Is /dev/hda hosed?

ewtrowbr · Posted: Tue Dec 14, 2004 5:57 pm Post subject: Is /dev/hda hosed?

A couple of weeks ago, I started getting errors like this in /var/log/messages.

Dec 2 07:42:37 littleboy hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
Dec 2 07:42:37 littleboy hda: dma_intr: error=0x40 { UncorrectableError }, LBAsect=3707147, sector=3707146
Dec 2 07:42:37 littleboy end_request: I/O error, dev hda, sector 3707146

As time went on, I started having trouble shutting down. I would init 0 and the com would hang when tryng to remount hda ro. I would inevitably crash the machine, and then fsck.jfs -f with a knoppix disk. This usually produced a BUNCH of UncorrectableErrors as above. My laptop seems to still boot, but shutdown is always an adventure. My questions...

1) am I screwed? Is there some magic I can do to fix my hd?

2) What is te FS type that survives crashes the best? I am not impressed with JFS. I have a laptop with a battery with about 10 minutes of life. I inevitably drop the system occasionally, and I require a fstype that can take it.

3) If I have to get a new HD (which is not a problem - I am under warantee), what is the best way to backup and restore my gentoo system as is? I have put many hours into customizing my desktop environment, and don't want to lose any of my work.

As Always, thanks for the help. This is the best linux forum in the universe.

erich

neilhwatson · l33t Joined: 06 Feb 2003 Posts: 719 Location: Canada

stettler · Posted: Tue Dec 14, 2004 6:17 pm Post subject:

smartctl probably tells you more about the HD problem. BTW, I would immediately backup my data.

Best regards,
stettler
_________________
Claiming that your operating system is the best in the world because more people use it, is like saying McDonalds makes the best food in the world.

jschellhaass · Guru Joined: 20 Jan 2004 Posts: 341

emerge smartmontools and configure.

At worst your hd is failing. It's also possible there are some bad sectors now that were originally marked as good. The SMART system of modern drives will try and correct for bad sectors.

If the data on the sector is readable (recoverable) the sector will be remapped automatically. If the data is not readable the sector will NOT be remapped by the SMART system. This avoids arbitrary data loss. If the system tries to write to a sector that was originally marked as good and has gone bad since, the sector will be remapped. i.e. if you were to try and rewrite the same data to the bad block it would be remapped, if the spares area is not already full.

jeff

ewtrowbr · Posted: Tue Dec 14, 2004 7:36 pm Post subject:

when I said "drop the system" I meant, unplanned shutdown. Not "spike it and do a dance in the endzone". Sorry about the confustion.

Thanks for smartmontools. I'm emerging now. I think it's gonna fix me.

erich

freak4u · n00b Joined: 04 Jul 2003 Posts: 69 Location: MA, USA

If you have another system, dd worked great for me.