Usenet 1994 January

home *** CD-ROM | disk | FTP | other *** search

/ Usenet 1994 January / usenetsourcesnewsgroupsinfomagicjanuary1994.iso / sources / unix / volume25 / ispin / part15 < prev next >

Wrap

Text File | 1992-02-01 | 13.7 KB | 318 lines

Newsgroups: comp.sources.unix From: sir-alan!ispin!lbartz@iuvax.cs.indiana.edu (Larry Bartz) Subject: v25i126: Indianapolis Standard Printer Interface for Networked printers, Part15/15 Sender: sources-moderator@pa.dec.com Approved: vixie@pa.dec.com Submitted-By: sir-alan!ispin!lbartz@iuvax.cs.indiana.edu (Larry Bartz) Posting-Number: Volume 25, Issue 126 Archive-Name: ispin/part15 #! /bin/sh # This is a shell archive. Remove anything before this line, then unpack # it by saving it into a file and typing "sh file". To overwrite existing # files, type "sh file -c". You can also feed this as standard input via # unshar, or by typing "sh <file", e.g.. If this archive is complete, you # will see the following message at the end: # "End of archive 15 (of 15)." # Contents: ISPIN/misc/PC-LANtoISPIN/README # ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN # Wrapped by socrates@indy6 on Tue Jan 28 15:27:23 1992 PATH=/bin:/usr/bin:/usr/ucb ; export PATH if test -f 'ISPIN/misc/PC-LANtoISPIN/README' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'ISPIN/misc/PC-LANtoISPIN/README'\" else echo shar: Extracting \"'ISPIN/misc/PC-LANtoISPIN/README'\" $7134 characters$ sed "s/^X//" >'ISPIN/misc/PC-LANtoISPIN/README' <<'END_OF_FILE' X11/21/91 X XAll the articles in the various PC and networking magazines on the topic Xof "Network Printing" are enough to make me gag myself with a printer ribbon. X XThey aren't talking about networked printers at all! They're talking hard- Xwired printers on networked systems. That's baby stuff. X XWith ISPIN, we have printers ON THE NETWORK. If you're this deep into the Xapplication, you already know how we do it. X XThis document outlines our recent (like last week) successful efforts to Xissue print jobs from MS-DOS PC clients of a Novell server to a UNIX box Xon the same LAN which (among other things) is acting as an ISPIN server. X XWe've been doing network printing from UNIX boxes, through X.25 apads and Xlocal switches, to printers for about three years. A 10BASE-T ethernet, XNovell server, and MS-DOS client PCs were recently introduced to our local Xenvironment. X XRight away, the Novellers (ex MS-DOSers) started talking about how expensive Xand difficult it is to do network printing. The PC and network mags said Xpretty much the same. X XHey, bud. I got yer network printing right here. X XAs with all things UNIX, there are/will be many possible ways to do this X(see OTHER WAYS TO SKIN THE CAT). X XHere's our setup: X X Hardware: X * X.25 network with asynchronous pads X - all of our printers, dumb ascii terminals, X and UNIX cpu serial i/o ttys are connected X - provides RS-232C serial i/o connectivity X - we use it for both local and wide area X (nationwide) connectivity X - on a smaller or local scale, a "smart switch" X or stat mux (like Tellabs, Equinox, Mitron, X David, etc) can do the same thing X X * 10BASE-T Ethernet, cards, transceivers, concentrator X - connects MS-DOS PCs, Novell server, UNIX box X X Software: X * UNIX X - of course X - some particular features of this UNIX: X * support Sys V lp/lpsched - for ISPIN X * support TCP/IP - for LAN communications X * support BSD lpr/lpd X * In our case, it's a Sequent, running a X dual universe ucb/att environment. Pyramid X does this too. Others? X * see "OTHER WAYS..." X X * ISPIN - to run on the UNIX box X X * packet driver software - This is a software X frontend for the ethernet card on the MS-DOS X PC. Our Cabletron cards came with a very effective X packet driver. If yours didn't, the Clarkson X drivers would fit here. The packet driver allows X the PC to talk to the UNIX box via TCP/IP while X also allowing the PC to talk IPX/SPX to Novell. X X * lpr client software for