Prefer highest value on duplicate container metrics

[dippynark]

What this PR does / why we need it:

If kubelet fails to cleanup cgroups from a previous instance of a container (e.g. due to some zombie process), cAdvisor will continue to expose its metrics (see kubernetes/kubernetes#134518 (comment) for more details).

The following shows an example of a kubelet v1.30.10 returning multiple metrics for the same container (same namespace/name but different UID):

$ kubectl get --raw /api/v1/nodes/[NODE_NAME]/proxy/metrics/resource | grep prometheus-prometheus-k8s-0
...
container_cpu_usage_seconds_total{container="prometheus",namespace="monitoring-system",pod="prometheus-prometheus-k8s-0"} 994631.562157 1773934038841
container_cpu_usage_seconds_total{container="prometheus",namespace="monitoring-system",pod="prometheus-prometheus-k8s-0"} 295207.606798 1773934038962
container_memory_working_set_bytes{container="prometheus",namespace="monitoring-system",pod="prometheus-prometheus-k8s-0"} 5.690306969e+10 1173934038841
container_memory_working_set_bytes{container="prometheus",namespace="monitoring-system",pod="prometheus-prometheus-k8s-0"} 3.3157259264e+10 1773934038962
container_start_time_seconds{container="prometheus",namespace="monitoring-system",pod="prometheus-prometheus-k8s-0"} 1.7691949565992033e+09

Currently Metrics Server will silently override these duplicates giving unpredictable results. This PR improves consistency by always picking the highest value when there are duplicates. In particular, if we are using these metrics to scale memory vertically with VPA, we prefer to overprovision than underpovision.

Note that doing this with CPU does not necessarily help since it is calculated as a rate, however we do this anyway just to make the response consistent (and once container_cpu_usage_seconds_total for a new instance catches up with the old one CPU utilisation for the latest container instance will be calculated properly).

This PR coupled with #1778 makes Metrics Server more resilient to the root cause which should be fixed by this PR.

[k8s-ci-robot]

This issue is currently awaiting triage.

If metrics-server contributors 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 @dippynark. 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.

Tip

We noticed you've done this a few times! Consider joining the org to skip this step and gain /lgtm and other bot rights. We recommend asking approvers on your previous PRs to sponsor you.

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.

[dippynark]

/assign @serathius

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: dippynark
Once this PR has been reviewed and has the lgtm label, please ask for approval from serathius. 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:

• OWNERS

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

[dippynark]

This is caused by a kubelet bug fixed in the following PR: kubernetes/kubernetes#137942