12 messages in net.sourceforge.lists.courier-usersRe: [courier-users] Re: Controlling w...
FromSent OnAttachments
Courier UserMay 5, 2003 6:41 am 
Sam VarshavchikMay 5, 2003 7:19 am 
Courier UserMay 5, 2003 7:36 am 
Jon NelsonMay 5, 2003 7:44 am 
Sam VarshavchikMay 5, 2003 8:35 am 
Courier UserMay 5, 2003 8:55 am 
James A BakerMay 5, 2003 10:25 am 
Brian CandlerMay 5, 2003 10:35 am 
Courier UserMay 5, 2003 11:54 am 
Courier UserMay 5, 2003 12:02 pm 
Anand BuddhdevMay 5, 2003 1:30 pm 
Courier UserMay 5, 2003 8:27 pm 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:Re: [courier-users] Re: Controlling when to send a DSNActions...
From:Brian Candler (B.Ca@pobox.com)
Date:May 5, 2003 10:35:41 am
List:net.sourceforge.lists.courier-users

On Mon, May 05, 2003 at 10:36:37AM -0400, Courier User wrote:

I forgot to mention that many of the spams are being sent to postmaster at my host, which is an existing mailbox. This large group of spams are being rejected by courier because their "from" address is illegal: <#@[]>

That's not an incoming message - that's a double-bounce, a message generated internally by your mailserver.

1. Incoming message: MAIL FROM: <bada@baddomain.com> RCPT TO: <some@yourdomain.com>

2. Fails your filters

3. Bounce is generated: MAIL FROM: <> RCPT TO: <bada@baddomain.com>

4. Bounce fails to be delivered because of invalid address

5. Double-bounce is generated: MAIL FROM: <#@[]> RCPT TO: <post@yourdomain.com> <<<<<<<<

So the spam wasn't necessarily to 'postmaster', it could have been to any user at your domain.

In my experience, in the absence of configuration errors, double-bounces are all attempts to return spam with invalid envelope-sender, so I just discard these errors silently.

However, whenever you are making changes to your MTA configs, it's advisable to keep and examine double-bounces for a while: they may point out a configuration error that you had overlooked.

Regards,

Brian.