[comp.mail.elm] Elm removing /tmp files

yountm@numenor.UUCP (Marvin Yount) (10/18/89)

We seem to have a couple of minor problems with Elm.  We're running Version
2.2 PL11 in a sys5 mail environment on an Apollo SR9.7 system.  I know that
Elm isn't supposed to run well on SR9.7, what I want to find out is whether
the problems we're seeing are because of SR9.7, a problem with our
configuration, or a problem with Elm itself.  We are moving to SR10, so if
that's the problem it'll take care of itself.  If it's our configuration,
we can deal with it when we configure Elm for SR10.  So... on to the
problems.


Whenever we startup Elm from the startup file (startup_dm.1280bw), we get the
following error after a mail message is sent:

  sh: rm: not found

Also, files similar to those below are left in the /tmp directory:

Unix: ls -C -F /tmp
mbx_$helper_lock*  snd.264.bak        snd.AABa00264
mbx_$sysmbx*       snd.96.bak         snd.AACa00264
snd.181.bak        snd.AAAa00264      snd.AADa00264
Unix:

The strange thing is that if we startup Elm from a key definition after
logging in we don't see the same problem.  I've looked through the
configuration document, and the transcript pad for the configuration and
everything looks fine.  It says it's found rm in /bin/rm.


The second problem also deals with /tmp files.  If our mail system crashes it
usually takes Elm down with it.  This isn't bad, except that Elm doesn't
clean-up the /tmp/mbox.<userid> file, forcing the user to do so manually
before restarting Elm.  Again, I'd like to know if this is a problem with
SR9.7, our configuration of Elm, or Elm.

Thanks is advance for any help.
-- 
======================================================================          
Marv Yount (yountm@gtephx)
UUCP: {ncar!noao!asuvax | uunet!zardoz!hrc}!gtephx!yountm
AG Communication Systems, Phoenix