Project

General

Profile

Actions

Bug #9308

closed

EBUSY on umount of osd volume after ceph-osd shutdown

Added by Sage Weil over 9 years ago. Updated about 7 years ago.

Status:
Can't reproduce
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Source:
Q/A
Tags:
Backport:
Regression:
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

2014-08-31T04:23:42.711 INFO:tasks.ceph:Unmounting /var/lib/ceph/osd/ceph-4 on ubuntu@plana21.front.sepia.ceph.com
2014-08-31T04:23:42.711 INFO:teuthology.orchestra.run.plana21:Running: 'sync && sudo umount -f /var/lib/ceph/osd/ceph-4'
2014-08-31T04:23:42.807 INFO:teuthology.orchestra.run.plana21.stderr:umount2: Device or resource busy
2014-08-31T04:23:42.808 INFO:teuthology.orchestra.run.plana21.stderr:umount: /var/lib/ceph/osd/ceph-4: device is busy.
2014-08-31T04:23:42.808 INFO:teuthology.orchestra.run.plana21.stderr:        (In some cases useful info about processes that use
2014-08-31T04:23:42.808 INFO:teuthology.orchestra.run.plana21.stderr:         the device is found by lsof(8) or fuser(1))
2014-08-31T04:23:42.808 INFO:teuthology.orchestra.run.plana21.stderr:umount2: Device or resource busy

ubuntu@teuthology:/a/sage-2014-08-30_20:39:25-rados-wip-sage-testing-testing-basic-multi/462763

have seen this a few times before. maybe we can make the ceph task do an lsof when this particular command fails?

Actions #1

Updated by Sage Weil over 9 years ago

  • Priority changed from Urgent to High
Actions #2

Updated by Loïc Dachary over 9 years ago

http://pulpito.ceph.com/loic-2015-01-10_11:54:25-rados-giant-backports-testing-basic-vps/694164/

2015-01-10T22:50:12.519 INFO:teuthology.orchestra.run.vpm196:Running: 'sync && sudo umount -f /var/lib/ceph/osd/ceph-2'
2015-01-10T22:50:12.559 INFO:teuthology.orchestra.run.vpm196.stderr:umount2: Device or resource busy
2015-01-10T22:50:12.559 INFO:teuthology.orchestra.run.vpm196.stderr:umount: /var/lib/ceph/osd/ceph-2: device is busy.
2015-01-10T22:50:12.559 INFO:teuthology.orchestra.run.vpm196.stderr:        (In some cases useful info about processes that use
2015-01-10T22:50:12.559 INFO:teuthology.orchestra.run.vpm196.stderr:         the device is found by lsof(8) or fuser(1))
2015-01-10T22:50:12.560 INFO:teuthology.orchestra.run.vpm196.stderr:umount2: Device or resource busy
2015-01-10T22:50:12.561 INFO:tasks.ceph:Checking for errors in any valgrind logs...

Actions #3

Updated by Loïc Dachary about 8 years ago

  • Status changed from New to Can't reproduce

I've not seen this in any infernalis / hammer suites in the past year.

Actions #4

Updated by David Zafman over 7 years ago

  • Status changed from Can't reproduce to 12
  • Release set to master
Actions #5

Updated by Samuel Just over 7 years ago

  • Priority changed from High to Normal
Actions #6

Updated by Sage Weil over 7 years ago

http://tracker.ceph.com/issues/17944 is probably the root cause

Actions #7

Updated by Sage Weil about 7 years ago

  • Status changed from 12 to Can't reproduce
Actions

Also available in: Atom PDF