ietf
[Top] [All Lists]

Re: The utilitiy of IP is at stake here

2003-05-27 18:08:12
Tony writes:

I get:
[<02>] The reason of the delivery failure was:

550-The IP address you are using to connect to AOL is a dynamic
(residential) 550-IP address.  AOL will not accept future e-mail
transactions from your 550-IP address until your ISP removes your IP
from its list of dynamic
550-(residential) IP addresses.  For additional information, please
visit 550-http://postmaster.info.aol.com.

Did you call AOL and ask them to remove you from their list of dynamic IP
addresses?  If so, what did they say?

I'm not on their list of dynamic addresses.  I'm not on their open-relay or
proxy lists, either.  The lady who answered the phone claimed that I need to
have a reverse lookup that matches my announced domain when I connect (at
least, as far as she seemed to understand what I was talking about).  But
when I reconfigure sendmail to send a HELO with the exact rDNS name for my
IP address, AOL still refuses the e-mail.  I think it is looking at MX
records or something, but the lady at the network center didn't know what an
MX record was, so I knew I was wasting my time with her, and she couldn't
help further or refer me to anyone else.  This prevented me from making any
progress.

Some have suggested that I smarthost through my upstream ISP's mail server.
But they bounce my e-mail, too ("relaying denied").  It seems that their
configuration believes my IP address to be a "foreign" address, even though
they assigned it to me.  They never updated their IP address database when
they added addresses.  When I e-mailed them about it, the person I talked to
didn't even understand the difference between a login name and password and
an IP address, and simply assured me over and over that my login and
password were valid.  So that was a waste of time as well.

It is not viable for me to refuse mail from AOL
since my wife's nephew works there so most of her
family uses AOL for email.

How does she e-mail back to them, in that case?

They may be able to get Internet access from other ISPs for less money.

As I've said, I'm bouncing all incoming AOL mail now, and, ironically, it
has eliminated a lot of spam, so a great deal of spam comes from AOL.COM, or
appears to, at least.

I was not AOL bashing, because they are not the
only ones doing this.

CS.COM and NETSCAPE.NET are doing it, but they are part of AOL.  I suppose
others are doing it.

FREEBSD.ORG bounces my e-mail because they don't like what I have to say,
but their e-mail administrator, in his infinite wisdom, apparently didn't
realize that if he bounces all my e-mail, I cannot unsubscribe from their
lists.  They are spamming my server with hundreds of messages a day.  I may
have to complain to their upstream ISP, since sending unsolicited e-mail and
not allowing me to put a stop to it is spamming.

As Vixie has pointed out, the current state of
affairs is the result of the protocol design. We
as the IETF need to step up and provide an alternate
design if we want the system to change.

I agree in principle, but I can't think of any changes in the protocol
offhand that would not require changes in every machine in the world.  That
isn't practical, so the only practical change is a backward-compatible
change that excludes only spam.  The problem is that spam doesn't look any
different from legitimate e-mail to a machine.

Some components of a new design need to be a viable
trust model, and irrefutable traceability.

Neither of these meet the criterion of "no changes to all the machines in
the world."  In order for them to work, every machine has to change,
everywhere.  That isn't going to happen; and if it doesn't happen, the
scheme won't work, and/or tons of legitimate e-mail will be lost.