[prev in list] [next in list] [prev in thread] [next in thread] 

List:       qmail
Subject:    Re: nolisting and netqmail 1.06
From:       Diego Pomatta <infosis () abelsonsa ! com ! ar>
Date:       2008-07-22 15:47:01
Message-ID: 488600F5.1090600 () abelsonsa ! com ! ar
[Download RAW message or body]

Markus Stumpf escribió:
> On Tue, Jul 22, 2008 at 12:16:08PM -0300, Diego Pomatta wrote:
>   
>> The debates I've read around the web, center around this issue. That 
>> qmail should not give up without trying the other MXs, because if I'm 
>> about to give up, I might as well do everything I can before; and the 
>> mail might/would be delivered with just giving the secondary MX a single 
>> try, instead of bumping my head against the primary MX for 3 days and 
>> quiting. ;)
>>     
>
> I have a better strategy than "nolisting":
>
> Set up 500 MXs with the same distance and make all but one return 4xx. I
> am sure spammers will give up, but every other mailserver surely has the
> resources and can happily spend a lot of connections and time sorting the
> correct one out.
> And hey ... the Internet is fast ... with a mean chance of 250 tries for each
> mail that shouldn't be that hard, right?
>
> And to make it even more effective make not one mailserver permanently
> accept messages but let them return the 4xx with a 499:500 probability
> and accept the message with a 1:500 probability. And of course use a
> initial timeout of 2 minutes before sending the greeting.
>
> WOW! That would eliminate so much spam. Maybe I should get a patent for this.
>   

Yes, irony aside, I also think "nolisting" stinks.
/D

[Attachment #3 (text/html)]

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Markus Stumpf escribi&oacute;:
<blockquote cite="mid:20080722153547.GD3330@server.thinkof.de"
 type="cite">
  <pre wrap="">On Tue, Jul 22, 2008 at 12:16:08PM -0300, Diego Pomatta wrote:
  </pre>
  <blockquote type="cite">
    <pre wrap="">The debates I've read around the web, center around this issue. That 
qmail should not give up without trying the other MXs, because if I'm 
about to give up, I might as well do everything I can before; and the 
mail might/would be delivered with just giving the secondary MX a single 
try, instead of bumping my head against the primary MX for 3 days and 
quiting. ;)
    </pre>
  </blockquote>
  <pre wrap=""><!---->
I have a better strategy than "nolisting":

Set up 500 MXs with the same distance and make all but one return 4xx. I
am sure spammers will give up, but every other mailserver surely has the
resources and can happily spend a lot of connections and time sorting the
correct one out.
And hey ... the Internet is fast ... with a mean chance of 250 tries for each
mail that shouldn't be that hard, right?

And to make it even more effective make not one mailserver permanently
accept messages but let them return the 4xx with a 499:500 probability
and accept the message with a 1:500 probability. And of course use a
initial timeout of 2 minutes before sending the greeting.

WOW! That would eliminate so much spam. Maybe I should get a patent for this.
  </pre>
</blockquote>
<br>
Yes, irony aside, I also think "nolisting" stinks.<br>
/D<br>
</body>
</html>


[prev in list] [next in list] [prev in thread] [next in thread] 

Configure | About | News | Add a list | Sponsored by KoreLogic