>> AppliedTheory [our network provider] has worked with Sprint [Applied
>> Theory's network provider] overnight on the issue,
>> as AT did not have a route to AOL networks on our backbone. Sprint
>> informed AT that AOL is making software and routing changes on
>> their network. This is causing significant changes in how Sprint
>> and AppliedTheory are receiving the routes to AOL and ICQ.
>
>This appears to have no relationship to the problem that I (and others)
>experienced. The information I provided clearly shows that mail was
>received by AOL's mail server(s), and clearly shows that it was returned,
>thus indicating a functional, if not optimal/robust network path.
>
>It's pretty obvious that AOL broke their mail servers; I'm waiting to
>see if they acknowledge their error and provide an explanation for it.
Rick, you may be right. I agree that on the surface it appears to have
no relationship to the problem described. However depending on how AOL's
e-mail is handled internally it is certainly possible that it does. I
have no idea how their e-mail systems are engineered, but if the mailboxes
are not kept on the systems that MX for the domain, and the routing problems
kept the MX systems from communicating with the mailbox systems, then we
have the makings for bouncing mail at the MX systems due to inability to
verify mailbox status.
-Mitch
