Path: utzoo!utgpu!water!watmath!clyde!bellcore!faline!thumper!ulysses!andante!mit-eddie!bu-cs!purdue!i.cc.purdue.edu!j.cc.purdue.edu!pur-ee!uiucdcs!uxc.cso.uiuc.edu!uxe.cso.uiuc.edu!sandrock
From: sandrock@uxe.cso.uiuc.edu.UUCP
Newsgroups: comp.os.vms
Subject: Re: HELP: symbiont dies under 4.7
Message-ID: <62500014@uxe.cso.uiuc.edu>
Date: 19 May 88 21:55:00 GMT
References: <62500009@uxe.cso.uiuc.edu>
Lines: 39
Nf-ID: #R:uxe.cso.uiuc.edu:62500009:uxe.cso.uiuc.edu:62500014:000:1879
Nf-From: uxe.cso.uiuc.edu!sandrock    May 19 16:55:00 1988
Posted: Thu May 19 17:55:00 1988


> /* Written 10:54 am  May 17, 1988 by sandrock@uxe.cso.uiuc.edu in uxe.cso.uiuc.edu:comp.os.vms */
> /* ---------- "HELP: symbiont dies under 4.7" ---------- */
>    Since we recently upgraded our VAX/780 from VMS 4.4 to 4.7, a rather
>  annoying problem is cropping up with TeX/LaTeX files spooled to our
>  Xerox 2700 laser printer.
>    The problem is that certain TeX/LaTeX files will print to a certain
>  point in the file, and then cause the symbiont process to terminate!
>  The result is that all three printer queues (all terminal queues) are
>  stopped until someone with the right privs executes a START/QUEUE for
>  each of the three printer queues.
>
Interestingly enough, the problem turned out to be due to a new "feature"
of VMS 4.7, wherein SMBSRVSHR.EXE checks for "non-printing characters" in
the output stream. The offending character in this case is 0x9D, something
DEC refers to as OSC (Operating System Command), and apparently does not
think should be sent to a printer/terminal port. This would be fine if all
we were printing were ASCII files, but the files generated by the Beebe
DVIX27 driver (from Utah) contain mostly non-ASCII data!

There may be a better way around this problem, but for a quick fix we simply
replaced the new SYS$SHARE:SMBSRVSHR.EXE image with the old one from 4.4,
and everything was back to normal. Ideally, there should (and perhaps is)
some way of telling VMS that a non-ASCII stream is being sent, so that it
doesn't try to interfere with the data being sent.

Also, the fact that the print symbiont was crashing seems to just be a
"bug" in their new "feature" so far as I can tell.

Regards,
Mark Sandrock

University of Illinois at Urbana-Champaign
School of Chemical Sciences Computing Services
505 S. Mathews Ave., Urbana, IL  61801

BITNET:   sandrock@uiucscs
Internet: sandrock@b.scs.uiuc.edu
Phone:    217-244-0560