merge with 0.24.1

deploy/helm/clickhouse-operator/templates/generated/ServiceAccount-clickhouse-operator.yaml

@@ -13,7 +13,6 @@ metadata:
   namespace: {{ .Release.Namespace }}
   labels: {{ include "altinity-clickhouse-operator.labels" . | nindent 4 }}
   annotations: {{ toYaml .Values.serviceAccount.annotations | nindent 4 }}
-
 # Template Parameters:
 #
 # NAMESPACE=kube-system

deploy/helm/clickhouse-operator/values.yaml

@@ -3,92 +3,67 @@ operator:
     # operator.image.repository -- image repository
     repository: altinity/clickhouse-operator
     # operator.image.tag -- image tag (chart's appVersion value will be used if not set)
-
     tag: ""
     # operator.image.pullPolicy -- image pull policy
-
     pullPolicy: IfNotPresent
   containerSecurityContext: {}
-  # operator.resources -- custom resource configuration, look `kubectl explain pod.spec.containers.resources` for details
-
+  # operator.resources -- custom resource configuration, check `kubectl explain pod.spec.containers.resources` for details
   resources: {}
   #  limits:
-
   #    cpu: 100m
-
   #    memory: 128Mi
-
   #  requests:
-
   #    cpu: 100m
-
   #    memory: 128Mi
 
   # operator.env -- additional environment variables for the clickhouse-operator container in deployment
-
-  # possible format value [{"name": "SAMPLE", "value": "text"}]
+  # possible format value `[{"name": "SAMPLE", "value": "text"}]`
   env: []
 metrics:
   enabled: true
   image:
     # metrics.image.repository -- image repository
     repository: altinity/metrics-exporter
     # metrics.image.tag -- image tag (chart's appVersion value will be used if not set)
-
     tag: ""
     # metrics.image.pullPolicy -- image pull policy
-
     pullPolicy: IfNotPresent
   containerSecurityContext: {}
   # metrics.resources -- custom resource configuration
-
   resources: {}
   #  limits:
-
   #    cpu: 100m
-
   #    memory: 128Mi
-
   #  requests:
-
   #    cpu: 100m
-
   #    memory: 128Mi
 
   # metrics.env -- additional environment variables for the deployment of metrics-exporter containers
-
-  # possible format value [{"name": "SAMPLE", "value": "text"}]
+  # possible format value `[{"name": "SAMPLE", "value": "text"}]`
   env: []
 # imagePullSecrets -- image pull secret for private images in clickhouse-operator pod
-
-#  possible value format [{"name":"your-secret-name"}]
-
-#  look `kubectl explain pod.spec.imagePullSecrets` for details
+# possible value format `[{"name":"your-secret-name"}]`,
+# check `kubectl explain pod.spec.imagePullSecrets` for details
 imagePullSecrets: []
 # podLabels -- labels to add to the clickhouse-operator pod
-
 podLabels: {}
-# podAnnotations -- annotations to add to the clickhouse-operator pod, look `kubectl explain pod.spec.annotations` for details
-
+# podAnnotations -- annotations to add to the clickhouse-operator pod, check `kubectl explain pod.spec.annotations` for details
+# @default -- check the `values.yaml` file
 podAnnotations:
   prometheus.io/port: '8888'
   prometheus.io/scrape: 'true'
   clickhouse-operator-metrics/port: '9999'
   clickhouse-operator-metrics/scrape: 'true'
 # nameOverride -- override name of the chart
-
 nameOverride: ""
 # fullnameOverride -- full name of the chart.
-
 fullnameOverride: ""
 serviceAccount:
   # serviceAccount.create -- specifies whether a service account should be created
   create: true
   # serviceAccount.annotations -- annotations to add to the service account
-
   annotations: {}
   # serviceAccount.name -- the name of the service account to use; if not set and create is true, a name is generated using the fullname template
-
   name:
 rbac:
   # rbac.create -- specifies whether cluster roles and cluster role bindings should be created
@@ -97,35 +72,26 @@ secret:
   # secret.create -- create a secret with operator credentials
   create: true
   # secret.username -- operator credentials username
-
   username: clickhouse_operator
   # secret.password -- operator credentials password
-
   password: clickhouse_operator_password
-# nodeSelector -- node for scheduler pod assignment, look `kubectl explain pod.spec.nodeSelector` for details
-
+# nodeSelector -- node for scheduler pod assignment, check `kubectl explain pod.spec.nodeSelector` for details
 nodeSelector: {}
-# tolerations -- tolerations for scheduler pod assignment, look `kubectl explain pod.spec.tolerations` for details
-
+# tolerations -- tolerations for scheduler pod assignment, check `kubectl explain pod.spec.tolerations` for details
 tolerations: []
-# affinity -- affinity for scheduler pod assignment, look `kubectl explain pod.spec.affinity` for details
-
+# affinity -- affinity for scheduler pod assignment, check `kubectl explain pod.spec.affinity` for details
 affinity: {}
-# podSecurityContext - operator deployment SecurityContext, look `kubectl explain pod.spec.securityContext` for details
-
+# podSecurityContext - operator deployment SecurityContext, check `kubectl explain pod.spec.securityContext` for details
 podSecurityContext: {}
-# topologySpreadConstraints - topologySpreadConstraints affinity for scheduler pod assignment, look `kubectl explain pod.spec.topologySpreadConstraints` for details
-
+# topologySpreadConstraints - topologySpreadConstraints affinity for scheduler pod assignment, check `kubectl explain pod.spec.topologySpreadConstraints` for details
 topologySpreadConstraints: []
 serviceMonitor:
-  # serviceMonitor.enabled -- ServiceMonitor Custom resource is created for a (prometheus-operator)[https://github.com/prometheus-operator/prometheus-operator]
+  # serviceMonitor.enabled -- ServiceMonitor Custom resource is created for a [prometheus-operator](https://github.com/prometheus-operator/prometheus-operator)
   enabled: false
   # serviceMonitor.additionalLabels -- additional labels for service monitor
-
   additionalLabels: {}
-# configs -- clickhouse-operator configs
-
-# @default -- check the values.yaml file for the config content, auto-generated from latest operator release
+# configs -- clickhouse operator configs
+# @default -- check the `values.yaml` file for the config content (auto-generated from latest operator release)
 configs:
   confdFiles: null
   configdFiles:
@@ -782,68 +748,43 @@ configs:
     readme: |-
       Templates in this folder are packaged with an operator and available via 'useTemplate'
   keeperUsersdFiles: null
-# additionalResources -- list of additional resources to create (are processed via `tpl` function), useful for create ClickHouse clusters together with clickhouse-operator, look `kubectl explain chi` for details
-
+# additionalResources -- list of additional resources to create (processed via `tpl` function),
+# useful for create ClickHouse clusters together with clickhouse-operator.
+# check `kubectl explain chi` for details
 additionalResources: []
 #  - |
-
 #    apiVersion: v1
-
 #    kind: ConfigMap
-
 #    metadata:
-
 #      name: {{ include "altinity-clickhouse-operator.fullname" . }}-cm
-
 #      namespace: {{ .Release.Namespace }}
-
 #  - |
-
 #     apiVersion: v1
-
 #     kind: Secret
-
 #     metadata:
-
 #       name: {{ include "altinity-clickhouse-operator.fullname" . }}-s
-
 #       namespace: {{ .Release.Namespace }}
-
 #     stringData:
-
 #       mykey: my-value
-
 #  - |
-
 #     apiVersion: clickhouse.altinity.com/v1
-
 #     kind: ClickHouseInstallation
-
 #     metadata:
-
 #       name: {{ include "altinity-clickhouse-operator.fullname" . }}-chi
-
 #       namespace: {{ .Release.Namespace }}
-
 #     spec:
-
 #       configuration:
-
 #         clusters:
-
 #         - name: default
-
 #           layout:
-
 #             shardsCount: 1
+
 dashboards:
   # dashboards.enabled -- provision grafana dashboards as configMaps (can be synced by grafana dashboards sidecar https://github.com/grafana/helm-charts/blob/grafana-8.3.4/charts/grafana/values.yaml#L778 )
   enabled: false
   # dashboards.additionalLabels -- labels to add to a secret with dashboards
-
   additionalLabels:
     grafana_dashboard: ""
   # dashboards.annotations -- annotations to add to a secret with dashboards
-
   annotations: {}
   grafana_folder: clickhouse

dev/generate_helm_chart.sh

@@ -4,9 +4,9 @@ function usage() {
   cat << EOT
  Script splits clickhouse-operator-install-bundle.yaml to separate files and adjusts them to conform the helm standards
  NOTE script requires some pre-installed tools:
- - yq ( https://mikefarah.gitbook.io/yq/ ) > v4.14.x. Do not use brew install yq in MacOS，Version is lower than it.
+ - yq ( https://mikefarah.gitbook.io/yq/ ) > v4.14.x
  - jq ( https://github.com/stedolan/jq )
- - helm-docs ( https://github.com/norwoodj/helm-docs )
+ - helm-docs ( https://github.com/norwoodj/helm-docs ) > v1.14.x
  - perl ( https://learn.perl.org/installing/ )
 
  Usage: ./generate_helm_chart.sh
@@ -70,7 +70,7 @@ function main() {
   done
 
   if [[ $(command -v helm-docs) ]]; then
-    helm-docs --chart-search-root="${chart_path}" --log-level=warning
+    helm-docs --skip-version-footer --chart-search-root="${chart_path}" --log-level=warning
   else
     echo "WARNING"
     echo "helm-docs is not available, skip docs generation"

pkg/announcer/announcer.go

@@ -351,7 +351,7 @@ func (a Announcer) tryToFindNamespaceNameEverywhere(m interface{}) (string, bool
 	return "", false
 }
 
-// findInObjectMeta
+// findNamespaceName
 func (a Announcer) findNamespaceName(m interface{}) (string, bool) {
 	if m == nil {
 		return "", false

pkg/controller/chk/worker-chk-reconciler.go

@@ -168,6 +168,21 @@ func (w *worker) reconcileCRAuxObjectsPreliminary(ctx context.Context, cr *apiCh
 		w.a.F().Error("failed to reconcile config map users. err: %v", err)
 	}
 
+	return w.reconcileCRAuxObjectsPreliminaryDomain(ctx, cr)
+}
+
+func (w *worker) reconcileCRAuxObjectsPreliminaryDomain(ctx context.Context, cr *apiChk.ClickHouseKeeperInstallation) error {
+	switch {
+	case cr.HostsCount() < cr.GetAncestor().HostsCount():
+		// Downscale
+		time.Sleep(120 * time.Second)
+	case cr.HostsCount() > cr.GetAncestor().HostsCount():
+		// Upscale
+		time.Sleep(30 * time.Second)
+	default:
+		// Same size
+		time.Sleep(10 * time.Second)
+	}
 	return nil
 }
 
@@ -577,9 +592,9 @@ func (w *worker) reconcileHostMain(ctx context.Context, host *api.Host) error {
 		reconcileStatefulSetOpts *statefulset.ReconcileOptions
 	)
 
-	if !host.IsLast() {
-		reconcileStatefulSetOpts = reconcileStatefulSetOpts.SetDoNotWait()
-	}
+	//if !host.IsLast() {
+	//	reconcileStatefulSetOpts = reconcileStatefulSetOpts.SetDoNotWait()
+	//}
 
 	if err := w.reconcileConfigMapHost(ctx, host); err != nil {
 		w.a.V(1).
@@ -607,6 +622,8 @@ func (w *worker) reconcileHostMain(ctx context.Context, host *api.Host) error {
 			Info("Data loss detected for host: %s. Will do force migrate", host.GetName())
 	}
 
+	_ = w.reconcileHostService(ctx, host)
+
 	if err := w.reconcileHostStatefulSet(ctx, host, reconcileStatefulSetOpts); err != nil {
 		w.a.V(1).
 			M(host).F().
@@ -620,8 +637,23 @@ func (w *worker) reconcileHostMain(ctx context.Context, host *api.Host) error {
 		storage.NewStoragePVC(kube.NewPVC(w.c.Client)),
 	).ReconcilePVCs(ctx, host, api.DesiredStatefulSet)
 
-	_ = w.reconcileHostService(ctx, host)
+	// _ = w.reconcileHostService(ctx, host)
+
+	return w.reconcileHostMainDomain(ctx, host)
+}
+
+func (w *worker) reconcileHostMainDomain(ctx context.Context, host *api.Host) error {
+	// Should we wait for host to startup
+	wait := false
 
