[Mimedefang] recipient filter and rbl's

[Kimmo Jaskari]

On Mon, 2007-12-03 at 15:49 +0000, Paul Houselander wrote:

> Im happy with my recipient verification script, I cache results etc.. to cut
> down on how many times I need to query the backup system, it works really
> well once the cache is built up.
> 
> After a bit of digging around I think ive pretty much decided not to use the
> rbl feature in sendmail but to intergrate spamhaus checking into my
> mime-defang script.

I can't positively state I'm the biggest MD fan around, but I have been
running it for years and would probably threaten physical violence if my
boss wanted me to switch it out for anything else ;) - but it is still
PERL, and that means lots of load, comparatively speaking. Minimizing
the stuff that has to be pumped through MD seems sane to me.

Having Sendmail block stuff on its own is much more efficient and less
load producing when we're talking stuff that you unequivocally do not
want, such as spam mail to "misspelled.recipient at company.com" as well as
stuff that Spamhaus marks as probable spam (unless you want to permit
that to reach users, only marked as probable spam, which is different.)

If you fetch a list of valid mail addresses every hour, or daily, or
four times a day etc, then there is literally no queries further in to
the Exchange servers at other times, Sendmail just does its thing and
checks the access db when mail comes in. Good stuff. I'm really thrilled
with how well my gateways work now, and they have been almost completely
hands-off since I set things up this way.

I'm sure your approach will work just as well, but it will definitely
require your mail gateways to do more heavy lifting when it comes to
processing mail, I'd guess.

/Kimmo