Sun-Spots Digest, v6n256

William LeFebvre Sun-Spots-Request at Rice.edu
Thu Oct 13 14:36:47 AEST 1988


SUN-SPOTS DIGEST        Wednesday, 12 October 1988    Volume 6 : Issue 256

Today's Topics:
                        RE: videoing a sun screen?
                           Re: .cshrc vs .login
                        SunOS 4.0 Suntools speedup
                Multiple cron jobs on Sun4 with SunOS 4.0
           Problems with SunOS 4.0 and BSD/SYSV group semantics
                Laserwriter hanging with framedevice error
            Help requested on understanding mysterious crash.
                      how 'bout that permisc. mode ?
                    DEC LANBRIDGE 100/Sun 4 problems?
              power supply specs in 19" monochrome monitor?
      printing on laserwriter connected to CMC terminal server box?
            2 questions: scsi tape drives and 3270 connections

Send contributions to:  sun-spots at rice.edu
Send subscription add/delete requests to:  sun-spots-request at rice.edu
Bitnet readers can subscribe directly with the CMS command:
    TELL LISTSERV AT RICE SUBSCRIBE SUNSPOTS My Full Name
Recent backissues are available via anonymous FTP from "titan.rice.edu".
For volume X, issue Y, "get sun-spots/vXnY".  They are also accessible
through the archive server:  mail the request "send sun-spots vXnY" to
"archive-server at rice.edu" or mail the word "help" to the same address
for more information.

----------------------------------------------------------------------

Date:    Thu, 6 Oct 88 17:35:24 PDT
From:    gene%filbert at sun.com (Gene Banman)
Subject: RE: videoing a sun screen?

>>I hope you don't have to get a genlock video camera! 

I don't know about a genlock video camera. Here are the two ways the Sun
marketing people do it.  There may be other approaches but this gives you
production quality output.

Good luck,
Gene Banman

Specifications For FILMING Sun Monitors:
_________________________________________

In order to eliminate phase interference when filming Sun monitors, a
motion picture camera must be used.  The 'phase bar' CANNOT be eliminated
using videotape as the origination medium.  The film can be transferred to
video.

Sun's monochrome monitor runs at a difference frequency than the color and
greyscale monitors.  When filming either of these monitors, the motion
picture camera must be controlled using an external shutter speed
controller called a 'precision speed controller'.  This device allows
control of the shutter speed down to .001/second.

    *When filming the Sun monochrome monitors use a
     shutter speed of 33.387 frames per second.

    *When filming the Sun color and grey scale monitors,
     use a shutter speed of 32.976 frames per second.

The shutter angle and mirror must be at 180 degrees.  These results were
obtained using both the Arriflex BL 35mm camera, and an Arriflex 16SR 16mm
camera.  These can be rented in any major city.

How to go direct to VIDEO TAPE:
________________________________

Another option is to use a scan converter which converts high resolution
graphics to NTSC video format. Images are converted in real time for
recording to a video tape recorder. These units enable you to save the 
ENTIRE screen to video. (You cannot have a person in the shot).
   Monarch CGC Color Graphics Converter, manufactured by Folsom
      Research Inc., 916-985-2481. Contact is Mike Morgan
   Products from RGB Technology, 415-284-4330. Contact is 
      Bob Marcus.

Cost of these units is approximately $15,000.  Some video production
facilities in high-tech areas have rental units.

------------------------------

Date:    Thu, 6 Oct 88 22:54:10 CDT
From:    Oliver Schoett <schoett at lan.informatik.tu-muenchen.dbp.de>
Subject: Re: .cshrc vs .login

Two previous replies to the question of what to put where have suggested
putting the setenv commands into .login.  However, .login isn't read by
shells started by rsh, and so all setenv commands needed by programs that
you might wish to execute via rsh need to go into .cshrc.

