On Fri, Mar 26, 2004 at 08:13:08AM +0800, Peter Santiago wrote:
> Subject: Does error msg indicate impeding disk crash?
>
> Mar 26 00:48:58 mail kernel: EXT3-fs error (device ide0(3,2)): ext3_readdir:
> directory #686444 contains a hole at offset 4096
> Mar 26 00:48:58 mail kernel: EXT3-fs error (device ide0(3,2)): ext3_readdir:
> directory #686444 contains a hole at offset 8192
....
It could be a bad disk ...but
As a minimum I would check backups FIRST.
Do not reboot, do not do anything until you backup as much as you can.
Do look for other follow ups to this ... if there are none.
Then next you might force a file system check thus:
# touch /forcefsck
# reboot
I say force an fsck because this error does not look totally hardware
to me. But could have its root cause in hardware. You may have a
useful hardware message in the logs well before these.
A Google search found some errors like these associated with broken
hardware but they clearly had corresponding hardware errors in the log
you do not have these hardware errors. The Google messages I found
were old so the error reporting may have changed. Some were "disk is
dead" and some "Your filesystem is wrecked. Did you get some I/O
errors? You need to unmount it, run `e2fsck -f' and fix it up."
i.e. It is possible that this was only a bad sector or track that the
disk has since recovered. So far it looks like only one dir is
broken... 686444... perhaps the OS still has the meta data cached and
you can do the backup thing.
After fsck is run look in lost+found for files and dirs that were
discovered.
I think that this find command will tell you what dir is ill.
# find /your-disk -inum 686444
Good luck,
tom
--
T o m M i t c h e l l
/dev/null the ultimate in secure storage.
