Add NodeSystemdServiceCrashlooping alert

Signed-off-by: Vitaly Zhuravlev <[email protected]>
grafana · Jun 3, 2024 · 395c7a1 · 395c7a1
1 parent dc6a1e8
commit 395c7a1
Showing 1 changed file with 14 additions and 0 deletions.
diff --git a/docs/node-observ-lib/linux/alerts.libsonnet b/docs/node-observ-lib/linux/alerts.libsonnet
@@ -414,6 +414,20 @@
                 description: 'Systemd service {{ $labels.name }} has entered failed state at {{ $labels.instance }}',
               },
             },
+            {
+              alert: 'NodeSystemdServiceCrashlooping',
+              expr: |||
+                increase(node_systemd_service_restart_total{%(filteringSelector)s}[5m]) > 2
+              ||| % this.config,
+              'for': '15m',
+              labels: {
+                severity: 'warning',
+              },
+              annotations: {
+                summary: 'Systemd service keeps restaring, possibly crash looping.',
+                description: 'Systemd service {{ $labels.name }} has being restarted too many times at {{ $labels.instance }} for the last 15 minutes. Please check if service is crash looping.',
+              },
+            },
           ]
           + if this.config.enableHardware then
             [{