when the service port changes, the ingress cannot detect the change and adjust the upstream port.

[zzzzoy]

What happened:

To illustrate this problem in a simple example,

1. I created an nginx service, and used the wrong targetPort port：

deploy.yaml：

apiVersion: apps/v1
kind: Deployment
metadata:
name: tony
namespace: test
spec:
replicas: 1
selector:
matchLabels:
app: tony
version: v1
template:
metadata:
labels:
app: tony
version: v1
spec:
containers:
- image: nginx:latest
imagePullPolicy: IfNotPresent
name: container-0
imagePullSecrets:
- name: default-secret
restartPolicy: Always
schedulerName: default-scheduler

service.yaml:

apiVersion: v1
kind: Service
metadata:
labels:
app: tony
name: tony
namespace: test
spec:
internalTrafficPolicy: Cluster
ports:

• name: cce-service-0
  port: 80
  protocol: TCP
  targetPort: 81
  selector:
  app: tony
  type: ClusterIP

2. create ingress for this service, now we can try to access www.test.com but ingress will response 502 Bad Gateway

ingress.yaml:

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: test
namespace: test
spec:
rules:

• host: www.test.com
  http:
  paths:
  • backend:
    service:
    name: tony
    port:
    number: 80
    path: /
    pathType: ImplementationSpecific

3. modify the targetPort of service and apply

service.yaml:

apiVersion: v1
kind: Service
metadata:
labels:
app: tony
name: tony
namespace: test
spec:
internalTrafficPolicy: Cluster
ports:

• name: cce-service-0
  port: 80
  protocol: TCP
  targetPort: 81
  selector:
  app: tony
  type: ClusterIP

4. try to access www.test.com but still get response 502 Bad Gateway, i have to rebuild ingress by kubectl delete -f ingress.yaml && kubectl apply -f ingress.yaml and get correct response message

What you expected to happen:

when I modify a service, the ingress should reload the service to ensure the correctness of the forwarding result.
i wonder, is this a normal situation? or ingress cannot be supported this feature for some reason

maybe the ingress cannot identify service changes

NGINX Ingress controller version (exec into the pod and run nginx-ingress-controller --version.):

Kubernetes version (use kubectl version):

Environment:

• Cloud provider or hardware configuration:

• OS (e.g. from /etc/os-release): Centos 7.6

• Kernel (e.g. uname -a):

• Linux 192-168-0-174 3.10.0-862.14.4.el7.x86_64 Basic structure  #1 SMP Wed Sep 26 15:12:11 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

• Install tools:
  kubeadm

• Basic cluster related info:

  • kubectl version
  • v1.23.15
  • kubectl get nodes -o wide
  • NAME STATUS ROLES AGE VERSION INTERNAL-IP EXTERNAL-IP OS-IMAGE KERNEL-VERSION CONTAINER-RUNTIME
    192.168.0.121 Ready 3d21h v1.23.15 192.168.0.121 CentOS Linux 7 (Core) 3.10.0-862.14.4.el7.x86_64 docker://18.9.0
    192.168.0.174 Ready etcd,master 4d1h v1.23.15 192.168.0.174 CentOS Linux 7 (Core) 3.10.0-862.14.4.el7.x86_64 docker://18.9.0
    192.168.0.185 Ready etcd,master 4d1h v1.23.15 192.168.0.185 CentOS Linux 7 (Core) 3.10.0-957.el7.x86_64 docker://18.9.0
    192.168.0.57 Ready etcd,master 4d1h v1.23.15 192.168.0.57 CentOS Linux 7 (Core) 3.10.0-957.el7.x86_64 docker://18.9.0

• How was the ingress-nginx-controller installed: helm

• Current State of the controller: running

[k8s-ci-robot]

This issue is currently awaiting triage.

If Ingress contributors determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

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-sigs/prow repository.

[longwuyuan]

• The information you have provided is very very vague as it can not be used for reproducing a problem in a way that helps
• You can help the readers by first looking at the template of a new bug report and answer the questions that are asked in the template, here, by editing the issue description of this issue
• You can also help by ensuring markdown format because right now its harder to read the data
• Looks like you changes some specs and want to see response change but its not even clear if you get a response from a cache like a browser cache
• After you have provided the info as suggested above, please try to help the readers here by providing precise instructions that someone can copy/paste from , to reproduce the problem you are having. Do not make it vague like refreshing a browser

Once the information in the issue description is helpful to readers here, then you can re-open this issue. I will close it for now as there is no action item here.

/remove-kind bug
/close

[k8s-ci-robot]

@longwuyuan: Closing this issue.

In response to this:

• The information you have provided is very very vague as it can not be used for reproducing a problem in a way that helps
• You can help the readers by first looking at the template of a new bug report and answer the questions that are asked in the template, here, by editing the issue description of this issue
• You can also help by ensuring markdown format because right now its harder to read the data
• Looks like you changes some specs and want to see response change but its not even clear if you get a response from a cache like a browser cache
• After you have provided the info as suggested above, please try to help the readers here by providing precise instructions that someone can copy/paste from , to reproduce the problem you are having. Do not make it vague like refreshing a browser

Once the information in the issue description is helpful to readers here, then you can re-open this issue. I will close it for now as there is no action item here.

/remove-kind bug
/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-sigs/prow repository.