Github action job tests for xks on push to master and release branch

.github/workflows/xks-push-ci-automation.yaml

@@ -0,0 +1,215 @@
+name: Build, Deploy and Test on EKS
+
+on:
+  push:
+    branches:
+      - "master"
+      - "v*"
+
+env:
+  IMG: gitops-operator:test
+  CLUSTER_NAME: gitops-ci-${{ github.run_id }}
+  AWS_REGION: us-east-1
+
+jobs:
+  ci-build:
+    name: Build image, deploy to EKS cluster
+    runs-on: ubuntu-latest
+    permissions:
+      id-token: write
+      contents: read
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
+
+      - name: Setup Go
+        uses: actions/setup-go@40f1582b2485089dde7abd97c1529aa768e1baff # v5
+        with:
+          go-version-file: 'go.mod'
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@ececac1a45f3b08a01d2dd070d28d111c5fe6722 # v4
+        with:
+          role-to-assume: ${{ secrets.AWS_ROLE_ARN }}
+          aws-region: ${{ env.AWS_REGION }}
+
+      - name: Install eksctl
+        run: |
+          ARCH=amd64
+          PLATFORM=$(uname -s)_$ARCH
+          curl -sLO "https://github.com/eksctl-io/eksctl/releases/latest/download/eksctl_$PLATFORM.tar.gz"
+          tar -xzf eksctl_$PLATFORM.tar.gz -C /tmp
+          sudo mv /tmp/eksctl /usr/local/bin
+
+      - name: Create EKS cluster
+        run: |
+          hack/scripts/eks-cluster.sh create "${{ env.CLUSTER_NAME }}" "${{ env.AWS_REGION }}"
+
+      - name: Login to Amazon ECR
+        id: ecr-login
+        uses: aws-actions/amazon-ecr-login@062b18b96a7aff071d4dc91bc00c4c1a7945b076 # v2
+
+      - name: Create ECR repository
+        run: |
+          aws ecr create-repository \
+            --repository-name gitops-operator \
+            --region ${{ env.AWS_REGION }} 2>/dev/null || true
+
+      - name: Build and push image to ECR
+        env:
+          ECR_REGISTRY: ${{ steps.ecr-login.outputs.registry }}
+        run: |
+          ECR_IMG="${ECR_REGISTRY}/gitops-operator:${{ github.sha }}"
+          make docker-build IMG="${ECR_IMG}"
+          docker push "${ECR_IMG}"
+          echo "ECR_IMG=${ECR_IMG}" >> "$GITHUB_ENV"
+
+      # TODO: check if porting is required for non-OCP clusters
+      - name: Disable webhook and conversion for non-OCP cluster
+        run: |
+          sed -i 's|^- ../prometheus|#- ../prometheus|' config/default/kustomization.yaml
+          sed -i 's|^- ../webhook|#- ../webhook|' config/default/kustomization.yaml
+          sed -i 's|^- patches/webhook_in_argocds.yaml|#- patches/webhook_in_argocds.yaml|' config/crd/kustomization.yaml
+          sed -i 's|^- patches/cainjection_in_argocds.yaml|#- patches/cainjection_in_argocds.yaml|' config/crd/kustomization.yaml
+          echo "=== Verify sed applied ==="
+          grep -n 'webhook\|prometheus' config/default/kustomization.yaml
+          grep -n 'patches/' config/crd/kustomization.yaml
+
+      - name: Install CRDs
+        run: |
+          make install
+
+      - name: Deploy operator
+        run: |
+          make deploy IMG="${ECR_IMG}"
+
+      - name: Verify Controller Manager deployment is available
+        run: |
+          kubectl get deployment -n openshift-gitops-operator
+          kubectl describe deployment -n openshift-gitops-operator
+          kubectl wait --for=condition=available --timeout=300s \
+            deployment/openshift-gitops-operator-controller-manager \
+            -n openshift-gitops-operator
+
+      - name: Create ArgoCD instance
+        run: |
+          kubectl create ns test-argocd
+          kubectl apply -f - <<'EOF'
+          apiVersion: argoproj.io/v1beta1
+          kind: ArgoCD
+          metadata:
+            name: argocd
+            namespace: test-argocd
+          EOF
+
+      - name: Wait for ArgoCD component pods to exist
+        run: |
+          EXPECTED_LABELS=("argocd-application-controller" "argocd-redis" "argocd-repo-server" "argocd-server")
+          TIMEOUT=300
+          INTERVAL=10
+          ELAPSED=0
+
+          echo "Waiting for ArgoCD component pods to exist in test-argocd..."
