Bug 6844 - Device or resource busy after unmounting filesystems
Summary: Device or resource busy after unmounting filesystems
Status: RESOLVED OLD
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: 2
Hardware: x86_64 Linux
Priority: Normal major
Target Milestone: ---
Assignee: Thomas Backlund
QA Contact:
URL:
Whiteboard:
Keywords:
: 6964 (view as bug list)
Depends on:
Blocks:
 
Reported: 2012-07-21 21:43 CEST by François Andriot
Modified: 2013-11-23 16:14 CET (History)
3 users (show)

See Also:
Source RPM: util-linux
CVE:
Status comment:


Attachments

Description François Andriot 2012-07-21 21:43:03 CEST
Description of problem:
Stale directory after umounting filesystem.

During Mageia boot, all filesystems declared in fstab are automatically mounted.
E.g I have this line in fstab:
LABEL=MomentusXT /disk/MomentusXT ext4 noatime,commit=60,barrier=0,data=writeback 1 2

I'm at runlevel 5, logged as normal user in a KDE4 session.

I can obviously see the mount point inside a konsole :
# grep MomentusXT /proc/mounts 
/dev/sdb1 /disk/MomentusXT ext4 rw,noatime,nobarrier,commit=600,stripe=128,data=writeback 0 0

# umount /dev/sdb1
(no output)

# grep MomentusXT /proc/mounts 
(no output)

# rmdir /disk/MomentusXT/
rmdir: échec de suppression de « /disk/MomentusXT/ »: Périphérique ou ressource occupé

# fuser /disk/MomentusXT
(no output)

# lsof /disk/MomentusXT
(no output)

# lsof /dev/sdb
(no output)

# lsof /dev/sdb1
(no output)



This is getting a major problem when using a raid array ... I need to reconfigure some MD devices, but I can't !

# grep md0 /proc/mounts 
/dev/md0 /mnt/storage/raid ext4 rw,noatime,user_xattr,acl,barrier=1,stripe=1792,data=ordered 0 0

# umount /dev/md0
(no output)

# grep md0 /proc/mounts 
(no output)

# rmdir /mnt/storage/raid
rmdir: échec de suppression de « /disk/MomentusXT/ »: Périphérique ou ressource occupé
(In english: Device or resource busy)


# mdadm -S /dev/md0
mdadm: Cannot get exclusive access to /dev/md0:Perhaps a running process, mounted filesystem or active volume group?

[...] again, nothing found with either 'lsof' or 'fuser' ...


In order to correctly free my block devices, I have either to switch to single user mode, or kill running processes randomly (!) until I can remove the mount directory or stop the raid array.

Problem happens in runlevel 3 too, so I'm sure this is not caused by some graphical utility running in the user session.

I found that the "colord" process was one of the culprit, but it's not the only one.


Version-Release number of selected component (if applicable):
Mageia 2, everything up to date.

How reproducible:
Happens constantly on my 2 computers.
Just umount an EXT4 filesystem that was mounted during boot, then try to delete the directory it was mounted on. It always result in "device or resource busy".
Comment 1 Franz Holzinger 2012-08-08 08:30:23 CEST
I can confirm this bug with my USB stick. Since the upgrade to Mageia 2 it can never be unmounted again. I have to go into the hibernate mode to plug it off.

/dev/sdb1 on /run/media/franz/1D39-7202 type vfat (rw,nosuid,nodev,relatime,uid=500,gid=500,fmask=0022,dmask=0077,codepage=cp437,iocharset=iso8859-1,shortname=mixed,showexec,utf8,errors=remount-ro)



[root@localhost html]# umount  /dev/sdb1
démontage : /run/media/franz/1D39-7202 : périphérique occupé.
       (Dans certains cas, des infos sur les processus l'utilisant
        sont récupérables par lsof(8) ou fuser(1))

CC: (none) => flink

Comment 2 Franz Holzinger 2012-08-08 08:42:35 CEST
[root@localhost html]# lsof /dev/sdb1 
lsof: WARNING: can't stat() fuse.gvfs-fuse-daemon file system /run/user/franz/gvfs
      Output information may be incomplete.
COMMAND     PID  USER   FD   TYPE DEVICE SIZE/OFF NODE NAME
evince-th 24508 franz    6u   REG   8,17   170504 1486 /run/media/franz/1D39-7202/Windows/USt-Voranmeldung-2009-Q2.xps


I have had opened the usb stick content in Nautilus. So the unmount did not close it how it should. The unmount should automatically close all processes having formerly used the folder and files.
Comment 3 Todd Pfaff 2012-10-25 15:28:02 CEST
I'm also experiencing this problem.  I am unable to stop an md device.  There are no volume groups involved, the filesystem on that device has been unmounted and the device is not associated with any userland processes.  However an lsof|grep shows several kernel processes associated with the device.

