Hubert.Feyrer@rz.uni-regensburg.de writes:
> Hi!
> 
> What does this mean, and do i have to worry about my harddisk:
> 
> Feb 28 07:21:21 jabberwock /netbsd: wdc0: lost interrupt: status
> 58<drdy,dsc,drq
> > error 0
> Feb 28 07:21:21 jabberwock /netbsd: wdc0: lost interrupt: writing 8@wd0:35104
> 
> This is on a IDE-equipped i386 V1.2 system.

We've got a Pentium box here (NetBSD 1.2Beta (released a few weeks before 1.2 
proper), w/ IDE drive) that just started giving similar messages.  We're
getting:

Feb 28 11:24:42 jasmine /netbsd: wdc0: lost interrupt: status 50<drdy,dsc> 
error
 0
Feb 28 11:24:42 jasmine /netbsd: wdc0: lost interrupt: reading 2@wd0:1175230
Feb 28 11:44:38 jasmine /netbsd: wdc0: lost interrupt: status 50<drdy,dsc> 
error
 0
Feb 28 11:44:39 jasmine /netbsd: wdc0: lost interrupt: reading 2@wd0:1175230
...

The block number (1175230) is the same for all the errors.  (We've had 8 of 
these errors so far)

This machine has been been down exactly once (power failure) in the last 140 
days or so, and has been quite rock solid.  I've been running a bunch of 
simultaneous "dd's" on the drive for the better part of the afternoon, and 
I've been unable to get it to fail in any way... 

The *only* thing that has changed on the machine recently is that we started 
NFS serving parts of the disk to another machine...  

Are you NFS serving anything from your machine?

Could there be some sort of weird interrupt lossage between the WD driver and 
the ethernet driver (3Com 3C509, in this case)?  The block that it's failing 
on 
lives in the partition that is NFS exported.  (I tried beating on the NFS 
portion of the machine thisafternoon as well, but again was unable to make it 
break :-/ )

Just speculation...  Am I completely out to lunch?

I'd appreciate finding out if our drive is failing too... :-/

Thanks.

Later...

Greg Oster

oster@cs.usask.ca
Department of Computer Science
University of Saskatchewan, Saskatoon, Saskatchewan, CANADA