🚀 go-pugleaf

In article <1486@gazette.bcm.tmc.edu> jmm@skivs.ski.org writes:
>Has anyone noticed the following bug in the version of mailx distributed
>with RTU-3.1A: When a message is sent to an internet address, eg:

Sorry, but I didn't notice this when we were running 3.1 or now under
4.0, but I have noticed another interesting mail bug I could use help
with...

Our sendmail seems to periodically get confused while recieving SMTP
mail, but instead of reporting its error to a log file or the console,
if writes it back out over the network connection, causing the other
end to think an error has occured. Note that the mail IS received
normally - the message is just a warning.

The end result is that the other end assumes mail has failed since
it detected a SMTP protocol error - i.e. rtu has messed up, and will
retry after some time-delay. This means that the masscomp user recieves
a copy of the letter on a regular basis (say every 1/2 hour) for days
(until sendmail is restarted).

Our sysop has no idea what is causing this; in fact there seems to be
no obvious trigger that starts sendmail on this death spin!

So, has anyone else seen this problem? The network gurus here claim
that RTU is violating the STMP and tcp/ip protocols by reporting
errors to the network socket (instead of a logfile); they say masscomp
should fix it.

Thanks for any help,
/Jamie
--------
  UUCP: {rutgers,ames}!cit-vax!bek-owl!mazer
  ARPA: mazer@bek-owl.caltech.edu
BITNET: jmazer@caltech.bitnet      "Michael J. Fox has no Elvis in him."

  UUCP: {rutgers,ames}!cit-vax!bek-owl!mazer
  ARPA: mazer@bek-owl.caltech.edu
BITNET: jmazer@caltech.bitnet      "Michael J. Fox has no Elvis in him."

The MASSCOMP USERS' SOCIETY newsgroup is comp.sys.masscomp.
Articles to: masscomp@soma.bcm.tmc.edu or uunet!soma.bcm.tmc.edu!masscomp
Administrative stuff: masscomp-request@soma.bcm.tmc.edu
Stan Barber, Moderator