on 6/18/00 3:44 PM, Topher Sandalwood at sideband@phoenixrising.cac.net
wrote something like:

> Got a really strange one:
> 
> DECstation 5000/200, 64M RAM, 3 SCSI disks, 1 SCSI tape
> NetBSD 1.4.2
> 
> from dmesg:
> asc_timeout: cmd 0xB0198468 drive 3
> rebooting...
> 
> from /var/log/messages:
> Jun 18 14:52:49 maximillian /netbsd: asc_timeoute: cmd 0xB0198468 drive 3
> Jun 18 14:52:49 maximillian /netbsd: rebooting...
> 
> Here's the deal.. drive 3 is the boot drive, and where /var is stored...
> How in the heck did it write the timeout to /var/log/messages if the drive
> had timed out?
> 
> Now how did the system write to /var if the drive had timed out?
> 
> Told ya it was a good one...
> 
> Any ideas??

    Um, well as you might already know, the 5000/200 doesn't have an asc! So
I don't understand how it could time out! I would try removing the Tape
drive and see if you still run into the same problem. You're using the
4.4BSD driver right (rzX disks), and not the 53C9x driver (sdX)? Something
very fishy is going on. It's possible that the tape drive is leaving the
SCSI bus in a strange state, or your chain is just unhappy. Either way it's
pretty weird. You might want to send-pr it so Jason Thorpe or someone who
knows more about the driver could help you.


    chris

    Chris