Project

General

Profile

Actions

Bug #67628

open

mds goes read-only during segment expiry

Added by Venky Shankar over 1 year ago. Updated 9 months ago.

Status:
New
Priority:
Normal
Assignee:
Category:
Correctness/Safety
Target version:
-
% Done:

0%

Source:
Q/A
Backport:
reef,squid
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
MDS
Labels (FS):
Pull request ID:
Tags (freeform):
Merge Commit:
Fixed In:
Released In:
Upkeep Timestamp:

Description

/a/vshankar-2024-07-08_07:21:13-fs-wip-vshankar-testing-20240705.150505-debug-testing-default-smithi/7791798

2024-07-09T19:51:18.700+0000 7fdb4d750640 10 MDSContext::complete: 21C_MaybeExpiredSegment
2024-07-09T19:51:18.700+0000 7fdb4d750640 -1 mds.0.14 unhandled write error (2) No such file or directory, force readonly...
2024-07-09T19:51:18.700+0000 7fdb4d750640  1 mds.0.cache force file system read-only
2024-07-09T19:51:18.700+0000 7fdb4d750640  0 log_channel(cluster) log [WRN] : force file system read-only
2024-07-09T19:51:18.700+0000 7fdb4d750640 10 mds.0.server force_clients_readonly

This was seen in a test branch which had https://github.com/ceph/ceph/pull/55421, but the mds going read only doesn't seem to be happening due to backtrace update. II'll put https://github.com/ceph/ceph/pull/55421 to another round of test (with some additional debugging) to be extra sure that it isn't caused by the change.


Related issues 1 (0 open1 closed)

Related to CephFS - Bug #63259: mds: failed to store backtrace and force file system read-onlyResolvedVenky Shankar

Actions
Actions #1

Updated by Venky Shankar over 1 year ago

Leaving this unassigned until test results for https://github.com/ceph/ceph/pull/55421 are analyzed.

Actions #2

Updated by Venky Shankar over 1 year ago

  • Category set to Correctness/Safety
  • Assignee set to Venky Shankar
  • Backport set to quincy,reef,squid

Venky Shankar wrote in #note-1:

Leaving this unassigned until test results for https://github.com/ceph/ceph/pull/55421 are analyzed.

Didn't see this issue again with https://github.com/ceph/ceph/pull/55421, so this issue is likely a new issue.

Actions #3

Updated by Venky Shankar over 1 year ago

  • Related to Bug #63259: mds: failed to store backtrace and force file system read-only added
Actions #4

Updated by Venky Shankar about 1 year ago

Hmmm another one: /a/vshankar-2024-12-19_10:00:45-fs-wip-vshankar-testing-20241219.063429-debug-testing-default-smithi/8044410

2024-12-19T12:38:29.533+0000 7fc3a2134640 10 MDSContext::complete: 21C_MaybeExpiredSegment
2024-12-19T12:38:29.533+0000 7fc3a2134640 10 mds.0.log finish: ls=LogSegment(8056/0x19994da~19d1de4 events=107), r=-2
2024-12-19T12:38:29.533+0000 7fc3a2134640 -1 mds.0.14 unhandled write error (2) No such file or directory, force readonly...
2024-12-19T12:38:29.533+0000 7fc3a2134640  1 mds.0.cache force file system read-only
2024-12-19T12:38:29.533+0000 7fc3a2134640  0 log_channel(cluster) log [WRN] : force file system read-only
2024-12-19T12:38:29.533+0000 7fc3a2134640 10 mds.0.server force_clients_readonly
Actions #5

Updated by Konstantin Shalygin about 1 year ago

  • Backport changed from quincy,reef,squid to reef,squid
Actions #6

Updated by Patrick Donnelly 9 months ago

  • Target version deleted (v20.0.0)
Actions

Also available in: Atom PDF