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

authelia/helm-values.yaml

@@ -10,10 +10,10 @@ pod:
   replicas: 1
   resources:
     requests:
-      cpu: 25m
+      cpu: 15m
       memory: 256Mi
     limits:
-      memory: 512Mi
+      memory: 384Mi
   extraVolumes:
     - name: users-database
       configMap:

falco/helm-values.yaml

@@ -15,11 +15,10 @@ image:
 # Resource requests
 resources:
   requests:
-    cpu: 30m
-    memory: 256Mi
+    cpu: 49m
+    memory: 263Mi
   limits:
-    cpu: null  # Remove CPU limit to prevent throttling (chart default is 1 core)
-    memory: 512Mi  # Prevent OOM
+    memory: 263Mi
 
 # Falco configuration
 falco:

trivy/helm-values.yaml

@@ -12,10 +12,10 @@ operator:
   # Operator 리소스
   resources:
     requests:
-      cpu: 10m
-      memory: 128Mi
+      cpu: 50m
+      memory: 256Mi
     limits:
-      memory: 512Mi  # Prevent OOM
+      memory: 384Mi
 
   # 스캔 설정
   scanJobTimeout: 10m
@@ -58,10 +58,9 @@ trivy:
   resources:
     requests:
       cpu: 50m
-      memory: 128Mi
+      memory: 256Mi
     limits:
-      cpu: null  # Remove CPU limit to prevent throttling (chart default is 500m)
-      memory: 768Mi  # Increased from 500M to prevent OOM during scans
+      memory: 384Mi
 
 # 스캔 작업 설정
 trivyOperator:

vault/helm-values.yaml

@@ -46,10 +46,10 @@ server:
   # 리소스 제한
   resources:
     requests:
-      cpu: 50m  # Reduced from 100m (actual usage: 24-30m)
-      memory: 256Mi
+      cpu: 35m
+      memory: 263Mi
     limits:
-      memory: 512Mi
+      memory: 263Mi
 
   # Ingress 설정
   ingress: