On Jan 16, Dave Burgess wrote
> The other day, I noticed something similar.
> 
> I was building a new machine and only had 8 Meg of memory available.
> I've built and used dozens of 8M NetBSD machines in the past, so I knew
> it was possible.
> 
> I was working away and the system froze (similar to the way it's
> described above).
> 
> Figuring it was a swap problem, I increated the amount of swap space
> available to 100M (I like big hard drives :-).  I put swapctl into a
> loop and let it run every second.  I started up all of the programs that
> consistently froze the machine up before.  I froze up again - with
> swapctl showing that I had used 'almost' 16M of swap before the freeze.

I have a 8 Mb machine here (mail/ftp/http server) running 1.3_BETA and
I didn't notice any problems. 

antioche:/home/bouyer>uptime
 9:34AM  up 39 days, 2 users, load averages: 1.67, 1.19, 0.99
antioche:/home/bouyer>swapctl -lk
Device      1K-blocks     Used    Avail Capacity  Priority
/dev/sd2b       32370     6228    26142    19%    0
/dev/sd0b       49612     6100    43512    12%    0
/dev/sd1b       32370     6196    26174    19%    0
Total          114352    18524    95828    16%

(I've seen it using more swap than that).

--
Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
--