client: always refresh mds feature bits on session open

[vshankar]

Fixes: http://tracker.ceph.com/issues/63188

Contribution Guidelines

• To sign and title your commits, please refer to Submitting Patches to Ceph.

• If you are submitting a fix for a stable branch (e.g. "pacific"), please refer to Submitting Patches to Ceph - Backports for the proper workflow.

Checklist

• Tracker (select at least one)
  • References tracker ticket
  • Very recent bug; references commit where it was introduced
  • New feature (ticket optional)
  • Doc update (no ticket needed)
  • Code cleanup (no ticket needed)
• Component impact
  • Affects Dashboard, opened tracker ticket
  • Affects Orchestrator, opened tracker ticket
  • No impact that needs to be tracked
• Documentation (select at least one)
  • Updates relevant documentation
  • No doc update is appropriate
• Tests (select at least one)
  • Includes unit test(s)
  • Includes integration test(s)
  • Includes bug reproducer
  • 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

[mchangir]

could you rephrase commit title to:
client: always refresh mds feature bits on session open

[vshankar]

could you rephrase commit title to: client: always refresh mds feature bits on session open

done.

[dparmar18]

in a cluster with multiple active MDSs, is it possible that different MDSs are on different versions?

[vshankar]

in a cluster with multiple active MDSs, is it possible that different MDSs are on different versions?

If you mean different ceph versions, then yes.

[dparmar18]

in a cluster with multiple active MDSs, is it possible that different MDSs are on different versions?

If you mean different ceph versions, then yes.

and the client must be maintaining some table (or some sort of data structure) to keep records of the MDSs and their feature bits right? So that client doesn't get baffled with what and how to communicate with mds.a, mds.b and mds.c if all three are on different ceph versions with possibly different set of feature bits to offer.

[vshankar]

in a cluster with multiple active MDSs, is it possible that different MDSs are on different versions?

If you mean different ceph versions, then yes.

and the client must be maintaining some table (or some sort of data structure) to keep records of the MDSs and their feature bits right? So that client doesn't get baffled with what and how to communicate with mds.a, mds.b and mds.c if all three are on different ceph versions with possibly different set of feature bits to offer.

yes - those are maintained per (mds) session.

[dparmar18]

Can we test this out apart from the upgrade tests? Maybe add a dummy feature bit to the MDS and expect client to not crash?

[vshankar]

Can we test this out apart from the upgrade tests? Maybe add a dummy feature bit to the MDS and expect client to not crash?

This yaml

fs/upgrade/featureful_client/old_client/{bluestore-bitmap centos_latest clusters/1-mds-2-client-micro conf/{client mds mon osd} overrides/{ignorelist_health ignorelist_wrongly_marked_down multimds/no pg-warn} tasks/{0-octopus 1-client 2-upgrade 3-compat_client/pacific}}

always reproduces this, so I think we can depend on that.

[vshankar]

@batrick @dparmar18 fixed and updated. please check.

[mchangir]

LGTM

[vshankar]

https://pulpito.ceph.com/vshankar-2023-10-18_10:34:52-fs-wip-vshankar-testing-20231018.065603-testing-default-smithi/

[vshankar]

jenkins test make check

[dparmar18]

https://pulpito.ceph.com/vshankar-2023-10-18_10:34:52-fs-wip-vshankar-testing-20231018.065603-testing-default-smithi/

i hope this includes #53734 which found out this issue

[vshankar]

https://pulpito.ceph.com/vshankar-2023-10-18_10:34:52-fs-wip-vshankar-testing-20231018.065603-testing-default-smithi/

i hope this includes #53734 which found out this issue

The issue was already seen in #50730 but I got to it rather late esp. when I looked at the core in your run, I figured that its the same issue.

[vshankar]

https://tracker.ceph.com/projects/cephfs/wiki/Main#18-Oct-2023

[vshankar]

rebase and updated for jenkins tests...

[vshankar]

jenkins test api