Max Children Issue and SA timeouts

[Jason Burzenski]

Sorry about the split subject but I think (somehow) these problems might be
related.

I am running 2 identical MS boxes that receive identical load and have
identical configuration.  Running MS 4.29.7, SA 2.55.

I have recently had a problem with a very sporadic number of child processes
running on box1 but since it doesn't' appear to be affecting message flow, I
haven't been too quick to fix it.  I suspect I has something to do with
bayes locking but I haven't been able to test it yet.

Recently, I set my Max Children = 14 from 10 in MailScanner.conf and
reloaded my services almost a full week ago and I am still running only 10
processes on box2 (which has a process count that stays very consistent).
On box1, I am always running somewhere between 12 and 19 processes, but it
jumps every few seconds.  I don't think either box is accepting the Max
Children = 14 parameter.  Am I missing something?

Here is where things get even weirder.  On a quarter million message per
week per box, I was averaging ~4K SA timeouts on both boxes (this was as of
last week).  After making the change to max children 14 over the weekend, I
see ZERO SA timeouts on box1 (sporadic process count) and ~1400 on box2.

What is going on here?

Thanks for any insight you can provide.

Jason


-------------------------- MailScanner list ----------------------
To leave, send    leave mailscanner    to jiscmail at jiscmail.ac.uk
For further info about MailScanner, please see the Most Asked
Questions at    http://www.mailscanner.biz/maq/     and the archives
at    http://www.jiscmail.ac.uk/lists/mailscanner.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.mailscanner.info/pipermail/mailscanner/attachments/20040423/8b3879f2/attachment.html