Merge pull request #134 from gianlucam76/main

Merge dev to main

.github/workflows/main.yaml

@@ -16,9 +16,9 @@ jobs:
     - name: checkout
       uses: actions/checkout@v4
     - name: Set up Go
-      uses: actions/setup-go@v3
+      uses: actions/setup-go@v4
       with:
-        go-version: 1.20.0
+        go-version: 1.20.8
     - name: Build
       run: make build
     - name: FMT
@@ -35,9 +35,9 @@ jobs:
     - name: checkout
       uses: actions/checkout@v4
     - name: Set up Go
-      uses: actions/setup-go@v3
+      uses: actions/setup-go@v4
       with:
-        go-version: 1.20.0
+        go-version: 1.20.8
     - name: ut
       run: make test
       env:
@@ -48,10 +48,23 @@ jobs:
     - name: checkout
       uses: actions/checkout@v4
     - name: Set up Go
-      uses: actions/setup-go@v3
+      uses: actions/setup-go@v4
       with:
-        go-version: 1.20.0
+        go-version: 1.20.8
     - name: fv
       run: make create-cluster fv
       env:
         FV: true
+  FV-SHARDING:
+    runs-on: ubuntu-latest
+    steps:
+    - name: checkout
+      uses: actions/checkout@v4
+    - name: Set up Go
+      uses: actions/setup-go@v4
+      with:
+        go-version: 1.20.8
+    - name: fv
+      run: make create-cluster fv-sharding
+      env:
+        FV: true

.gitignore

@@ -31,3 +31,6 @@ manager_image_patch.yaml-e
 manager_pull_policy.yaml-e
 
 version.txt
+
+ac-deployment-shard.yaml
+tmp-ac-deployment-shard.yaml

Makefile

@@ -27,15 +27,6 @@ K8S_LATEST_VER ?= $(shell curl -s https://storage.googleapis.com/kubernetes-rele
 export CONTROLLER_IMG ?= $(REGISTRY)/$(IMAGE_NAME)
 TAG ?= main
 
-# Get cluster-api version and build ldflags
-clusterapi := $(shell go list -m sigs.k8s.io/cluster-api)
-clusterapi_version := $(lastword  ., ,$(clusterapi))
-clusterapi_version_tuple := $(subst ., ,$(clusterapi_version:v%=%))
-clusterapi_major := $(word 1,$(clusterapi_version_tuple))
-clusterapi_minor := $(word 2,$(clusterapi_version_tuple))
-clusterapi_patch := $(word 3,$(clusterapi_version_tuple))
-CLUSTERAPI_LDFLAGS := "-X 'sigs.k8s.io/cluster-api/version.gitMajor=$(clusterapi_major)' -X 'sigs.k8s.io/cluster-api/version.gitMinor=$(clusterapi_minor)' -X 'sigs.k8s.io/cluster-api/version.gitVersion=$(clusterapi_version)'"
-
 .PHONY: all
 all: build
 
@@ -78,6 +69,7 @@ KIND := $(TOOLS_BIN_DIR)/kind
 KUBECTL := $(TOOLS_BIN_DIR)/kubectl
 
 GOLANGCI_LINT_VERSION := "v1.52.2"
+CLUSTERCTL_VERSION := "v1.5.3"
 
 $(CONTROLLER_GEN): $(TOOLS_DIR)/go.mod # Build controller-gen from tools folder.
 	cd $(TOOLS_DIR); $(GOBUILD) -tags=tools -o $(subst $(TOOLS_DIR)/hack/tools/,,$@) sigs.k8s.io/controller-tools/cmd/controller-gen
@@ -101,7 +93,8 @@ $(KIND): $(TOOLS_DIR)/go.mod
 	cd $(TOOLS_DIR) && $(GOBUILD) -tags tools -o $(subst $(TOOLS_DIR)/hack/tools/,,$@) sigs.k8s.io/kind
 
 $(CLUSTERCTL): $(TOOLS_DIR)/go.mod ## Build clusterctl binary
-	cd $(TOOLS_DIR); $(GOBUILD) -trimpath  -ldflags $(CLUSTERAPI_LDFLAGS) -o $(subst $(TOOLS_DIR)/hack/tools/,,$@) sigs.k8s.io/cluster-api/cmd/clusterctl
+	curl -L https://github.com/kubernetes-sigs/cluster-api/releases/download/$(CLUSTERCTL_VERSION)/clusterctl-$(OS)-$(ARCH) -o $@
+	chmod +x $@
 	mkdir -p $(HOME)/.cluster-api # create cluster api init directory, if not present	
 
