ietf-smtp
[Top] [All Lists]

Mixed Return Codes

2008-04-11 12:58:31

Not for nothing, but I did bring this up and was shot down. Now we have the "World's Large Email ISP" AOL exactly what I said will begin to happen:

S: 220-rly-ma02.mx.aol.com ESMTP mail_relay_in-ma02.3; Fri,
       11 Apr 2008 -0400
S: 220-America Online (AOL) and its affiliated companies do not
S: 220-     authorize the use of its proprietary computers and computer
S: 220-     networks to accept, transmit, or distribute unsolicited bulk
S: 220-     e-mail sent from the internet.  Effective immediately:  AOL
S: 220-     may no longer accept connections from IP addresses which
S: 220      have no reverse-DNS (PTR record) assigned.
C: EHLO fbgcollection.com
S: 250-rly-ma02.mx.aol.com peer name unknown
S: 250 HELP
C: MAIL FROM:<XXXXXXX(_at_)XXXXXXXXXXX(_dot_)com>
S: 250 OK
C: RCPT TO:<XXXXXXXX(_at_)aol(_dot_)com>
S: 250 OK
C: DATA
S: 354 START MAIL INPUT, END WITH "." ON A LINE BY ITSELF
S: 421-:  (DNS:NR)  http://postmaster.info.aol.com/errors/421dnsnr.html
S: 421 SERVICE NOT AVAILABLE
C: QUIT
S: 221 SERVICE CLOSING CHANNEL

The issue started with my brother reporting he was having outbound mail issues with AOL and COMCAST.NET and wanted me to check it out.

As you can see, they are requiring a PTR record. Brother Vince has domain DNS setup at Godaddy and after 1 hour of talking with them they have no way to setup a PTR record - obviously they don't know what they are talking about. This is a business account with static IPs and I see this PTR crap is that - crap.

But more interesting, is the mixed response codes for the DATA command:

 C: DATA
 S: 354 START MAIL INPUT, END WITH "." ON A LINE BY ITSELF
 S: 421-:  (DNS:NR) http://postmaster.info.aol.com/errors/421dnsnr.html
 S: 421 SERVICE NOT AVAILABLE

Lisa, now do you see this as been BIG enough and not just isolated to just "small vendors?"

Or is a AOL person lurking here with the powers to fix the above, do so?

Of course, it may be too late to address this, but this was all predicted as more and more DATA level processing is happening. While I was more concern with timeout issues, the issue is the same - MIXED reply codes which was "semantically" possible in 821 but now specifically made clear in 2821bis as a NO-NO.

--
Sincerely

Hector Santos, CTO
http://www.santronics.com
http://santronics.blogspot.com

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