load average

[dave madden <dhm_(_at_)_mersenne_(_dot_)_com>]

Hi.  I have an OpenBSD-2.6 box that's acting as firewall/gateway/NAT
for my private network.  It's been up 13 days, and is usually almost
completely idle.  However, I just checked it, and the load average is
just over 1:

    root_(_at_)_detox(27)# w
     4:28PM  up 13 days,  2:01, 1 user, load averages: 1.07, 1.08, 1.08
    USER    TTY FROM              LOGIN@  IDLE WHAT
    root     p0 vheissu.mersenne Sun06PM     0 w 

"ps -agx" shows only the "ps" process in RUN state, and "top" doesn't
show anything running furiously:

	load averages:  1.06,  1.07,  1.08                              16:34:43
	28 processes:  1 running, 27 idle
	CPU states:  0.3% user,  0.0% nice,  0.0% system,  0.2% interrupt, 99.5% idle
	Memory: Real: 1568K/21M act/tot  Free: 216M  Swap: 4K/256M used/tot

	  PID USERNAME PRI NICE  SIZE   RES STATE WAIT     TIME    CPU COMMAND
	24067 root       2    0  644K 1984K sleep select   0:04  0.39% xterm
	 8042 root      18  -12  328K  604K sleep pause    0:34  0.00% xntpd
	13927 root      10    0  140K  300K sleep nanosl   0:31  0.00% ipmon
	28763 root       2    0 1040K 1144K sleep select   0:26  0.00% httpd
    [...]

So I don't see anything running, but the load average isn't going
down.  Is this some sort of statistics-collection error, or is there
a stealth process running (and doing all sorts of nefarious things)?

regards,
d.