jansteen@cwi.nl (Jan van der Steen) (06/27/90)
We want one of our printer queues to be shared between two
identical PostScript printers.
So, if both printers are operational they will process the
queue with double speed, and if one of them is down, then the
operational one will do all the work.
Does anybody have experience with such a setup?
The printers are remotely driven by a VAX 750 running BSD 4.3 UNIX.
--
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Jan van der Steen jansteen@cwi.nl
Centre for Mathematics and Computer Science (CWI)
Kruislaan 413, 1098 SJ Amsterdam, The Netherlandsjr@oglvee.UUCP (Jim Rosenberg) (07/03/90)
jansteen@cwi.nl (Jan van der Steen) writes: >We want one of our printer queues to be shared between two >identical PostScript printers. >So, if both printers are operational they will process the >queue with double speed, and if one of them is down, then the >operational one will do all the work. >Does anybody have experience with such a setup? The public domain BRL spooler, called MDQS (Multi-Device Queueing System) will do this and much more. I installed it on an AT&T 3b1 with no problem at all, but attempting to do so on an Altos 2000 drove me nuts. First the code broke the compiler. Specifically, one of the routines caused the compiler to emit bad code, emerging from a function call with the frame pointer messed up. Altos verified this and is supposedly working on it. When I blasted my way around that one I got the software to come up, sort of, but it was very sick and mdqsdaemon kept dying. When you can trust your C compiler you just wanna put your head down and weep. (Anyone have gcc for V.3.1 on an Altos 2000 up and running?) I don't know how healthy MDQS is on a VAX, but you might look into it. MDQS is a *neat* package. It allows one queue to feed multiple devices and multiple queues to feed one device and multiple queues to feed multiple devices. Sigh. On my 3b1 it seems healthy as a tank, which is nice because the 3b1 comes with a bit of a yucky spooler -- for all of my one users on the 3b1 -- as opposed to my many many users on the Altos ... -- Jim Rosenberg pitt Oglevee Computer Systems >--!amanue!oglvee!jr 151 Oglevee Lane cgh Connellsville, PA 15425 #include <disclaimer.h>
gwyn@smoke.BRL.MIL (Doug Gwyn) (07/04/90)
In article <559@oglvee.UUCP> jr@oglvee.UUCP (Jim Rosenberg) writes: >I don't know how healthy MDQS is on a VAX, but you might look into it. It works fine on various flavors of 4BSD, at least, on VAXes. Early releases of MDQS didn't work very well in System V or NFS environments, but we think those problems are fixed in the current release, which is available via anonymous FTP from host VGR.BRL.MIL as file arch/mdqs.tar.Z (a compressed tar archive of the sources).
prs@tcsc3b2.tcsc.com (Paul Stath) (07/06/90)
jansteen@cwi.nl (Jan van der Steen) writes: >We want one of our printer queues to be shared between two >identical PostScript printers. >So, if both printers are operational they will process the >queue with double speed, and if one of them is down, then the >operational one will do all the work. >Does anybody have experience with such a setup? >The printers are remotely driven by a VAX 750 running BSD 4.3 UNIX. Does BSD 4.3 Unix vary from AT&T SV so much that there are no such entities as printer groups in the LP spooler?? I have had very little experiance w/BSD, but it seems like most of the ideas that held true in one had a similar concept in the other. In AT&T SV you can spool your requests to a group called, say postscript, and then have printer ps1 and ps2 as members of that group. The spooler will use both printers in the group to process the spooled print jobs. If this is not the case, I'm sure glad I'm working on an AT&T SV box. >-- > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- > Jan van der Steen jansteen@cwi.nl > Centre for Mathematics and Computer Science (CWI) > Kruislaan 413, 1098 SJ Amsterdam, The Netherlands -- =============================================================================== Paul R. Stath The Computer Solution Co., Inc. Voice: 804-794-3491 ------------------------------------------------+------------------------------ INTERNET: prs@tcsc3b2.tcsc.com | "There was no diety involved,
guy@auspex.auspex.com (Guy Harris) (07/08/90)
>Does BSD 4.3 Unix vary from AT&T SV so much that there are no such entities >as printer groups in the LP spooler?? The 4.3BSD print spooler subsystem and the S5 print spooler subsystem differ so much that there are no such entities as printer groups in the 4.3BSD print spooler subsystem. With the 4.3BSD print spooler subsystem you do, however, get built-in support for printing to printers attached to other machines. (I.e., it's not some gross hack wrapped around UUCP, or something equally horrid.) If you get S5R4, you get both, at least according to the documentation; the documentation claims that the daemon on one machine can talk to a daemon on another machine using either some native S5R4 protocol or the protocol used by the BSD spooler (although at times they misleadingly speak of the SunOS spooler - the latter is based on the BSD one and didn't change the protocol, so I'd expect the S5R4 spooler to be able to talk to any BSD-flavored "lpd").
det@hawkmoon.MN.ORG (Derek E. Terveer) (07/24/90)
In article <1725@charon.cwi.nl> jansteen@cwi.nl (Jan van der Steen) writes: > We want one of our printer queues to be shared between two > identical PostScript printers. > [...] > Does anybody have experience with such a setup? Yes, but on system V -- they have something called "classes" of printers, which does exactly what you are describing. I don't think vanilla bsd has an equivalent feature. derek -- Derek Terveer det@hawkmoon.MN.ORG