[bjencks]

Add a new built-in problem daemon plugin (apiserver-monitor) that checks connectivity to the Kubernetes API server.

It specifically creates a new client on every poll in order to test the full stack from TCP connection up.

This required moving kube_rest.Config creation from problemclient into pkg/util (util.NewKubeClientConfig) and initializing util.GlobalKubeClientConfig in npdMain for both problemclient and apiserver-monitor to consume.

[kubernetes-prow]

Hi @bjencks. 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.

[wangzhen127]

/ok-to-test

[wangzhen127]

Could you add e2e coverage? The unit tests only cover inClusterConfig=false over plain/insecure HTTP, so the real in-cluster path (CA + SA token + protobuf Accept against the real /version) never runs. Cheapest options: (a) a kustomize patch in test/e2e/manifests/ that adds --config.apiserver-monitor=/config/apiserver-monitor.json for the GCE cluster jobs — a smoke test of the in-cluster path on a real API server (DaemonSet stays Ready, problem_counter{reason="APIServerUnreachable"} stays 0); (b) in test/e2e/metriconly, a context that starts NPD with --apiserver-override=http://127.0.0.1:1?inClusterConfig=false and a short-interval config and asserts problem_gauge{type="APIServerUnreachable",reason="APIServerUnreachable"} == 1. Both work in metricsOnly mode since they check metrics; either needs its own short-interval config, which is another reason to settle the defaults.

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: bjencks / name: Ben Jencks (b73be69)

[wangzhen127]

/lgtm
/hold

Feel free to unhold to merge when you are ready.

[kubernetes-prow]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: bjencks, wangzhen127

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:

• OWNERS [wangzhen127]

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

[bjencks]

/hold cancel