Hi all...

Apologies if this is a FAQ - I don't think I've seen it anywhere
though (and my man pages are out of service for reasons explained
below!)


My desktop NetBSD machine (1.4 was -current a couple of months
ago!) is completely knackered thanks to an IDE disk that has self
destructed taking the /usr partition with it.
I am currently trying to rescue as much of the 5GB of data as possible
using cpio to get it from the old knackered disk to a new replacement
disk.

My question is...  is there any way (without recompiling the kernel,
which I am in no position to do at the moment) of reducing the number
of retries the WD driver makes when it encounters a bad read?  It seems
to retry about 5 times at the moment, which is making the exercise
painfully slow...  so I thought there may be a sysctl I could use
or something!


Thanks in advance,
Tim.

-- 
Tim Walls