squid: mds/quiesce: db: quiesce-await should EPERM if a set is past QS_QUIESCED

[leonid-s-usov]

Backport

Fixes: https://tracker.ceph.com/issues/66034

Original-issue: https://tracker.ceph.com/issues/65669
Original-PR: #57099

Show available Jenkins commands

• jenkins retest this please
• jenkins test classic perf
• jenkins test crimson perf
• jenkins test signed
• jenkins test make check
• jenkins test make check arm64
• jenkins test submodules
• jenkins test dashboard
• jenkins test dashboard cephadm
• jenkins test api
• jenkins test docs
• jenkins render docs
• jenkins test ceph-volume all
• jenkins test ceph-volume tox
• jenkins test windows
• jenkins test rook e2e

[zdover23]

@anthonyeleven: When this is merged, we'll make another PR that disentangles the docs from the code changes, and we'll tidy up the English.

[leonid-s-usov]

@zdover23 please note that this is a backport of a commit that had been approved by the docs for merging to main.

As for the disentanglement, I'm not very happy breaking such PR into two - one for the code, and one for the docs. This one should be an atomic patch, keeping the docs and the functionality coherent. We could merge two PRs in short succession, but I find that suboptimal and error-prone

[zdover23]

@zdover23 please note that this is a backport of a commit that had been approved by the docs for merging to main.

As for the disentanglement, I'm not very happy breaking such PR into two - one for the code, and one for the docs. This one should be an atomic patch, keeping the docs and the functionality coherent. We could merge two PRs in short succession, but I find that suboptimal and error-prone

@leonid-s-usov, I know that this is a backport.

I can understand your reluctance to disentangle code from docs. I understand the benefits of keeping the two together, in lockstep, so that code changes are explained in the documentation that accompanies them in the same PR. However, I'm not going to approve anything that touches code--I'm a docs guy, and in this matter and all other matters like it, I want to stay in my (docs-only) lane.

This is what I was planning: merge this PR (because #57099, from which this backport was created, has been merged), and then create a new latest-targeted PR in which the English is cleaned up. After that PR is merged, it will be backported.

I understand the error-prone nature of this kind of approach. My entire professional life is directed at keeping a list so that the kind of errors that can crop up in this situation don't crop up.

If you're willing to give it a go, I'll make a run at it this evening. You'll be consulted and satisfied throughout the process.