Feature/Proposal: Accept hook names in helm.sh/hook-delete-policy

[jgiles]

Currently, you can easily delete a hook resource immediately after its execution:
"helm.sh/hook-delete-policy": hook-succeeded
"helm.sh/hook-delete-policy": hook-failed
Or, though this is currently missing in the docs at docs/charts_hooks.md:
"helm.sh/hook-delete-policy": "hook-succeeded,hook-failed"
(according to comments in #2692)

However, it's a pretty common case to want to delete the resource later - most often, on running helm delete (which otherwise does not delete hook resources).

This is addressed in the documentation here, but the proposed workaround is painful: Users need to create some container image to do resource deletion, package it into a Job, and make that Job a hook. In practice, this leads to issues like #2777.

Proposal: support the specification of hook names (e.g. post-install,pre-delete,post-delete) which would do the deletion during the named hook.

The downside to specifying these in "helm.sh/hook-delete-policy" is that there would be no good way to specify a weight. It could always use 0, or some high weight to always go "last" (since the deletion is an async clean-up operation, it probably doesn't make sense to rely on it being scheduled "before" anything else). Or, we could add some new field(s) to more properly support hook specification.

I expect this could be tricky to implement and might be more properly addressed with Stages (#2243), but that looks like it might not happen/be a long ways off.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten
/remove-lifecycle stale

[jgiles]

@bacongobbler any chance of getting some eyes on this?

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close