Since I upgraded my sparc64 box from 1.6E to 1.6I, I've been finding it
at the 'ok' prompt at random time when I would expect it to be chugging
along doing something (or even nothing 8-).  It doesn't appear to be
a panic, or else I'd expect to be in ddb (which this kernel has compiled
in), and the logs haven't been forthcoming with any interesting events
around the time of the halt.

This box is a 360Mhz U5; before 1.6I it had no problems keeping several-
month uptimes (which were usually interrupted by my moving the machine 
or rebooting with new kernel).  Notable recent changes to the machine and
kernel config have been the addition of the sab driver and using sabtty as
console, but I'm not 100% sure if that predated the recent instability.


Anyone else seeing this?  Is there a way to try and track this down once
the system is at the PROM?  Is my hardware simply crapping out?  I just
had this happen while I was connected and was reading some mail;  I had
swapped to doing something on the laptop and when I pulled up the ssh
window to the U5 again, it was dead... telnet'ing to the term server got
me a 'ok' without any interesting info.

I'm curious if there's anything strange or informative that appears on
the console when this happens, so I may try and leave a logged telnet,
etc. session connected to the term server this box is one for a while...
I don't have a 1.6E kernel laying around anymore, but maybe I'll try
going back to the 1.6G /netbsd.old if I can't get any useful info from
the console before the halt or the PROM after.

Thanks,
--rafal

----
Rafal Boni                                                     rafal@attbi.com
  We are all worms.  But I do believe I am a glowworm.  -- Winston Churchill