LB not working with externalTrafficPolicy set to Local

[JrCs]

/kind bug

What happened:
If we create a service of Type LoadBalancer and ### externalTrafficPolicy with value Local the LB that will be create will not work as expected.

What you expected to happen:
Only LB traffic to nodes that have the pod running.

How to reproduce it (as minimally and precisely as possible):
Create a service of type LoadBalancer with externalTrafficPolicy set to Local.

Anything else we need to know?:
When using externalTrafficPolicy with value Local, the LB that will be create must have only members (nodes) where the pod of the service we want to access is running.
In case of externalTrafficPolicy set to Cluster, the LB that will be create must have all the worker nodes as members.

Environment:

• openstack-cloud-controller-manager version: v1.15.0
• OS (e.g. from /etc/os-release): Container Linux by CoreOS 2191.4.1
• Kernel (e.g. uname -a): 4.19.66-coreos

[lingxiankong]

The current OCCM implementation(based on OpenStack) doesn't support this feature, as per https://kubernetes.io/docs/concepts/services-networking/service/#aws-nlb-support and https://kubernetes.io/docs/tasks/access-application-cluster/create-external-load-balancer/#preserving-the-client-source-ip, seems this feature implementation is different from cloud to cloud.

If your purpose of using this param is to preserve the client source IP, you can use loadbalancer.openstack.org/x-forwarded-for: "true" in the annotation of LoadBalancer type Service.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[bgagnon]

/remove-lifecycle rotten

[bgagnon]

The option serves two mostly unrelated concerns:

• preserve the client ip
• remove nodes that do not have an endpoint from the load balancer rotation

The second point, according to docs:

[...] if you’re running on Google Kubernetes Engine/GCE, setting the same service.spec.externalTrafficPolicy field to Local forces nodes without Service endpoints to remove themselves from the list of nodes eligible for loadbalanced traffic by deliberately failing health checks.

This feature is valuable in very large cluster running an ingress controller. In this scenario, every few nodes have the ingress proxy (nginx, envoy, etc.), so there's no gain in including them in the cloud load balancer.

If we take health check hint from GCE above, I believe this can be implemented similarly in Openstack.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/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.

[bgagnon]

/reopen
/remove-lifecycle rotten

[k8s-ci-robot]

@bgagnon: You can't reopen an issue/PR unless you authored it or you are a collaborator.

In response to this:

/reopen
/remove-lifecycle rotten

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.

[bgagnon]

@lingxiankong can you please reopen? thanks!

[JrCs]

/reopen
/remove-lifecycle rotten

[k8s-ci-robot]

@JrCs: Reopened this issue.

In response to this:

/reopen
/remove-lifecycle rotten

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.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/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.

[jichenjc]

/reopen

reopen for bug track

[k8s-ci-robot]

@jichenjc: Reopened this issue.

In response to this:

/reopen

reopen for bug track

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.