Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 4.3bsd-beta 6/6/85; site ucbvax.BERKELEY.EDU
Path: utzoo!watmath!clyde!burl!ulysses!cbosgd!ucbvax!cs.ucl.ac.uk!gdmr%cstvax.edinburgh.ac.uk
From: gdmr%cstvax.edinburgh.ac.uk@CS.UCL.AC.UK (George D M Ross)
Newsgroups: mod.computers.vax
Subject: Re: print symbionts (and vax C rtl)
Message-ID: <4273.8603111305@cstvax.edinburgh.ac.uk>
Date: Tue, 11-Mar-86 08:05:15 EST
Article-I.D.: cstvax.4273.8603111305
Posted: Tue Mar 11 08:05:15 1986
Date-Received: Sun, 16-Mar-86 10:32:35 EST
References: <8603081141.AA15136@ucbvax.berkeley.edu>
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: Comp. Sci., Edinburgh Univ.
Lines: 24
Approved: info-vax@sri-kl.arpa

From Eric Gisin, egisin@waterloo.csnet:
>It currently stops the job controller, which says it received an EOF.
>..... (the job controller and symbiont seem to use a mailbox
>assigned to sys$input and sys$output to communicate).

Ah.  That would explain why the job controller shut down when my modified
symbiont died, having claimed it (the job controller) had received a number
of dud requests and an EOF.  Presumably the symbiont's traceback information
would go to SYS$OUTPUT.

Anyway, while we're on the subject of print symbionts, does anyone know of
a clean way of a user-output routine detecting the end of a print job, or
alternatively detecting when its queue has drained?  The scenario is that
we have a Gandalf PACX 2000 interposed between the printers and the various
hosts which might want to talk to them.  The symbionts and output filters in
the various hosts toggle DTR and are connected to the printer by means of a
programmed call.  At present they decide when to relinquish the printer on the
basis of a timeout.  It works very nicely, but it's rather inelegant....


George D M Ross, Dept. of Computer Science, Univ. of Edinburgh, Scotland
Phone: +44 31-667 1081 x2730
JANET: gdmr@UK.AC.ed.cstvax  --> ARPA: gdmr@cstvax.ed.AC.UK
UUCP:  <UK>!ukc!cstvax!gdmr