Oliver Schoett	schoett at lan.informatik.tu-muenchen.dbp.de
		schoett%lan.informatik.tu-muenchen.dbp.de at relay.cs.net

------------------------------

Date:    Thu, 6 Oct 88 19:00:17 edt
From:    dg-rtp!meissner at mcnc.org (Michael Meissner)
Subject: SunOS 4.0 Suntools speedup

This is relayed from our Sunyvale people who helped Sun find it, but for
some reason, Sun isn't publishing it.  The core of the problem is a file
needed for error logging doesn't exist, and some program (I'm not clear
which one), loops if it gets an error opening the error daemon, and
therefor presents a drain on the system.  The above explaination is third
hand (Sun -> Sunyvale DG -> North Carolina DG).  Anyway, the following is
the patch:

Edit /etc/syslog.conf and enter the following line before the first
non-comment line, starting in column 1 (note the comma space):
define(LOGFILE, 1)
then do a:
touch /var/log/syslog
and reboot.

Michael Meissner, Data General Corporation
meissner at dg-rtp.dg.com          (internet) 
...!mcnc!rti!xyzzy!meissner     (uucp)

------------------------------

Date:    Fri, 07 Oct 88 08:55:56 EST
From:    trinkle at purdue.edu
Subject: Multiple cron jobs on Sun4 with SunOS 4.0

Since converting some of our Sun4's to SunOS 4.0, I have noticed cron jobs
getting executed multiple times.  We have a daily admin script (similar to
4.3 BSD) that is run by cron overnight.  After a couple days I start
getting two reports.  It is not a matter of duplicate mail messages.  I am
definitely getting two executions of the script.  If I run crontab -e on
the system (make some minor change to the file), the problem will clear up
for a few days.  The clocks are within a few milliseconds of each other,
so I don't think it is a tie problems.

This has NEVER happened with SunOS 3.4 on Sun3s.  I am sure it has not
happened on any of our Sun3s running SunOS 4.0.

Has anyone else witnessed this behavior?  Do you know of anything else to
check?

Daniel Trinkle			trinkle at cs.purdue.edu			ARPA
Department of Computer Sciences	trinkle%purdue.edu at relay.cs.net		CSNET
Purdue University		{ucbvax,decvax}!purdue!trinkle		UUCP
West Lafayette, IN 47907	(317) 494-7844				PHONE

------------------------------

Date:    Thu, 06 Oct 88 13:52:10 +1000
From:    Craig Bishop <munnari!lupus.cc.deakin.oz.au!craig at uunet.uu.net>
Subject: Problems with SunOS 4.0 and BSD/SYSV group semantics

When we installed SunOS 4.0 one of the problems we had to grapple with is
Sun's new concept of how to allow BSD and SYSV group semantics to coexist
on the filesystems.

Anyway if you don't know, to get BSD group semantics a directory must have
the setgid bit set. (Which I also believe is arse-about (back-to-front).
SunOS is still basically a BSD system the default ie. no setgid bit set,
should follow BSD semantics not SYSV.)

Our experience with SunOS BSD/SYSV group semantics follows,

    1.	We wanted BSD semantics so ensured that every directory
	had the setgid bit set. This was a problem because if you
	use something like rdist and all your suns are not
	running SunOS 4.0 or God forbid you are running other
	hardware. The setgid bits are propagated to the other
	machines. YUCK.

    2.	Upon reading the manuals for mount the option "grpid" was
	found, which enforces BSD semantics on the filesystem without
	reference to the dirtectory setgid bit. We rebooted all
	our Suns with the grpid option set (Yeah, no problems
	with rdist.) However. Now this is the problem we have.

The root filesystem ignores the "grpid" option to mount. It doesn't matter
whether the root filesystem is on a local disk or mounted across the
network via NFS, Sun3 or Sun4, it all screws up the same. 

The root filesystem with the "grpid" option set reverts to the setgid bit
rule for BSD/SYSV group semantics. If you turn on the setgid bit on a
directory in the root filesystem it will follow BSD semantics otherwise it
will exhibit SYSV semantics by default (ARGHHH!!).

