Remove kvm provider (#1058)

CHANGELOG.md

@@ -18,10 +18,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Removed
 
-- Remove Azure provider.
-- Remove Tiller related alerts.
-- Remove GCP alerts and configs.
-- Remove Openstack alerts and configs.
+= Remove `kvm` provider alerts.
+- Remove `azure` provider alerts.
+- Remove `tiller` alerts.
+- Remove `gcp` provider alerts.
+- Remove `openstack` provider alerts.
 
 ## [2.153.1] - 2024-02-28
 

helm/prometheus-rules/templates/_helpers.tpl

@@ -29,7 +29,7 @@ giantswarm.io/service-type: {{ .Values.serviceType }}
 {{- end -}}
 
 {{- define "providerTeam" -}}
-{{- if has .Values.managementCluster.provider.kind (list "kvm" "cloud-director" "vsphere") -}}
+{{- if has .Values.managementCluster.provider.kind (list "cloud-director" "vsphere") -}}
 rocket
 {{- else if has .Values.managementCluster.provider.kind (list "capa" "capz") -}}
 {{- /* hydra alerts merged into phoenix business hours on-call */ -}}

helm/prometheus-rules/templates/alerting-rules/calico.rules.yml

@@ -37,35 +37,3 @@ spec:
         severity: notify
         team: {{ include "providerTeam" . }}
         topic: kubernetes
-{{- if eq .Values.managementCluster.provider.kind "kvm" }}
-    - alert: CalicoNodeFailingToSaveIptables
-      annotations:
-        description: '{{`calico-node {{ $labels.namespace}}/{{ $labels.pod_name }} on {{ $labels.cluster_id}}/{{ $labels.cluster }} is failing to save iptables rules.`}}'
-        opsrecipe: calico-iptables-failing/
-      expr: increase(felix_iptables_save_errors[10m]) > 0
-      for: 10m
-      labels:
-        area: kaas
-        cancel_if_cluster_status_creating: "true"
-        cancel_if_cluster_status_deleting: "true"
-        cancel_if_cluster_status_updating: "true"
-        cancel_if_outside_working_hours: "true"
-        severity: notify
-        team: {{ include "providerTeam" . }}
-        topic: kubernetes
-    - alert: CalicoNodeFailingToRestoreIptables
-      annotations:
-        description: '{{`calico-node {{ $labels.namespace}}/{{ $labels.pod_name }} on {{ $labels.cluster_id}}/{{ $labels.cluster }} is failing to restore iptables rules.`}}'
-        opsrecipe: calico-iptables-failing/
-      expr: increase(felix_iptables_restore_errors[10m]) > 0
-      for: 10m
-      labels:
-        area: kaas
-        cancel_if_cluster_status_creating: "true"
-        cancel_if_cluster_status_deleting: "true"
-        cancel_if_cluster_status_updating: "true"
-        cancel_if_outside_working_hours: "true"
-        severity: notify
-        team: {{ include "providerTeam" . }}
-        topic: kubernetes
-{{- end }}

helm/prometheus-rules/templates/alerting-rules/certificate.management-cluster.rules.yml

@@ -13,18 +13,6 @@ spec:
   groups:
   - name: certificate.management-cluster
     rules:
-    - alert: ManagementClusterKVMCertificateWillExpireInLessThanOneMonth
-      annotations:
-        description: '{{`Certificate {{ $labels.path }} on {{ $labels.node }} will expire in less than a month.`}}'
-        opsrecipe: renew-certificates/
-      expr: (cert_exporter_not_after{cluster_type="management_cluster", provider="kvm", path!="/etc/kubernetes/ssl/service-account-crt.pem"} - time()) < 4 * 7 * 24 * 60 * 60
-      for: 5m
-      labels:
-        area: kaas
-        cancel_if_outside_working_hours: "true"
-        severity: page
-        team: rocket
-        topic: security
     - alert: ManagementClusterCertificateIsMissing
       annotations:
         description: '{{`Cannot renew Certificate for Secret {{ $labels.namespace }}/{{ $labels.certificatename }} because it is missing.`}}'

helm/prometheus-rules/templates/alerting-rules/deployment.management-cluster.rules.yml

@@ -150,32 +150,6 @@ spec:
         team: phoenix
         topic: managementcluster
     {{- end }}
-    {{- if eq .Values.managementCluster.provider.kind "kvm" }}
-    - alert: KVMManagementClusterDeploymentScaledDownToZero
-      annotations:
-        description: '{{`Deployment {{ $labels.namespace}}/{{ $labels.deployment }} on KVM has been scaled down to zero for prolonged period of time.`}}'
-      expr: kube_deployment_status_replicas_available{cluster_type="management_cluster", deployment=~"([a-z]*)-operator([a-z,-]*)",provider="kvm"} + kube_deployment_status_replicas_unavailable{cluster_type="management_cluster", deployment=~"([a-z]*)-operator([a-z],-]*)",provider="kvm"} == 0
-      for: 4h
-      labels:
-        area: kaas
-        severity: notify
-        team: rocket
-        topic: managementcluster
-    - alert: DeploymentNotSatisfiedRocket
-      annotations:
-        description: '{{`Deployment {{ $labels.namespace}}/{{ $labels.deployment }} is not satisfied.`}}'
-        opsrecipe: deployment-not-satisfied/
-      expr: kube_deployment_status_replicas_unavailable{cluster_type="management_cluster", deployment=~"worker-.+|master-.+"} > 0
-      for: 30m
-      labels:
-        area: kaas
-        cancel_if_cluster_status_creating: "true"
-        cancel_if_cluster_status_deleting: "true"
-        cancel_if_cluster_status_updating: "true"
-        severity: page
-        team: rocket
-        topic: managementcluster
-    {{- end }}
     - alert: DeploymentNotSatisfiedCabbage
       annotations:
         description: '{{`Deployment {{ $labels.namespace}}/{{ $labels.deployment }} is not satisfied.`}}'

