Cadoles
/
eole-prometheus
10
0
Fork 0
Code Issues 3 Pull Requests Releases Wiki Activity

ajout de règles supplémentaires pour les alertes

This commit is contained in:
vfebvre 2021-10-21 16:09:53 +02:00
parent 4ef331f6a1
commit a3472eaea5
2 changed files with 36 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
dicos/70_prometheus.xml
View File

@ -5,6 +5,7 @@
<file filelist='prometheus' name='/etc/prometheus/prometheus.yml' mkdir='True' rm='True'/>
<file filelist='prometheus-alertmanager' name='/etc/prometheus/alertmanager.yml' mkdir='True' rm='True'/>
<file filelist='prometheus-alertmanager' name='/etc/prometheus/rules.d/alert-rules.yml' mkdir='True' rm='True'/>
<file filelist='prometheus-alertmanager' name='/etc/prometheus/rules.d/alert-rules-node-exporter.yml' mkdir='True' rm='True'/>
<file filelist='prometheus-alertmanager' name='/etc/prometheus/rules.d/predict-rules.yml' mkdir='True' rm='True'/>
<file filelist='grafana' name='/etc/grafana/grafana.ini' mkdir='True' rm='True'/>
<file filelist='grafana' name='/etc/grafana/provisioning/dashboards/eole.yml' source='grafana-dashboards.yml' mkdir='True' rm='True'/>

35
tmpl/alert-rules-node-exporter.yml Normal file
View File

@ -0,0 +1,35 @@
#
# Alert Rules
#
groups:
- name: GeneralNodeExporterRules
rules:
# TooMuch Data IN
- alert: TooMuchNetworkThroughputIn
expr: sum by (instance) (rate(node_network_receive_bytes_total[2m])) / 1024 / 1024 > 100
for: 5m
labels:
severity: warning
annotations:
summary: Host unusual network throughput in (instance {{ $labels.instance }})
description: "Host network interfaces are probably receiving too much data (> 100 MB/s)\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"
- alert: AlertInodes
expr: node_filesystem_files_free{mountpoint ="/"} / node_filesystem_files{mountpoint="/"} * 100 < 10 and ON (instance, device, mountpoint) node_filesystem_readonly{mountpoint="/"} == 0
for: 2m
labels:
severity: warning
annotations:
summary: Host out of inodes (instance {{ $labels.instance }})
description: "Disk is almost running out of available inodes (< 10% left)\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"
- alert: HostSwapIsFillingUp
expr: (1 - (node_memory_SwapFree / node_memory_SwapTotal)) * 100 > 80
for: 2m
labels:
severity: warning
annotations:
summary: Host swap is filling up (instance {{ $labels.instance }})
description: "Swap is filling up (>80%)\n VALUE = {{ $value }}\n LABELS = {{ $labels }}"