 $(KUBECTL):
@@ -131,6 +124,7 @@ manifests: $(CONTROLLER_GEN) $(KUSTOMIZE) $(ENVSUBST) fmt generate ## Generate W
 	$(CONTROLLER_GEN) rbac:roleName=manager-role crd webhook paths="./..." output:crd:artifacts:config=config/crd/bases
 	MANIFEST_IMG=$(CONTROLLER_IMG)-$(ARCH) MANIFEST_TAG=$(TAG) $(MAKE) set-manifest-image
 	$(KUSTOMIZE) build config/default | $(ENVSUBST) > manifest/manifest.yaml
+	./scripts/extract_deployment.sh manifest/manifest.yaml manifest/deployment-shard.yaml
 
 .PHONY: generate
 generate: $(CONTROLLER_GEN) ## Generate code containing DeepCopy, DeepCopyInto, and DeepCopyObject method implementations.
@@ -186,6 +180,14 @@ kind-test: test create-cluster fv ## Build docker image; start kind cluster; loa
 fv: $(GINKGO) ## Run Sveltos Controller tests using existing cluster
 	cd test/fv; $(GINKGO) -nodes $(NUM_NODES) --label-filter='FV' --v --trace --randomize-all
 
+.PHONY: fv-sharding
+fv-sharding: $(KUBECTL) $(GINKGO) ## Run Sveltos Controller tests using existing cluster
+	$(KUBECTL) patch cluster clusterapi-workload  -n default --type json -p '[{ "op": "add", "path": "/metadata/annotations/sharding.projectsveltos.io~1key", "value": "shard1" }]'
+	sed -e "s/{{.SHARD}}/shard1/g"  manifest/deployment-shard.yaml > test/hc-deployment-shard.yaml
+	$(KUBECTL) apply -f test/hc-deployment-shard.yaml
+	rm -f test/hc-deployment-shard.yaml
+	cd test/fv; $(GINKGO) -nodes $(NUM_NODES) --label-filter='FV' --v --trace --randomize-all
+
 .PHONY: create-cluster
 create-cluster: $(KIND) $(CLUSTERCTL) $(KUBECTL) $(ENVSUBST) ## Create a new kind cluster designed for development
 	$(MAKE) create-control-cluster
@@ -272,9 +274,13 @@ deploy-projectsveltos: $(KUSTOMIZE) $(KUBECTL)
 	# Install addon-compliance-controller
 	$(KUBECTL) apply -f https://raw.githubusercontent.com/projectsveltos/addon-compliance-controller/$(TAG)/manifest/manifest.yaml
 
-	# Install sveltos-manager
-	$(KUBECTL) apply -f https://raw.githubusercontent.com/projectsveltos/sveltos-manager/$(TAG)/manifest/manifest.yaml
-
+	# Install projectsveltos addon-controller
+	$(KUBECTL) apply -f https://raw.githubusercontent.com/projectsveltos/addon-controller/$(TAG)/manifest/manifest.yaml
+	curl https://raw.githubusercontent.com/projectsveltos/addon-controller/$(TAG)/manifest/deployment-shard.yaml -o ac-deployment-shard.yaml
+	sed -e "s/{{.SHARD}}/shard1/g"  ac-deployment-shard.yaml > tmp-ac-deployment-shard.yaml
+	$(KUBECTL) apply -f tmp-ac-deployment-shard.yaml
+	rm ac-deployment-shard.yaml
+	rm tmp-ac-deployment-shard.yaml
 
