Hi,
I've got a ultra5 running a 3.99.15 32bit kernel in a strange state.
some processes where hung, sleeping on various channels. The problem seems
to be, again, the clock not going on.
Indeed after login on the console here is what I noticed:
# date
Wed Mar  1 16:54:46 MET 2006
# date
Wed Mar  1 16:54:48 MET 2006
# date
Wed Mar  1 16:54:41 MET 2006
# date
Wed Mar  1 16:54:47 MET 2006
# date
Wed Mar  1 16:54:48 MET 2006
# date
Wed Mar  1 16:54:49 MET 2006
# date
Wed Mar  1 16:54:40 MET 2006
# date
Wed Mar  1 16:54:42 MET 2006
# date
Wed Mar  1 16:54:44 MET 2006
# date
Wed Mar  1 16:54:45 MET 2006
# date
Wed Mar  1 16:54:46 MET 2006

waiting a bit and looking at the date again show the clock still going
between 16:54:40 and 16:54:50
Setting the date manually doesn't help, it still loops in the same 10 seconds:
# date 031220    
cc_microset[0]: delta 1700717344, resetting state
Fri Mar  3 12:20:00 MET 2006
Mar  3 12:20:00 samba date: date set by root
# date                 
Fri Mar  3 12:20:06 MET 2006
# date 
Fri Mar  3 12:20:07 MET 2006
# date 
Fri Mar  3 12:20:08 MET 2006
# date 
Fri Mar  3 12:20:09 MET 2006
# date 
Fri Mar  3 12:20:10 MET 2006
# date 
Fri Mar  3 12:20:00 MET 2006
# date 
Fri Mar  3 12:20:01 MET 2006
If I set the date to e.g. 12:21:33 it will loop between 12:21:33 and
12:21:43
This started after an unwanted break was sent to the concole and entering 'c'
in ddb. This system is not running ntpd.

Any idea ? I can keep the system in this state for some more time, if
someone has an idea at what to look next.

-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 26 ans d'experience feront toujours la difference
--