Improvements from scale testing (#1812)

Poolmanager when tested at high load had some issues and this PR fixes one set of them which were found so far. 

Co-authored-by: Vishal <vishal-biyani@users.noreply.github.com>

charts/fission-all/templates/deployment.yaml

@@ -221,6 +221,8 @@ spec:
           value: "{{ .Values.pullPolicy }}"
         - name: ADOPT_EXISTING_RESOURCES
           value: {{ .Values.executor.adoptExistingResources | default false | quote }}
+        - name: POD_READY_TIMEOUT
+          value: {{ .Values.executor.podReadyTimeout | default false | quote }}
         - name: ENABLE_ISTIO
           value: "{{ .Values.enableIstio }}"
         - name: TRACE_JAEGER_COLLECTOR_ENDPOINT

charts/fission-all/templates/router.yaml

@@ -53,6 +53,8 @@ spec:
             value: {{ .Values.router.svcAddressMaxRetries | default 5 | quote }}
           - name: ROUTER_SVC_ADDRESS_UPDATE_TIMEOUT
             value: {{ .Values.router.svcAddressUpdateTimeout | default "30s" | quote }}
+          - name: ROUTER_UNTAP_SERVICE_TIMEOUT
+            value: {{ .Values.router.unTapServiceTimeout | default "3600s" | quote }}
           - name: TRACE_JAEGER_COLLECTOR_ENDPOINT
             value: "{{ .Values.traceCollectorEndpoint }}"
           - name: TRACING_SAMPLING_RATE

charts/fission-all/values.yaml

@@ -90,13 +90,14 @@ logger:
 
 executor:
   adoptExistingResources: false
+  podReadyTimeout: 300s
 
 ## Router config
 router:
   deployAsDaemonSet: false
   svcAddressMaxRetries: 5
   svcAddressUpdateTimeout: 30s
-
+  unTapServiceTimeout: 3600s
   ## Display endpoint access logs
   ## To be aware of enabling logging endpoint access log, it increases
   ## router resource utilization when under heavy workloads.

charts/fission-core/templates/deployment.yaml

@@ -221,6 +221,8 @@ spec:
           value: {{ .Values.traceSamplingRate | default "0.5" | quote }}
         - name: ADOPT_EXISTING_RESOURCES
           value: {{ .Values.executor.adoptExistingResources | default false | quote }}
+        - name: POD_READY_TIMEOUT
+          value: {{ .Values.executor.podReadyTimeout | default false | quote }}
         - name: ENABLE_ISTIO
           value: "{{ .Values.enableIstio }}"
         - name: FETCHER_MINCPU

charts/fission-core/templates/router.yaml

@@ -53,6 +53,8 @@ spec:
             value: {{ .Values.router.svcAddressMaxRetries | default 5 | quote }}
           - name: ROUTER_SVC_ADDRESS_UPDATE_TIMEOUT
             value: {{ .Values.router.svcAddressUpdateTimeout | default "30s" | quote }}
+          - name: ROUTER_UNTAP_SERVICE_TIMEOUT
+            value: {{ .Values.router.unTapServiceTimeout | default "3600s" | quote }}
           - name: TRACE_JAEGER_COLLECTOR_ENDPOINT
             value: "{{ .Values.traceCollectorEndpoint }}"
           - name: TRACING_SAMPLING_RATE

charts/fission-core/values.yaml

@@ -58,13 +58,14 @@ fetcher:
 
 executor:
   adoptExistingResources: false
+  podReadyTimeout: 300s
 
 ## Router config
 router:
   deployAsDaemonSet: false
   svcAddressMaxRetries: 5
   svcAddressUpdateTimeout: 30s
-
+  unTapServiceTimeout: 3600s
   ## Display endpoint access logs
   ## To be aware of enabling logging endpoint access log, it increases
   ## router resource utilization when under heavy workloads.