feat(chart): make s3proxy probes configurable (tolerant liveness)

chart/templates/s3proxy/deployment.yaml

@@ -84,29 +84,11 @@ spec:
           resources:
             {{- toYaml .Values.resources | nindent 12 }}
           startupProbe:
-            httpGet:
-              path: /readyz
-              port: http
-            initialDelaySeconds: 2
-            periodSeconds: 2
-            timeoutSeconds: 3
-            failureThreshold: 60
+            {{- toYaml .Values.startupProbe | nindent 12 }}
           livenessProbe:
-            httpGet:
-              path: /healthz
-              port: http
-            initialDelaySeconds: 5
-            periodSeconds: 10
-            timeoutSeconds: 5
-            failureThreshold: 3
+            {{- toYaml .Values.livenessProbe | nindent 12 }}
           readinessProbe:
-            httpGet:
-              path: /readyz
-              port: http
-            initialDelaySeconds: 5
-            periodSeconds: 5
-            timeoutSeconds: 3
-            failureThreshold: 3
+            {{- toYaml .Values.readinessProbe | nindent 12 }}
           securityContext:
             runAsNonRoot: true
             runAsUser: 1000

chart/values.yaml

@@ -150,6 +150,36 @@ resources:
     cpu: "500m"
     memory: "512Mi"
 
+# s3proxy container probes. Overridable so deployments can tolerate a busy
+# single event loop: under heavy upload load a probe can take seconds to be
+# served, and a tight liveness timeout restarts a live pod (drops in-flight
+# uploads -> clients retry -> load spiral). Maps merge, so a deployment can set
+# just e.g. livenessProbe.timeoutSeconds without restating the whole probe.
+startupProbe:
+  httpGet:
+    path: /readyz
+    port: http
+  initialDelaySeconds: 2
+  periodSeconds: 2
+  timeoutSeconds: 3
+  failureThreshold: 60
+livenessProbe:
+  httpGet:
+    path: /healthz
+    port: http
+  initialDelaySeconds: 5
+  periodSeconds: 10
+  timeoutSeconds: 5
+  failureThreshold: 3
+readinessProbe:
+  httpGet:
+    path: /readyz
+    port: http
+  initialDelaySeconds: 5
+  periodSeconds: 5
+  timeoutSeconds: 3
+  failureThreshold: 3
+
 service:
   type: ClusterIP
   port: 4433