Document deflaking process

[ivanvc]

What would you like to be added?

This is a follow-up to a comment from our latest triage meeting (September 26th, 2024).

We've been ensuring that tests are flakes by running stress against tests. However, no official documentation exists on how a contributor must run it. There are references in the community meeting document and traces in some pull request comments.

We should add a document in Documentation/contributor-guide so we can refer contributors who want to help identify flakes.

Why is this needed?

To improve the contributor experience and our documentation stance.

[lucasrod16]

Just responded to a comment on this topic: #18585 (comment)

@ghouscht expressed interest in adding this information to the contributing guide

[ivanvc]

Oh, I didn't realize we had a section in CONTRIBUTING.md. We just need to improve it. Thanks, @lucasrod16.

@ghouscht, can you comment on this issue so we can assign it to you? (You can also comment /assign, and the k8s bot will assign it to you).

Thanks, team.

[ivanvc]

@ghouscht, refer to:

• https://docs.google.com/document/d/16XEGyPBisZvmmoIHSZzv__LoyOeluC5a4x353CX0SIM/edit?pli=1#heading=h.e48w90nud901
• Flaky test TestAuthLeaseTimeToLive #18585 (comment)
• etcdserver: add server range duration metrics #17983 (comment)
• https://gist.github.com/liggitt/6a3a2217fa5f846b52519acfc0ffece0#running-unit-tests-to-reproduce-flakes

[ghouscht]

/assign

Thank you, will look into this today!