Merge pull request #629 from red-hat-storage/sync_us--master

Syncing latest changes from upstream master for rook

.github/workflows/canary-test-config/action.yaml

@@ -19,7 +19,7 @@ runs:
     - name: Setup Minikube
       shell: bash --noprofile --norc -eo pipefail -x {0}
       run: |
-        tests/scripts/github-action-helper.sh install_minikube_with_none_driver v1.29.2
+        tests/scripts/github-action-helper.sh install_minikube_with_none_driver v1.30.0
 
     - name: install deps
       shell: bash --noprofile --norc -eo pipefail -x {0}

.github/workflows/integration-test-helm-suite.yaml

@@ -25,7 +25,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4

.github/workflows/integration-test-mgr-suite.yaml

@@ -24,7 +24,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.29.2"]
+        kubernetes-versions: ["v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4

.github/workflows/integration-test-multi-cluster-suite.yaml

@@ -25,7 +25,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.29.2"]
+        kubernetes-versions: ["v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4

.github/workflows/integration-test-object-suite.yaml

@@ -25,7 +25,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4

.github/workflows/integration-test-smoke-suite.yaml

@@ -25,7 +25,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4

.github/workflows/integration-test-upgrade-suite.yaml

@@ -25,7 +25,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4
@@ -69,7 +69,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4

.github/workflows/integration-tests-on-release.yaml

@@ -18,7 +18,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.27.11", "v1.28.7", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.27.13", "v1.28.9", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4
@@ -58,7 +58,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.27.11", "v1.28.7", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.27.13", "v1.28.9", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4
@@ -99,7 +99,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.27.11", "v1.28.7", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.27.13", "v1.28.9", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4
@@ -137,7 +137,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.27.11", "v1.28.7", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.27.13", "v1.28.9", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4
@@ -175,7 +175,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.27.11", "v1.28.7", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.27.13", "v1.28.9", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4
@@ -216,7 +216,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        kubernetes-versions: ["v1.25.16", "v1.29.2"]
+        kubernetes-versions: ["v1.25.16", "v1.30.0"]
     steps:
       - name: checkout
         uses: actions/checkout@v4

.mergify.yml

@@ -238,37 +238,37 @@ pull_request_rules:
       - "check-success=crds-gen"
       - "check-success=docs-check"
       - "check-success=pylint"
-      - "check-success=canary (quay.io/ceph/ceph:v18)"
-      - "check-success=raw-disk-with-object (quay.io/ceph/ceph:v18)"
-      - "check-success=two-osds-in-device (quay.io/ceph/ceph:v18)"
-      - "check-success=osd-with-metadata-partition-device (quay.io/ceph/ceph:v18)"
-      - "check-success=osd-with-metadata-device (quay.io/ceph/ceph:v18)"
-      - "check-success=encryption (quay.io/ceph/ceph:v18)"
-      - "check-success=lvm (quay.io/ceph/ceph:v18)"
-      - "check-success=pvc (quay.io/ceph/ceph:v18)"
-      - "check-success=pvc-db (quay.io/ceph/ceph:v18)"
-      - "check-success=pvc-db-wal (quay.io/ceph/ceph:v18)"
-      - "check-success=encryption-pvc (quay.io/ceph/ceph:v18)"
-      - "check-success=encryption-pvc-db (quay.io/ceph/ceph:v18)"
-      - "check-success=encryption-pvc-db-wal (quay.io/ceph/ceph:v18)"
-      - "check-success=encryption-pvc-kms-vault-token-auth (quay.io/ceph/ceph:v18)"
-      - "check-success=encryption-pvc-kms-vault-k8s-auth (quay.io/ceph/ceph:v18)"
-      - "check-success=lvm-pvc (quay.io/ceph/ceph:v18)"
-      - "check-success=multi-cluster-mirroring (quay.io/ceph/ceph:v18)"
-      - "check-success=rgw-multisite-testing (quay.io/ceph/ceph:v18)"
-      - "check-success=encryption-pvc-kms-ibm-kp (quay.io/ceph/ceph:v18)"
-      - "check-success=multus-cluster-network (quay.io/ceph/ceph:v18)"
-      - "check-success=csi-hostnetwork-disabled (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / canary (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / raw-disk-with-object (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / two-osds-in-device (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / osd-with-metadata-partition-device (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / osd-with-metadata-device (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / encryption (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / lvm (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / pvc (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / pvc-db (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / pvc-db-wal (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / encryption-pvc (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / encryption-pvc-db (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / encryption-pvc-db-wal (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / encryption-pvc-kms-vault-token-auth (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / encryption-pvc-kms-vault-k8s-auth (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / lvm-pvc (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / multi-cluster-mirroring (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / rgw-multisite-testing (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / encryption-pvc-kms-ibm-kp (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / multus-cluster-network (quay.io/ceph/ceph:v18)"
+      - "check-success=canary-tests / csi-hostnetwork-disabled (quay.io/ceph/ceph:v18)"
       - "check-success=TestCephSmokeSuite (v1.25.16)"
-      - "check-success=TestCephSmokeSuite (v1.29.2)"
+      - "check-success=TestCephSmokeSuite (v1.30.0)"
       - "check-success=TestCephHelmSuite (v1.25.16)"
-      - "check-success=TestCephHelmSuite (v1.29.2)"
-      - "check-success=TestCephMultiClusterDeploySuite (v1.29.2)"
-      - "check-success=TestCephObjectSuite (v1.29.2)"
+      - "check-success=TestCephHelmSuite (v1.30.0)"
+      - "check-success=TestCephMultiClusterDeploySuite (v1.30.0)"
+      - "check-success=TestCephObjectSuite (v1.30.0)"
       - "check-success=TestCephUpgradeSuite (v1.25.16)"
-      - "check-success=TestCephUpgradeSuite (v1.29.2)"
+      - "check-success=TestCephUpgradeSuite (v1.30.0)"
       - "check-success=TestHelmUpgradeSuite (v1.25.16)"
-      - "check-success=TestHelmUpgradeSuite (v1.29.2)"
+      - "check-success=TestHelmUpgradeSuite (v1.30.0)"
     actions:
       merge:
         method: merge

