Add mimir.enabled flag

giantswarm · Mar 28, 2024 · 12e81b2 · 12e81b2
1 parent 31301ab
commit 12e81b2
Show file tree

Hide file tree

Showing 3 changed files with 8 additions and 5 deletions.
diff --git a/helm/prometheus-rules/templates/alerting-rules/mimir.rules.yml b/helm/prometheus-rules/templates/alerting-rules/mimir.rules.yml
@@ -1,3 +1,4 @@
+{{- if .Values.mimir.enabled }}
 apiVersion: monitoring.coreos.com/v1
 kind: PrometheusRule
 metadata:
@@ -16,14 +17,11 @@ spec:
       expr: up{app="mimir"} > 0
       labels:
         area: "empowerment"
-        # TODO(@team-atlas): do we need this label? Let's test once we use mimir alertmanager
         installation: {{ .Values.managementCluster.name }}
+        # TODO(@team-atlas): We need this label as long as we have the old and new heartbeats. Let's remove once the legacy monitoring is gone
+        type: "mimir-heartbeat"
         team: "atlas"
         topic: "observability"
-        # TODO(@team-atlas): do we need this label? Let's test once we use mimir alertmanager
-        type: "heartbeat"
-        # TODO(@team-atlas): remove once we use mimir alertmanager
-        namespace: "monitoring" # Needed due to https://github.com/prometheus-operator/prometheus-operator/issues/3737
     # Coming from https://github.com/giantswarm/giantswarm/issues/30124
     # This alert ensures Mimir containers are not restarting too often (flappiness).
     # If it is not the the case, this can incur high costs by cloud providers (s3 api calls are quite expensive).
@@ -86,3 +84,4 @@ spec:
         severity: page
         team: atlas
         topic: observability
+{{- end }}
diff --git a/helm/prometheus-rules/templates/alerting-rules/prometheus-meta-operator.rules.yml b/helm/prometheus-rules/templates/alerting-rules/prometheus-meta-operator.rules.yml
@@ -9,6 +9,7 @@ spec:
   groups:
   - name: observability
     rules:
+    {{- if not .Values.mimir.enabled }}
     - alert: "Heartbeat"
       expr: up{app="prometheus",instance!="prometheus-agent"}
       labels:
@@ -20,6 +21,7 @@ spec:
         namespace: "monitoring" # Needed due to https://github.com/prometheus-operator/prometheus-operator/issues/3737
       annotations:
         description: This alert is used to ensure the entire alerting pipeline is functional.
+    {{- end }}
     - alert: "MatchingNumberOfPrometheusAndCluster"
       annotations:
         description: This alert is used to ensure we have as many workload cluster prometheus as we have workload cluster CR.

diff --git a/helm/prometheus-rules/templates/recording-rules/mimir-mixins.rules.yml b/helm/prometheus-rules/templates/recording-rules/mimir-mixins.rules.yml
@@ -1,3 +1,4 @@
+{{- if .Values.mimir.enabled }}
 apiVersion: monitoring.coreos.com/v1
 kind: PrometheusRule
 metadata:
@@ -577,3 +578,4 @@ spec:
     - expr: |
         sum by(cluster, namespace, pod) (rate(cortex_ingester_ingested_samples_total[1m]))
       record: cluster_namespace_pod:cortex_ingester_ingested_samples_total:rate1m
+{{- end }}