XeLabs/awesome-prometheus-alerts
2
0
Fork 0
mirror of https://github.com/samber/awesome-prometheus-alerts.git synced 2026-06-24 10:26:58 +08:00
Code Issues Projects Releases Packages Wiki Activity

Publish

Browse source
This commit is contained in:
samber 2024-02-09 19:25:17 +00:00
parent 854688d17a
commit 0dba950ccc
1 changed files with 50 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

50
dist/rules/s.m.a.r.t-device-monitoring/smartctl-exporter.yml vendored Normal file
View file

@ -0,0 +1,50 @@
groups:
- name: SmartctlExporter
rules:
- alert: SmartDeviceTemperatureWarning
expr: 'smartctl_device_temperature > 60'
for: 2m
labels:
severity: warning
annotations:
summary: Smart device temperature warning (instance {{ $labels.instance }})
description: "Device temperature warning (instance {{ $labels.instance }})\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"
- alert: SmartDeviceTemperatureCritical
expr: 'smartctl_device_temperature > 80'
for: 2m
labels:
severity: critical
annotations:
summary: Smart device temperature critical (instance {{ $labels.instance }})
description: "Device temperature critical (instance {{ $labels.instance }})\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"
- alert: SmartCriticalWarning
expr: 'smartctl_device_critical_warning > 0'
for: 15m
labels:
severity: critical
annotations:
summary: Smart critical warning (instance {{ $labels.instance }})
description: "device has critical warning (instance {{ $labels.instance }})\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"
- alert: SmartMediaErrors
expr: 'smartctl_device_media_errors > 0'
for: 15m
labels:
severity: critical
annotations:
summary: Smart media errors (instance {{ $labels.instance }})
description: "device has media errors (instance {{ $labels.instance }})\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"
- alert: SmartNvmeWearoutIndicator
expr: 'smartctl_device_available_spare{device=~"nvme.*"} < smartctl_device_available_spare_threshold{device=~"nvme.*"}'
for: 15m
labels:
severity: critical
annotations:
summary: Smart NVME Wearout Indicator (instance {{ $labels.instance }})
description: "NVMe device is wearing out (instance {{ $labels.instance }})\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"