Skip to content

Migrate all but one of the other jobs to use cncf-hosted gha runners#17943

Merged
rohit-nayak-ps merged 1 commit intovitessio:mainfrom
jeefy:cncf-gha-runners
Mar 19, 2025
Merged

Migrate all but one of the other jobs to use cncf-hosted gha runners#17943
rohit-nayak-ps merged 1 commit intovitessio:mainfrom
jeefy:cncf-gha-runners

Conversation

@jeefy
Copy link
Copy Markdown
Contributor

@jeefy jeefy commented Mar 10, 2025

Description

Finishing the work from #17879

CNCF has hosted ephemeral GitHub runners in Oracle that we're wanting projects to use rather than the GitHub hosted ones, which are now incur a cost to use. This PR is currently a WIP to work through any tests that break or dependencies that may be missing. <3

There is one final test that was not migrated, which was the Unit Test Race. That will be migrated at a later date to unblock the rest of these.

Please direct any questions to myself, @krook and @RobertKielty

Related Issue(s)

Checklist

  • "Backport to:" labels have been added if this change should be back-ported to release branches
  • If this change is to be back-ported to previous releases, a justification is included in the PR description
  • Tests were added or are not required
  • Did the new or modified tests pass consistently locally and on CI?
  • Documentation was added or is not required

Deployment Notes

@vitess-bot
Copy link
Copy Markdown
Contributor

vitess-bot bot commented Mar 10, 2025

Review Checklist

Hello reviewers! 👋 Please follow this checklist when reviewing this Pull Request.

General

  • Ensure that the Pull Request has a descriptive title.
  • Ensure there is a link to an issue (except for internal cleanup and flaky test fixes), new features should have an RFC that documents use cases and test cases.

Tests

  • Bug fixes should have at least one unit or end-to-end test, enhancement and new features should have a sufficient number of tests.

Documentation

  • Apply the release notes (needs details) label if users need to know about this change.
  • New features should be documented.
  • There should be some code comments as to why things are implemented the way they are.
  • There should be a comment at the top of each new or modified test to explain what the test does.

New flags

  • Is this flag really necessary?
  • Flag names must be clear and intuitive, use dashes (-), and have a clear help text.

If a workflow is added or modified:

  • Each item in Jobs should be named in order to mark it as required.
  • If the workflow needs to be marked as required, the maintainer team must be notified.

Backward compatibility

  • Protobuf changes should be wire-compatible.
  • Changes to _vt tables and RPCs need to be backward compatible.
  • RPC changes should be compatible with vitess-operator
  • If a flag is removed, then it should also be removed from vitess-operator and arewefastyet, if used there.
  • vtctl command output order should be stable and awk-able.

@vitess-bot vitess-bot bot added NeedsBackportReason If backport labels have been applied to a PR, a justification is required NeedsDescriptionUpdate The description is not clear or comprehensive enough, and needs work NeedsIssue A linked issue is missing for this Pull Request NeedsWebsiteDocsUpdate What it says labels Mar 10, 2025
@jeefy jeefy force-pushed the cncf-gha-runners branch from 3a5ec4e to 389ce93 Compare March 10, 2025 14:31
@frouioui frouioui added Type: Internal Cleanup Component: Build/CI and removed NeedsDescriptionUpdate The description is not clear or comprehensive enough, and needs work NeedsWebsiteDocsUpdate What it says NeedsIssue A linked issue is missing for this Pull Request NeedsBackportReason If backport labels have been applied to a PR, a justification is required labels Mar 10, 2025
@github-actions github-actions bot added this to the v22.0.0 milestone Mar 10, 2025
@jeefy jeefy changed the title migrate all other jobs to use cncf-hosted gha runners WIP: migrate all other jobs to use cncf-hosted gha runners Mar 10, 2025
@jeefy jeefy force-pushed the cncf-gha-runners branch 2 times, most recently from d2a6b68 to ca26096 Compare March 10, 2025 17:11
@jeefy jeefy force-pushed the cncf-gha-runners branch from ca26096 to 989377b Compare March 10, 2025 19:14
@frouioui frouioui self-requested a review March 10, 2025 23:14
@jeefy jeefy force-pushed the cncf-gha-runners branch 4 times, most recently from 9f48de2 to c8be9f3 Compare March 14, 2025 04:12
@jeefy jeefy requested a review from derekperkins as a code owner March 14, 2025 04:12
@jeefy jeefy force-pushed the cncf-gha-runners branch 2 times, most recently from 7e67ab9 to d3fba07 Compare March 14, 2025 18:05
@rohit-nayak-ps
Copy link
Copy Markdown
Member

@jeefy, the errors on the cluster workflows you see here are also happening on our other PRs, for example https://github.com/vitessio/vitess/actions/runs/13862724676/job/38795933540?pr=17967. I tried switching one to the gh runners and that worked: https://github.com/vitessio/vitess/actions/runs/13862724732/job/38794873563?pr=17967.

We are currently blocked on this one and upcoming urgent PRs due today. I was thinking of temporarily switching back to the gh runners for the failing tests while we figure out the issue with the oracle ones. wdyt?

@jeefy
Copy link
Copy Markdown
Contributor Author

jeefy commented Mar 14, 2025

If possible I'd like another 12 hours of debugging things this way. I'll submit a PR to revert back to the hosted GH runners if I haven't resolved the issue by tonight and resume work next week. Thanks for your patience during this.

