feat: add clusterName configuration

Signed-off-by: lenglet-k <[email protected]>

api/v1/backup_types.go

@@ -33,6 +33,10 @@ type BackupSpec struct {
 	// +optional
 	PromURL string `json:"promURL,omitempty"`
 
+	// ClusterName sets athe kubernetes cluster name to send to pushgateway for grouping metrics
+	// +optional
+	ClusterName string `json:"clusterName,omitempty"`
+
 	// StatsURL sets an arbitrary URL where the restic container posts metrics and
 	// information about the snapshots to. This is in addition to the prometheus
 	// pushgateway.

api/v1/check_types.go

@@ -18,6 +18,10 @@ type CheckSpec struct {
 	// +optional
 	PromURL string `json:"promURL,omitempty"`
 
+	// ClusterName sets athe kubernetes cluster name to send to pushgateway for grouping metrics
+	// +optional
+	ClusterName string `json:"clusterName,omitempty"`
+
 	// KeepJobs amount of jobs to keep for later analysis.
 	//
 	// Deprecated: Use FailedJobsHistoryLimit and SuccessfulJobsHistoryLimit respectively.

cmd/operator/main.go

@@ -73,7 +73,7 @@ var (
 			&cli.StringFlag{Destination: &cfg.Config.GlobalStatsURL, Name: "globalstatsurl", EnvVars: []string{"BACKUP_GLOBALSTATSURL"}, Usage: "set the URL to post metrics globally"},
 			&cli.StringFlag{Destination: &cfg.Config.MetricsBindAddress, Name: "metrics-bindaddress", EnvVars: []string{"BACKUP_METRICS_BINDADDRESS"}, Value: ":8080", Usage: "set the bind address for the prometheus endpoint"},
 			&cli.StringFlag{Destination: &cfg.Config.PromURL, Name: "promurl", EnvVars: []string{"BACKUP_PROMURL"}, Value: "http://127.0.0.1/", Usage: "set the operator wide default prometheus push gateway"},
-
+			&cli.StringFlag{Destination: &cfg.Config.ClusterName, Name: "clusterName", EnvVars: []string{"CLUSTER_NAME"}, Value: "default", Usage: "set the operator wide kubernetes cluster name to send to push gateway for grouping metrics"},
 			&cli.StringFlag{Destination: &cfg.Config.RestartPolicy, Name: "restartpolicy", EnvVars: []string{"BACKUP_RESTARTPOLICY"}, Value: "OnFailure", Usage: "set the RestartPolicy for the backup jobs. According to https://kubernetes.io/docs/concepts/workloads/controllers/jobs-run-to-completion/, this should be 'OnFailure' for jobs that terminate"},
 			&cli.StringFlag{Destination: &cfg.Config.PodFilter, Name: "podfilter", EnvVars: []string{"BACKUP_PODFILTER"}, Value: "backupPod=true", Usage: "the filter used to find the backup pods"},
 			&cli.StringFlag{Destination: &cfg.Config.ServiceAccount, Name: "podexecaccountname", Aliases: []string{"serviceaccount"}, EnvVars: []string{"BACKUP_PODEXECACCOUNTNAME"}, Value: "pod-executor", Usage: "set the service account name that should be used for the pod command execution"},

cmd/restic/integration_test.go

@@ -91,9 +91,10 @@ func initTest(t *testing.T) *testEnvironment {
 	ctx := context.Background()
 
 	cfg.Config = &cfg.Configuration{
-		Hostname:   os.Getenv("HOSTNAME"),
-		PromURL:    os.Getenv("PROM_URL"),
-		WebhookURL: os.Getenv("STATS_URL"),
+		Hostname:    os.Getenv("HOSTNAME"),
+		PromURL:     os.Getenv("PROM_URL"),
+		ClusterName: os.Getenv("CLUSTER_NAME"),
+		WebhookURL:  os.Getenv("STATS_URL"),
 
 		RestoreS3Endpoint:  os.Getenv("RESTORE_S3ENDPOINT"),
 		RestoreS3AccessKey: os.Getenv("RESTORE_ACCESSKEYID"),
@@ -105,7 +106,7 @@ func initTest(t *testing.T) *testEnvironment {
 	}
 
 	mainLogger := zapr.NewLogger(zaptest.NewLogger(t))
-	statHandler := stats.NewHandler(cfg.Config.PromURL, cfg.Config.Hostname, cfg.Config.WebhookURL, mainLogger)
+	statHandler := stats.NewHandler(cfg.Config.PromURL, cfg.Config.ClusterName, cfg.Config.Hostname, cfg.Config.WebhookURL, mainLogger)
 	resticCli := cli.New(ctx, mainLogger, statHandler)
 
 	cleanupDirs(t)

cmd/restic/main.go

@@ -55,6 +55,7 @@ var (
 			&cli.BoolFlag{Destination: &cfg.Config.SkipPreBackup, Name: "skipPreBackup", EnvVars: []string{"SKIP_PREBACKUP"}, Usage: "If the job should skip the backup command and only backup volumes."},
 
 			&cli.StringFlag{Destination: &cfg.Config.PromURL, Name: "promURL", EnvVars: []string{"PROM_URL"}, Usage: "Sets the URL of a prometheus push gateway to report metrics."},
+			&cli.StringFlag{Destination: &cfg.Config.ClusterName, Name: "clusterName", EnvVars: []string{"CLUSTER_NAME"}, Usage: "Sets the Kubernetes cluster name for grouping metrics in push gateway"},
 			&cli.StringFlag{Destination: &cfg.Config.WebhookURL, Name: "webhookURL", Aliases: []string{"statsURL"}, EnvVars: []string{"STATS_URL"}, Usage: "Sets the URL of a server which will retrieve a webhook after the action completes."},
 
 			&cli.StringFlag{Destination: &cfg.Config.Hostname, Name: "hostname", EnvVars: []string{"HOSTNAME"}, Usage: "Sets the hostname to use in reports.", Hidden: true, Required: true},
@@ -120,7 +121,7 @@ func resticMain(c *cli.Context) error {
 	ctx, cancel := context.WithCancel(c.Context)
 	cancelOnTermination(cancel, resticLog)
 
-	statHandler := stats.NewHandler(cfg.Config.PromURL, cfg.Config.Hostname, cfg.Config.WebhookURL, resticLog)
+	statHandler := stats.NewHandler(cfg.Config.PromURL, cfg.Config.ClusterName, cfg.Config.Hostname, cfg.Config.WebhookURL, resticLog)
 
 	resticCLI := resticCli.New(ctx, resticLog.WithName("restic"), statHandler)
 

config/crd/apiextensions.k8s.io/v1/k8up.io_backups.yaml

@@ -436,6 +436,10 @@ spec:
                       type: object
                     type: array
                 type: object
+              clusterName:
+                description: ClusterName sets athe kubernetes cluster name to send
+                  to pushgateway for grouping metrics
+                type: string
               failedJobsHistoryLimit:
                 description: |-
                   FailedJobsHistoryLimit amount of failed jobs to keep for later analysis.

config/crd/apiextensions.k8s.io/v1/k8up.io_checks.yaml

@@ -431,6 +431,10 @@ spec:
                       type: object
                     type: array
                 type: object
+              clusterName:
+                description: ClusterName sets athe kubernetes cluster name to send
+                  to pushgateway for grouping metrics
+                type: string
               failedJobsHistoryLimit:
                 description: |-
                   FailedJobsHistoryLimit amount of failed jobs to keep for later analysis.

config/crd/apiextensions.k8s.io/v1/k8up.io_schedules.yaml

@@ -1742,6 +1742,10 @@ spec:
                           type: object
                         type: array
                     type: object
+                  clusterName:
+                    description: ClusterName sets athe kubernetes cluster name to
+                      send to pushgateway for grouping metrics
+                    type: string
                   concurrentRunsAllowed:
                     type: boolean
                   failedJobsHistoryLimit:
@@ -2574,6 +2578,10 @@ spec:
                           type: object
                         type: array
                     type: object
+                  clusterName:
+                    description: ClusterName sets athe kubernetes cluster name to
+                      send to pushgateway for grouping metrics
+                    type: string
                   concurrentRunsAllowed:
                     type: boolean
                   failedJobsHistoryLimit:

config/samples/k8up_v1_schedule.yaml

@@ -41,6 +41,7 @@ spec:
     failedJobsHistoryLimit: 2
     successfulJobsHistoryLimit: 2
     promURL: http://minio.minio:9000
+    clusterName: default
     resources:
       requests:
         memory: "64Mi"
@@ -55,6 +56,7 @@ spec:
         cpu: "250m"
     schedule: '@hourly-random'
     promURL: http://minio.minio:9000
+    clusterName: default
   prune:
     schedule: '*/4 * * * *'
     retention:

operator/cfg/config.go

@@ -75,6 +75,7 @@ type Configuration struct {
 	PodExecRoleName                  string
 	PodFilter                        string
 	PromURL                          string
+	ClusterName                      string
 	RestartPolicy                    string
 	SkipWithoutAnnotation            bool
 

restic/cfg/config.go

@@ -35,8 +35,9 @@ type Configuration struct {
 
 	SkipPreBackup bool
 
-	PromURL    string
-	WebhookURL string
+	PromURL     string
+	ClusterName string
+	WebhookURL  string
 
 	Hostname   string
 	KubeConfig string

restic/stats/handler.go

@@ -23,14 +23,16 @@ type Handler struct {
 	promURL      string
 	promHostname string
 	webhookURL   string
+	clusterName  string
 	log          logr.Logger
 }
 
-func NewHandler(promURL, promHostname, webhookURL string, log logr.Logger) *Handler {
+func NewHandler(promURL, clusterName, promHostname, webhookURL string, log logr.Logger) *Handler {
 	return &Handler{
 		promHostname: promHostname,
 		promURL:      promURL,
 		webhookURL:   webhookURL,
+		clusterName:  clusterName,
 		log:          log.WithName("statsHandler"),
 	}
 }
@@ -55,7 +57,7 @@ func (h *Handler) SendPrometheus(promStats cli.PrometheusProvider) error {
 
 func (h *Handler) updatePrometheus(collector prometheus.Collector) error {
 	return push.New(h.promURL, subsystem).Collector(collector).
-		Grouping("instance", h.promHostname).
+		Grouping("instance", h.promHostname).Grouping("cluster", h.clusterName).
 		Add()
 }