Documentation/Getting-Started/Prerequisites/prerequisites.md

@@ -7,7 +7,7 @@ and Rook is granted the required privileges (see below for more information).
 
 ## Kubernetes Version
 
-Kubernetes versions **v1.25** through **v1.29** are supported.
+Kubernetes versions **v1.25** through **v1.30** are supported.
 
 ## CPU Architecture
 

Documentation/Getting-Started/quickstart.md

@@ -12,7 +12,7 @@ This guide will walk through the basic setup of a Ceph cluster and enable K8s ap
 
 ## Kubernetes Version
 
-Kubernetes versions **v1.25** through **v1.29** are supported.
+Kubernetes versions **v1.25** through **v1.30** are supported.
 
 ## CPU Architecture
 

deploy/charts/rook-ceph-cluster/templates/cephecblockpool.yaml

@@ -33,8 +33,8 @@ provisioner: {{ $root.Values.operatorNamespace }}.rbd.csi.ceph.com
 {{- end }}
 parameters:
   clusterID: {{ $ecblockpool.parameters.clusterID }}
-  dataPool: {{ $ecblockpool.name }}-metadata
-  pool: {{ $ecblockpool.name }}
+  pool: {{ $ecblockpool.name }}-metadata
+  dataPool: {{ $ecblockpool.name }}
   imageFormat: "{{ $ecblockpool.parameters.imageFormat }}"
   imageFeatures: {{ $ecblockpool.parameters.imageFeatures }}
 
@@ -58,4 +58,4 @@ parameters:
 allowVolumeExpansion: {{ $ecblockpool.storageClass.allowVolumeExpansion }}
 reclaimPolicy: {{ $ecblockpool.storageClass.reclaimPolicy }}
 {{ end }}
-{{ end }}
+{{ end }}

pkg/operator/ceph/cluster/cluster_external.go

@@ -108,12 +108,6 @@ func (c *ClusterController) configureExternalCephCluster(cluster *cluster) error
 		}
 	}
 
-	// Create CSI config map
-	err = csi.CreateCsiConfigMap(c.OpManagerCtx, c.namespacedName.Namespace, c.context.Clientset, cluster.ownerInfo)
-	if err != nil {
-		return errors.Wrap(err, "failed to create csi config map")
-	}
-
 	// update the msgr2 flag
 	for _, m := range cluster.ClusterInfo.Monitors {
 		// m.Endpoint=10.1.115.104:3300

pkg/operator/ceph/cluster/osd/spec.go

@@ -580,8 +580,9 @@ func (c *Cluster) makeDeployment(osdProps osdProperties, osd OSDInfo, provisionC
 				initContainers = append(initContainers, c.getExpandEncryptedPVCInitContainer(osdDataDirPath, osdProps))
 			}
 			initContainers = append(initContainers, c.getActivatePVCInitContainer(osdProps, osdID))
+			// The expand init container fails for legacy LVM-based OSDs, so only supported expansion for raw mode OSDs
+			initContainers = append(initContainers, c.getExpandPVCInitContainer(osdProps, osdID))
 		}
-		initContainers = append(initContainers, c.getExpandPVCInitContainer(osdProps, osdID))
 	} else {
 		// Add the volume to the spec and the mount to the daemon container
 		// so that it can pick the already mounted/activated osd metadata path