Say a NFS client creates a file in it's root partition, by default it will
have group ownership equal to the group field in the passwd file (assuming
no newgrp). If on the server you change to the /exports/roots directory
which corresponds to that particular client and create a file, it follows
BSD semantics because it is not the root filesystem of the server.

The above is not a serious problem but annoying all the same, has anyone
seen this, fixed this. I have reported this to Sun but to this time I have
not had a solution. They have check out my fstab file and believe them to
be correct.

Craig Bishop    ARPA:   craig%lupus.cc.deakin.oz.au at uunet.uu.net
                UUCP:   ...!uunet!munnari!lupus.cc.deakin.oz!craig

------------------------------

Date:    Thu, 6 Oct 88 15:20:25 PDT
From:    Jonathan Eisenhamer <jon at mira.astro.ucla.edu>
Subject: Laserwriter hanging with framedevice error

To All Spots,

We are having a problem with our Apple Laserwriter.  The sun is a 3/260
server running 3.5 and Transcipt 2.0.  The Apple Laserwriter says its at
version 38 (discovered by interactive session with it).  The problem is
thus:  For a number of reasons (discussed below), the Laserwriter gets
into a mode where no file is accepted for printing and it continually
returns one or more off the following errors:

	psif: accounting error 1 (%%[ Error: limitcheck; OffendingCommand: 
		framedevice \]%%

or just

	%%[ Error: limitcheck; OffendingCommand: framedevice ]%%

In trying to talk to the laserwriter interactively when it is in this
mode, we found that it will return the limitcheck error on the first
character sent, then ignore all input until a Control-D is received
(flushing the rest of the job), and then on the very next character return
the limitcheck error again.  The only way to remedy the problem is to
cycle power on the laserwriter.  We have tried two laserwriters, both at
version 38, and both do the samething.

Most often, this occurs when people are trying out postscript files and
they make mistakes.  It seems connected with errors involving the image
and matric commands, but it does occur after other errors (like syntax
errors).  It usually doesn't occure immediately after the error though; a
couple of jobs will go through and then it will enter this limitcheck
error mode.  It also occurs spontaneously; i.e. several days of error-free
printing (according to the printer log), then suddenly it goes into the
limitcheck error mode on a one page text file.

This is apparently a bug, but whether its in the transcript or the
laserwriter itself is not clear.  If anyone has info on this or could
suggest where to get more info, I would be grateful.

	Thank you for your time,
	Jonathan Eisenhamer
	mira.astro.ucla.edu (128.97.64.198)
	jon at uclastro.bitnet
	bonnie::jon (span 5.708)

Summaries will be posted if I get any responses.

------------------------------

Date:    7 Oct 88 06:55:43 GMT
From:    madhu at hubcap.clemson.edu
Subject: Help requested on understanding mysterious crash.

Greetings:

We have a SUN4/260 acting as a gateway between local SUN4/110's and the
campus Ethernet. Recently our gateway which is also the NFS server for the
local SUN's went down mysteriously.  I am totally clueless as to the real
reason, looks like ie1 our ethernet interface to the campus had something
to do with it...  but what could have happened...? Has anyone else seen a
similar type of crash ? ... we have an ALM board and are running 4.0 ..
Please e-mail to madhu at hubcap.clemson.edu. Transcript of pre and post
-crash session follows: 
Note: "%" lines are comments which have been insered.
%%%%%%%%%%%%%%%%%%%%%%%Pre- Post crash %%%%%%%%%%%%%%%%%%%%%%%
% Pre-crash messages
Oct  1 07:32:42 omni vmunix: ie1: spurious interrupt
Oct  1 07:32:42 omni vmunix: ie1: giant packet
Oct  1 07:32:42 omni vmunix: ie1: spurious interrupt
Oct  1 07:32:42 omni vmunix: ie1: spurious interrupt
Oct  1 13:27:39 omni vmunix: ie1: spurious interrupt
Oct  1 13:27:39 omni vmunix: ie1: spurious interrupt
% CRASH 
> b
% post-crash messages
......blah blah....
% we have two ethernet boards ie0, ie1 and the alm board ..
% 
Oct  1 14:11:48 omni vmunix: ie0 at obio 0xf6000000 pri 6 
Oct  1 14:11:48 omni vmunix: ie1 at vme24d16 0xe88000 vec 0x75 
Oct  1 14:11:48 omni vmunix: mcp0 at vme32d32 0x1000000 vec 0x8b 
...blah blah ....
% Offending message during boot up....
Oct  1 14:11:48 omni vmunix: spurious VME interrupt at processor level 5
Oct  1 14:11:48 omni vmunix: VME level 3, VME vector 0xffffffff
...blah blah...
Omni login:
-------       
Thanks in advance for solving the mystery.

