Restart nginx in case of errors in the master process

[aledbf]

https://github.com/kubernetes/ingress/blob/master/controllers/nginx/pkg/cmd/controller/nginx.go#L101

[SleepyBrett]

Here is the pod yaml you asked for:

apiVersion: v1
kind: Pod
metadata:
  annotations:
  generateName: ingress-controller-
  labels:
    k8s-app: ingress-controller
spec:
  containers:
  - args:
    - /nginx-ingress-controller
    - --default-backend-service=kube-system/default-http-backend
    - --nginx-configmap=kube-system/ingress-controller-conf
    - --tcp-services-configmap=kube-system/ingress-controller-tcp
    env:
    - name: POD_NAME
      valueFrom:
        fieldRef:
          apiVersion: v1
          fieldPath: metadata.name
    - name: POD_NAMESPACE
      valueFrom:
        fieldRef:
          apiVersion: v1
          fieldPath: metadata.namespace
    image: gcr.io/google_containers/nginx-ingress-controller:0.8.3
    imagePullPolicy: Always
    livenessProbe:
      failureThreshold: 3
      httpGet:
        path: /healthz
        port: 18080
        scheme: HTTP
      initialDelaySeconds: 10
      periodSeconds: 10
      successThreshold: 1
      timeoutSeconds: 1
    name: nginx
    ports:
    - containerPort: 80
      hostPort: 8000
      protocol: TCP
    - containerPort: 443
      hostPort: 8443
      protocol: TCP
    - containerPort: 18080
      protocol: TCP
    readinessProbe:
      failureThreshold: 3
      httpGet:
        path: /healthz
        port: 18080
        scheme: HTTP
      periodSeconds: 10
      successThreshold: 1
      timeoutSeconds: 1
    resources:
      limits:
        cpu: 200m
        memory: 200Mi
      requests:
        cpu: 200m
        memory: 200Mi
    terminationMessagePath: /dev/termination-log
    volumeMounts:
    - mountPath: /etc/nginx/template
      name: ingress-controller-template-volume
      readOnly: true
    - mountPath: /var/run/secrets/kubernetes.io/serviceaccount
      name: default-token-neshp
      readOnly: true
  dnsPolicy: ClusterFirst
  nodeName: ip-172-24-224-171.us-west-2.compute.internal
  nodeSelector:
    beta.nordstrom.net/cluster-role: worker
  restartPolicy: Always
  securityContext: {}
  serviceAccount: default
  serviceAccountName: default
  terminationGracePeriodSeconds: 30
  volumes:
  - configMap:
      defaultMode: 420
      items:
      - key: nginx.tmpl
        path: nginx.tmpl
      name: ingress-controller-nginx-template
    name: ingress-controller-template-volume
  - name: default-token-neshp
    secret:
      defaultMode: 420
      secretName: default-token-neshp

[bprashanth]

can you expose that as a liveness probe instead, and get the kubelet to do it automatically? I mean, there's no way you're going to detect all errors that might arise at runtime without spinning off a goroutine, and if you do, you're basically re-implementing something we already expose throug the api

[aledbf]

fixed in #31