fix cert-manager down alert for old clusters (#1372)

giantswarm · Sep 23, 2024 · 124dc12 · 124dc12
1 parent afa68df
commit 124dc12
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 1 deletion.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -14,6 +14,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Fixed
 
 - Dashboard links in alertmanager and mimir rules
+- Fix cert-manager down alert.
 - Remove deprecated app labels for `external-dns` and `ingress-nginx` alerts.
 - Remove deprecated app labels for `kube-state-metrics` alerts.
 - Fix falco events alerts node label to hostname as node does not exist.

diff --git a/helm/prometheus-rules/templates/kaas/bigmac/alerting-rules/cert-manager.rules.yml b/helm/prometheus-rules/templates/kaas/bigmac/alerting-rules/cert-manager.rules.yml
@@ -29,7 +29,7 @@ spec:
       annotations:
         description: '{{`cert-manager in namespace {{ $labels.namespace }} is down.`}}'
         opsrecipe: cert-manager-down/
-      expr: label_replace(up{job=~"cert-manager-(app|controller)"}, "ip", "$1.$2.$3.$4", "node", "ip-(\\d+)-(\\d+)-(\\d+)-(\\d+).*") == 0
+      expr: label_replace(up{container=~"cert-manager(-app-controller)?"}, "ip", "$1.$2.$3.$4", "node", "ip-(\\d+)-(\\d+)-(\\d+)-(\\d+).*") == 0
       for: 15m
       labels:
         area: kaas