Gardener alerts revisited

charts/controlplane-operations/Chart.yaml

@@ -1,6 +1,6 @@
 apiVersion: v2
 name: controlplane-operations
-version: 1.0.29
+version: 1.1.0
 description: A set of Plutono dashboards and Prometheus alerting rules combined with playbooks to ensure effective operations of Controlplane clusters.
 maintainers:
   - name: Vladimir Videlov (d051408)

charts/controlplane-operations/alerts/controlplane-gardener.yaml

@@ -4,51 +4,95 @@ groups:
 
 ### Gardener ###
 
-{{- if not (.Values.prometheusRules.disabled.ShootUnavailability | default false) }}
-  - alert: ShootUnavailability
-    expr: shoot:availability{instance=~"https:\\/\\/api\\..+",container=""} == 0
-    for: {{ dig "ShootUnavailability" "for" "15m" .Values.prometheusRules }}
+{{- if not (.Values.prometheusRules.disabled.ShootReconciliationFailed | default false) }}
+  - alert: ShootReconciliationFailed
+    expr: min_over_time((garden_shoot_operation_states{operation="Reconcile"} != 1 or garden_shoot_operation_states{operation="Reconcile"} != 2)[{{ dig "ShootReconciliationFailed" "for" "30m" .Values.prometheusRules }}:]) != 1
     labels:
       {{ include "controlplane-operations.additionalRuleLabels" . }}
-      severity: {{ dig "ShootUnavailability" "severity" "warning" .Values.prometheusRules }}
-      playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/ShootUnavailability.md
-      service: {{ dig "ShootUnavailability" "service" .Values.prometheusRules.defaultService .Values.prometheusRules }}
-      support_group: {{ dig "ShootUnavailability" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules }}
+      severity: {{ dig "ShootReconciliationFailed" "severity" "warning" .Values.prometheusRules }}
+      playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/ShootReconciliationFailed.md
+      service: gardener
+      support_group: {{ include "controlplane-operations.supportGroup" "`{{ $labels.project }}`" | default (dig "ShootReconciliationFailed" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules) }}
     annotations:
-      description: Shoot {{`{{ $labels.shoot_name }}`}} is unavailable. Need manual investigation of the root cause. Check the shoot and underlying infrastructure for issues.
-      summary: Shoot {{`{{ $labels.shoot_name }}`}} is unavailable.
+      description: Shoot {{`{{ $labels.name }}`}} from project {{`{{ $labels.project }}`}} on {{`{{ $labels.landscape }}`}} is not being reconciled successfully for {{ dig "ShootReconciliationFailed" "for" "30m" .Values.prometheusRules }} minutes. Check the shoot's conditions and events for more details.
+      summary: Shoot {{`{{ $labels.name }}`}} from project {{`{{ $labels.project }}`}} on {{`{{ $labels.landscape }}`}} is not being reconciled successfully.
+{{- end }}
+
+{{- if not (.Values.prometheusRules.disabled.ShootConditionNotTrue | default false) }}
+  - alert: ShootConditionNotTrue
+    expr: max by(project,landscape,seed,name) (garden_shoot_condition != 1)
+    for: {{ dig "ShootConditionNotTrue" "for" "30m" .Values.prometheusRules }}
+    labels:
+      {{ include "controlplane-operations.additionalRuleLabels" . }}
+      severity: {{ dig "ShootConditionNotTrue" "severity" "warning" .Values.prometheusRules }}
+      playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/ShootConditionNotTrue.md
+      service: gardener
+      support_group: {{ include "controlplane-operations.supportGroup" "`{{ $labels.project }}`" | default (dig "ShootConditionNotTrue" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules) }}
+    annotations:
+      description: Shoot {{`{{ $labels.name }}`}} of project {{`{{ $labels.project }}`}} seeded from {{`{{ $labels.landscape }}`}}/{{`{{ $labels.seed }}`}} has a condition that is not True. Check the Shoot's conditions and events for more details.
+      summary: Shoot {{`{{ $labels.name }}`}} of project {{`{{ $labels.project }}`}} seeded from {{`{{ $labels.landscape }}`}}/{{`{{ $labels.seed }}`}} has a condition that is not True.
+{{- end }}
+
+{{- if not (.Values.prometheusRules.disabled.SeedConditionNotTrue | default false) }}
+  - alert: SeedConditionNotTrue
+    expr: max by(landscape,name) (garden_seed_condition != 1)
+    for: {{ dig "SeedConditionNotTrue" "for" "30m" .Values.prometheusRules }}
+    labels:
+      {{ include "controlplane-operations.additionalRuleLabels" . }}
+      severity: {{ dig "SeedConditionNotTrue" "severity" "warning" .Values.prometheusRules }}
+      playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/SeedConditionNotTrue.md
+      service: gardener
+      support_group: {{ dig "SeedConditionNotTrue" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules }}
+    annotations:
+      description: Seed {{`{{ $labels.name }}`}} on {{`{{ $labels.landscape }}`}} has a condition that is not True. Check the Seed's conditions and events for more details.
+      summary: Seed {{`{{ $labels.name }}`}} on {{`{{ $labels.landscape }}`}} has a condition that is not True.
+{{- end }}
+
+{{- if not (.Values.prometheusRules.disabled.GardenletNotReady | default false) }}
+  - alert: GardenletNotReady
+    expr: max by(landscape) (garden_garden_condition{status="True"}) != 1
+    for: {{ dig "GardenletNotReady" "for" "30m" .Values.prometheusRules }}
+    labels:
+      {{ include "controlplane-operations.additionalRuleLabels" . }}
+      severity: {{ dig "GardenletNotReady" "severity" "warning" .Values.prometheusRules }}
+      playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/GardenletNotReady.md
+      service: gardener
+      support_group: {{ dig "GardenletNotReady" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules }}
+    annotations:
+      description: Gardenlet on {{`{{ $labels.landscape }}`}} is not Ready. Check the Gardenlet's conditions and events for more details.
+      summary: Gardenlet on {{`{{ $labels.landscape }}`}} is not Ready.
 {{- end }}
 
