Re: [exim] Has anyone done this?

Top Page
Delete this message
Reply to this message
Author: Ian Eiloart
Date:  
To: Matt Fretwell, exim
CC: 
New-Topics: [exim] Port Tests to Verify the Sending Hosts
Subject: Re: [exim] Has anyone done this?


--On March 6, 2005 16:29:39 +0000 Matt Fretwell <mattf@???> wrote:

> Marc Haber wrote:
>
>> > Be prepared to explain to your upstream why you are port scanning other
>> > people's mail servers.
>>
>> Trying to establish TCP connections to other machines that are
>> connected to the internet is not illegal. At least it isn't in
>> Germany, and that's a feature.
>
> No, but it is extremely annoying in most cases, and ends up getting the
> client firewalled, (in my book anyway),


That's ridiculous. Why would an SMTP server that's trying to send me mail
object that I try to make a TCP connection to it. I already do ident
callouts to the sending host. Surely its completely fair to make a callout
to a sending host as frequently as it tries to send mail to me.

It looks to me like the list is playing the game of trying to find the most
spurious objections to Marc's proposals that it can - again.

> if the results are not cached.


I think he already suggested something like that.

> Persistent SAV probes for every mail, even when minutes apart, are a prime
> example. If, BIG IF, however, he does use it to feed a whitelist as he
> later mentioned, that would not be a problem|annoyance.
>
> I will be honest though, even after Peter pointed out the plus point if
> implemented correctly, I still cannot see an overall benefit from this
> method. The concept would be simple to circumvent if it became generally
> deployed.


As are most of the methods that we use to prevent spam. Its an arms race.

>
> Matt



--
Ian Eiloart
Servers Team
Sussex University ITS