fix removing pods from podTopologyHints mapping

[aheng-ch]

What type of PR is this?

/kind bug

What this PR does / why we need it:

podTopologyHints records TopologyHints and is indexed by PodUID to ContainerName

kubernetes/pkg/kubelet/cm/topologymanager/scope.go

Line 84 in 81dd9d7

s.podTopologyHints[podUID][containerName] = th

but when remove a container, it deletes the record according to the containerID which will result in the record never being deleted

kubernetes/pkg/kubelet/cm/topologymanager/scope.go

Line 115 in 81dd9d7

delete(s.podTopologyHints[podUIDString], containerID)

Which issue(s) this PR fixes:

Fixes #

Special notes for your reviewer:

Does this PR introduce a user-facing change?

NONE

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

[k8s-ci-robot]

Thanks for your pull request. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please follow instructions at https://git.k8s.io/community/CLA.md#the-contributor-license-agreement to sign the CLA.

It may take a couple minutes for the CLA signature to be fully registered; after that, please reply here with a new comment and we'll verify. Thanks.

---

• If you've already signed a CLA, it's possible we don't have your GitHub username or you're using a different email address. Check your existing CLA data and verify that your email is set on your git commits.
• If you signed the CLA as a corporation, please sign in with your organization's credentials at https://identity.linuxfoundation.org/projects/cncf to be authorized.
• If you have done the above and are still having issues with the CLA being reported as unsigned, please log a ticket with the Linux Foundation Helpdesk: https://support.linuxfoundation.org/
• Should you encounter any issues with the Linux Foundation Helpdesk, send a message to the backup e-mail support address at: login-issues@jira.linuxfoundation.org

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. I understand the commands that are listed here.

[k8s-ci-robot]

Welcome @aheng-ch!

It looks like this is your first PR to kubernetes/kubernetes 🎉. Please refer to our pull request process documentation to help your PR have a smooth ride to approval.

You will be prompted by a bot to use commands during the review process. Do not be afraid to follow the prompts! It is okay to experiment. Here is the bot commands documentation.

You can also check if kubernetes/kubernetes has its own contribution guidelines.

You may want to refer to our testing guide if you run into trouble with your tests not passing.

If you are having difficulty getting your pull request seen, please follow the recommended escalation practices. Also, for tips and tricks in the contribution process you may want to read the Kubernetes contributor cheat sheet. We want to make sure your contribution gets all the attention it needs!

Thank you, and welcome to Kubernetes. 😃

[k8s-ci-robot]

Hi @aheng-ch. 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.

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.

[aheng-ch]

/assign @derekwaynecarr @Random-Liu @nolancon

[aheng-ch]

/kind bug

[aheng-ch]

/cc @jfbai @derekwaynecarr @nolancon

/cc @klueska

[klueska]

Hi @aheng-ch thanks for finding this issue and submitting your PR. This seems like a legit issue that should actually be backported to 1.20 and 1.21 as well. I will review it more formally early next week.

[klueska]

Looks good in general. A few comments.

[klueska]

Since this always returns nil, let's remove the error return value and make this function consistent with the other AddContainer() calls from other components.

[aheng-ch]

Thanks for the review, I will update it

[klueska]

Some of this feels redundant. I think the following would be sufficient:

if podUIDString, containerName, exists := s.podMap.GetContainerRef(containerID); exists {
	s.podMap.RemoveByContainerID(containerID)
	delete(s.podTopologyHints[podUIDString], containerName)
	if len(s.podTopologyHints[podUIDString]) == 0 {
		delete(s.podTopologyHints, podUIDString)
	}
}

[aheng-ch]

It seems ok to delete(s.podTopologyHints[podUIDString], containerName) immediately after s.podMap.RemoveByContainerID(containerID) because we don't use it after the pod is created. But podMap is the mapping of (PodUid, ContainerName) to ContainerID, when a container restart (ContainerID changed), a new record will be added to podMap. Then the contents in podMap would be like ContainerIDOld: {PodUID1, ContainerName1}, ContainerIDNew:{PodUID1, ContainerName1}. If we delete ContainerName1 in podTopologyHints when remove ContainerIDOld from podMap, it will cause the info between podMap and podTopologyHints is inconsistent. So I think it may be better to remove the container from podTopologyHints after deleting all records about this container in podMap.

[klueska]

I see. You are handling the case where a container with the same podUUID and containerName is started with a different ContainerID (i.e. after a restart). We clearly don't want to delete that new containers topology hints inadvertently.

I think this deserves a comment, since it is not obvious why we would need to re-call into the podMap in this way. How about:

// Get the podUID and containerName associated with the containerID to be removed and remove it
podUIDString, containerName, err := s.podMap.GetContainerRef(containerID)
if err != nil {
	return nil
}
s.podMap.RemoveByContainerID(containerID)

// In cases where a container has been restarted, it's possible that the same podUID and
// containerName are already associated with a *different* containerID now. Only remove
// the TopologyHints associated with that podUID and containerName if this is not true.
if _, err := s.podMap.GetContainerID(podUIDString, containerName); err != nil {
	delete(s.podTopologyHints[podUIDString], containerName)
	if len(s.podTopologyHints[podUIDString]) == 0 {
		delete(s.podTopologyHints, podUIDString)
	}
}

[aheng-ch]

Thanks a lot for your suggestion

[klueska]

Can we break this into two checks with different errors based on which one was incorrect?

[aheng-ch]

Yeah, you are right, I will update it

[klueska]

Looks good. Let's backport this to 1.20 and 1.21 once this is merged.

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: aheng-ch, klueska

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:

• pkg/kubelet/cm/OWNERS [klueska]

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

[aheng-ch]

/test pull-kubernetes-e2e-kind
/test pull-kubernetes-integration

[ffromani]

/triage accepted
/priority priority/important-soon
(while important, this doesn't seem to require critical-urgent)

[k8s-ci-robot]

@fromanirh: The label(s) priority/priority/important-soon cannot be applied, because the repository doesn't have them.

In response to this:

/triage accepted
/priority priority/important-soon
(while important, this doesn't seem to require critical-urgent)

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.

[ffromani]

/priority important-soon