 ### Calico ###
 
 {{- if not (.Values.prometheusRules.disabled.CalicoBgpNeighborSessionDown | default false) }}
   - alert: CalicoBgpNeighborSessionDown
-    expr: sum by (node) (bird_protocol_up{proto="BGP",state="Established"}) < {{ .Values.prometheusRules.calico.bgpNeighborCount }}
+    expr: sum by (cluster,node,pod) (bird_protocol_up{proto="BGP",state="Established"}) < {{ .Values.prometheusRules.calico.bgpNeighborCount }}
     for: {{ dig "CalicoBgpNeighborSessionDown" "for" "30m" .Values.prometheusRules }}
     labels:
       {{ include "controlplane-operations.additionalRuleLabels" . }}
       severity: {{ dig "CalicoBgpNeighborSessionDown" "severity" "warning" .Values.prometheusRules }}
       playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/CalicoBgpNeighborSessionDown.md
-      service: {{ dig "CalicoBgpNeighborSessionDown" "service" .Values.prometheusRules.defaultService .Values.prometheusRules }}
+      service: calico
       support_group: {{ dig "CalicoBgpNeighborSessionDown" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules }}
     annotations:
-      description: Node {{`{{ $labels.node }}`}} has less than {{ .Values.prometheusRules.calico.bgpNeighborCount }} BGP neighbors. BGP peer is not established. Network datapath threatened! Switch upgrades or misconfiguration?
-      summary: Node {{`{{ $labels.node }}`}} has less than {{ .Values.prometheusRules.calico.bgpNeighborCount }} BGP neighbors.
+      description: Calico Node Pod {{`{{ $labels.pod }}`}} on Shoot/Node {{`{{ trimPrefix "shoot--cp--" $labels.cluster }}`}}/{{`{{ $labels.node }}`}} has less than {{ .Values.prometheusRules.calico.bgpNeighborCount }} BGP neighbors. BGP peer is not established. Network datapath threatened! Switch upgrades or misconfiguration?
+      summary: Calico Node Pod {{`{{ $labels.pod }}`}} on Shoot/Node {{`{{ trimPrefix "shoot--cp--" $labels.cluster }}`}}/{{`{{ $labels.node }}`}} has less than {{ .Values.prometheusRules.calico.bgpNeighborCount }} BGP neighbors.
 {{- end }}
 