+          while true; do
+            ALL_EXIST=true
+            for label in "${EXPECTED_LABELS[@]}"; do
+              if ! kubectl get pod -n test-argocd -l "app.kubernetes.io/name=${label}" --no-headers 2>/dev/null | grep -q .; then
+                ALL_EXIST=false
+                break
+              fi
+            done
+
+            if $ALL_EXIST; then
+              echo "All ArgoCD component pods exist after ${ELAPSED}s."
+              break
+            fi
+
+            if [ $ELAPSED -ge $TIMEOUT ]; then
+              echo "Timed out after ${TIMEOUT}s waiting for ArgoCD pods."
+              kubectl get pods -n test-argocd
+              kubectl get argocd -n test-argocd -o yaml
+              exit 1
+            fi
+
+            sleep $INTERVAL
+            ELAPSED=$((ELAPSED + INTERVAL))
+          done
+
+      - name: Verify ArgoCD components are ready
+        run: |
+          kubectl get pods -n test-argocd
+          kubectl wait --for=condition=Ready -n test-argocd pod --timeout=300s \
+            -l 'app.kubernetes.io/name in (argocd-application-controller,argocd-redis,argocd-repo-server,argocd-server)'
+          echo "All ArgoCD components are ready."
+          kubectl get pods -n test-argocd
+
+      - name: Run e2e tests
+        run: |
+          make e2e-tests-parallel-ginkgo
+
+      - name: Run non-default e2e tests
+        run: |
+          make e2e-tests-sequential-ginkgo
+
+      - name: Collect e2e debug info on failure
+        if: failure()
+        run: |
+          echo "=== All pods across namespaces ==="
+          kubectl get pods -A -o wide || true
+          echo ""
+          echo "=== ArgoCD instances ==="
+          kubectl get argocds -A -o yaml 2>/dev/null || true
+          echo ""
+          echo "=== GitopsService ==="
+          kubectl get gitopsservices -A -o yaml 2>/dev/null || true
+
+      - name: Collect operator debug info on failure
+        if: failure()
+        run: |
+          echo "=== Deployment status ==="
+          kubectl get deployment -n openshift-gitops-operator -o wide || true
+          echo ""
+          echo "=== Pod status ==="
+          kubectl get pods -n openshift-gitops-operator -o wide || true
+          echo ""
+          echo "=== Pod descriptions ==="
+          kubectl describe pods -n openshift-gitops-operator || true
+          echo ""
+          echo "=== Controller manager logs ==="
+          kubectl logs deployment/openshift-gitops-operator-controller-manager \
+            -n openshift-gitops-operator --all-containers=true --tail=200 || true
+          echo ""
+          echo "=== Events in operator namespace ==="
+          kubectl get events -n openshift-gitops-operator --sort-by='.lastTimestamp' || true
+          echo ""
+          echo "=== CRD conversion config ==="
+          kubectl get crd argocds.argoproj.io -o jsonpath='{.spec.conversion}' || true
+          echo ""
+
+      - name: Collect ArgoCD debug info on failure
+        if: failure()
+        run: |
+          echo "=== ArgoCD resources ==="
+          kubectl get argocds -n test-argocd -o yaml 2>/dev/null || true
+          echo ""
+          echo "=== Pods in test-argocd ==="
+          kubectl get pods -n test-argocd -o wide 2>/dev/null || true
+          echo ""
+          echo "=== Pod descriptions in test-argocd ==="
+          kubectl describe pods -n test-argocd 2>/dev/null || true
+          echo ""
+          echo "=== Events in test-argocd ==="
+          kubectl get events -n test-argocd --sort-by='.lastTimestamp' 2>/dev/null || true
+
+      - name: Delete EKS cluster
+        if: always()
+        run: |
+          FORCE=true hack/scripts/eks-cluster.sh delete "${{ env.CLUSTER_NAME }}" "${{ env.AWS_REGION }}"
+
+      - name: Delete ECR repository
+        if: always()
+        run: |
+          aws ecr delete-repository \
+            --repository-name gitops-operator \
+            --region ${{ env.AWS_REGION }} \
+            --force 2>/dev/null || true

hack/scripts/eks-cluster.sh

@@ -0,0 +1,96 @@
+#!/usr/bin/env bash
+#
+# Manage an AWS EKS cluster (3 nodes) suitable for Argo CD HA.
