[linux-lvm] EXT2-fs panic (device lvm(58,0)):

[Andreas Dilger]

Bill Clark writes:
> Andreas Dilger wrote:
> > > cp: win2000.dsk: Input/output error
> > 
> > Any error messages in /var/log/messages?
> > 
> 
> The one thing I do see in the messages file is that there are some SCSI
> errors, but they only start after the cp bombs, so the question is which
> one causes the other?

This would _probably_ be the cause of the error you got above (EIO).
Sounds like your disk is dying.

> Here is the output from the debugfs:
> 
> [root at zale log]# debugfs
> debugfs 1.18, 11-Nov-1999 for EXT2 FS 0.5b, 95/08/09
> debugfs:  open /dev/vg00/vmware01
> debugfs:  stat win2000.dsk
> Inode: 12   Type: regular    Mode:  0664   Flags: 0x0  
> Version/Generation: 592822
> User:     0   Group:     0   Size: 1108485120
> File ACL: 0    Directory ACL: 0
> Links: 1   Blockcount: 2167144
> Fragment:  Address: 0    Number: 0    Size: 0
> ctime: 0x3aa54634 -- Tue Mar  6 15:19:00 2001
> atime: 0x3aa79587 -- Thu Mar  8 09:21:59 2001
> mtime: 0x3aa54634 -- Tue Mar  6 15:19:00 2001
> BLOCKS:
> 521 531 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549
> 550 551 552 553 554 555 556 557 558 559 560 
> 
>  ...Plus a LOT more blocks...it is a 1.1G file

> > > Filesystem state:         not clean with errors
> > 
> > This means that there is something wrong with the filesystem (could be a
> 
> Ran e2fsck and it rans for 5 minutes spitting out errors.  I beginning
> to think the whole filesystem is hosed...glad it is only test data.

Sorry, I think with dumpe2fs 1.19 and later it has a much more compact
output format.  In any case, I think it is either a bad disk, or the
filesystem is corrupt, or both.  I think the patch I sent fixes the bug
that was causing the first problem, and the rest of it is the disk.  If
you don't need the data on that disk, I would try running "badblocks -w"
on the device (please see man page first!) to see if it is really dying.

Cheers, Andreas
-- 
Andreas Dilger  \ "If a man ate a pound of pasta and a pound of antipasto,
                 \  would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/               -- Dogbert