 {{- if not (.Values.prometheusRules.disabled.CalicoBgpNeighborSessionAllDown | default false) }}
   - alert: CalicoBgpNeighborSessionAllDown
-    expr: sum by (node) (bird_protocol_up{proto="BGP",state="Established"}) == 0
+    expr: sum by (cluster,node,pod) (bird_protocol_up{proto="BGP",state="Established"}) == 0
     for: {{ dig "CalicoBgpNeighborSessionAllDown" "for" "10m" .Values.prometheusRules }}
     labels:
       {{ include "controlplane-operations.additionalRuleLabels" . }}
       severity: {{ dig "CalicoBgpNeighborSessionAllDown" "severity" "warning" .Values.prometheusRules }}
       playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/CalicoBgpNeighborSessionAllDown.md
-      service: {{ dig "CalicoBgpNeighborSessionAllDown" "service" .Values.prometheusRules.defaultService .Values.prometheusRules }}
+      service: calico
       support_group: {{ dig "CalicoBgpNeighborSessionAllDown" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules }}
     annotations:
-        description: Node {{`{{ $labels.node }}`}} has no BGP neighbors. Network datapath is down! Switch upgrades or misconfiguration?
-        summary: Node {{`{{ $labels.node }}`}} has no BGP neighbors.
+        description: Calico Node Pod {{`{{ $labels.pod }}`}} on Shoot/Node {{`{{ trimPrefix "shoot--cp--" $labels.cluster }}`}}/{{`{{ $labels.node }}`}} has no BGP neighbors. Network datapath is down! Switch upgrades or misconfiguration?
+        summary: Calico Node Pod {{`{{ $labels.pod }}`}} on Shoot/Node {{`{{ trimPrefix "shoot--cp--" $labels.cluster }}`}}/{{`{{ $labels.node }}`}} has no BGP neighbors.
 {{- end }}
 
 {{- if not (.Values.prometheusRules.disabled.CalicoNodeMissing | default false) }}
@@ -59,11 +103,11 @@ groups:
       {{ include "controlplane-operations.additionalRuleLabels" . }}
       severity: {{ dig "CalicoNodeMissing" "severity" "warning" .Values.prometheusRules }}
       playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/CalicoNodeMissing.md
-      service: {{ dig "CalicoNodeMissing" "service" .Values.prometheusRules.defaultService .Values.prometheusRules }}
+      service: calico
       support_group: {{ dig "CalicoNodeMissing" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules }}
     annotations:
-        description: Calico is not running on all bare metal nodes. Network datapath threatened!
-        summary: Calico is not running on all nodes.
+        description: Calico is not running on all BareMetal nodes. Network datapath threatened!
+        summary: Calico is not running on all BareMetal nodes.
 {{- end }}
 
 {{- if not (.Values.prometheusRules.disabled.CalicoNodeNotReady | default false) }}
@@ -74,9 +118,9 @@ groups:
       {{ include "controlplane-operations.additionalRuleLabels" . }}
       severity: {{ dig "CalicoNodeNotReady" "severity" "warning" .Values.prometheusRules }}
       playbook: https://github.com/cobaltcore-dev/controlplane-operations/playbooks/CalicoNodeNotReady.md
-      service: {{ dig "CalicoNodeNotReady" "service" .Values.prometheusRules.defaultService .Values.prometheusRules }}
+      service: calico
       support_group: {{ dig "CalicoNodeNotReady" "support_group" .Values.prometheusRules.defaultSupportGroup .Values.prometheusRules }}
     annotations:
-      description: Calico-Node Pod {{`{{ $labels.pod }}`}} on shoot {{`{{ $labels.shoot_name }}`}} is not Ready. Network datapath threatened!
-      summary: Calico-Node Pod {{`{{ $labels.pod }}`}} on shoot {{`{{ $labels.shoot_name }}`}} is not Ready.
+      description: Calico Node Pod {{`{{ $labels.pod }}`}} on Shoot {{`{{ $labels.shoot_name }}`}} is not Ready. Network datapath threatened!
+      summary: Calico Node Pod {{`{{ $labels.pod }}`}} on Shoot {{`{{ $labels.shoot_name }}`}} is not Ready.
 {{- end }}

charts/controlplane-operations/plugindefinition.yaml

@@ -3,15 +3,15 @@ kind: PluginDefinition
 metadata:
   name: controlplane-operations
 spec:
-  version: 1.0.29
+  version: 1.1.0
   displayName: Controlplane operations bundle
   description: Operations bundle for Controlane clusters
   docMarkDownUrl: https://raw.githubusercontent.com/cloudoperators/controlplane-operations/main/README.md
   icon: https://raw.githubusercontent.com/cloudoperators/controlplane-operations/main/charts/controlplane-operations/kubernetes-logo.png
   helmChart:
     name: controlplane-operations
     repository: oci://ghcr.io/cloudoperators/controlplane-operations/charts
-    version: 1.0.29
+    version: 1.1.0
   options:
     - name: prometheusRules.create
       description: Create Prometheus rules

charts/controlplane-operations/templates/_helpers.tpl

@@ -38,3 +38,13 @@ plugin: {{ $root.Release.Name }}
 {{- end }}
 {{- end }}
 {{- end }}
+
+{{- define "controlplane-operations.supportGroup" }}
+{{- if eq "cp" . }}
+{{- "containers" }}
+{{- else if eq "compute" . }}
+{{- "compute" }}
+{{- else if eq "storage" . }}
+{{- "storage" }}
+{{- end }}
+{{- end }}