[comp.mail.sendmail] sendmail on SCO Xenix

jjr@siva.UUCP ( Joe Rackelmann ) (08/29/90)

Has anyone gotten sendmail working under SCO Xenix?  sendmail
comes with the SCO TCP/IP package.  I have 1.0.1h version of
TCP/IP for _Xenix_, supposedly the latest.  I have been hacking
(torturing myself) at the sendmail.cf file, and have just one
obscure problem:  Certain messages from off-site machines that
arrive via uucp get stuck in "mailq".  When I run sendmail -q,
it says that it is Connecting to  (local)
and hangs.  From another terminal, I see that it is stuck
(using ps):   mail.local -r site!site!site.dom!username
where username also has dots in it.   I don't know if that's
the problem, but calling mail.local from a shell works fine.
I can also receive mail from other sites via uucp just fine.
Our tcp connection is local for now, but will be on Internet
by end of year -- mail sent via tcp (ether) also works fine.
Any ideas?
-- 
jjr@cdffp                        Someone's had too much to think!
or mebbe ucdavis!csusac!sactoh0!siva!cdffp!jjr

david@actsn.fay.ar.us (David Summers) (08/29/90)

I got a version of Sendmail 5.61 + IDA 1.3.4 last February (?) to work with
Xenix, and have been advertising it off and on since then.  With the IDA
config file it is fairly easy to configure with the m4 macro processing package.
I put this on our BBS (shown above in the "Summary:".  I have found out that the
config file STILL does not work "right" (or at least do what I want) and I got
tired of messing with it (after three years(!?)) and I've read that it isn't
POSSIBLE to be completely correct when acting as gateway between the Internet
and UUCP because there are no STANDARDS...just suggestions, and everyone picks
one that suits them.  Also there may be SEMANTIC differences between routes and
domain addresses so it MAY not even be possible to have a completely correct
config file.   Arrrgggghhh!

Anyway, last night I finally got around to compiling SMAIL 3.1.19 and boy is it
nice!  Readable config files!  Wow!  It compiled with only one hitch.  It tried
to compile pathalias in the "pd" directory (I think) with a MEMSET constant that
was already declared or maybe it was incorrect return type or something...I
forget.   But since that file is not REQUIRED then everything compiled 
BEAUTIFULLY, the first time and I seem to be running it with no problems right
now.  I'm working on beginning to understand the configuration files but that
should be no problem as everything seems very well documented.

  In summary, I'm well pleased with SMAIL 3 and suggest to everyone that if
you have the chance, TRY IT!  (I haven't tried it on the INTERNET yet, just
a Xenix system with UUCP).

Hats off to the SMAIL 3 designers/implementors!  Thanks!

   - David Summers
-- 
I'm sick and tired of this machine, I wish that I could sell it.
It never does just what I want, but only what I tell it!
- David Summers (david@actsn.fay.ar.us)

rickert@mp.cs.niu.edu (Neil Rickert) (08/30/90)

In article <557@actsn.fay.ar.us> david@actsn.fay.ar.us (David Summers) writes:
>
>I got a version of Sendmail 5.61 + IDA 1.3.4 last February (?) to work with

 That was rather clever, since I didn't start designing IDA-1.3.x until
March.  The first public release was in May or June.  1.3.4 dates from
July 23 this year.  May I have a copy of the plans for your time machine?

 You probably have IDA-1.2.8 which likes to convert all addresses into
%-hack addresses.  You will find 1.3.4 quite different - it tries to not
monkey around with address formats any more than needed for reliable
mail transmission.

-- 
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
  Neil W. Rickert, Computer Science               <rickert@cs.niu.edu>
  Northern Illinois Univ.
  DeKalb, IL 60115.                                  +1-815-753-6940

david@actsn.fay.ar.us (David Summers) (08/30/90)

In article <1990Aug29.180124.16838@mp.cs.niu.edu>, rickert@mp.cs.niu.edu (Neil Rickert) writes:
> In article <557@actsn.fay.ar.us> david@actsn.fay.ar.us (David Summers) writes:
> >
> >I got a version of Sendmail 5.61 + IDA 1.3.4 last February (?) to work with
> 
>  That was rather clever, since I didn't start designing IDA-1.3.x until
> March.  The first public release was in May or June.  1.3.4 dates from
> July 23 this year.  May I have a copy of the plans for your time machine?
> 
>  You probably have IDA-1.2.8 which likes to convert all addresses into
> %-hack addresses.  You will find 1.3.4 quite different - it tries to not
> monkey around with address formats any more than needed for reliable
> mail transmission.

Ok.  What I MEANT to say (I said it but not clearly), was that I got a 
version of Sendmail 5.61 + IDA patches and used the IDA 1.3.4 (or whatever the
latest) configuration file.  Sorry for the mis-understanding.

   - David Summers







-- 
I'm sick and tired of this machine, I wish that I could sell it.
It never does just what I want, but only what I tell it!
- David Summers (david@actsn.fay.ar.us)

avg@hq.demos.su (Vadim G. Antonov) (08/30/90)

In article <86@siva.UUCP> jjr@siva.UUCP ( Joe Rackelmann ) writes:
>Has anyone gotten sendmail working under SCO Xenix?

	I've ported the sendmail from 4.3 BSD tape to XENIX 386 (2.3.2).\
	It was not a trivial action because of bugs (incompatibilites?)
	in SCO STREAMS TCP (by Lachman & Convergent Tech.).
	Yet another feature I've added - a full 8-bit transparency
	(may be useful for European guys). This code is ported now
	under Interactive ix/386 and back to VAX under BSD.

	I could send it to you (or to anybody who wants to get it).

	Vadim.