Ext3 users - Dec 2005 - EXT3-fs error

Hello friends,

I am experiencing some weird beheaviour on one of my computers.
I appologize for the laconic subject, but I really don't know anything 
more than that. I'll tell the story in chronological order.

I have an HP server with five SCSI disks.
The disks are _not_ in RAID. I installed FC2 on one disk. I installed a 
2.6.11.7 and used it for boot. The server runs some software that I wrote, 
and it worked flawlessly for more than three months.

On Thursday I noticed that some of the files under / are not 
accessible, reporting Input/Output error.
The file-system itself was in read-only mode.

Soon no file under / was accessible.

fdisk /dev/cciss/c0d0 (the disk containing /) reported that the device is 
not valid.

The console was full of this message repeating:
"EXT3-fs error (device /dev/cciss/c0d0) in start_transaction: Journal has 
aborted"

I booted the machine.

fsck did not report anything, and the computer seemed to work fine.

After a day the problem repeated.

I booted again.

The BIOS reported an error: "POST Error: 1727-Drive Array - New Logical 
Drive(s) Attachment Detected"

The boot failed as GRUB complained not to find the disk containing the 
kernel file.

I pressed reset again.

This time the machine booted normaly, but somehow I get a feeling that 
something is wrong with the machine.

A technition from HP said that everything looks OK hardware-wise.


Can anyone help me understand what is going on here?


Thanks,
   Hayim.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160

Hayim Shaul schrieb:
> fdisk /dev/cciss/c0d0 (the disk containing /) reported that the device
> is not valid.
if the device is "not valid", this has to be fixed. concrete error
messages would help here though.
> The console was full of this message repeating:
> "EXT3-fs error (device /dev/cciss/c0d0) in start_transaction: Journal
> has aborted"
if it really is a hardware error (bad disk/controller/cable/driver), then
ext3 can't do anything about it (except perhaps mounting the device ro)
> The BIOS reported an error: "POST Error: 1727-Drive Array - New
Logical
> Drive(s) Attachment Detected"
[...]
> A technition from HP said that everything looks OK hardware-wise.
hm, this POST error looks "OK"?
> Can anyone help me understand what is going on here?
please send useful error messages.

Christian.
- --
BOFH excuse #335:

the AA battery in the wallclock sends magnetic interference

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFDqBSxC/PVm5+NVoYRA7CgAKCN3xiAqTeT+gRKzfGXccYwdW5RSACeKxhf
7Onirmpv7bMN/70t4mbFQ+A=TY8w
-----END PGP SIGNATURE-----