Skip to content

Kubernetes CI Policy: create and enforce policy of removing continuously unhealthy jobs #18601

@spiffxp

Description

@spiffxp

Part of #18551

Once we have:

It's time to craft a policy to maintain a reasonable state of job health going forward!

A rough suggestion is: "continuously failing for O(weeks) means removal"

Perhaps there is more room for nuance here:

  • maybe infrequently scheduled jobs are permissible for longer?
  • do we care about continuously failing jobs not run on prow.k8s.io?
  • what are the logistics of outreach, deadlines, "hang on we can't fix it now" etc.

Metadata

Metadata

Assignees

Labels

area/jobshelp wantedDenotes an issue that needs help from a contributor. Must meet "help wanted" guidelines.kind/documentationCategorizes issue or PR as related to documentation.lifecycle/frozenIndicates that an issue or PR should not be auto-closed due to staleness.sig/testingCategorizes an issue or PR as relevant to SIG Testing.

Type

No type

Projects

No projects

Milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions