EndpointSlice support for IP target groups

[harivall]

Description

Added support for EndpointSlices for TargetGroupBindings/pod endpoint resolution for IP targets!

Components added:

• new event handler for EndpointSlices, unit tests
• method for pod endpoint resolution from EndpointSlices, unit tests
• command line flag for enabling the use of EndpointSlices. Based on this flag:
  • conditional use of Endpoints vs. EndpointSlices event handlers in TGB controller
  • conditional use of Endpoints vs. EndpointSlices pod endpoint resolution methods in TGB resource manager
• update aws-load-balancer-controller-role to be able to access EndpointSlices

Checklist

• Added tests that cover your change (if possible)
• Added/modified documentation as required (such as the README.md, or the docs directory)
• Manually tested
• Made sure the title of the PR is a good description that can go into the release notes

BONUS POINTS checklist: complete for good vibes and maybe prizes?! 🤯

• Backfilled missing tests for code in same general area 🎉
• Refactored something and made the world a better place 🌟

[k8s-ci-robot]

Hi @harivall. Thanks for your PR.

I'm waiting for a kubernetes-sigs member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

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.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 66.10169% with 20 lines in your changes missing coverage. Please review.

Project coverage is 53.29%. Comparing base (2a1a0a3) to head (eaa723f).
Report is 551 commits behind head on main.

Files with missing lines	Patch %	Lines
pkg/backend/endpoint_resolver.go	81.25%	5 Missing and 4 partials ⚠️
pkg/targetgroupbinding/resource_manager.go	0.00%	9 Missing ⚠️
pkg/config/controller_config.go	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2169      +/-   ##
==========================================
+ Coverage   53.18%   53.29%   +0.11%     
==========================================
  Files         135      135              
  Lines        7446     7503      +57     
==========================================
+ Hits         3960     3999      +39     
- Misses       3178     3192      +14     
- Partials      308      312       +4     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[kishorj]

@harivall, thanks a lot for your contribution.

Please document the following tests in the description

• Scale deployment to at least 1000 replicas, check the endpoint slices, verify the target group got updated as expected for both ALB and NLB
• Specify at least 5 ports in the service spec, verify NLB gets provisioned with the listeners and target groups get populated as expected
• Create an app deployment, wait until the target group reconcile is complete and targets are healthy, then induce readiness probe failure in half of the replicas. Verify target group gets reconciled as expected. When the readiness check pass again, target group should be updated.
• Induce liveness check failure, and the container gets restarted. Verify the behavior.
• Setup pod readiness gate on a namespace, verify all endpoints get registered as expected. Also verify endpoints with container in not-ready state.
• Run the e2e tests on your cluster and make sure all the e2e tests pass
  • ginkgo -v -r test/e2e -- --kubeconfig=$KUBECONFIG --cluster-name=<Cluster_name> --aws-region=<region> --aws-vpc-id=<vpc_id>
• Verify upgrade works as expected. For example, create ALB/NLB with the old controller version (or with endpoint slices disabled), then upgrade to the new controller version (or enable endpoint slices). Existing ALB/NLB target group should continue to work as expected.

[kishorj]

You need to specify this via the kube-builder annotation. Please refer to the following

aws-load-balancer-controller/controllers/elbv2/targetgroupbinding_controller.go

Line 83 in f861a8e

// +kubebuilder:rbac:groups="",resources=endpoints,verbs=get;list;watch

Once you specify the kubebuilder annotation, make generate will automatically generate this file

[kishorj]

/ok-to-test

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: harivall, kishorj

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

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [kishorj]

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

[kishorj]

/lgtm