This is a discussion on RE: DNS-BL for e-mail addresses within the mailing.postfix.users forums, part of the Mail Servers and Related category; Robert Felber writes... > On Sat, Jun 04, 2005 at 05:42:28PM +0200, Ralf Hildebrandt wrote: > > * Robert ...
|
|||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
|
|||
|
Robert Felber writes...
> On Sat, Jun 04, 2005 at 05:42:28PM +0200, Ralf Hildebrandt wrote: > > * Robert Felber <r.felber@ek-muc.de>: > > > Hi ho, > > >=20 > > > are there DNS based blacklists for e-mail addresses? > >=20 > > They're called RHSBLs. >=20 > Ah, I recall - I am sure someone asked me to implemet RHSBLs=20 > *looking at my todo list* :) >=20 > > > I would like to catch those abused yahoo and hotmail accounts. > > ? You could try sender address verification, since yahoo & hotmail > > quickly close abused accounts. >=20 > Thats expensive (or slow), isn't it? Yes, but if you cache the results, it's not bad. However, my current experience is that sender address verification = ("SAV") is not as effective now as it was even 6 months ago. Many spammers now = seem to be using real, valid sender addresses, or to have discovered sites = that will confirm any address. Of course, the putative senders have nothing = to do with the spam or the spammer, they are just there to deflect = complaints. That they defeat SAV is probably just a side-benefit for the spammers. Currently, I'm running a few Postfix-internal checks, a whitelist, a = couple of local blacklists, then a bunch of RBLs and RHSBLs, then a greylister, then SAV. Putting SAV very late in the process means that only the "senders" who pass all the other checks inititate an SAV probe, making = the speed and expense of the probes manageable. I also auto-whitelist anyone to whom my users send mail, which further = cuts down on the expense of accepting mail, since that whitelist comes before = any off-premises checks. -Shel DISCLAIMER: etc. |