[sclug] Recovering gracefully from lost interrupt

[Will Dickson]

Tim Sutton wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi
> 
> I use a zip drive for backups on my debian server. Occassionally it loses 
> track of hdd and I get (lots of) messages like this printed to 
> /var/log/messages :
> 
> Aug  3 06:47:36 appc63 syslogd 1.4.1#11: restart.
> Aug  3 06:47:41 appc63 kernel: hdd: lost interrupt
> 
> Does anyone know of a graceful way of sorting this out short of rebooting?

Not an answer to your question, but be careful with this. My Debian 
installation corrupted large backups written to zip disk - they came out 
the correct length, but garbled content. I strongly suggest reading back 
your backups after writing them - unmount the disk and remount before 
checking otherwise it'll give you the cached copy and not what's on the 
media.

md5sum can be useful here.

Will.