Decouple test image bumping from prow bumping

[spiffxp]

https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/99655/pull-kubernetes-unit/1367006331575209984
=> prowjob yaml button =>
https://prow.k8s.io/prowjob?prowjob=cb9cbe1e-7bed-11eb-97bd-22fa521bf179
=> look for image: =>
gcr.io/k8s-testimages/kubekins-e2e:v20210226-c001921-master

yeah that's not going to have the go1.16, looks like we haven't had a prow bump since 02/26

we need to decouple these update loops (images for prow itself, test images) as we discussed at the last SIG testing. cc @spiffxp

Originally posted by @BenTheElder in kubernetes/kubernetes#99655 (comment)

[spiffxp]

/sig-testing
/area jobs
/area artifacts
/sig-release
/area release-eng
/priority important-soon
/milestone v1.21

[BenTheElder]

cc @chaodaiG

[chaodaiG]

cc @mpherman2 as well

[chaodaiG]

/assign

[spiffxp]

Pulling out of #21144 (review)

One question for followup: is there some way we could have merges of just these autobump PRs alert to slack?

A longer term thing: someplace to see all deploys and all config changes pushed to prow.k8s.io to aid "what changed when"

But for now being filter down just to "big PRs that change a lot" matches the current experience people are used to. I'd rather not take away what they're used to before having something better.

[spiffxp]

Let's figure out what it would look like to hand this off to community. My sketch:

• if they can, community members in appropriate OWNERS can approve these now
• wait until after code freeze, formally handoff ASAP
• use existing groups (ci-signal or release-eng) or use this to bootstrap a prowjob-oncall
• making sure the group knows a PR is available to merge
  • start with empty oncall field and hardcode ping to appropriate github team?
  • move to something generated by and stored in community
  • consider moving to auto-merge
• make sure there are visible notifications when these PRs merge
  • continue practice of manually posting on slack
  • post-merge alerts sent to #prow-alerts?
  • update config-updater to send to slack?
• document that this is how jobs configs are bumped for prow.k8s.io (a linkable subheading of https://github.com/kubernetes/test-infra/tree/master/config/jobs if there isn't a better place)

[spiffxp]

/reopen
the job landed, but there's still more to do

[k8s-ci-robot]

@spiffxp: Reopened this issue.

In response to this:

/reopen
the job landed, but there's still more to do

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.

[chaodaiG]

Also #21144 probably had left out some of the images in prow jobs that's part of prow but not gcr.io/k8s-testimages. I will take a look at the new bumping PRs generated and see what's missing

[spiffxp]

/sig testing

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-contributor-experience at kubernetes/community.
/lifecycle stale

[BenTheElder]

/remove-lifecycle stale

[BenTheElder]

@chaodaiG this was completed right?

[chaodaiG]

Yep it was completed

[chaodaiG]

/close

[k8s-ci-robot]

@chaodaiG: Closing this issue.

In response to this:

/close

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.

[BenTheElder]

Thank you!

[spiffxp]

/reopen
Handoff never happened as described in #21137 (comment)

[k8s-ci-robot]

@spiffxp: Reopened this issue.

In response to this:

/reopen
Handoff never happened as described in #21137 (comment)

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.

[spiffxp]

FYI @kubernetes/release-engineering

[justaugustus]

For SIG Release: I'm happy to have @kubernetes/release-managers in this approval group

[puerco]

/cc

[spiffxp]

/assign

[spiffxp]

/milestone v1.23

[spiffxp]

Thanks to the following PRs:

• config/jobs: add ci-test-infra-autobump-prowjobs-for-auto-deploy #23602
• config/prow: bump all community test-infra images #23610
• sig-testing: update test-infra teams org#2979
• config/prow: allow k8s-infra-ci-robot to skip-review #23644
• config/jobs: autobump prowjobs always #23736

We now autobump job images and automatically deploy the change twice a day (if there are any changes to deploy):

• https://github.com/kubernetes/test-infra/pulls?q=is%3Apr+author%3Ak8s-infra-ci-robot+is%3Aclosed++label%3Askip-review
• https://testgrid.k8s.io/sig-testing-prow#autobump-prowjobs-for-auto-deploy

[spiffxp]

/close
Announcement: https://groups.google.com/g/kubernetes-dev/c/aBzQQw8KgUU/m/wnOtP5vjDAAJ

[k8s-ci-robot]

@spiffxp: Closing this issue.

In response to this:

/close
Announcement: https://groups.google.com/g/kubernetes-dev/c/aBzQQw8KgUU/m/wnOtP5vjDAAJ

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.