the PC - We're using X an application called "LPR, LPQ, LPRM Network X Printing Utilities (for MS-DOS Computers)" by X Paul Hilchey, University of British Colombia. We X got it as part of Clarkson University's CUTCP X suite. X XHOW WE DID IT X XISPIN on the Sequent has been cranking for years. The PC/Novell/Ethernet thing Xwas set up recently. Then we: X X 1. Hooked up the Sequent to the Ethernet and did the X necessary BSD networking configuration stuff. X X 2. Cranked up the Sequent's lpd subsystem. We had X previously used only the Sys V side's lp/lpsched. X X 3. Configured lpr queue members which pass their work X to corresponding Sys V lp queue members. See the file X "lpr~lpd_ISPIN", a document by John Withers (IRS, X Springfield District Office) which describes this X procedure. X X 4. Installed the client LPR software on the PC(s). X Some config work here - not bad - good documentation. X We'll probably put the executable on the Novell X server so the diskless people can use it too. X X 5. Started printing. The PC users who are on the LAN X can print anything which resides on their own box X or on the Novell to any networked printer on the X X.25. X X LAN users, LAN files, WAN printing. X XHOW IT WORKS X X 1. The BSD lpd is a network server daemon and a X local print request server daemon. On the network X side, it's waiting for print requests and data X to come in over the LAN. X X 2. The lpr client program on the PC establishes a X TCP/IP connection with the UNIX's lpd, then sends X the request and the data. X X 3. John's lpr/lpd-to-lp/lpsched interface hands off X the job to the Sys V scheduler, which ultimately X calls ISPIN. X X XOTHER WAYS TO SKIN THE CAT X X There are just a few components of this scenario which are absolutely X required to make it work. X X REQUIREMENTS: X 1. UNIX X 2. Sys V lp/lpsched subsystem X - ISPIN doesn't do BSD lpr/lpd - yet X 3. ISPIN X X ALTERNATIVES: X 1. As previously mentioned, if you don't have X.25 X apads for cpu to printer virtual connections, X smart switches or stat muxes will work. ISPIN X is so flexible you could even do it with dialup X modems (feasible, but not recommended). X X 2. The only reason we have the BSD lpr/lpd subsystem X involved is because it supports the networked X client/server model and the software is available. X X If you don't have a dual universe UNIX (ucb/att), X finding both lp/lpsched and lpr/lpd on the same X system would be rare. X X A network client/server pair could be cooked-up X to support Sys V lp/lpsched directly, thus avoiding X the extra step of passing through BSD lpr/lpd. X I haven't seen a UNIX System V Release 4 box yet, X but I have heard that this is a supported feature. X XMore later, XLSB END_OF_FILE if test 7134 -ne `wc -c <'ISPIN/misc/PC-LANtoISPIN/README'`; then echo shar: \"'ISPIN/misc/PC-LANtoISPIN/README'\" unpacked with wrong size! fi # end of 'ISPIN/misc/PC-LANtoISPIN/README' fi if test -f 'ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN'\" else echo shar: Extracting \"'ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN'\" $4195 characters$ sed "s/^X//" >'ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN' <<'END_OF_FILE' XTHIS PROCEDURE WAS DEVELOPED & DOCUMENTED BY JOHN WITHERS, SYSTEMS ADMINISTRATOR XAT INTERNAL REVENUE SERVICE, SPRINGFIELD DISTRICT OFFICE X XDate: Thu Jan 10 08:51 CST 1991 XFrom: pharaoh!withers (withers) XMessage-Id: <101240010132614EMU0000@pharaoh> XTo: lbartz@ispin XSubject: Revised 'How To' document! X XThis reflects my 'reflections' and our discussion: X XUsing ISPIN in the UCB universe (Pyramid, Sequent, et. al) X(Requires no modification to ISPIN) X XThis 'hack' evolved from our use of the UCB line printer spooler (lpr/lpd) Xon our Pyramid MIServer4/1 instead of the ATT line printer spooler X(lp/lpsched). We had implemented lpr/lpd in all of our applications and Xdidn't really want to change just to run ISPIN. X XGeneral procedure. X XIn the ATT universe: X X1) Compile