+	if host.GetReconcileAttributes().IsAdd() {
+		wait = true
+	}
+
+	// Wait for host to startup
+	if wait {
+		time.Sleep(7 * time.Second)
+	}
 	return nil
 }
 

pkg/controller/common/statefulset/statefulset-reconciler.go

@@ -334,7 +334,7 @@ func (r *Reconciler) createStatefulSet(ctx context.Context, host *api.Host, regi
 func (r *Reconciler) waitForConfigMapPropagation(ctx context.Context, host *api.Host) bool {
 	// No need to wait for ConfigMap propagation on stopped host
 	if host.IsStopped() {
-		r.a.V(1).M(host).F().Info("No need to wait for ConfigMap propagation - on stopped host")
+		r.a.V(1).M(host).F().Info("No need to wait for ConfigMap propagation - host is stopped")
 		return false
 	}
 
@@ -347,7 +347,7 @@ func (r *Reconciler) waitForConfigMapPropagation(ctx context.Context, host *api.
 	// What timeout is expected to be enough for ConfigMap propagation?
 	// In case timeout is not specified, no need to wait
 	if !host.GetCR().GetReconciling().HasConfigMapPropagationTimeout() {
-		r.a.V(1).M(host).F().Info("No need to wait for ConfigMap propagation - not applicable")
+		r.a.V(1).M(host).F().Info("No need to wait for ConfigMap propagation - not applicable due to missing timeout value")
 		return false
 	}
 
@@ -357,18 +357,19 @@ func (r *Reconciler) waitForConfigMapPropagation(ctx context.Context, host *api.
 	// May be there is no need to wait already
 	elapsed := time.Now().Sub(r.task.CmUpdate())
 	if elapsed >= timeout {
-		r.a.V(1).M(host).F().Info("No need to wait for ConfigMap propagation - already elapsed. %s/%s", elapsed, timeout)
+		r.a.V(1).M(host).F().Info("No need to wait for ConfigMap propagation - already elapsed. [elapsed/timeout: %s/%s]", elapsed, timeout)
 		return false
 	}
 
 	// Looks like we need to wait for Configmap propagation, after all
 	wait := timeout - elapsed
-	r.a.V(1).M(host).F().Info("Wait for ConfigMap propagation for %s %s/%s", wait, elapsed, timeout)
+	r.a.V(1).M(host).F().Info("Going to wait for ConfigMap propagation for: %s [elapsed/timeout: %s/%s]", wait, elapsed, timeout)
 	if util.WaitContextDoneOrTimeout(ctx, wait) {
 		log.V(2).Info("task is done")
 		return true
 	}
 
+	r.a.V(1).M(host).F().Info("Wait completed for: %s  of timeout: %s]", wait, timeout)
 	return false
 }
 

pkg/model/chk/creator/probe.go

@@ -36,7 +36,8 @@ func (m *ProbeManager) CreateProbe(what interfaces.ProbeType, host *api.Host) *c
 	case interfaces.ProbeDefaultLiveness:
 		return m.createDefaultLivenessProbe(host)
 	case interfaces.ProbeDefaultReadiness:
-		return m.createDefaultReadinessProbe(host)
+		return nil
+		//return m.createDefaultReadinessProbe(host)
 	}
 	panic("unknown probe type")
 }
@@ -53,9 +54,9 @@ func (m *ProbeManager) createDefaultLivenessProbe(host *api.Host) *core.Probe {
 				},
 			},
 		},
-		InitialDelaySeconds: 60,
-		PeriodSeconds:       3,
-		FailureThreshold:    10,
+		InitialDelaySeconds: 5,
+		PeriodSeconds:       5,
+		FailureThreshold:    12,
 	}
 }
 
@@ -77,7 +78,8 @@ func (m *ProbeManager) createDefaultReadinessProbe(host *api.Host) *core.Probe {
 				Port: intstr.Parse("9182"),
 			},
 		},
-		InitialDelaySeconds: 10,
-		PeriodSeconds:       3,
+		InitialDelaySeconds: 5,
+		PeriodSeconds:       5,
+		FailureThreshold:    12,
 	}
 }