[15295] in Athena Bugs
Re: Who Are you?
daemon@ATHENA.MIT.EDU (Matt Braun)
Sun Jul 27 13:45:59 1997
From: Matt Braun <matt@MIT.EDU>
To: Greg Hudson <ghudson@MIT.EDU>
Cc: Matt Braun <matt@MIT.EDU>, testers@MIT.EDU, postmaster@MIT.EDU,
bugs@MIT.EDU
In-Reply-To: Your message of "Sat, 26 Jul 1997 12:38:27 EDT."
<199707261638.MAA16516@the-light-fantastic.MIT.EDU>
Date: Sun, 27 Jul 1997 13:45:52 EDT
> > OK...let me rephrase the question. What is rel-eng doing to remedy
> > this problem?
>
> rel-eng does not appreciate being held directly responsible for
> developing fixes for every problem in the environment.
Well, postmaster is held accountable for all the mail problem on the internet
that vaguely have something to do with MIT, including, but not limitted to aol
not accepting mail for weeks and mail queueing on Athena workstations for long
periods of time, so consider yourself lucky.
> However, since rel-eng continues to be an approximation of Athena
> development, let me ask for more information about the problem. Do we
> know what is going on such that a piece of mail bounces to postmaster?
> My understanding is that mail simply sitting in the sendmail queue on
> a workstation should be delivered without problem, so I don't really
> understand where these bounces are coming from.
OK...flaming aside, I do not know what is actually going on on the
workstations when this happens. I verified that the mail is coming from 8.1
machines in the cluster shortly after the release and thought that was
sufficient to believe it was a release problem not a 3psw of some sort problem
(including SIPB). I believe what is happening is that the user is logging out
before the mail is fully queued, but that is pure speculation. I am not sure
how to provide more information here. Obviously, postmaster has the full
bounces with sender, recipient and message, but I am not willing to forward
them to world-readable lists. I believe it would be appropriate for me to
provide either username / workstation / timestamp information (so you can or
someone can follow up) or a couple of full bounces minus the message body
personally to whomever in rel-eng is working on this. Of course, if there is
a better option, feel free to suggest it.
Matt
