A while back I posted a question about some problems with the PINT scanner
drivers under NetBSD 1.1. Charles suggested that there were some fixes in 
-current that might make a difference to me. I'll quickly rehash the problem
here then describe the new one.

The machine in question crashes (ranging from once a week to four times a day)
occasionally I'll get an mb_map full error on the console (which I'm not sure
what to do about - I found some source references in the vm code, but 
couldn't see what parameter to tweak) but more oftenthe machine gets into a 
state where accessing the scanner will return errors like:
ss0: device already open
(That wording may not be exact)
Even though there are no processes running while have the device open - I
assume the driver's state is getting confused somehow. Any suggestions?

With the upgrade to 1.1B complete (-current from about 2 weeks ago) I built
a new kernel with 'options SCANJET' and 'ss*' lines, and I have two problems 
now. Firstly, when the new kernel is booting, the drive lights on the various
SCSI devices blink during the aha probe (1542 controller in the machine) and
the light on the Exabyte tape drive starts blinking and gets faster, then the
aha probe times out and says the controller is not taking commands. This
doesn't seem likely to be scanner related. Anyone with 1.1B - a 1542 and an
Exabyte out there? Seen something similar?

If I eject the tape and leave the Exabyte open the system boots okay, and
gets past the aha probe fine - but doesn't register an 'ss' device, so the
scanner doesn't get configured.

As usual thoughts on this matter are appreciated - thanks in advance.

							David Maxwell
							david@mobtech.com