easylinux/kube-prometheus
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[bot] [main] Automated version update

Browse Source
This commit is contained in:
Prometheus Operator Bot
2022-05-30 07:45:33 +00:00
parent 1c19d2a261
commit 730d8a4365
12 changed files with 46 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
manifests/kubernetesControlPlane-prometheusRule.yaml
View File

@@ -30,11 +30,11 @@ spec:
runbook_url: https://runbooks.prometheus-operator.dev/runbooks/kubernetes/kubepodnotready
summary: Pod has been in a non-ready state for more than 15 minutes.
expr: |
sum by (namespace, pod) (
max by(namespace, pod) (
sum by (namespace, pod, cluster) (
max by(namespace, pod, cluster) (
kube_pod_status_phase{job="kube-state-metrics", phase=~"Pending|Unknown"}
) * on(namespace, pod) group_left(owner_kind) topk by(namespace, pod) (
1, max by(namespace, pod, owner_kind) (kube_pod_owner{owner_kind!="Job"})
) * on(namespace, pod, cluster) group_left(owner_kind) topk by(namespace, pod, cluster) (
1, max by(namespace, pod, owner_kind, cluster) (kube_pod_owner{owner_kind!="Job"})
)
) > 0
for: 15m
@@ -174,7 +174,7 @@ spec:
runbook_url: https://runbooks.prometheus-operator.dev/runbooks/kubernetes/kubecontainerwaiting
summary: Pod container waiting longer than 1 hour
expr: |
sum by (namespace, pod, container) (kube_pod_container_status_waiting_reason{job="kube-state-metrics"}) > 0
sum by (namespace, pod, container, cluster) (kube_pod_container_status_waiting_reason{job="kube-state-metrics"}) > 0
for: 1h
labels:
severity: warning
@@ -209,7 +209,7 @@ spec:
runbook_url: https://runbooks.prometheus-operator.dev/runbooks/kubernetes/kubejobnotcompleted
summary: Job did not complete in time
expr: |
time() - max by(namespace, job_name) (kube_job_status_start_time{job="kube-state-metrics"}
time() - max by(namespace, job_name, cluster) (kube_job_status_start_time{job="kube-state-metrics"}
and
kube_job_status_active{job="kube-state-metrics"} > 0) > 43200
labels:
@@ -488,7 +488,7 @@ spec:
runbook_url: https://runbooks.prometheus-operator.dev/runbooks/kubernetes/kubeversionmismatch
summary: Different semantic versions of Kubernetes components running.
expr: |
count(count by (git_version) (label_replace(kubernetes_build_info{job!~"kube-dns|coredns"},"git_version","$1","git_version","(v[0-9]*.[0-9]*).*"))) > 1
count by (cluster) (count by (git_version, cluster) (label_replace(kubernetes_build_info{job!~"kube-dns|coredns"},"git_version","$1","git_version","(v[0-9]*.[0-9]*).*"))) > 1
for: 15m
labels:
severity: warning
@@ -594,7 +594,7 @@ spec:
runbook_url: https://runbooks.prometheus-operator.dev/runbooks/kubernetes/kubeaggregatedapierrors
summary: Kubernetes aggregated API has reported errors.
expr: |
sum by(name, namespace)(increase(aggregator_unavailable_apiservice_total[10m])) > 4
sum by(name, namespace, cluster)(increase(aggregator_unavailable_apiservice_total[10m])) > 4
labels:
severity: warning
- alert: KubeAggregatedAPIDown
@@ -604,7 +604,7 @@ spec:
runbook_url: https://runbooks.prometheus-operator.dev/runbooks/kubernetes/kubeaggregatedapidown
summary: Kubernetes aggregated API is down.
expr: |
(1 - max by(name, namespace)(avg_over_time(aggregator_unavailable_apiservice[10m]))) * 100 < 85
(1 - max by(name, namespace, cluster)(avg_over_time(aggregator_unavailable_apiservice[10m]))) * 100 < 85
for: 5m
labels:
severity: warning