 	# Install projectsveltos healthcheck-manager components
 	@echo 'Install projectsveltos controller-manager components'

cmd/main.go

@@ -57,7 +57,7 @@ import (
 var (
 	setupLog                     = ctrl.Log.WithName("setup")
 	metricsAddr                  string
-	enableLeaderElection         bool
+	shardKey                     string
 	probeAddr                    string
 	workers                      int
 	concurrentReconciles         int
@@ -95,8 +95,6 @@ func main() {
 		MetricsBindAddress:     metricsAddr,
 		Port:                   9443,
 		HealthProbeBindAddress: probeAddr,
-		LeaderElection:         enableLeaderElection,
-		LeaderElectionID:       "7964f530.projectsveltos.io",
 	})
 	if err != nil {
 		setupLog.Error(err, "unable to start manager")
@@ -122,6 +120,7 @@ func main() {
 		ConcurrentReconciles: concurrentReconciles,
 		Mux:                  sync.Mutex{},
 		Deployer:             d,
+		ShardKey:             shardKey,
 		ClusterMap:           make(map[corev1.ObjectReference]*libsveltosset.Set),
 		CHCToClusterMap:      make(map[types.NamespacedName]*libsveltosset.Set),
 		ClusterHealthChecks:  make(map[corev1.ObjectReference]libsveltosv1alpha1.Selector),
@@ -139,6 +138,7 @@ func main() {
 		Client:                mgr.GetClient(),
 		Scheme:                mgr.GetScheme(),
 		HealthCheckReportMode: reportMode,
+		ShardKey:              shardKey,
 	}).SetupWithManager(mgr); err != nil {
 		setupLog.Error(err, "unable to create controller", "controller", "HealthCheck")
 		os.Exit(1)
@@ -187,9 +187,8 @@ func initFlags(fs *pflag.FlagSet) {
 	fs.StringVar(&probeAddr, "health-probe-bind-address", ":8081",
 		"The address the probe endpoint binds to.")
 
-	fs.BoolVar(&enableLeaderElection, "leader-elect", false,
-		"Enable leader election for controller manager. "+
-			"Enabling this will ensure there is only one active controller manager.")
+	fs.StringVar(&shardKey, "shard-key", "",
+		"If set, and report-mode is set to collect, this deployment will fetch only from clusters matching this shard")
 
 	fs.IntVar(&workers, "worker-number", defaultWorkers,
 		"Number of worker. Workers are used to verify health checks in managed clusters")

config/default/kustomization.yaml

@@ -31,10 +31,6 @@ patchesStrategicMerge:
 - manager_image_patch.yaml
 - manager_pull_policy.yaml
 
-# Mount the controller config file for loading manager configurations
-# through a ComponentConfig type
-#- manager_config_patch.yaml
-
 # [WEBHOOK] To enable webhook, uncomment all the sections with [WEBHOOK] prefix including the one in
 # crd/kustomization.yaml
 #- manager_webhook_patch.yaml

config/default/manager_auth_proxy_patch.yaml

@@ -36,5 +36,5 @@ spec:
         args:
         - "--health-probe-bind-address=:8081"
         - "--metrics-bind-address=127.0.0.1:8080"
-        - "--leader-elect"
+        - "--shard-key="
         - "--v=5"

config/manager/kustomization.yaml

@@ -4,10 +4,6 @@ resources:
 generatorOptions:
   disableNameSuffixHash: true
 
-configMapGenerator:
-- files:
-  - controller_manager_config.yaml
-  name: manager-config
 apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 images:

config/manager/manager.yaml

@@ -35,7 +35,6 @@ spec:
       - command:
         - /manager
         args:
-        - --leader-elect
         image: controller:latest
         name: manager
         securityContext:

config/rbac/kustomization.yaml

@@ -7,8 +7,6 @@ resources:
 - service_account.yaml
 - role.yaml
 - role_binding.yaml
-- leader_election_role.yaml
-- leader_election_role_binding.yaml
 # Comment the following 4 lines if you want to disable
 # the auth proxy (https://github.com/brancz/kube-rbac-proxy)
 # which protects your /metrics endpoint.

controllers/clusterhealthcheck_controller.go

@@ -77,7 +77,7 @@ type ClusterHealthCheckReconciler struct {
 	Scheme               *runtime.Scheme
 	ConcurrentReconciles int
 	Deployer             deployer.DeployerInterface
-
+	ShardKey             string // when set, only clusters matching the ShardKey will be reconciled
 	// use a Mutex to update Map as MaxConcurrentReconciles is higher than one
 	Mux sync.Mutex
 
@@ -182,6 +182,7 @@ func (r *ClusterHealthCheckReconciler) Reconcile(ctx context.Context, req ctrl.R
 	// changes.
 	defer func() {
 		if err := clusterHealthCheckScope.Close(ctx); err != nil {
+			logger.V(logs.LogInfo).Info(fmt.Sprintf("failed to update: %v", err))
 			reterr = err
 		}
 	}()