Massive queue buildup

Jay Chandler chandler at
Mon Nov 13 23:39:02 GMT 2006

On Nov 13, 2006, at 2:42 PM, Brent Addis wrote:

> Check your not running one of those massive blacklists from SARE.
> I was running one for a while while testing and a similar thing  
> happened. removing it dropped my average scan time from ~2 1/2  
> minutes to 11 seconds per message.
Where does one determine how long the average scan time is?

> Other ideas:
> - Check your dns servers are capable of standing up to the amount  
> of dns requests you are making. Running something like nscd locally  
> is a good idea.
I suspect they are, but I'll verify this.

> - Are you running very many RBL's within mailscanner? Try disabling  
> these and see if it helps.
Three or four-- nothing insane.
> - Are you running any type of recipient verification? (as in,  
> checking that the person being sent the mail actually exists). If  
> not, try turning it on. I am unsure what it is called within  
> postfix as I don't use it.
We are.  Messages to undefined users fault to a 5xx error.
> - Check out 
> id=maq:index#optimization_tips which has a few others as well.
> How much mail are you handling a day? I have a couple of single cpu  
> 3.0 ghz machines comfortably handling many thousands of messages a  
> day.

Right now, about 100K a day.

Thanks for the help!

Jay Chandler
Network Administrator, Chapman University
714-628-7249 / chandler at
"Bother," said Pooh as he struggled with /etc/, "it never  
does quite what I want.  I wish Christopher Robin was here." -- Peter  
Da Silva in a.s.r.

-------------- next part --------------
An HTML attachment was scrubbed...

More information about the MailScanner mailing list