From de752dbca0aef0cd9ea16987ce5bfae7cdcd6b31 Mon Sep 17 00:00:00 2001 From: Kevin Veen-Birkenbach Date: Thu, 16 Jul 2026 03:28:50 +0200 Subject: [PATCH] fix(prometheus): raise cadvisor mem_limit to 512m At 256m cadvisor OOM-loops every 7-20 minutes on nodes running 60+ containers (run 29448092737 nc0: five kills at exactly the 262144kB cap), burning 26% CPU on restart churn that feeds the memory pressure it is supposed to observe. Co-Authored-By: Claude Opus 4.8 (1M context) --- roles/web-app-prometheus/meta/services.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/roles/web-app-prometheus/meta/services.yml b/roles/web-app-prometheus/meta/services.yml index 82a59eb19e..6849cd779a 100644 --- a/roles/web-app-prometheus/meta/services.yml +++ b/roles/web-app-prometheus/meta/services.yml @@ -101,7 +101,7 @@ cadvisor: name: prometheus-cadvisor cpus: "0.25" mem_reservation: 64m - mem_limit: 256m + mem_limit: 512m pids_limit: 64 min_storage: 1GB ports: