Cut v1.33.0-rc.0 release

[mbianchidev]

Scheduled to happen: Tue, 2025-04-08

Release Blocking Issues

N/A

Release Steps

• Create a thread on #release-management: https://kubernetes.slack.com/archives/CJH2GBF7Y/p1744125003875769
• Screenshot unhealthy release branch testgrid boards
• Check the health of the publishing-bot
• Mock Run
  • Stage
  • Release
• NoMock Run
  • Stage
  • Image Promotion: releng: Image promotion for kubernetes v1.33.0-rc.0 k8s.io#7971 and releng: Image promotion for kubernetes v1.34.0-alpha.0 k8s.io#7972
  • Release
• Notify #release-management:
• Send notification:
• Update schedule.yaml file with the latest release using schedule-builder
• Collect metrics and add them to the Release steps table below

Release Tools Version

GitVersion:    v0.18.1-0.20250408060241-9a9572f7c48f
GitCommit:     9a9572f7c48f637de8499a201fb8e3ff52f8d4ba
GitTreeState:  clean
BuildDate:     2025-04-08T06:02:41
GoVersion:     go1.24.2
Compiler:      gc
Platform:      darwin/arm64

Release Jobs History

Step	Command	Link	Start	Duration	Succeeded?
mock stage	krel stage --type rc --branch release-1.33 --build-version v1.33.0-rc.0.24+2f7a115a691ce0	https://console.cloud.google.com/cloud-build/builds/4bc8d329-e92b-434b-9208-aa368606cf4a?project=648026197307	2025-04-15T14:46:42.049437006Z	00:56:39	Yes
mock release	krel release --type rc --branch release-1.33 --build-version v1.33.0-rc.0.24+2f7a115a691ce0	https://console.cloud.google.com/cloud-build/builds/7cc8c0c2-3806-4b43-8be9-0b2d76da850d?project=648026197307	2025-04-15T16:08:39.871102725Z	00:15:25	Yes
stage	krel stage --type rc --branch release-1.33 --build-version v1.33.0-rc.0.24+2f7a115a691ce0 --nomock	https://console.cloud.google.com/cloud-build/builds/a82f4d74-590e-45ca-aafa-53008bd11da6?project=648026197307	2025-04-15T16:39:40.603991326Z	01:21:33	Yes
release	krel release --type rc --branch release-1.33 --build-version v1.33.0-rc.0.24+2f7a115a691ce0 --nomock	https://console.cloud.google.com/cloud-build/builds/f5535ba0-0f59-4014-83e5-15a059ff5339?project=648026197307	2025-04-15T20:39:06.425011800Z	00:09:39	No, Canceled
release	krel release --type rc --branch release-1.33 --build-version v1.33.0-rc.0.24+2f7a115a691ce0 --nomock	https://console.cloud.google.com/cloud-build/builds/7d7f8e7d-370f-4491-926c-a4ec50d2056a?project=648026197307	2025-04-15T21:23:45.503888339Z	00:18:49	Yes

Action Items

• Docs PR to be merged Adding release cut onboarding handbook #2731
• KPromo Docs PR to be merged Update kpromo docs kubernetes-sigs/promo-tools#1517
• Script improvement PR Improved release branch prep python script test-infra#34690
• Document post rc.0 release process, now unclear and with undocumented magic - opened Documentation for post rc.0 cut tasks #2776

Open Questions (related to 1.33.0)

• CI jobs occasionally hitting node not ready issue presented again with 1.33.0 how do we tackle it?
• Rerunning a failing job was a bit chaotic and generated 3 different re-runs of the failing test sig-release-1.33-blocking#Conformance-GCE-1.33-kubetest2 (failure due to "Node not ready, deemed to be infra being flaky) - how do we ensure reruns are actually queued against each other if they are multiple for the same test? Opened [feature] Rerun queue discard extra runs kubernetes-sigs/prow#436
• Add monitoring for OBS resources usage extracting usage from https://build.opensuse.org/monitor, opened [feature] Krel command to check OBS resource usage/status release#3992
• Too many requests / quota limits during image promo resolution (see proposal doc) - issue tracking this image-promotion hits 429 quota limits kubernetes-sigs/promo-tools#1271

/milestone v1.33
/assign @mbianchidev
/cc @kubernetes/release-managers
/priority important-soon
/kind documentation

[mbianchidev]

Testgrid dashboards for master

Boards checked for FAILING:

• sig-release-master-blocking
• sig-release-master-informing

No FAILING Jobs

comment generated by krel

[mbianchidev]

Documenting the whole process to follow after any rc.0 cut:

1. https://github.com/kubernetes/test-infra/pull/34650/files
2. https://github.com/kubernetes/test-infra/pull/34651/files
3. Most stuff is done by this PR from Jim https://github.com/kubernetes/test-infra/pull/34668/files (cmake command)
   3.1 read and write about this Slack thread for detailed steps for this: (note: most stuff is autogenerated by the cmake command)
4. ./hack/run-in-python-container.sh ./releng/generate_tests.py --yaml-config-path ./releng/test_config.yaml (see this PR)
5. Publishing bot staging/publishing: add release-1.33 rules kubernetes#131250
6. Look at this DM from Jim with some more info on the CLI side of things
7. Perf branch Need 1.33 branch? perf-tests#3290 (comment)

---

6. Remove 1.29 jobs drop 1.29 from image variants, 1.29 is EOL test-infra#34674 (version to be removed or not varies depending on where in the overall release cycle we are in terms of EOL support) - this could actually be step 0, first drop old jobs and then add new ones so the prow cluster is never overwhelmed - but it could be actually a patch release job that has to be performed instead of doing it before/during rc.1 (see Slack convo with @BenTheElder )

Extra: fix line 307 of this script to avoid having a "release job" message for nomock stage commands being run (could cause random panic) and generally error messages could use some more polish