diff --git a/jsonnet/kube-prometheus/alerts/alertmanager.libsonnet b/jsonnet/kube-prometheus/alerts/alertmanager.libsonnet index bf58862dd26a16b718a2b3d931faa0ef2819d744..bda69d00f3dde050a1b458412a2a8e8a341c2118 100644 --- a/jsonnet/kube-prometheus/alerts/alertmanager.libsonnet +++ b/jsonnet/kube-prometheus/alerts/alertmanager.libsonnet @@ -10,7 +10,7 @@ message: 'The configuration of the instances of the Alertmanager cluster `{{$labels.service}}` are out of sync.', }, expr: ||| - count_values("config_hash", alertmanager_config_hash{%(alertmanagerSelector)s}) BY (service) / ON(service) GROUP_LEFT() label_replace(prometheus_operator_spec_replicas{%(prometheusOperatorSelector)s,controller="alertmanager"}, "service", "alertmanager-$1", "name", "(.*)") != 1 + count_values("config_hash", alertmanager_config_hash{%(alertmanagerSelector)s}) BY (service) / ON(service) GROUP_LEFT() label_replace(max(prometheus_operator_spec_replicas{%(prometheusOperatorSelector)s,controller="alertmanager"}) by (name, job, namespace, controller), "service", "alertmanager-$1", "name", "(.*)") != 1 ||| % $._config, 'for': '5m', labels: { @@ -31,8 +31,8 @@ }, }, { - alert:'AlertmanagerMembersInconsistent', - annotations:{ + alert: 'AlertmanagerMembersInconsistent', + annotations: { message: 'Alertmanager has not found all other members of the cluster.', }, expr: ||| diff --git a/manifests/prometheus-rules.yaml b/manifests/prometheus-rules.yaml index f35510b53feb0df789fad8cead803cb5c4ce1ee0..2f0eaa756004dc7f18674c9b7f1062a7a08c594a 100644 --- a/manifests/prometheus-rules.yaml +++ b/manifests/prometheus-rules.yaml @@ -1054,7 +1054,7 @@ spec: message: The configuration of the instances of the Alertmanager cluster `{{$labels.service}}` are out of sync. expr: | - count_values("config_hash", alertmanager_config_hash{job="alertmanager-main",namespace="monitoring"}) BY (service) / ON(service) GROUP_LEFT() label_replace(prometheus_operator_spec_replicas{job="prometheus-operator",namespace="monitoring",controller="alertmanager"}, "service", "alertmanager-$1", "name", "(.*)") != 1 + count_values("config_hash", alertmanager_config_hash{job="alertmanager-main",namespace="monitoring"}) BY (service) / ON(service) GROUP_LEFT() label_replace(max(prometheus_operator_spec_replicas{job="prometheus-operator",namespace="monitoring",controller="alertmanager"}) by (name, job, namespace, controller), "service", "alertmanager-$1", "name", "(.*)") != 1 for: 5m labels: severity: critical