test commit

.github/workflows/tests.yaml

@@ -1,61 +1,73 @@
-name: Build and Deploy
-
 on: [push, pull_request]
 
 jobs:
   build-and-deploy:
     runs-on: ubuntu-latest
+    env:
+      CHECKPOINT_DIR: /var/lib/kubelet/checkpoints
     steps:
       - name: Checkout code
         uses: actions/checkout@v3
 
       - name: Set up Go
         uses: actions/setup-go@v5
         with:
-          go-version: 1.22
+          go-version: 1.21
 
       - name: Install dependencies
         run: |
           # Install kind
-          curl -sLo kind "$(curl -sL https://api.github.com/repos/kubernetes-sigs/kind/releases/latest | jq -r '[.assets[] | select(.name == "kind-linux-amd64")] | first | .browser_download_url')"
+          curl -sLo kind "$(curl -sL https://api.github.com/repos/kubernetes-sigs/kind/releases/latest | jq -r '[.assets[] | select(.name == "kind-linux-arm64")] | first | .browser_download_url')"
           chmod +x kind
           sudo mv kind /bin/
-
+          
           # Install kubectl
-          curl -LO "https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl"
+          curl -LO "https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/arm64/kubectl"
           chmod +x kubectl
           sudo mv kubectl /bin/
 
-      - name: Create Kubernetes cluster
+      - name: Prepare Checkpoints Directory in Host
         run: |
-          kind delete cluster
-          kind create cluster
-          kind export kubeconfig
+          sudo mkdir -p ${{ env.CHECKPOINT_DIR }}
+          sudo chmod 700 ${{ env.CHECKPOINT_DIR }}
+          sudo chown $USER ${{ env.CHECKPOINT_DIR }}
+          sudo echo ${{ env.CHECKPOINT_DIR }}
+
+      - name: Generate Checkpoint Tar Files
+        run: ./test/generate_checkpoint_tar.sh
+
+      - name: Generate Kind Config
+        run: ./test/generate_kind_config.sh
+
+      - name: Delete Existing Kind Cluster
+        run: kind delete cluster || true
+
+      - name: Create Kubernetes cluster
+        run: kind create cluster --config=kind-config.yaml
 
       - name: Load Docker image into kind cluster
         run: |
+          docker rmi checkpoint-restore-operator:ci || true
           make docker-build IMG=checkpoint-restore-operator:ci
+          docker images | grep checkpoint-restore-operator
           kind load docker-image checkpoint-restore-operator:ci
 
+      - name: Replace Placeholder in Manager
+        run: |
+          sed -i "s#__CHECKPOINT_DIR__#${{ env.CHECKPOINT_DIR }}#g" ./config/manager/manager.yaml
+
       - name: Deploy to Kubernetes
         run: |
           make install
           make deploy IMG=checkpoint-restore-operator:ci
+          sudo ls ${{ env.CHECKPOINT_DIR }}
 
       - name: Wait for deployments to be ready
         run: ./test/wait_for_deployment.sh checkpoint-restore-operator-controller-manager
 
       - name: Check resources
         run: kubectl get all -n checkpoint-restore-operator-system
 
-      - name: Prepare Checkpoints Directory in Kind Node
-        run: |
-          sudo mkdir -p /var/lib/kubelet/checkpoints
-          sudo chmod 777 /var/lib/kubelet/checkpoints
-
-      - name: Generate Checkpoint Tar Files
-        run: ./test/generate_checkpoint_tar.sh
-
       - name: Wait for Checkpoint Tar Files to Reduce
         run: |
           kubectl apply -f ./test/test_checkpointrestoreoperator.yaml

internal/controller/checkpointrestoreoperator_controller.go

@@ -495,6 +495,16 @@ func (gc *garbageCollector) runGarbageCollector() {
 	gc.Lock()
 	defer gc.Unlock()
 
+	// Log the list of all files in the checkpointDirectory(dor debugging purpose)
+	files, err := os.ReadDir(checkpointDirectory)
+	if err != nil {
+		log.Error(err, "failed to read checkpointDirectory")
+	} else {
+		for _, file := range files {
+			log.Info("Existing file", "name", file.Name())
+		}
+	}
+
 	// based on fsnotify example code
 	watcher, err := fsnotify.NewWatcher()
 	if err != nil {

test/checkpoints/test.txt

@@ -0,0 +1 @@
+test

test/generate_kind_config.sh

@@ -0,0 +1,14 @@
+#!/bin/bash
+
+CONFIG_FILE="kind-config.yaml"
+CHECKPOINT_DIR="/var/lib/kubelet/checkpoints"
+
+{
+	echo "kind: Cluster"
+	echo "apiVersion: kind.x-k8s.io/v1alpha4"
+	echo "nodes:"
+	echo "- role: control-plane"
+	echo "  extraMounts:"
+	echo "  - hostPath: /var/lib/kubelet/checkpoints"
+	echo "    containerPath: $CHECKPOINT_DIR"
+} >$CONFIG_FILE

test/test_checkpointrestoreoperator.yaml

@@ -9,7 +9,7 @@ metadata:
     app.kubernetes.io/created-by: checkpoint-restore-operator
   name: checkpointrestoreoperator-sample
 spec:
-  checkpointDirectory: /var/lib/kubelet/checkpoints/
+  checkpointDirectory: /var/lib/kubelet/checkpoints
   applyPoliciesImmediately: true
   globalPolicy:
     maxCheckpointsPerNamespace: 10

test/wait_for_checkpoint_reduction.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-CHECKPOINTS_DIR="/var/lib/kubelet/checkpoints/"
+CHECKPOINTS_DIR="/var/lib/kubelet/checkpoints"
 EXPECTED_COUNT=2
 TIMEOUT=60
 start_time=$(date +%s)

test/wait_for_deployment.sh

@@ -2,7 +2,7 @@
 
 wait_for_deployment() {
 	local deployment_name="$1"
-	timeout=60 # 5 minutes (60 * 5 sec)
+	timeout=20 # 5 minutes (60 * 5 sec)
 	i=1
 	echo "Checking if the ${deployment_name} deployment is ready"
 	until kubectl -n checkpoint-restore-operator-system get deployment "${deployment_name}" -o jsonpath='{.status.conditions[?(@.status=="True")].type}' | grep "Available" 2>/dev/null; do
@@ -14,9 +14,14 @@ wait_for_deployment() {
 			kubectl -n checkpoint-restore-operator-system describe deployment "${deployment_name}"
 			kubectl -n checkpoint-restore-operator-system get pods
 			kubectl -n checkpoint-restore-operator-system describe pods
+			echo "Fetching detailed logs from the pods for further debugging:"
+			for pod in $(kubectl -n checkpoint-restore-operator-system get pods -o jsonpath='{.items[*].metadata.name}'); do
+				echo "Logs for pod ${pod}:"
+				kubectl -n checkpoint-restore-operator-system logs "${pod}"
+			done
 			exit 1
 		fi
-		echo "Waiting for ${deployment_name} deployment to report a ready status"
+		echo "Waiting for ${deployment_name} deployment to report a ready status (Attempt ${i}/${timeout})"
 		sleep 5
 	done
 	echo "The ${deployment_name} deployment is ready"