[Sheflug] aargh, disk crash

[matt fairtlough <xxxxx [at] xxxxxxxxxxxxxx>]

I think I'm in some bother.  Mozilla locked up on my suddenly, and the
hdd started making
strange zzzt zzzt zzzt noises.  Looks like I've lost all my email :-()
because my .mozilla directory is on a damaged sector I suppose.

Now remarkably, linux kept on running, though at one point in switching
from a console to X-windows the screen turned red which made me think:
panic, virus and switched off.  So then I thought, well, where do I go
from here, things can't get much worse.  So booted up again and I'm able
to carry on as before but without my mozilla inbox and stuff.  Ain't
linux great?  I can't imagine windoze carrying on like this after a disk
crash. 

Now why am I telling you all this?  Guess I'm looking for advice. 
Firstly, do people reckon this is definitely a disk crash and not some
nasty virus posing as one?  Secondly, assuming it is a crash, my
priority should obviously be saving all the work I've done before any
more of the disk goes down. Of course, I should have already got a nice
automated backup system, but I haven't. I'd planned to burn stuff to CD
but never managed to get CD burning working (tried the simple stuff,
think I need to recompile the kernel as per the CD-HOWTO to get it to
work now).  Are there any tools than can automate a backup over an
internet connection (cable modem)?  I've decided to tar up and compress
/home (0.7G) (omitting the dodgy directory) and /usr (1.5G) and try to
ftp them to work but if that fails for any reason (namely the excessive
size) I may need more help.

Assuming I've managed to back everything I need up, how do I find out
just how much of the disk is damaged, and is there any prospect of
recovering the damaged data?

I'm running reiserfs on the affected disk, might that help?

Finally, any suggestions about repair/replacement?  It is a 45Gig IBM
Maxtor disk so I'm reluctant to scrap it, but once bitten etc.

>From /var/log/messages:

Oct  1 09:49:14 linux kernel: hde: lost interrupt
Oct  1 09:49:14 linux kernel: hde: dma_intr: status=0x51 { DriveReady
SeekComplete Error }
Oct  1 09:49:14 linux kernel: hde: dma_intr: error=0x40 {
UncorrectableError }, LBAsect=21625877, sector=82648
Oct  1 09:49:14 linux kernel: end_request: I/O error, dev 21:08 (hde),
sector 82648
Oct  1 09:49:14 linux kernel: vs-13070: reiserfs_read_inode2: i/o
failure occurred trying to find stat data of [4038 3771 0x0 SD]

my other problem is that, in rather desperately trying to save my data
(for some reason,
in spite of keeping backups of most things, I don't backup my email...)
I tried to copy
my mozilla Inbox to a zip file; now there appears to be nothing on the
disk.  I guess it may be time for a re-format then?

bash-2.05# fsck /media/zip
Parallelizing fsck version 1.19a (13-Jul-2000)
fsck: fsck.auto: not found
fsck: Error 2 while executing fsck.auto for /dev/hdd4
bash-2.05# 

btw, thanks to those of you who responded re UPS. Isolating the room my
machines are in won't help much because they do not cause the
circuit-breaker to trip, and when it does trip the main switch is closed
meaning that the whole house is powered down
___________________________________________________________________

Sheffield Linux User's Group - http://www.sheflug.co.uk . 
To unsubscribe from this list send mail to 
shef-lug-request@list.sheflug.org.uk with the word
"unsubscribe" in the body of the message. 

  GNU the choice of a complete generation.