dedup ram alert and increase threshold to 95% [ci skip]

This commit is contained in:
Viktor Barzin 2026-01-17 22:42:22 +00:00
parent 4ccf2298fa
commit 5609bbbaf3
No known key found for this signature in database
GPG key ID: 4056458DBDBF8863

View file

@ -199,13 +199,6 @@ serverFiles:
severity: page
annotations:
summary: No iDRAC amperage reading. Can signal that prometheus is not scraping
- alert: HighRAMUsage
expr: clamp_min((1 - (node_memory_MemAvailable_bytes{instance="pve-node-r730"} / node_memory_MemTotal_bytes{instance="pve-node-r730"})) * 100, 0) > 90
for: 30m
labels:
severity: page
annotations:
summary: "High memory usage: {{ $value }}. Risk of OOM-ing."
- alert: HighSystemLoad
expr: scalar(node_load1{instance="pve-node-r730"}) * 100 / count(count(node_cpu_seconds_total{instance="pve-node-r730"}) by (cpu)) > 50
for: 30m
@ -310,7 +303,7 @@ serverFiles:
annotations:
summary: "High CPU usage on {{ $labels.node }} - {{ $value }}"
- alert: NodeLowFreeMemory
expr: (1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) or on() vector(1)) > 0.9
expr: (1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) or on() vector(1)) > 0.95
for: 10m
labels:
severity: page