Qmail
[Top] [All Lists]

Re: Bind outgoing SMTP traffic to specific IP address

To: qmail List <qmail@list.cr.yp.to>
Subject: Re: Bind outgoing SMTP traffic to specific IP address
From: Phil Breskey <pbreskey@dotcomholdingsofbuffalo.com>
Date: Thu, 09 Aug 2007 12:38:25 -0400
Delivered-to: sp-com-lists@consult.net
Delivered-to: gmail-qmail@securepoint.com
Delivered-to: sp.com.list@gmail.com
Delivered-to: mailing list qmail@list.cr.yp.to
In-reply-to: <20070808152841.GK11024@s876651.local>
Mailing-list: contact qmail-help@list.cr.yp.to; run by ezmlm
References: <46B9C83E.4060806@dotcomholdingsofbuffalo.com> <20070808152841.GK11024@s876651.local>
User-agent: Thunderbird 2.0.0.4 (X11/20070604)
Kyle Wheeler wrote:
On Wednesday, August  8 at 09:42 AM, quoth Phil Breskey:
Is the Markus Stumpf patch included in netqmail 1.05?

No. If you read the netqmail readme, it will tell you exactly what patches are included and what the philosophy was for including them (hint: the idea wasn't to add a bunch of features).


I did read it. It states that it is comprised of qmail-1.03...<and stuff>. On qmail.org, under "Yet More Qmail Addons", it says that Andy Repton had ported the patch to qmail 1.03. I looked at the sources from netqmail 1.05, and didn't see it but realized that I could have easily missed it. Understanding that 1.03 might not equal 1.03, I figured it be best to ask.

If the patch is included, would it be optimal to use the patch, or iptables?

Depends on your definition of "optimal".

Using iptables isn't portable, but doesn't require patching (which is of course a bonus if you're worried about bugs introduced by patching). It also doesn't require recompiling. On the other hand, the patch has lower overhead, and is sufficiently simple to be easy to validate yourself.


Optimal from a performance perspective.

There's an equivalent patch by Felix von Leitner, available here: http://www.fefe.de/qmail/


I'll check it out.  Thanks.

<Prev in Thread] Current Thread [Next in Thread>