Backport #53339: pacific: src/osd/scrub_machine.cc: FAILED ceph_assert(state_cast<const NotActive*>()) - RADOS - Ceph

Actions

Copy link

Backport #53339

closed

pacific: src/osd/scrub_machine.cc: FAILED ceph_assert(state_cast<const NotActive*>())

Added by Backport Bot over 2 years ago. Updated over 1 year ago.

Status:

Resolved

Priority:

Urgent

Assignee:

Ronen Friedman

Target version:

Release:

pacific

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description

https://github.com/ceph/ceph/pull/46767

Related issues 1 (0 open — 1 closed)

Actions

Copy link

Updated by Backport Bot over 2 years ago

Copied from Bug #51942: src/osd/scrub_machine.cc: FAILED ceph_assert(state_cast<const NotActive*>()) added

Actions

Copy link

Updated by Neha Ojha about 2 years ago

Assignee set to Ronen Friedman

Actions

Copy link

Updated by André Cruz about 2 years ago

Could this be the reason I'm seeing a spam of "handle_scrub_reserve_grant: received unsolicited reservation grant" messages on OSD logs and scrubs/deep-scrubs don't make progress?

I'm using 16.2.7.

Actions

Copy link

Updated by Maximilian Stinsky about 2 years ago

Hey I got here through the following mailing list post: https://lists.ceph.io/hyperkitty/list/ceph-users@ceph.io/thread/A45BWXLWC2PKLGA5G7GXKCZDNHEOL2LL/#RHHJXQCBV2GGTGD7GAX5PACEY6TFSWRQ

We upgraded one of our ceph clusters to pacific 16.2.7 and are getting a lot of "handle_scrub_reserve_grant: received unsolicited reservation grant" messages.
In our setup some osd's also crash after the upgrade with logs like "1 heartbeat_map is_healthy 'OSD::osd_op_tp thread 0x7f010c49b700' had timed out after 15.000000954s" right after the handle_scrub_reserve_grant message, we think that all this is related to each other.

Are there any known workarounds to fix this or is a backport expected in the near future which would fix the mentioned issues.

Actions

Copy link