PipelineRun Timeouts

[jerop]

In TEP-0046: Finally Tasks Execution Post PipelineRun Timeout, we introduced new timeouts fields to replace the timeout field.

# Before
kind: PipelineRun
spec:
  timeout: "0h0m60s"

# After
kind: PipelineRun
spec:
  timeouts:
    pipeline: "0h4m0s"
    tasks: "0h1m0s"
    finally: "0h3m0s"

As a follow up to that work, we need to:

• Promote PipelineRun.Timeouts from Alpha to Beta
• Deprecate PipelineRun.Timeout

This work should be included in our V1 effort.

/kind feature
/cc @lbernick @souleb @pritidesai

[pritidesai]

thank you @jerop 👍

Yes, definitely, the pipelineRun.timeouts was introduced in 0.25 about 6 releases back - https://github.com/tektoncd/pipeline/blob/main/docs/install.md#alpha-features. We can work towards promoting it to beta and making it part of V1 efforts. We also updated our docs with a warning of deprecating pipelineRun.timeout in the same release which should buy us enough time for the deprecation.

[lbernick]

Before promoting pipelinerun.timeouts to beta, we should clearly document and validate what happens if a user wants no timeout for pipelinerun.timeouts.tasks and pipelinerun.timeouts.finally. Right now, it's possible to set those timeouts to 0 (no timeout) without setting pipelinerun.timeouts.pipeline to no timeout, which I would consider an invalid configuration.

[jerop]

/help wanted

[pritidesai]

/assign @vinamra28

Please feel free to unassign yourself if needed

[tekton-robot]

@pritidesai: GitHub didn't allow me to assign the following users: vinamra28.

Note that only tektoncd members, 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 @vinamra28

Please feel free to unassign yourself if needed

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.

[jerop]

@vinamra28 mentioned at Pipeline working group we discussed this and you're willing to take this on after #4611

[vinamra28]

/assign

[lbernick]

#4071 is relevant here-- not sure if it's a blocker for moving timeouts to beta but it's a pretty gnarly bug

[lbernick]

Just talked with Vinamra; I think it would be OK to promote timeouts to beta and deprecate Timeout despite this bug. Reasoning here is that this bug doesn't affect Timeouts.Pipeline (which replaces Timeout), and we are confident we want Timeouts to be part of the API rather than Timeout.