unhealthy backend services, with 400, 412, 409 errors

[chaima-ennar]

I am trying to deploy an application composed of two services front and back, I am using ingress and "gce" as a controller, Every time I modify the docker image of the service ( I add modifications in the code and nothing beoyond that), the service turns to be unhealthy. and I got these errors ,

loadbalancer-controller received errors when updating backend service: googleapi: Error 400: The resource 'projects/fl../global/backendServices/k8s-be-32112--17c72.....ece101' is not ready, resourceNotReady

loadbalancer-controller googleapi: Error 412: Invalid fingerprint., conditionNotMet

error 404 services already exists,

here is my ingress ressource

piVersion: extensions/v1beta1
kind: Ingress
metadata:
  labels:
    app: app
    part: ingress
  name: my-irool-ingress
  annotations:
    kubernetes.io/ingress.class: "gce"
    kubernetes.io/ingress.global-static-ip-name: my-ip
spec:
 backend:
    serviceName: client-svc
    servicePort: 80
 rules:
  - http:
        paths:
        - path: /back
          backend:
            serviceName: back-svc
            servicePort: 9000
  - http:
        paths:
        - path: /back/*
          backend:
            serviceName: back-svc
            servicePort: 9000

apiVersion: v1
kind: Service
metadata:
  labels:
    app: app
    part: back
  name: back-svc
  namespace: default
spec:
  type: NodePort
  ports:
  - port: 9000
    nodePort: 30049
    protocol: TCP
  selector:
    app: app
    part: back

And I the result is ?
ingress.kubernetes.io/backends: {"k8s-be-30049--17c7.....e101":"UNHEALTHY","k8s-be-30651--17c7......e101":"HEALTHY"}

So could you please to find the issue here ?

[rramkumar1]

@chaima-ennar Do you know what version of ingress-gce you are running?

[chaima-ennar]

@rramkumar1 I didn't really specify a version of the gce controller, so I guess it's the latest by default when I specified the controller class is "gce"

[chaima-ennar]

Update: I have a 2 node cluster, for the healthy service I noticed that 2/2 vm instances are running but with the unhealthy service it's marked 0/2 instances are operational

[rramkumar1]

@chaima-ennar Why are your instances not operational? I suspect that fixing your vm instances will fix this issue.

[chaima-ennar]

@rramkumar1 I don't know the same instances are used by both services front and back, they are working fine when it comes to the front service though, ( and I am pretty sure that the back service works fine) . I think it is something wrong with the instance group created by the ingress

[rramkumar1]

@chaima-ennar Are you still encountering this?

[rramkumar1]

@chaima-ennar Going to close this for now given the inactivity. Please re-open if you feel the issue still occurring.

/close

[k8s-ci-robot]

@rramkumar1: Closing this issue.

In response to this:

@chaima-ennar Going to close this for now given the inactivity. Please re-open if you feel the issue still occurring.

/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.