Promote LoadBalancerIPMode to GA

[RyanAoh]

What type of PR is this?

/kind feature

What this PR does / why we need it:

Promote LoadBalancerIPMode feature to GA.

Which issue(s) this PR fixes:

Part of kubernetes/enhancements#1860

Special notes for your reviewer:

Does this PR introduce a user-facing change?

LoadBalancerIPMode feature is now marked as GA.

Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

• KEP
• Usage

[k8s-ci-robot]

This issue is currently awaiting triage.

If a SIG or subproject 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.

Details

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.

[RyanAoh]

/sig network

[aojea]

I always forget, if is locked to default don't we remove the features gates from the code too @thockin ?

[aojea]

I just realized there is no e2e test covering this feature, we should think in how to add one

[thockin]

e2e is challenging, indeed.

[thockin]

I always forget, if is locked to default don't we remove the features gates from the code too @thockin?

We used to, but now with the multi-version compatibility, I think this is different - @jpbetz did we write up the new gate lifecycle? When do we actually remove the gate usage from the code (not the gate, but the use of it to disable features) ?

[thockin]

I don't really know how to e2e this. Would it be sufficient to test that iptables does not have a short-circuit rule? That seems super brittle.

[aojea]

I don't really know how to e2e this. Would it be sufficient to test that iptables does not have a short-circuit rule? That seems super brittle.

We already have unit and integration test https://github.com/search?q=repo%3Akubernetes%2Fkubernetes+LoadBalancerIPMode+path%3A%2F%5Epkg%5C%2Fproxy%5C%2F%2F&type=code and cloud-provider-kind implements this mode, so on my side I'm fine with the existing testing coverage for this feature

/lgtm

@thockin defer approval to you

[aojea]

/lgtm cancel

@RyanAoh please remove all the references to the feature gate from the code base

https://github.com/kubernetes/community/blob/master/contributors/devel/sig-architecture/feature-gates.md#ga-features

When we set LockToDefault to true, we also remove all references to the feature gate from the codebase.

[liggitt]

@RyanAoh please remove all the references to the feature gate from the code base

cc @Jefftree @jpbetz on normal process of graduating a gate removing all conditionals ... this means 1.32 would not be able to emulate <=1.31 where the feature was beta and disableable

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: aojea, RyanAoh, thockin

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• pkg/apis/OWNERS [thockin]
• pkg/features/OWNERS [aojea,thockin]
• pkg/proxy/OWNERS [aojea,thockin]
• pkg/registry/core/service/OWNERS [aojea,thockin]
• test/OWNERS [aojea,thockin]
• test/featuregates_linter/test_data/OWNERS [aojea,thockin]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[Jefftree]

/lgtm

(Promotion has already been approved #127348 (comment), #127348 (comment), and this review is for the feature gate testing changes da5738d)

Thanks! Apologies for all the new changes required with the feature gates.

[k8s-ci-robot]

LGTM label has been added.

Details

Git tree hash: 3cae7b75a082a05fe7131871c0598db784cab91d