feature toggle for using maintenance windows (#3074)

* feature toggle for using maintenance windows

charts/postgres-operator/crds/operatorconfigurations.yaml

@@ -79,6 +79,9 @@ spec:
               enable_lazy_spilo_upgrade:
                 type: boolean
                 default: false
+              enable_maintenance_windows:
+                type: boolean
+                default: true
               enable_pgversion_env_var:
                 type: boolean
                 default: true

charts/postgres-operator/values.yaml

@@ -27,6 +27,8 @@ configGeneral:
   - "all"
   # update only the statefulsets without immediately doing the rolling update
   enable_lazy_spilo_upgrade: false
+  # toogle to use maintenance windows feature
+  enable_maintenance_windows: true
   # set the PGVERSION env var instead of providing the version via postgresql.bin_dir in SPILO_CONFIGURATION
   enable_pgversion_env_var: true
   # start any new database pod without limitations on shm memory

docs/administrator.md

@@ -95,11 +95,11 @@ Thus, the `full` mode can create drift between desired and actual state.
 
 ### Upgrade during maintenance windows
 
-When `maintenanceWindows` are defined in the Postgres manifest the operator
+When `maintenanceWindows` are defined in the Postgres manifest or in the global
-will trigger major-version-related pod rotation and the major version upgrade
+config the operator will trigger major-version-related pod rotation and the
-only during these periods. Make sure they are at least twice as long as your
+major version upgrade only during these periods. Make sure they are at least
-configured `resync_period` to guarantee
+twice as long as your configured `resync_period` to guarantee that operator
-that operator actions can be triggered.
+actions can be triggered.
 
 ### Upgrade annotations
 

docs/reference/cluster_manifest.md

@@ -118,7 +118,9 @@ These parameters are grouped directly under  the `spec` key in the manifest.
   a list which defines specific time frames when certain maintenance operations
   such as automatic major upgrades or master pod migration are allowed to happen.
   Accepted formats are "01:00-06:00" for daily maintenance windows or
-  "Sat:00:00-04:00" for specific days, with all times in UTC.
+  "Sat:00:00-04:00" for specific days, with all times in UTC. Note, when the
+  global config option `enable_maintenance_windows` is false, the specified
+  windows will be ignored.
 
 * **users**
   a map of usernames to user flags for the users that should be created in the

docs/reference/operator_parameters.md

@@ -173,6 +173,9 @@ Those are top-level keys, containing both leaf keys and groups.
   the thresholds. The value must be `"true"` to be effective. The default is empty
   which means the feature is disabled.
 
+* **enable_maintenance_windows**
+  toggle for using the maintenance windows feature. Default is `"true"`.
+
 * **maintenance_windows**
   a list which defines specific time frames when certain maintenance
   operations such as automatic major upgrades or master pod migration are

manifests/configmap.yaml

@@ -46,6 +46,7 @@ data:
   enable_ebs_gp3_migration_max_size: "1000"
   enable_init_containers: "true"
   enable_lazy_spilo_upgrade: "false"
+  enable_maintenance_windows: "true"
   enable_master_load_balancer: "false"
   enable_master_pooler_load_balancer: "false"
   enable_password_rotation: "false"

manifests/operatorconfiguration.crd.yaml

@@ -77,6 +77,9 @@ spec:
               enable_lazy_spilo_upgrade:
                 type: boolean
                 default: false
+              enable_maintenance_windows:
+                type: boolean
+                default: true
               enable_pgversion_env_var:
                 type: boolean
                 default: true

manifests/postgresql-operator-default-configuration.yaml

@@ -8,6 +8,7 @@ configuration:
   # crd_categories:
   # - all
   # enable_lazy_spilo_upgrade: false
+  enable_maintenance_windows: true
   enable_pgversion_env_var: true
   # enable_shm_volume: true
   enable_spilo_wal_path_compat: false

pkg/apis/acid.zalan.do/v1/crds.go

@@ -105,6 +105,9 @@ var OperatorConfigCRDResourceValidation = apiextv1.CustomResourceValidation{
 					"enable_lazy_spilo_upgrade": {
 						Type: "boolean",
 					},
+					"enable_maintenance_windows": {
+						Type: "boolean",
+					},
 					"enable_shm_volume": {
 						Type: "boolean",
 					},

pkg/apis/acid.zalan.do/v1/operator_configuration_type.go

@@ -266,6 +266,7 @@ type OperatorConfigurationData struct {
 	Workers                       uint32                             `json:"workers,omitempty"`
 	ResyncPeriod                  Duration                           `json:"resync_period,omitempty"`
 	RepairPeriod                  Duration                           `json:"repair_period,omitempty"`
+	EnableMaintenanceWindows      *bool                              `json:"enable_maintenance_windows,omitempty"`
 	MaintenanceWindows            []MaintenanceWindow                `json:"maintenance_windows,omitempty"`
 	SetMemoryRequestToLimit       bool                               `json:"set_memory_request_to_limit,omitempty"`
 	ShmVolume                     *bool                              `json:"enable_shm_volume,omitempty"`

pkg/apis/acid.zalan.do/v1/zz_generated.deepcopy.go

@@ -433,6 +433,11 @@ func (in *OperatorConfigurationData) DeepCopyInto(out *OperatorConfigurationData
 		*out = make([]string, len(*in))
 		copy(*out, *in)
 	}
+	if in.EnableMaintenanceWindows != nil {
+		in, out := &in.EnableMaintenanceWindows, &out.EnableMaintenanceWindows
+		*out = new(bool)
+		**out = **in
+	}
 	if in.MaintenanceWindows != nil {
 		in, out := &in.MaintenanceWindows, &out.MaintenanceWindows
 		*out = make([]MaintenanceWindow, len(*in))

pkg/cluster/util.go

@@ -675,7 +675,9 @@ func isStandbyCluster(spec *acidv1.PostgresSpec) bool {
 }
 
 func (c *Cluster) isInMaintenanceWindow(specMaintenanceWindows []acidv1.MaintenanceWindow) bool {
-	if len(specMaintenanceWindows) == 0 && len(c.OpConfig.MaintenanceWindows) == 0 {
+	ignoreMaintenanceWindows := c.OpConfig.EnableMaintenanceWindows != nil && !*c.OpConfig.EnableMaintenanceWindows
+	noWindowsDefined := len(specMaintenanceWindows) == 0 && len(c.OpConfig.MaintenanceWindows) == 0
+	if noWindowsDefined || ignoreMaintenanceWindows {
 		return true
 	}
 	now := time.Now()

pkg/cluster/util_test.go

@@ -660,6 +660,7 @@ func TestIsInMaintenanceWindow(t *testing.T) {
 	cluster := New(
 		Config{
 			OpConfig: config.Config{
+				EnableMaintenanceWindows: util.True(),
 				Resources: config.Resources{
 					ClusterLabels:        map[string]string{"application": "spilo"},
 					ClusterNameLabel:     "cluster-name",
@@ -683,12 +684,27 @@ func TestIsInMaintenanceWindow(t *testing.T) {
 		name          string
 		windows       []acidv1.MaintenanceWindow
 		configWindows []string
+		windowsFlag   bool
 		expected      bool
 	}{
 		{
 			name:          "no maintenance windows",
 			windows:       nil,
 			configWindows: nil,
+			windowsFlag:   true,
+			expected:      true,
+		},
+		{
+			name: "maintenance windows diabled",
+			windows: []acidv1.MaintenanceWindow{
+				{
+					Everyday:  true,
+					StartTime: mustParseTime("00:00"),
+					EndTime:   mustParseTime("23:59"),
+				},
+			},
+			configWindows: nil,
+			windowsFlag:   false,
 			expected:      true,
 		},
 		{
@@ -701,6 +717,7 @@ func TestIsInMaintenanceWindow(t *testing.T) {
 				},
 			},
 			configWindows: nil,
+			windowsFlag:   true,
 			expected:      true,
 		},
 		{
@@ -713,6 +730,7 @@ func TestIsInMaintenanceWindow(t *testing.T) {
 				},
 			},
 			configWindows: nil,
+			windowsFlag:   true,
 			expected:      true,
 		},
 		{
@@ -724,24 +742,35 @@ func TestIsInMaintenanceWindow(t *testing.T) {
 					EndTime:   mustParseTime(futureTimeEndFormatted),
 				},
 			},
-			expected: false,
+			windowsFlag: true,
+			expected:    false,
 		},
 		{
 			name:          "global maintenance windows with future interval time",
 			windows:       nil,
 			configWindows: []string{fmt.Sprintf("%s-%s", futureTimeStartFormatted, futureTimeEndFormatted)},
+			windowsFlag:   true,
 			expected:      false,
 		},
 		{
 			name:          "global maintenance windows all day",
 			windows:       nil,
 			configWindows: []string{"00:00-02:00", "02:00-23:59"},
+			windowsFlag:   true,
+			expected:      true,
+		},
+		{
+			name:          "global maintenance windows ignored",
+			windows:       nil,
+			configWindows: []string{"00:00-02:00", "02:00-23:59"},
+			windowsFlag:   false,
 			expected:      true,
 		},
 	}
 
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
+			cluster.OpConfig.EnableMaintenanceWindows = &tt.windowsFlag
 			cluster.OpConfig.MaintenanceWindows = tt.configWindows
 			cluster.Spec.MaintenanceWindows = tt.windows
 			if cluster.isInMaintenanceWindow(cluster.Spec.MaintenanceWindows) != tt.expected {

pkg/controller/operator_config.go

@@ -51,6 +51,7 @@ func (c *Controller) importConfigurationFromCRD(fromCRD *acidv1.OperatorConfigur
 	result.ShmVolume = util.CoalesceBool(fromCRD.ShmVolume, util.True())
 	result.SidecarImages = fromCRD.SidecarImages
 	result.SidecarContainers = fromCRD.SidecarContainers
+	result.EnableMaintenanceWindows = util.CoalesceBool(fromCRD.EnableMaintenanceWindows, util.True())
 	if len(fromCRD.MaintenanceWindows) > 0 {
 		result.MaintenanceWindows = make([]string, 0, len(fromCRD.MaintenanceWindows))
 		for _, window := range fromCRD.MaintenanceWindows {

pkg/util/config/config.go

@@ -173,14 +173,15 @@ type Config struct {
 	LogicalBackup
 	ConnectionPooler
 
-	WatchedNamespace        string            `name:"watched_namespace"` // special values: "*" means 'watch all namespaces', the empty string "" means 'watch a namespace where operator is deployed to'
+	WatchedNamespace         string            `name:"watched_namespace"` // special values: "*" means 'watch all namespaces', the empty string "" means 'watch a namespace where operator is deployed to'
-	KubernetesUseConfigMaps bool              `name:"kubernetes_use_configmaps" default:"false"`
+	KubernetesUseConfigMaps  bool              `name:"kubernetes_use_configmaps" default:"false"`
-	EtcdHost                string            `name:"etcd_host" default:""` // special values: the empty string "" means Patroni will use K8s as a DCS
+	EtcdHost                 string            `name:"etcd_host" default:""` // special values: the empty string "" means Patroni will use K8s as a DCS
-	MaintenanceWindows      []string          `name:"maintenance_windows"`
+	EnableMaintenanceWindows *bool             `name:"enable_maintenance_windows" default:"true"`
-	DockerImage             string            `name:"docker_image" default:"ghcr.io/zalando/spilo-18:4.1-p1"`
+	MaintenanceWindows       []string          `name:"maintenance_windows"`
-	SidecarImages           map[string]string `name:"sidecar_docker_images"` // deprecated in favour of SidecarContainers
+	DockerImage              string            `name:"docker_image" default:"ghcr.io/zalando/spilo-18:4.1-p1"`
-	SidecarContainers       []v1.Container    `name:"sidecars"`
+	SidecarImages            map[string]string `name:"sidecar_docker_images"` // deprecated in favour of SidecarContainers
-	PodServiceAccountName   string            `name:"pod_service_account_name" default:"postgres-pod"`
+	SidecarContainers        []v1.Container    `name:"sidecars"`
+	PodServiceAccountName    string            `name:"pod_service_account_name" default:"postgres-pod"`
 	// value of this string must be valid JSON or YAML; see initPodServiceAccount
 	PodServiceAccountDefinition              string            `name:"pod_service_account_definition" default:""`
 	PodServiceAccountRoleBindingDefinition   string            `name:"pod_service_account_role_binding_definition" default:""`