diff --git a/roles/prometheus/templates/prometheus/alert.rules.yml.j2 b/roles/prometheus/templates/prometheus/alert.rules.yml.j2 index 8df9288f..477ed057 100644 --- a/roles/prometheus/templates/prometheus/alert.rules.yml.j2 +++ b/roles/prometheus/templates/prometheus/alert.rules.yml.j2 @@ -16,8 +16,9 @@ groups: summary: "{{ $labels.instance }} ({{ $labels.job }}) est invisible depuis plus de 3 minutes !" # Alert for out of memory + # Do not take into account memory not used by apps - alert: OutOfMemory - expr: (node_memory_MemFree_bytes + node_memory_Cached_bytes + node_memory_Buffers_bytes) / node_memory_MemTotal_bytes * 100 < 10 + expr: (node_memory_MemFree_bytes + node_memory_Cached_bytes + node_memory_Buffers_bytes + node_memory_PageTables_bytes + node_memory_VmallocUsed_bytes + node_memory_SwapCached_bytes + node_memory_Slab_bytes) / node_memory_MemTotal_bytes * 100 < 10 for: 5m labels: severity: warning