[Raw Msg Headers][Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Zmailer SMTP: Anal retentive or correct?
> We have a few sites that get mail via UUCP, typically the uucp name is
> foo and the domain name is foo.com or foo.org. Recently I've noticed
> that some smtp receivers have been rejecting mail from these sites
> because the name is not fully qualified. I grumbled, because I assumed
> it was another revision to the ever imperfect sendmail.
Message transport envelope origins should always be
in FQDN, if they are not, receiving system is IMO
qualified to complain. I don't do it, because at
utu.fi there are systems using our mailhost as blinding
hub -- uname@host --> uname@utu.fi
> However, tonight I had occasion to try to manually send one of these
> pieces of mail (to parc.xerox.com), and guess what, it's a zmailer
> site.
Was it like this ?
<-- MAIL FROM:<foo@foo>
--> 553 Bad FROM address ...
Now presume the sender hostname to be:
polaris
At utu.fi there is an old server with that same name,
if we receive email from outside claiming origins from:
user@polaris
are we right to assume it to be user@polaris.utu.fi ?
> I can't believe this is the correct behavior - has this been added to
> zmailer recently - our smtpserver certianly has *never* done this!
The PARC server uses extremely conservative approach,
their server is way old (and stable..). It also has
some local changes.
> In any case, is there something I can do here as a hack to fix this
> problem? My success rate in getting over worked admins to change what
> looks like a working config to them is less than nil...
Perhaps a bit education to the UUCP feeders ?
The UUCP is fine for store & forward, however its
use does not mandate the use of UUCP names for
the message origination.
In fact the only UUCP link I have experience with
does use FQDN for the source, and destination
addresses, while using UUCP link names for the
store & forward service.
> Thanks...
> sdb
> --
> sdb@ssr.com
/Matti Aarnio <mea@utu.fi> <mea@nic.funet.fi>