pkg/operator/ceph/cluster/osd/spec_test.go

@@ -198,12 +198,11 @@ func testPodDevices(t *testing.T, dataDir, deviceName string, allDevices bool) {
 	deployment, err = c.makeDeployment(osdProp, osd, dataPathMap)
 	assert.Nil(t, err)
 	assert.NotNil(t, deployment)
-	assert.Equal(t, 5, len(deployment.Spec.Template.Spec.InitContainers), deployment.Spec.Template.Spec.InitContainers[2].Name)
+	assert.Equal(t, 4, len(deployment.Spec.Template.Spec.InitContainers), deployment.Spec.Template.Spec.InitContainers[2].Name)
 	assert.Equal(t, "config-init", deployment.Spec.Template.Spec.InitContainers[0].Name)
 	assert.Equal(t, "copy-bins", deployment.Spec.Template.Spec.InitContainers[1].Name)
 	assert.Equal(t, "blkdevmapper", deployment.Spec.Template.Spec.InitContainers[2].Name)
-	assert.Equal(t, "expand-bluefs", deployment.Spec.Template.Spec.InitContainers[3].Name)
-	assert.Equal(t, "chown-container-data-dir", deployment.Spec.Template.Spec.InitContainers[4].Name)
+	assert.Equal(t, "chown-container-data-dir", deployment.Spec.Template.Spec.InitContainers[3].Name)
 	assert.Equal(t, 1, len(deployment.Spec.Template.Spec.Containers))
 	initCont = deployment.Spec.Template.Spec.InitContainers[0]
 	assert.Equal(t, 5, len(initCont.VolumeMounts), initCont.VolumeMounts)

pkg/operator/ceph/csi/cluster_config.go

@@ -254,12 +254,50 @@ func CreateCsiConfigMap(ctx context.Context, namespace string, clientset kuberne
 		if !k8serrors.IsAlreadyExists(err) {
 			return errors.Wrapf(err, "failed to create initial csi config map %q (in %q)", configMap.Name, namespace)
 		}
+		// CM already exists; update owner refs to it if needed
+		// this corrects issues where the csi config map was sometimes created with CephCluster
+		// owner ref, which would result in the cm being deleted if that cluster was deleted
+		if err := updateCsiConfigMapOwnerRefs(ctx, namespace, clientset, ownerInfo); err != nil {
+			return errors.Wrapf(err, "failed to ensure csi config map %q (in %q) owner references", configMap.Name, namespace)
+		}
 	}
 
 	logger.Infof("successfully created csi config map %q", configMap.Name)
 	return nil
 }
 
+// check the owner references on the csi config map, and fix incorrect references if needed
+func updateCsiConfigMapOwnerRefs(ctx context.Context, namespace string, clientset kubernetes.Interface, expectedOwnerInfo *k8sutil.OwnerInfo) error {
+	cm, err := clientset.CoreV1().ConfigMaps(namespace).Get(ctx, ConfigName, metav1.GetOptions{})
+	if err != nil {
+		return errors.Wrapf(err, "failed to fetch csi config map %q (in %q) which already exists", ConfigName, namespace)
+	}
+
+	existingOwners := cm.GetOwnerReferences()
+	var currentOwner *metav1.OwnerReference = nil
+	if len(existingOwners) == 1 {
+		currentOwner = &existingOwners[0] // currentOwner is nil unless there is exactly one owner on the cm
+	}
+	// if there is exactly one owner, and it is correct --> no fix needed
+	if currentOwner != nil && (currentOwner.UID == expectedOwnerInfo.GetUID()) {
+		logger.Debugf("csi config map %q (in %q) has the expected owner; owner id: %q", ConfigName, namespace, currentOwner.UID)
+		return nil
+	}
+
+	// must fix owner refs
+	logger.Infof("updating csi configmap %q (in %q) owner info", ConfigName, namespace)
+	cm.OwnerReferences = []metav1.OwnerReference{}
+	if err := expectedOwnerInfo.SetControllerReference(cm); err != nil {
+		return errors.Wrapf(err, "failed to set updated owner reference on csi config map %q (in %q)", ConfigName, namespace)
+	}
+	_, err = clientset.CoreV1().ConfigMaps(namespace).Update(ctx, cm, metav1.UpdateOptions{})
+	if err != nil {
+		return errors.Wrapf(err, "failed to update csi config map %q (in %q) to update its owner reference", ConfigName, namespace)
+	}
+
+	return nil
+}
+
 // SaveClusterConfig updates the config map used to provide ceph-csi with
 // basic cluster configuration. The clusterNamespace and clusterInfo are
 // used to determine what "cluster" in the config map will be updated and
@@ -292,10 +330,7 @@ func SaveClusterConfig(clientset kubernetes.Interface, clusterNamespace string,
 	configMap, err := clientset.CoreV1().ConfigMaps(csiNamespace).Get(clusterInfo.Context, ConfigName, metav1.GetOptions{})
 	if err != nil {
 		if k8serrors.IsNotFound(err) {
-			err = CreateCsiConfigMap(clusterInfo.Context, csiNamespace, clientset, clusterInfo.OwnerInfo)
-			if err != nil {
-				return errors.Wrap(err, "failed creating csi config map")
-			}
+			return errors.Wrap(err, "waiting for CSI config map to be created")
 		}
 		return errors.Wrap(err, "failed to fetch current csi config map")
 	}