PERF(resources): remove CPU limits - keep memory limits only

- CPU throttling prevents app startup, not crashes - Memory OOM is the real cascading failure cause - CPU request ensures fair scheduling
2026-01-07 23:48:35 +09:00
parent 69dc3b34be
commit 4286296591
10 changed files with 42 additions and 33 deletions
--- a/grafana/helm-values.yaml
+++ b/grafana/helm-values.yaml
@@ -24,8 +24,10 @@ podSecurityContext:

 resources:
  requests:
-    cpu: 25m
-    memory: 128Mi
+    cpu: 11m
+    memory: 425Mi
+  limits:
+    memory: 425Mi

 service:
  type: ClusterIP