[Mimedefang] Milter: connect: host= temp failing commands (busy servers tuning)

[Stephane Lentz]

Hi,

Lately on some Linux mailserver running Trend Interscan, SA 2.55 and 
MIMEDefang 2.28 (waiting for 2.35 to upgrade) I've noticed a bunch of
Milter or MIMEDefang temporary errors reported :

Jul  1 19:48:41 mx1 sendmail[3723]: h61HmfOu003723: Milter: data, reject=451 4.7.1 Please try again later
Jul  1 19:48:41 mx1 sendmail[3716]: h61HmfOu003716: Milter: data, reject=451 4.7.1 Please try again later
...

and most often also 
Jul  1 19:48:41 mx1 sendmail[3725]: h61HmfOu003725: Milter: connect: host=aps43.muc.domeus.com, addr=62.67.54.143, temp failing commands
Jul  1 19:48:41 mx1 sendmail[3726]: h61HmfOu003726: Milter: connect: host=aps43.muc.domeus.com, addr=62.67.54.143, temp failing commands
Jul  1 19:48:41 mx1 sendmail[3736]: h61HmfOu003736: Milter: connect: host=aps43.muc.domeus.com, addr=62.67.54.143, temp failing commands
Jul  1 19:48:41 mx1 sendmail[3727]: h61HmfOu003727: Milter: connect: host=aps43.muc.domeus.com, addr=62.67.54.143, temp failing commands
Jul  1 19:48:43 mx1 sendmail[3739]: h61HmhOu003739: Milter: connect: host=teng3.be3a.com, addr=217.27.90.134, temp failing commands

I had 932 messages 'mfconnect: No free slave' in a day.

These messages have pop-up when adding SA analysis (and addition of headers) for all messages passing through (below 40kbytes limit). SA looks really like the 
culprit for these errors.
It seems that I should increase MX_MINIMUM & MX_MAXIMUM (too low : default values). 
Any recommandations for busy servers about these parameters ?
I should consider a RAM-based spool directory too I guess.
Any advice welcome.

regards, 

SL/
---
Stephane Lentz 
Alcanet International, Internet Services