Well, I updated netbsd-2-0 on my Ultra 1, and it seems to have either
hit the sleep sleeps forever bug and deadlocked sometime while I was
asleep, or just deadlocked for no known reason.

Unlike the previous time the SSF bug bit it, it wasn't even up for two
days before locking this time.

Does anybody know of a way to reproduce the bug so that it might be
backtraced from DDB? I can make serial console access to the system
available to a NetBSD developer if they would like to check it out...
I was hoping to use it for IPF/IPNat at some point, but that's
obviously out of the question as long as I can't trust it to stay up.

Oddly enough, the SSF bug has only hit my Ultra 5 once, a couple
months after I first got it, which was well over a year ago. The Ultra
5 is running current, built a few hours after the 2-0 build on the
Ultra 1.

I might have hit PR/26328, dunno. Serial console wasn't attached at
the time the machine stopped responding. Sending a serial BREAK didn't
do anything - I had to manually flip the power. I'm building a debug
kernel now, in the hopes that if it happens again, I'll get some
information out of it...

Any suggestions would be greatly appreciated. I really want to keep
NetBSD on this machine.

-- 
Sean