From d823ad72c40e7bff952ce1a14a55c39576f0b769 Mon Sep 17 00:00:00 2001
From: Alejandro Pedraza <alejandro@buoyant.io>
Date: Wed, 16 Aug 2023 13:34:50 -0500
Subject: [PATCH] Fixed service-mirror metrics warning (#11246)

Whenever the service mirror's main loop was triggered again, the following warnings were generated:

```
time="2023-08-14T20:16:29Z" level=warning msg="failed to register Prometheus gauge Desc{fqName: \"service_cache_size\", help: \"Number of items in the client-go service cache\", constLabels: {cluster=\"remote\"}, variableLabels: []}: duplicate metrics collector registration attempted"
time="2023-08-14T20:16:29Z" level=warning msg="failed to register Prometheus gauge Desc{fqName: \"endpoints_cache_size\", help: \"Number of items in the client-go endpoints cache\", constLabels: {cluster=\"remote\"}, variableLabels: []}: duplicate metrics collector registration attempted"
```

To fix, this adds into the cluster watcher's `Stop()` method a directive to unregister the prometheus cache metrics associated to the cluster's client API.
---
 controller/k8s/api.go                          | 5 +++++
 controller/k8s/prometheus.go                   | 6 ++++++
 multicluster/service-mirror/cluster_watcher.go | 4 ++++
 3 files changed, 15 insertions(+)

diff --git a/controller/k8s/api.go b/controller/k8s/api.go
index cdd123a70890b..64902e31aeff8 100644
--- a/controller/k8s/api.go
+++ b/controller/k8s/api.go
@@ -288,6 +288,11 @@ func (api *API) Sync(stopCh <-chan struct{}) {
 	waitForCacheSync(api.syncChecks)
 }
 
+// UnregisterGauges unregisters all the prometheus cache gauges associated to this API
+func (api *API) UnregisterGauges() {
+	api.promGauges.unregister()
+}
+
 // NS provides access to a shared informer and lister for Namespaces.
 func (api *API) NS() coreinformers.NamespaceInformer {
 	if api.ns == nil {
diff --git a/controller/k8s/prometheus.go b/controller/k8s/prometheus.go
index aed1739db8628..b4daad5f58616 100644
--- a/controller/k8s/prometheus.go
+++ b/controller/k8s/prometheus.go
@@ -20,3 +20,9 @@ func (p *promGauges) addInformerSize(kind string, labels prometheus.Labels, inf
 		return float64(len(inf.GetStore().ListKeys()))
 	}))
 }
+
+func (p *promGauges) unregister() {
+	for _, gauge := range p.gauges {
+		prometheus.Unregister(gauge)
+	}
+}
diff --git a/multicluster/service-mirror/cluster_watcher.go b/multicluster/service-mirror/cluster_watcher.go
index cb6727bc52957..67fadc4ead6b8 100644
--- a/multicluster/service-mirror/cluster_watcher.go
+++ b/multicluster/service-mirror/cluster_watcher.go
@@ -974,6 +974,10 @@ func (rcsw *RemoteClusterServiceWatcher) Stop(cleanupState bool) {
 			rcsw.log.Warnf("error removing service informer handler: %s", err)
 		}
 	}
+
+	if rcsw.remoteAPIClient != nil {
+		rcsw.remoteAPIClient.UnregisterGauges()
+	}
 }
 
 func (rcsw *RemoteClusterServiceWatcher) resolveGatewayAddress() ([]corev1.EndpointAddress, error) {