Okay, I solve this problem with couple of external scripts. Interesting ?
2 hours of work = 250 hosts in my RBL :))
------------- Original message -----------
> From: exim@???
> To: exim-users@???
> Subject: [Exim] Re: Request for anti-spam feature in exim4
> Date: Wednesday, December 05, 2001 at 14:39
MBM> On Wed, Dec 05, 2001 at 12:18:24PM +0000, Drav Sloan wrote:
>> Suresh Ramasubramanian wrote:
>> > There are plenty of third party blocklists available for this - which can be
>> > queried using exim.
>> Of which I'd argue the majority are mostly useless/out of date/have
>> very bad policies/will not listen to the public.
MBM> Yes. I agree with this.
>> I also would agree with Sharun; I'd like to see the ability for exim
>> to maintain it's own db of 'open relays' by verifying this at connection
>> time from other hosts - (very much like the verify_recipients option).
MBM> Erm? You mean doing a callback to test for open relay? Are you completely
MBM> insane, or completely clueless? There are several reasons not to do this.
MBM> 1) it is hostile to mailadmins who maintain clean machines.
MBM> 2) you're assuming that a relay can be delivered straight away, or that
MBM> acceptance of responsibility for the message is an indication of that
MBM> machine being about to relay it.
MBM> 3) you will probably get the tests wrong, think of the time when 127.0.0.1
MBM> hit the MAPS RSS, and loads of people started generating X-RBL-Warning
MBM> headers, which were (incorrectly) rejected by other people's mail
MBM> filters.
MBM> 4) you will waste enormous amounts of bandwidth.
>> Is there similar, has anyone heard of anyone managing to impliment this?
MBM> If you just delay messages[1] initially, and change the delay parameters
MBM> depending on what lists they are on, rather than outright rejecting them
MBM> things work a bit better. Often this means that other people complain
MBM> about the spam run and the machine gets pulled.
MBM> [1] by issuing a 450 in response to the RCPT TO:< line in the SMTP dialogue.
MBM> I've found this trick works extremely well...
MBM> of course,
MBM> 5) you will have problems with anybody who implements such a system on their
MBM> mailserver, as you'll have to wait to be able to test it.
MBM> MBM