# mdadm --stop /dev/md99
mdadm: Cannot get exclusive access to /dev/md99:Perhaps a running process, mounted filesystem or active volume group?

# lsof /dev/md99

# lsof | grep md99
md99_raid  2561             root  cwd       DIR                8,1     4096          2 /
md99_raid  2561             root  rtd       DIR                8,1     4096          2 /
md99_raid  2561             root  txt   unknown                                        /proc/2561/exe
jbd2/md99 31622             root  cwd       DIR                8,1     4096          2 /
jbd2/md99 31622             root  rtd       DIR                8,1     4096          2 /
jbd2/md99 31622             root  txt   unknown                                        /proc/31622/exe

CC: (none) => pfaff

Comment 4 Todd Pfaff 2012-10-25 15:31:38 CEST
more information about my environment

# cat /etc/release 
Mageia release 2 (Official) for x86_64

# uname -a
Linux localhost 3.3.8-server-2.mga2 #1 SMP Mon Jul 30 22:06:50 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux

# lsb_release -a
LSB Version:    core-2.0-amd64:core-2.0-noarch:core-3.0-amd64:core-3.0-noarch:core-3.1-amd64:core-3.1-noarch:core-3.2-amd64:core-3.2-noarch:core-4.0-amd64:core-4.0-noarch:core-4.1-amd64:core-4.1-noarch:cxx-3.1-amd64:cxx-3.1-noarch:cxx-3.2-amd64:cxx-3.2-noarch:graphics-3.1-amd64:graphics-3.1-noarch:graphics-3.2-amd64:graphics-3.2-noarch:lsb-2.0-amd64:lsb-2.0-noarch:lsb-3.0-amd64:lsb-3.0-noarch:lsb-3.1-amd64:lsb-3.1-noarch:lsb-3.2-amd64:lsb-3.2-noarch:lsb-4.0-amd64:lsb-4.0-noarch:lsb-4.1-amd64:lsb-4.1-noarch
Distributor ID: Mageia
Description:    Mageia 2
Release:        2
Codename:       thornicroft

# mdadm --version
mdadm - v3.2.3 - 23rd December 2011
Comment 5 Manuel Hiebel 2012-10-26 22:58:34 CEST
thomas, something for you ?

See Also: (none) => https://bugs.mageia.org/show_bug.cgi?id=6964
Assignee: bugsquad => tmb
Source RPM: base system and utilities => util-linux

Comment 6 Kim Garren 2013-03-25 14:52:45 CET
I can confirm, although I don't have the code here to contribute. The other day I had this problem, nothing was going on with the server on either side, and I tried to umount it from console and mcc both, with "device is busy" error. I ended up having to delete it in mcc on the server side (then re-added it back later), then it finally umounted on the user side. At the time, it was an NFS share that was crashing Nautilus (a different bug here) that I was working on; had to umount it from my netbook, and couldn't.

Thanks.

CC: (none) => skeeter1029

Comment 7 Samuel Verschelde 2013-08-29 17:32:24 CEST
*** Bug 6964 has been marked as a duplicate of this bug. ***
Comment 8 Manuel Hiebel 2013-10-22 12:10:42 CEST
This message is a reminder that Mageia 2 is nearing its end of life.
Approximately one month from now Mageia will stop maintaining and issuing updates for Mageia 2. At that time this bug will be closed as WONTFIX (EOL) if it remains open with a Mageia 'version' of '2'.

Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Mageia version prior to Mageia 2's end of life.

Bug Reporter: Thank you for reporting this issue and we are sorry that we may not be able to fix it before Mageia 2 is end of life.  If you would still like to see this bug fixed and are able to reproduce it against a later version of Mageia, you are encouraged to click on "Version" and change it against that version of Mageia.

Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Mageia release includes newer upstream software that fixes bugs or makes them obsolete.

-- 
The Mageia Bugsquad
Comment 9 Manuel Hiebel 2013-11-23 16:14:04 CET
Mageia 2 changed to end-of-life (EOL) status on ''22 November''. Mageia 2 is no
longer maintained, which means that it will not receive any further security or
bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of Mageia
please feel free to click on "Version" change it against that version of Mageia
and reopen this bug.

Thank you for reporting this bug and we are sorry it could not be fixed.

--
The Mageia Bugsquad

Status: NEW => RESOLVED
Resolution: (none) => OLD


Note You need to log in before you can comment on or make changes to this bug.