blob: 621b6ec2c1e7c9258a650d24e864d59e606359c7 [file] [log] [blame]
# Alerts for things in the skolo-rack4 cluster only.
#
groups:
- name: general
rules:
- alert: CrashLoop
expr: max_over_time(liveness_uptime_s{kubernetes_pod_name=""}[6m]) < 60 * 3
for: 5m
labels:
category: infra
severity: critical
annotations:
abbr: '{{ $labels.instance }}'
description: '{{ $labels.instance }} of job {{ $labels.job }} is crashing on
startup. Logs:
kubectl logs -l app={{ reReplaceAll `:[0-9]+` `` $labels.instance }} -c {{ $labels.job }}
https://console.cloud.google.com/logs/viewer?project={{ $labels.project }}&advancedFilter=logName%3D"projects%2F{{ $labels.project }}%2Flogs%2F{{ $labels.job }}"
'
- alert: InstanceDown
expr: up{kubernetes_pod_name=""} == 0
for: 5m
labels:
category: infra
severity: critical
annotations:
abbr: '{{ $labels.instance }}'
description: '{{ $labels.instance }} of job {{ $labels.job }} has been down
for more than 5 minutes. Logs:
kubectl logs -l app={{ reReplaceAll `:[0-9]+` `` $labels.instance }} -c {{ $labels.job }}
'
- alert: TooManyGoRoutines
expr: go_goroutines{app=~".+"} > 3000
for: 2m
labels:
category: infra
severity: warning
annotations:
abbr: '{{ $labels.app }}'
description: 'Too many Go routines in {{ $labels.kubernetes_pod_name }} for app
{{ $labels.app }}. Logs: `kubectl logs {{ $labels.kubernetes_pod_name }}`
https://console.cloud.google.com/logs/viewer?project={{ $labels.project }}&advancedFilter=resource.type%3D%22container%22%0Aresource.labels.pod_id%3D%22{{ $labels.kubernetes_pod_name }}%22'
- alert: TooManyOpenFDs
expr: process_open_fds{app=~".+"} > 5000
labels:
category: infra
severity: warning
annotations:
abbr: '{{ $labels.app }}'
description: 'Too many open file handles on {{ $labels.kubernetes_pod_name }} for app
{{ $labels.app }}. Logs: `kubectl logs {{ $labels.kubernetes_pod_name }}`
https://console.cloud.google.com/logs/viewer?project={{ $labels.project }}&advancedFilter=resource.type%3D%22container%22%0Aresource.labels.pod_id%3D%22{{ $labels.kubernetes_pod_name }}%22'