Migrate to Community CI By August 1st

[BenTheElder]

Node problem detector CI still needs migrating, per https://groups.google.com/a/kubernetes.io/g/dev/c/p6PAML90ZOU

You can find node problem detector jobs listed in:

https://github.com/kubernetes/test-infra/blob/master/docs/job-migration-todo.md

(see e.g. jobs with name containing "npd")

[wangzhen127]

/cc @wangzhen127

[wangzhen127]

The following tests still need migration:

• CI:
  • ci-npd-build
• presubmit:
  • pull-npd-e2e-node
  • pull-npd-e2e-kubernetes-gce-gci
  • pull-npd-e2e-kubernetes-gce-gci-custom-flags
  • pull-npd-e2e-kubernetes-gce-ubuntu
  • pull-npd-e2e-kubernetes-gce-ubuntu-custom-flags

kubernetes/kubernetes#119211 tracked the issues we encountered previously.

[wangzhen127]

/cc @DigitalVeer, @AnishShah, @cwangVT

[wangzhen127]

/assign @DigitalVeer

[k8s-ci-robot]

@wangzhen127: GitHub didn't allow me to assign the following users: digitalveer.

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

In response to this:

/assign @DigitalVeer

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]

I will work with @DigitalVeer to look into this.

[wangzhen127]

There is also similar work tracked in kubernetes/test-infra#31794

[BenTheElder]

Yeah, kubernetes/test-infra#31794 was tracking the community migrating it, raised here for visibility that the deadline is approaching.

I spoke with @wangzhen127 and another colleague about this, we'll get this pushing to a community bucket and migrate over.

/assign

#925 will make the problem a bit simpler.

[BenTheElder]

OK so after #925, I spoke to my fellow SIG K8s Infra TLs.

There is a gs://k8s-staging-npd already created for NPD in k8s infra, since we're not using it for publishing binaries (thank you!) we can add IAM permission to prow to upload there from the community cluster and switch the scripts over.

That, plus no longer pushing to GCR from CI, should fix everything.

The IAM is managed in https://github.com/kubernetes/k8s.io, can take a look at this later and then we can see about switching buckets.