[BlueOnyx:05394] EXT3-fs error (device dm-3): ext3_get_inode_loc: unable to read inode block - inode=21454983, block=85819397

[Reginald]

Seems I'm having a problem with the first HD, where the system is. The
box has 4 disks: 3 x 1 TB + 1 x 320 GB disks. It went offline at 4:28
AM. BlueOnyx Automated backups are taken every day at 4 AM and they
take an hour or two. This is what i saw when i came in the datacenter
and attached a screen to the box:

EXT3-fs error (device dm-3): ext3_get_inode_loc: unable to read inode
block - inode=21454983, block=85819397
EXT3-fs error (device dm-3): ext3_get_inode_loc: unable to read inode
block - inode=21454983, block=85819397
EXT3-fs error (device dm-3): ext3_get_inode_loc: unable to read inode
block - inode=21454983, block=85819397
EXT3-fs error (device dm-3): ext3_get_inode_loc: unable to read inode
block - inode=21454983, block=85819397
EXT3-fs error (device dm-3): ext3_get_inode_loc: unable to read inode
block - inode=21454983, block=85819397
etc.

It just hanged so I rebooted it. It took a while but it came back online.

In the Raqbackup email report of the previous day there was an error:

********************************* ERROR ********************************
Failed to transfer test-file to localhost!
Check username, password and hostname of the target machine!
You might find more information in your logs, see
/home/solarspeed/solbackup/server_backup/logs/output-ftptest.log and/or
/home/solarspeed/solbackup/server_backup/logs/errors-ftptest.log
You can find the complete backup on this RaQ under
/home/solarspeed/solbackup/server_backup/data


Once again, this is from the day before (12 Sep) but it tells me the
backup takes about 1h47m so if the server went offline at 4:28 it was
during the BO automated backup:
INFO: /usr/sbin/cmuExport started on Sun Sep 12 04:01:51 2010
INFO: /usr/sbin/cmuExport completed at Sun Sep 12 05:48:03 2010

I Googled "EXT3-fs error (device dm-3): ext3_get_inode_loc: unable to
read inode block", and it gave me a couple of results about faulty
harddisks. I disabled BlueOnyx backup and for now the server is up and
running. I guess I'll have to replace that disk ASAP, right? Very
strange because it is brandnew and less than a month old.