figure out how to test clusterTraifficPolicy=cluster vs local, in KPNG - Githubissues

kubernetes-retired / kpng

[EOL] Reworking kube-proxy's architecture

Apache License 2.0

246 stars 90 forks source link

figure out how to test clusterTraifficPolicy=cluster vs local, in KPNG #305

Closed jayunit100 closed 1 year ago

jayunit100 commented 2 years ago

Looking at https://github.com/kubernetes-sigs/kpng/pull/287/

possibly add new tests upstream (or see if theyre there already)

Talking to @mcluseau , he's trying to support the ability of the kpng proxy to make sure that it only uses local endpoints when "clusterTrafficPolicy=local".
to test this, we can make a service w/ externalIP or NodePort to pull a servieEndpoint=local, where EACH ENDPOINT serves its nodename ... and you NEVER get more then one value. This means theres zero node bounce forwarding occuring
similar to https://github.com/kubernetes/kubernetes/pull/110967 , we might need a new upstream e2e not sure yet

lets also ask the question, generally to upstream k8s sig

Services [It] should preserve source pod IP for traffic thru service cluster
should we perserve source addresses for NODEPORTs
... when externalTrafficPolicy=local ????

( DONT DO THIS but, its just a note, theres another way test this..... ( setup a cluster where nodes cant forward traffic somehow , and see connection failures when using ) )

jayunit100 commented 2 years ago

i.e. "services should not source NAT unnecessarily" - mc

astoycos commented 2 years ago

We should be able to test this with an extra "remote" container

mcluseau commented 2 years ago

@astoycos yes, I was thinking of a daemonset returning its nodeName ie over HTTP

astoycos commented 2 years ago

@mcluseau I can take a look at this :) we do something similar in ovn-kubernetes

/assign

astoycos commented 2 years ago

Opening this up for folks

astoycos commented 2 years ago

/unassign

astoycos commented 2 years ago

I can still help here though please reach out if you need help!

k8s-triage-robot commented 1 year ago

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Mark this issue or PR as fresh with /remove-lifecycle stale
Mark this issue or PR as rotten with /lifecycle rotten
Close this issue or PR with /close
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

k8s-triage-robot commented 1 year ago

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Mark this issue as fresh with /remove-lifecycle rotten
Close this issue with /close
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten

emabota commented 1 year ago

Hi @astoycos , can you help/guide me on with this issue? Where should I start from?

astoycos commented 1 year ago

Hey @emabota That'd be great, thanks for the help

I think what we can do is

Spin up an external client container
Make a Nodeport Service and set ETP=local
Make sure the External client container -> Nodeport service ends up on the correct node

For the backend pod using https://pkg.go.dev/k8s.io/kubernetes/test/images/agnhost/netexec often is super helpful here since it has a /hostname endpoint

emabota commented 1 year ago

/assign

k8s-triage-robot commented 1 year ago

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues according to the following rules:

After 90d of inactivity, lifecycle/stale is applied
After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

Reopen this issue with /reopen
Mark this issue as fresh with /remove-lifecycle rotten
Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/close not-planned

k8s-ci-robot commented 1 year ago

@k8s-triage-robot: Closing this issue, marking it as "Not Planned".

In response to [this](https://github.com/kubernetes-sigs/kpng/issues/305#issuecomment-1490774456): >The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs. > >This bot triages issues according to the following rules: >- After 90d of inactivity, `lifecycle/stale` is applied >- After 30d of inactivity since `lifecycle/stale` was applied, `lifecycle/rotten` is applied >- After 30d of inactivity since `lifecycle/rotten` was applied, the issue is closed > >You can: >- Reopen this issue with `/reopen` >- Mark this issue as fresh with `/remove-lifecycle rotten` >- Offer to help out with [Issue Triage][1] > >Please send feedback to sig-contributor-experience at [kubernetes/community](https://github.com/kubernetes/community). > >/close not-planned > >[1]: https://www.kubernetes.dev/docs/guide/issue-triage/ Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.