On Sun, 13 Jul 2003 00:02:59 +0200 Matthias Andree
<ma(_at_)dt(_dot_)e-technik(_dot_)uni-dortmund(_dot_)de> wrote:
Robert Vazan <robertvazan(_at_)privateweb(_dot_)sk> writes:
I tried to send myself such broken Return-Path by telneting to Smtp
server. I have been successful. Exchange is not the (only) problem.
The question is: can fetchmail recover from such a condition at all, or
has information been lost?
Headers and body survived. There is only one mailbox and only one local
recipient and fetchmail knows about it (it runs under my account).
In the latter case, Exchange must be fixed to
ALWAYS drop a syntactically correct Return-Path: header.
This isn't going to be done by Exchange vendor and I understand it.
Fixing bugs works at smaller scale, but robustness (tolerance against
flaws of neighbours) is necessary in large scale systems or internet
will never work.
It carries
ESSENTIAL information.
What would happen if fetchmail pretended that Return-Path contains
postmaster(_at_)local-domain?
But you are getting a Return-Path: for regular mail, i. e. after
MAIL FROM: <robertvazan(_at_)privateweb(_dot_)sk>
RCPT TO: <strange(_dot_)exch(_at_)ang(_dot_)example(_dot_)org>
DATA
blah
.
?
Hmm, this generated error message. What is it supposed to demonstrate?
As you've presented the original headers, that's OK. Can you leave the
example files available for a while?
Theoretically forever.