and install ISPIN. X2) Configure the ATT line printer spooler for network printers as X described in the ISPIN documentation. TEST ISPIN directly from X the ATT universe to make sure your rtab entries are good. (This X one bit me BIG-TIME!) X X Just one word of warning. NEVER EVER define the same X serial port as output for both lpsched and lpd. Why? X LPSCHED opens his output devices and holds them open X (non-exclusive) set as it thinks appropriate. LPD X opens the device each time it wants to print to it and X assumes that it is in some sort of 'virgin' state before X it sets it up. The net result is that LPD output to the X shared device is garbled and un readable. I did this X when I was testing my basic configuration of 'lpsched' X when ISPIN didn't work the way I expected and it took X a lot of head scratching to figure out what happened. X XIn the UCB universe: X X1) In a convenient directory (I used /usr/local/lib/filters) create X a shell script like the following (I called it FILTER): X X#!/bin/att /bin/sh X# X# This shell script acts as a 'universe bridge' to collect information X# from the UCB/LPR print spooler and pass it the ATT/LP print spooler X# so it can be processed by ISPIN (Thanks Larry) X# X XQUE=`basename $0` XNOW=`date "+%T %D"` Xecho "$NOW - Printer $QUE requested" >> /usr/local/lib/filters/filter.log Xexec lp -d$QUE - X X2) Link this file (FILTER) to a file with the same name as the ISPIN X controlled Queue under control of lpsched. (IMPORTANT!) X X For example, if you created a queue called 'spr_iss', use the X command 'ln FILTER spr_iss' to create the filter for the lpd. X X3) Edit /etc/printcap to define the new print spool for lpd. X This is how I did it for the que 'spr_iss' mentioned above: X X# X# Definition of remote printers via the 'bridge' filter to ISPIN X# Xspr_iss|ISS printer #1 (Shared Centronics):\ X :tr=\f:lf=/usr/adm/lpd-errs:pl=66:pw=255:\ X :of=/usr/local/lib/filters/spr_iss:\ X :sd=/usr/spool/lpd/remote:\ X :lo=remoteLOCK:\ X :lp=/dev/null: X X Notes: a) I used the same sd (spool directory) for all of the ISPIN queues. X This simplifies life a lot, otherwise you would have to have X a directory for each queue on both the UCB and ATT sides. X Besides, as far as I can tell, files 'cross the bridge' very X rapidly and the chances of them stacking up on the UCB side X is quite slim, if everything is working. (RECOMENDATION) X/* editorial comment: X I disagree with John on this. I believe each queue X member should have its own spool directory. This X would prevent confusion of sequence numbers, data X files, and control files. X X LSB X*/ X b) I used the same lo (lock file) for all the ISPIN queues. X This controls the number of files which can cross at the same X time. It should be fast enough, one a time. (RECOMENDATION) X c) Set of (output filter) to the full path of the script you created X in step two (2) above. (REQUIRED) X d) The lp (line printer) entry should always point to /dev/null. X For a more complete explanation, see the ISPIN documentation, X the reasoning is the same. (REQUIRED) X e) I kept the queue names the same on both sides of the bridge, but X you don't need to. The only requirement is that the output X filter name be the same as the ATT queue name. (RECOMENDATION) X END_OF_FILE if test 4195 -ne `wc -c <'ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN'`; then echo shar: \"'ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN'\" unpacked with wrong size! fi # end of 'ISPIN/misc/PC-LANtoISPIN/lpr~lpd_ISPIN' fi echo shar: End of archive 15 $of 15$. cp /dev/null ark15isdone MISSING="" for I in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ; do if test ! -f ark${I}isdone ; then MISSING="${MISSING} ${I}" fi done if test "${MISSING}" = "" ; then echo You have unpacked all 15 archives. rm -f ark[1-9]isdone ark[1-9][0-9]isdone else echo You still need to unpack the following archives: echo " " ${MISSING} fi ## End of shell archive. exit 0