[Users] How to fsck a container's ploop disk image?

Scott Dowdle dowdle at montanalinux.org
Fri Nov 7 08:15:03 PST 2014


Greetings,

Yesterday I had the need to remove the drives that make up my /vz partition and replace them.  I basically stopped all containers and then rsycn'ed the private dirs to a separate drive, got the new drives installed, and then copied the container's filesystems back over.

I do have one container I'm having a problem with.  It will run for a while and then give disk errors and go read-only.  How does one fsck the disk image of a container?  And yes, it is ploop-based.  Below are some sample error messages I see in /var/log/messages on the host:

Nov  7 08:40:34 ovz kernel: [53378.803703] CT: 202: started
Nov  7 08:40:36 ovz kernel: [53380.273006] ploop_set_error=-5 on ploop49691
Nov  7 08:40:36 ovz kernel: [53380.277608] EXT4-fs error (device ploop49691p1): __ext4_get_inode_loc: unable to read inode block - inode=6422529, block=25690144
Nov  7 08:40:36 ovz kernel: [53380.286925] Aborting journal on device ploop49691p1-8.
Nov  7 08:40:36 ovz kernel: [53380.291554] Buffer I/O error on device ploop49691p1, logical block 19431424
Nov  7 08:40:36 ovz kernel: [53380.296237] lost page write due to I/O error on ploop49691p1
Nov  7 08:40:36 ovz kernel: [53380.300886] JBD2: I/O error detected when updating journal superblock for ploop49691p1-8.

Yes, I do have fairly recent backups... and I've already tried copying the container's filesystem from the directory-based backup three times... but it eventually gives me errors after running a while.

Is this a sign of underlying disk issues... or is there a way I can fsck and get the disk happy again?

Help!

TYL,
-- 
Scott Dowdle
704 Church Street
Belgrade, MT 59714
(406)388-0827 [home]
(406)994-3931 [work]


More information about the Users mailing list