mds: try to choose a new batch head in request_clientup()

[lxbsz]

This will happen only for the client requests, not peer requests.
The 'mdr->killed' and 'mdr->dead' will always be set at the same
time when killing the client requests.

Fixes: https://tracker.ceph.com/issues/66124

Checklist

• Tracker (select at least one)
  • References tracker ticket
• Component impact
  • No impact that needs to be tracked
• Documentation (select at least one)
  • No doc update is appropriate
• Tests (select at least one)
  • No tests

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

[rishabh-d-dave]

@lxbsz Signed-off-by line is missing for commit Revert "mds: find a new head for the batch ops when the head is dead"

[lxbsz]

@lxbsz Signed-off-by line is missing for commit Revert "mds: find a new head for the batch ops when the head is dead"

@rishabh-d-dave Done. Thanks!

[leonid-s-usov]

Unlike in the old places, in the request_cleanup we should move on to the cleanup after we choose a new batch head, shouldn't we?

[lxbsz]

Unlike in the old places, in the request_cleanup we should move on to the cleanup after we choose a new batch head, shouldn't we?

BTW, what cleanup ? Could you point it out ?

[leonid-s-usov]

Unlike in the old places, in the request_cleanup we should move on to the cleanup after we choose a new batch head, shouldn't we?

BTW, what cleanup ? Could you point it out ?

well, everything below your new block in the request_cleanup method, starting with the line https://github.com/ceph/ceph/pull/57553/files#diff-ea7cb1a6ba9fa08363b14dd00a86bc6b79e01673e93af84ffdfcdbd0d3f26b19R9956 and until the end

[lxbsz]

Unlike in the old places, in the request_cleanup we should move on to the cleanup after we choose a new batch head, shouldn't we?

BTW, what cleanup ? Could you point it out ?

well, everything below your new block in the request_cleanup method, starting with the line https://github.com/ceph/ceph/pull/57553/files#diff-ea7cb1a6ba9fa08363b14dd00a86bc6b79e01673e93af84ffdfcdbd0d3f26b19R9956 and until the end

Currently doesn't it what you mean ?

request_cleanup() {
   1, choose new batch head
   2, cleanup
}

Or do you mean we should :

request_cleanup() {
   1, cleanup
   2, choose new batch head
}

???

[leonid-s-usov]

No no, with your latest update, it's all good. Sorry for the confusion. I commented on the code lines but I also had to leave the general review comment because it wouldn't allow me to submit it otherwise - so I just rephrased the same thing.

[batrick]

This PR is under test in https://tracker.ceph.com/issues/66261.

[batrick]

jenkins test make check

[batrick]

jenkins test make check arm64

[batrick]

jenkins test make check