Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!cica!iuvax!ux1.cso.uiuc.edu!mp.cs.niu.edu!rickert
From: rickert@mp.cs.niu.edu (Neil Rickert)
Newsgroups: comp.mail.sendmail
Subject: Re: Sendmail Internals Question
Message-ID: <1990Sep27.011020.14876@mp.cs.niu.edu>
Date: 27 Sep 90 01:10:20 GMT
References: <1990Sep26.170255.10577@odetics.com>
Distribution: na
Organization: Northern Illinois University
Lines: 35

In article <1990Sep26.170255.10577@odetics.com> frank@odetics.UUCP (Frank Merrow) writes:
>Hi,
>
>I have a question about the way sendmail works.  I have been toying with
>(....)
>Lets suppose I send the same message to 20 users on the net.  For this
>example lets say 10 of the users are on cpu1.com and 5 of the users are
>on cpu2.com and 5 of the users are on cpu3.com.  My question is how many
>times is my message sent to uunet:
>
>1. 20 times (once for each user) because sendmail is a pretty dense
>2. 3 times (once for each CPU) because sendmail is kind of smart
>3. Once (and copies made as needed on uunet) because sendmail is very smart

  It all depends on the MAILER definition in 'sendmail.cf'.  Look at the
F= string of flags.  If the 'm' flag is present, the message will be sent once.
If the 'm' flag is not present, it will be sent 20 times.

  I presume that you are using the UUCP mailer.  Usually the UUCP mailer does
not have the 'm' flag set, because there are some UUCP hosts out there which
cannot handle it.  It is very likely that uunet can handle the 'm' flag.
If all of your UUCP neighbors can handle it, then just change the mailer
definition to define the 'm' flag.  If some neighbors cannot handle it,
you should be able to define a new mailer which is almost identical to
the UUCP mailer.  Call it something different such as the UUCPM mailer.  Make
it identical except that it has the 'm' flag, then modify 'sendmail.cf' so
that mail thru 'uunet' selects the UUCPM mailer, while other UUCP mail selects
the standard mailer.

  Obviously you would need to test this thoroughly.
-- 
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
  Neil W. Rickert, Computer Science               <rickert@cs.niu.edu>
  Northern Illinois Univ.
  DeKalb, IL 60115.                                  +1-815-753-6940