Project

General

Profile

Actions
Copy link

Bug #8118

closed

MDS crashes

Added by Kenneth Waegeman about 10 years ago. Updated almost 10 years ago.

Status:
Closed
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Source:
Community (user)
Tags:
Backport:
Regression:
Severity:
2 - major
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
Labels (FS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

Active MDS crashes (v0.79).
log file attached.
Host did not ran out of memory, Standby MDS took over successfully.

dmesg also has such entries:
INFO: task ceph-mds:56548 blocked for more than 120 seconds.
Not tainted 2.6.32-431.5.1.el6.ug.x86_64 #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
ceph-mds D 0000000000000004 0 56548 1 0x00000000
ffff880048fe7c98 0000000000000082 dead000000200200 ffff88006210e400
ffff880048fe7c68 ffffffff810af320 ffff880000000000 ffff880048fe7b88
ffff8803cdad45f8 ffff880048fe7fd8 000000000000fbc8 ffff8803cdad45f8
Call Trace:
[<ffffffff810af320>] ? exit_robust_list+0x90/0x160
[<ffffffff81076ad5>] exit_mm+0x95/0x180
[<ffffffff81076f1f>] do_exit+0x15f/0x870
[<ffffffff811942c7>] ? pipe_read+0x2a7/0x4e0
[<ffffffff81077688>] do_group_exit+0x58/0xd0
[<ffffffff8108cff6>] get_signal_to_deliver+0x1f6/0x460
[<ffffffff8100a265>] do_signal+0x75/0x800
[<ffffffff8112ffd9>] ? free_pages+0x49/0x50
[<ffffffff81226346>] ? security_file_permission+0x16/0x20
[<ffffffff81189635>] ? vfs_read+0xb5/0x1a0
[<ffffffff8100aa80>] do_notify_resume+0x90/0xc0
[<ffffffff8100b341>] int_signal+0x12/0x17

Files

mds-ceph002.log.gz (231 KB) mds-ceph002.log.gz mds log file Kenneth Waegeman, 04/16/2014 01:21 AM
Actions
Copy link
 #1

Updated by Ian Colle about 10 years ago

  • Project changed from Ceph to CephFS
  • Category deleted (1)
Actions
Copy link
 #2

Updated by Greg Farnum almost 10 years ago

  • Status changed from New to Closed

This got a non-zero response from the OSD while writing out a directory. That's generally not an MDS bug, and if it was transient there's not much we can do about it now.

Actions
Copy link

Also available in: Atom PDF