+#
+# Usage:
+#   ./eks-cluster.sh create  [CLUSTER_NAME] [REGION]
+#   ./eks-cluster.sh delete  [CLUSTER_NAME] [REGION]
+#   ./eks-cluster.sh status  [CLUSTER_NAME] [REGION]
+#
+# Defaults:
+#   CLUSTER_NAME = argocd-ha
+#   REGION       = us-east-1
+
+set -euo pipefail
+
+usage() {
+  echo "Usage: $0 {create|delete|status} [CLUSTER_NAME] [REGION]"
+  exit 1
+}
+
+[[ $# -ge 1 ]] || usage
+
+ACTION="$1"; shift
+CLUSTER_NAME="${1:-argocd-ha}"
+REGION="${2:-us-east-1}"
+
+K8S_VERSION="1.30"
+NODE_TYPE="m5.xlarge"
+NODE_COUNT=3
+NODE_VOLUME_SIZE=100
+
+required_cmds=(aws eksctl)
+[[ "${ACTION}" == "create" || "${ACTION}" == "status" ]] && required_cmds+=(kubectl)
+for cmd in "${required_cmds[@]}"; do
+  if ! command -v "${cmd}" &>/dev/null; then
+    echo "ERROR: ${cmd} not found in PATH" >&2
+    exit 1
+  fi
+done
+
+cmd_create() {
+  echo "==> Creating EKS cluster: ${CLUSTER_NAME} in ${REGION}"
+  echo "    Kubernetes ${K8S_VERSION}, ${NODE_COUNT}x ${NODE_TYPE}, ${NODE_VOLUME_SIZE}GiB volumes"
+
+  eksctl create cluster \
+    --name "${CLUSTER_NAME}" \
+    --region "${REGION}" \
+    --version "${K8S_VERSION}" \
+    --nodegroup-name "argocd-workers" \
+    --node-type "${NODE_TYPE}" \
+    --nodes "${NODE_COUNT}" \
+    --nodes-min "${NODE_COUNT}" \
+    --nodes-max "${NODE_COUNT}" \
+    --node-volume-size "${NODE_VOLUME_SIZE}" \
+    --managed \
+    --with-oidc \
+    --ssh-access=false \
+    --asg-access
+
+  echo "==> Cluster ready"
+  kubectl get nodes -o wide
+}
+
+cmd_delete() {
+  echo "==> This will DELETE cluster '${CLUSTER_NAME}' in ${REGION} and all associated resources"
+  if [[ "${FORCE:-false}" != "true" ]]; then
+    read -rp "    Continue? [y/N] " confirm
+    [[ "${confirm}" =~ ^[Yy]$ ]] || { echo "Aborted."; exit 0; }
+  fi
+
+  echo "==> Deleting EKS cluster: ${CLUSTER_NAME}"
+  eksctl delete cluster \
+    --name "${CLUSTER_NAME}" \
+    --region "${REGION}" \
+    --wait
+
+  echo "==> Cleanup complete"
+}
+
+cmd_status() {
+  echo "==> Cluster: ${CLUSTER_NAME} (${REGION})"
+  if eksctl get cluster --name "${CLUSTER_NAME}" --region "${REGION}" 2>/dev/null; then
+    echo ""
+    echo "==> Nodes:"
+    kubectl get nodes -o wide 2>/dev/null || echo "    (kubectl not configured for this cluster)"
+  else
+    echo "    Cluster not found"
+  fi
+}
+
+case "${ACTION}" in
+  create) cmd_create ;;
+  delete) cmd_delete ;;
+  status) cmd_status ;;
+  *)      usage ;;
+esac