[Mimedefang] MD bottleneck problem

[Stephen L Johnson]

On Mon, 2003-09-29 at 11:22, David F. Skoll wrote:
> On Mon, 29 Sep 2003, Stephen L Johnson wrote:
> 
> > Over the week-end something happened on my test MD server. The MD slaves
> > seem to be taking a long time to process incoming e-mails.  And this is
> > causing MAXSLAVES to be running. There doesn't seems to be any sort of
> > excess incoming e-mail. And I can't seem to find any problems in the
> > normal Milter and MD syslog entries. I've also logging MX_STATS and I
> > still can't see an problems.
> 
> Could it be a network-related issue?  Are you using any DNS RBL's,
> Razor, Pyzor, DCC, etc in your filter?
> 
> If you trace a busy slave (truss/ptrace/ktrace, depending on OS), what is
> it stuck doing?

It was RBL related. I saw the Start and End times for all of the slaves
were 30 seconds. And when I rememberee some of the RBL sites shutting
down. it finally clicked. 

I enabled 'SALocalTests = 1;' until I can reconfigure my SA RBL tests.


-- 
Stephen L Johnson                       stephen.johnson at mail.state.ar.us
Unix Systems Administrator              sjohnson at monsters.org
Department of Information Systems
State of Arkansas
501-682-4339