helm/prometheus-rules/templates/alerting-rules/etcdbackup.rules.yml

@@ -58,21 +58,6 @@ spec:
         team: {{ include "providerTeam" . }}
         topic: etcd-backup
 {{- end }}
-    - alert: ManagementClusterNotBackedUp24h
-      annotations:
-        description: '{{`{{ $labels.cluster_id }} management cluster''s ETCD backup was unsuccessful.`}}'
-        opsrecipe: etcd-backup-failed/
-      expr: time() - etcd_backup_latest_success{etcd_version="V2",tenant_cluster_id="Control Plane", provider="kvm"} > 60*60*24 or time() - etcd_backup_latest_success{etcd_version="V3",tenant_cluster_id="Control Plane"} > 60*60*24
-      for: 5m
-      labels:
-        area: kaas
-        cancel_if_cluster_status_creating: "true"
-        cancel_if_cluster_status_deleting: "true"
-        cancel_if_cluster_status_updating: "true"
-        cancel_if_outside_working_hours: "true"
-        severity: page
-        team: {{ include "providerTeam" . }}
-        topic: etcd-backup
     - alert: ETCDBackupMetricsMissing
       annotations:
         description: '{{`ETCD backup metrics are missing`}}'

helm/prometheus-rules/templates/alerting-rules/inhibit.management-cluster.rules.yml

@@ -59,17 +59,6 @@ spec:
         cluster_status_deleting: "true"
         team: phoenix
         topic: status
-    {{- if eq .Values.managementCluster.provider.kind "kvm" }}
-    - alert: InhibitionKVMClusterHasNoRunningCPNodes
-      annotations:
-        description: '{{`KVM Cluster {{ $labels.cluster_id }} has nodes down.`}}'
-      expr: label_replace(up{app=~"master|worker", cluster_type="management_cluster"} == 0, "cluster_id", "$1", "namespace", "(.*)")
-      labels:
-        area: kaas
-        nodes_down: "true"
-        team: phoenix
-        topic: status
-    {{- end }}
     {{- if eq .Values.managementCluster.provider.kind "aws" }}
     - alert: InhibitionClusterWithNoNodePools
       annotations:

helm/prometheus-rules/templates/alerting-rules/inhibit.prometheus-agent.rules.yml

@@ -1,4 +1,4 @@
-{{- if and (ne .Values.managementCluster.provider.kind "kvm") ( eq .Values.managementCluster.provider.flavor "vintage") }}
+{{- if eq .Values.managementCluster.provider.flavor "vintage" }}
 apiVersion: monitoring.coreos.com/v1
 kind: PrometheusRule
 metadata:

helm/prometheus-rules/templates/alerting-rules/microendpoint.rules.yml

@@ -62,7 +62,7 @@ spec:
       annotations:
         description: '{{`CR version {{ $labels.version }} in cluster {{ $labels.cluster_id }} is reconciled by multiple apps including {{ $labels.app }}.`}}'
         opsrecipe: multiple-operators-running-same-version/
-      expr: sum(label_replace(giantswarm_build_info{app=~"kvm-operator|ignition-operator|cert-operator|node-operator"}, "version", "$1", "reconciled_version", "(.+)")) by (app, cluster_id, version) > 1
+      expr: sum(label_replace(giantswarm_build_info{app=~"ignition-operator|cert-operator|node-operator"}, "version", "$1", "reconciled_version", "(.+)")) by (app, cluster_id, version) > 1
       for: 5m
       labels:
         area: kaas

helm/prometheus-rules/templates/alerting-rules/node-exporter.all.rules.yml

@@ -14,7 +14,7 @@ spec:
       annotations:
         description: '{{`NodeExporter Collector {{ $labels.collector }} on {{ $labels.instance }} is failed.`}}'
         opsrecipe: node-exporter-device-error/
-      expr: node_scrape_collector_success{collector!~"conntrack|bonding|hwmon|powersupplyclass|mdadm|nfs|nfsd|tapestats|fibrechannel|nvme{{ if eq .Values.managementCluster.provider.kind "kvm" }}|pressure{{ end }}"} == 0
+      expr: node_scrape_collector_success{collector!~"conntrack|bonding|hwmon|powersupplyclass|mdadm|nfs|nfsd|tapestats|fibrechannel|nvme"} == 0
       for: 5m
       labels:
         area: kaas