I had my Zip drive mounted under /mnt/zip, forgot about it, and pulled the
plug, to plug in my HiFi :)
Later, I did a df. My zip drive is parallel SCSI proto, so initially the
command timed out. Fine I can cope with this.
A SCSI retry commenced. Again -- fine. When this failed, the failure
propagated to the FS layer, giving me a "breada" failed or some such (VFAT
file system). Very nifty.
HOWEVER, the "cool" retries/IO failures/etc. stopped there -- my df
process remain hung in the "D" state, and I had to reboot to get rid of
this. Even turning the Zip drive back on did not help.
I know the answer is "don't do this", but has 2.1.x been hacked a bit to
make systems more resilient to this form of thing? Would autofs have coped
more gracefully?
Cheers
Chri
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu