploy applications
+        run: |
+          kubectl apply -f hack/auditor.ci.yaml
+
+      - name: Wait for 15 minutes before starting pod checks
+        run: |
+          sleep 10s
+          kubectl wait --for=condition=ready -n istio-system pod -l istio=pilot
+          kubectl wait --for=condition=ready -n istio-system pod -l app=istio-ingressgateway
+          kubectl wait --for=condition=ready -n watch-auditor pod -l app=watch-auditor
+          kubectl wait --for=condition=ready -n pepr-system pod -l app=pepr-soak-ci-watcher
+
+      - name: Run the soak test and collect metrics
+        run: |
+          # Initialize the map to store pod counts
+          declare -A pod_map
+
+          update_pod_map() {
+            for pod in $(kubectl get pods -n pepr-demo -o jsonpath='{.items[*].metadata.name}'); do
+              count=${pod_map[$pod]}
+              if [ -z "$count" ]; then
+                pod_map[$pod]=1
+              else
+                pod_map[$pod]=$((count + 1))
+              fi
+            done
+          }
+          touch logs/auditor-log.txt
+          touch logs/informer-log.txt
+
+          update_pod_map
+
+          collect_metrics() {
+            kubectl exec metrics-collector -n watch-auditor -- curl watch-auditor:8080/metrics | grep watch_controller_failures_total > logs/auditor-log.txt || true
+            kubectl exec metrics-collector -n watch-auditor -- curl -k https://pepr-soak-ci-watcher.pepr-system.svc.cluster.local/metrics | egrep -E "pepr_cache_miss|pepr_resync_failure_count" > logs/informer-log.txt
+            kubectl logs -n pepr-system deploy/pepr-soak-ci-watcher > logs/watch-log.txt
+          }
+
+          # Start collecting metrics every 5 minutes and checking pod counts every 30 minutes
+          for i in {1..13}; do  # 13 iterations cover 65 minutes (1 hours and 5 minutes) (Every 5 mins x 13 = 65 mins = 1 hour 5 mins)
+            collect_metrics
+            cat logs/informer-log.txt
+            cat logs/auditor-log.txt
+            if [ $((i % 2)) -eq 0 ]; then  # Every 10 minutes 
+              update_pod_map
+
+              # get a list of pods every 10 mins
+              kubectl get pods -n pepr-demo 
+              kubectl top po -n pepr-system
+              kubectl get po -n pepr-system 
+
+              # Verify that no pod's count exceeds 1
+              for pod in "${!pod_map[@]}"; do
+                echo "$pod: ${pod_map[$pod]}"
+                if [ "${pod_map[$pod]}" -gt 1 ]; then
+                  echo "Test failed: Pod $pod has count ${pod_map[$pod]}"
+                  exit 1
+                fi
+              done
+
+              # Every 20 minutes, scale up or down the `watch-auditor` deployment
+              if [ $((i % 4)) -eq 0 ]; then
+                echo "Scaling down the watch-auditor deployment to 0 replicas"
+                kubectl scale deploy/watch-auditor -n watch-auditor --replicas=0
+              else
+                echo "Scaling up the watch-auditor deployment to 0 replicas"
+                kubectl scale deploy/watch-auditor -n watch-auditor --replicas=1
+              fi
+            fi
+            sleep 300s  # Sleep for 5 minutes before the next iteration
+          done
+
+          echo "Soak test passed successfully!"
+        shell: bash
+
+      - name: Upload logs
+        uses: actions/upload-artifact@50769540e7f4bd5e21e526ee35c689e35e0d6874 # v4.4.0
+        with:
+          name: soak-test-logs
+          path: logs