<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/3.28.3">
</HEAD>
<BODY>
On Tue, 2013-05-21 at 18:02 +1000, Karl Auer wrote:<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
<PRE>
> 5xx Reject unknown client host 45.71%
>
> That's a rather large chunk of trash that amavisd doesn't have to look
> at
Except, of course, that you don't know it's all trash. Which is the
point.
</PRE>
</BLOCKQUOTE>
<BR>
As I just indicated with Mark, I agree it has the potential to have FP's, just like we have the potential to be visited by ET <IMG SRC="cid:1369174102.4055.24.camel@tardis" ALIGN="middle" ALT=":)" BORDER="0"><BR>
<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
<PRE>
> I know, my bad for suggesting somebody actually do some
> work :)
I don't think anyone disagrees with the fact that it is, in the great
scheme of things, pretty easy to add a PTR record. Nonetheless, it may
be hard for some legitimate senders, and more so as IPv6 ramps up, who
you will then punish by dropping their emails.
</PRE>
</BLOCKQUOTE>
<BR>
So, you propose we put it in the too hard basket, and to not GAF? let the status quo go on? that;s the wrong appraach, nip it in the bud in the very early days is the way to fix it, that way, people will get accustomed to doing it right, what it we always say, fix the root cause, not try catch up patch it later.<BR>
<BR>
<BLOCKQUOTE TYPE=CITE>
<PRE>
Regards, K.
[1] This is not to say that rDNS has no use. It's cool from an
informational point of view to be able to see who an address range
"belongs" to, helpful in traceroutes and so forth. But it's not
*necessary*.
</PRE>
</BLOCKQUOTE>
<PRE>
</PRE>
I'm more interested in running traceroute with -A if I want to know who the RP is.<BR>
<BR>
<BR>
</BODY>
</HTML>