[Mimedefang] Process limits and loads

[listuser at numbnuts.net]

I noticed a lot of these types errors in our logs recently.

Apr 29 07:40:48 oak sm-msp-queue[25879]: h3S4eYxR001501: to=postmaster, 
delay=1+07:59:43, xdelay=00:00:00, mailer=relay, p
ri=3032036, relay=localhost.ourdomain.net. [127.0.0.1], dsn=4.0.0, 
stat=Deferred: 451 4.7.1 Please try again later

Now I know these are temp errors and that delivery will be attempted many
more times.  It's about a 50/50 split between errors for Sendmail and
errors for sm-msp-queue.  I believe the problem is that I have more
Sendmail processes than MIMEDefang processes.  I'm currently limiting
Sendmail with

define(`confMAX_DAEMON_CHILDREN',`60')
define(`confCONNECTION_RATE_THROTTLE',`3')dnl
define(`confBAD_RCPT_THROTTLE',`3')
define(`confMAX_RCPTS_PER_MESSAGE',`200')dnl
 (these are the only pertinent limits I can think of)

I am running the multiplexor.  I'm running with

MX_LOG=yes
MX_STATS=yes
MX_MINIMUM=5
MX_MAXIMUM=15
MX_IDLE=300
MX_BUSY=300

I just increased MX_MAXIMUM to 15 from 10 this morning.  Checking the logs 
I see that it has actually helped.  So it be set to the maximum possible 
number of Sendmail processes, 60?  MaxDaemonChildren isn't defined for my 
sm-msp-queue runner.  I supposed it will spawn however many it needs.  MD 
is not doing any AV checks.  Only SpamAssassin checks are being done 
currently.

The box is a dual PIII 1.26 with 1GB RAM running RH 7.3.  I'm planning on
maxing out its RAM one of these days and using the excess for MD's temp
files as well as the temp pop spools from qpopper.  The load on the
machine is usually quite low.  0-2.5 is common (usually under one).  It's
so low that I'm running Stanford's Folding @ Home client on it with ease
so I guess CPU load isn't a problem right now. :)

So is what I suspect is my problem really the problem?  Should I increase 
MX_MAXIMUM to 60?  Thanks

Justin