@jeefy
Copy link
Copy Markdown
Contributor Author

jeefy commented Mar 14, 2025

We are currently blocked on this one and upcoming urgent PRs due today.

I missed this. Go ahead and revert, I should be able to test this with a different PR.

@jeefy jeefy force-pushed the cncf-gha-runners branch 4 times, most recently from 7faaac5 to 987057f Compare March 15, 2025 00:46
@codecov
Copy link
Copy Markdown

codecov bot commented Mar 15, 2025

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 67.55%. Comparing base (0bf4276) to head (e1fee1a).
Report is 21 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main   #17943      +/-   ##
==========================================
- Coverage   67.56%   67.55%   -0.02%     
==========================================
  Files        1597     1597              
  Lines      259763   259845      +82     
==========================================
+ Hits       175506   175526      +20     
- Misses      84257    84319      +62     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

@jeefy
Copy link
Copy Markdown
Contributor Author

jeefy commented Mar 15, 2025

So the jobs in the prior PR should be stable and not blocking anything anymore.

It seems like this PR is mostly there, although the race-centric tests seem to fail at very inconsistent points. If these tests have harder performance requirements I'll bump up the resources for them.

@jeefy jeefy force-pushed the cncf-gha-runners branch from 987057f to fa7af4d Compare March 15, 2025 15:24
@jeefy
Copy link
Copy Markdown
Contributor Author

jeefy commented Mar 15, 2025

@rohit-nayak-ps Is there any issue with these race-specific tests running in a container? I cannot fathom why they're failing and I'm trying not to dive too deep into this.

@jeefy jeefy force-pushed the cncf-gha-runners branch 2 times, most recently from 0b2f0a8 to 345ecc3 Compare March 16, 2025 06:25
Signed-off-by: Jeffrey Sica <me@jeefy.dev>
@jeefy jeefy force-pushed the cncf-gha-runners branch from 345ecc3 to e1fee1a Compare March 16, 2025 15:16
@jeefy jeefy changed the title WIP: migrate all other jobs to use cncf-hosted gha runners Migrate all but one of the other jobs to use cncf-hosted gha runners Mar 16, 2025
@jeefy
Copy link
Copy Markdown
Contributor Author

jeefy commented Mar 16, 2025

Removed the changes for unit test e2e race tests. Everything else looks solid, so this is g2g. I'll work on migrating that final test at a later date. Thanks!

@jeefy
Copy link
Copy Markdown
Contributor Author

jeefy commented Mar 18, 2025

Ping! :)

Copy link
Copy Markdown
Member

@frouioui frouioui left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The code looks good to me. I am restarting some workflows to ensure there is no flakiness, or not more than before: https://github.com/vitessio/vitess/actions/runs/13884728877?pr=17943.

timeout-minutes: 60
name: Run Semi Sync Upgrade Downgrade Test
runs-on: gh-hosted-runners-16cores-1-24.04
runs-on: oracle-16cpu-64gb-x86-64
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there an alias that also guarantees a specific distro / version? We've been bitten many times before by floating names when they upgraded and we had to immediately rush and drop everything else to fix CI.

We had explicit distro versions before so we could upgrade at a specific moment of our choice and now when these runner pools are updated. It also allows for updating in a separate PR while not being blocked anywhere else.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We didn't set up a specific alias for distro, but our intent is for this to ONLY mirror ubuntu-latest unless/until we hit a specific instance of needing an older ubuntu distro.

The runner image used is defined here: https://github.com/cncf/automation/tree/main/ci/gha-runner-image

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We didn't set up a specific alias for distro, but our intent is for this to ONLY mirror ubuntu-latest unless/until we hit a specific instance of needing an older ubuntu distro.

Right, but that's what explicitly caused us pain every time it was updated in the past. The biggest problem is not even that we don't want to upgrade, it's that it blocks us from even pro-actively upgrading to avoid the problems in the first place. So we have no way to deal with it and are forced to let things break and then have to rush to fix it 😢.

So it's not that we need an older distro, it's that we need the explicit version so that we can control when to upgrade (which is often before -latest would change).

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@jeefy I don't want to hard block on this btw, but it's something that ideally we do have some solution for at some point. If that's more when Ubuntu 26.04 would be out next year that's also good but ideally there's something before then 😄.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'll update our container builds/tags to ensure the runners can be pinned to specific os/versions. That'll come with a follow-up PR, probably post-KubeCon.

Spit-balling, it'll probably be oracle-16cpu-64gb-x86-64-24.04 vs oracle-16cpu-64gb-x86-64

Copy link
Copy Markdown
Member

@rohit-nayak-ps rohit-nayak-ps left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks Jeffrey

@rohit-nayak-ps
Copy link
Copy Markdown
Member

@rohit-nayak-ps Is there any issue with these race-specific tests running in a container? I cannot fathom why they're failing and I'm trying not to dive too deep into this.

Sorry, haven't had the time to look at that. I don't think there should be an issue, but we have had tests being flaky due to performance of the underlying hardware for some our e2e and unit tests.

Once this PR is merged, maybe you can create a PR porting the remaining tests and we can take a look at identifying the root cause of the race-test failures.

@rohit-nayak-ps rohit-nayak-ps merged commit c6ddadb into vitessio:main Mar 19, 2025
103 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants