Canonicalize the IPv6 Addresses when they exist

[barbacbd]

Canonicalize the IPv6 Addresses when they exist.

providers/gce/gce_instances.go:
IPv6 Address short notation is treated differently by cluster components. Canonicalize the addresses to ensure
a consistent comparison.

Tests are created with Claude and updated by @barbacbd.

This addresses OCPBUGS-79354.

[k8s-ci-robot]

This issue is currently awaiting triage.

If the repository mantainers determine 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.

[k8s-ci-robot]

Hi @barbacbd. Thanks for your PR.

I'm waiting for a kubernetes 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.

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.

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: barbacbd
Once this PR has been reviewed and has the lgtm label, please assign mrhohn for approval. For more information see the Code Review Process.

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

Details

Needs approval from an approver in each of these files:

• providers/gce/OWNERS

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

[barbacbd]

Originally found by @tthvo

/cc @tthvo

[k8s-ci-robot]

@barbacbd: GitHub didn't allow me to request PR reviews from the following users: tthvo.

Note that only kubernetes members and repo collaborators can review this PR, and authors cannot review their own PRs.

Details

In response to this:

Originally found by @tthvo

/cc @tthvo

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.

[hdp617]

Thanks! Would you mind adding unit tests?

/ok-to-test

[barbacbd]

@hdp617 thank you, it should be updated to include tests now.

[k8s-ci-robot]

@barbacbd: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
cloud-provider-gcp-e2e-full	6f4e89b	link	true	/test cloud-provider-gcp-e2e-full
pull-cloud-provider-gcp-scenario-kops-simple	6f4e89b	link	false	/test pull-cloud-provider-gcp-scenario-kops-simple
cloud-provider-gcp-tests	6f4e89b	link	true	/test cloud-provider-gcp-tests

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

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. I understand the commands that are listed here.

[aojea]

this can create controller hotloops, we got bit by this in kubernetes multiple times , and most of the time is a bug in the client rather than in the server side.

I do not have access to the bug but I want to understand better this problem and why this solution.

/hold
/assign @marqc

[k8s-ci-robot]

@aojea: GitHub didn't allow me to assign the following users: marqc.

Note that only kubernetes members with read permissions, repo collaborators and people who have commented on this issue/PR can be assigned. Additionally, issues/PRs can only have 10 assignees at the same time.
For more information please see the contributor guide

Details

In response to this:

this can create controller hotloops, we got bit by this in kubernetes multiple times , and most of the time is a bug in the client rather than in the server side.

I do not have access to the bug but I want to understand better this problem and why this solution.

/hold
/assign @marqc

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.

[tthvo]

this can create controller hotloops, we got bit by this in kubernetes multiple times , and most of the time is a bug in the client rather than in the server side.

I do not have access to the bug but I want to understand better this problem and why this solution.

@aojea Indeed, we came across this "problem" when one of openshift cluster components assumes the node IPv6 addresses to be in canonical format. In our case, the etcd-operator made such an assumption; and ran into a loop of IP mismatch (compare full vs canonical form), causing a constant cert rotation --> constant pod rollout --> etcd unhealthy.

We should probably best fix it there. But, looking at cloud-provider-aws approach, which writes the canonical IPv6 address, I was just wondering if canonical/short form is the standard way and cloud-provider-gcp should also follow (hence this PR). WDYT?

[tthvo]

I was just wondering if canonical/short form is the standard way and cloud-provider-gcp should also follow (hence this PR).

hmm, just as I wrote this, I saw cloud-provider-azure does not canonicalize the ipv6 address at all. So, maybe we really should fix in our consumer side.

Though, do you know if there is/should-be a standard or enforced way for this at all?

[aojea]

Though, do you know if there is/should-be a standard or enforced way for this at all?

server side should not workaround clients problems, client write A it should expect to read A not f(A) ... kubernetes has this canonicalize hook on the apiserver, and it is not used because the few apis that used it ended in hotloop problems like the one you are describing,

[barbacbd]

Closed in favor of openshift/cluster-etcd-operator#1577. Rather than potentially upload a breaking change, we make the change on our "client" side.