10/9/1988 update: this mystery message has now appeared a number of times.
ANY and ALL mail related to this welcome. HELP!!

	-relatively CLUless

Internet address: madhu at hubcap.clemson.edu

------------------------------

Date:    7 Oct 88 05:12:40 GMT
From:    norm at shorty.cs.wisc.edu (Norman Jacobs)
Subject: how 'bout that permisc. mode ?

I need some help.  I need to grab raw packets off ethernet using a Sun
3/110 running 4.0.  I have read all of the man pages on nit and anything
that I could think of to help me.  Does anyone out there have some code
available for putting the nit device in Permisc. mode, and grabbing
packets?  Any help would be greatly appreciated.

	Thanx in advance,
	Norman Jacobs

norm at cs.wisc.edu

------------------------------

Date:    6 Oct 88 21:50:08 GMT
From:    tgarons at ucdavis.edu (Tom &)
Subject: DEC LANBRIDGE 100/Sun 4 problems?

We have a multi-segment ethernet connected by supposedly transparent
filtering bridges.  Recently we were unable to reach the lone node, a
Sun4/260  on the far side of a DEC LANBRIDGE 100.  Upon investigation we
found that the problem was due to the DEC bridge filtering out broadcast
packets, thus preventing arp requests from reaching the Sun4.  If the
Sun4's physical ethernet address was entered by hand there was no problem.
Monitoring the Sun4 side of the bridge revealed a bunch of packets that
the network analyser called "Lanbridge Management Packets".  Toggling
power to the Lanbridge fixed things.

This all sounds vaguely familar to me.  I seem to recall a discussion some
time back about Sun4's sending out broadcast packets "too quickly" on
bootup.  Has anyone else experience similar problems?  Does anyone have a
fix other than toggling power?
--
Tom Arons			Internet: arons at iris.ucdavis.edu
EE/CS Department		Bitnet:   arons at ucdavis
University of California	UUCP:     {lll-crg, ucbvax}!ucdavis!iris!arons
Davis, CA 95616			Phone:    (916) 752-1750

------------------------------

Date:    Thu, 6 Oct 88 16:32:21 EDT
From:    lai at vedge.UUCP (David Lai)
Subject: power supply specs in 19" monochrome monitor?

Hi, our sun3/50 monitor blew a power supply, and sun is unable to replace
it (they can replace the entire monitor though, at significant cost!).  So
the question is, does anyone know the voltage and current rating on the
supply (then I can get a replacement).  The monitor is made by Philips,
suns part number is 540-1062, and philips part number is M19P114A/6102.
The power supply itself is marked 94V-0, if that helps.

I'd measure it myself, if we had another one like it.  Any help will be
appreciated.  (Does anyone have a service manual for the thing?)

David Lai  Visual Edge Software, 3870 Cote Vertu, St. Laurent, PQ CANADA
David Lai (vedge!lai at larry.mcrcim.mcgill.edu || ...watmath!onfcanim!vedge!lai)

------------------------------

Date:    6 October 1988 1528-PDT (Thursday)
From:    gordon at nprdc.arpa (Jeff Gordon)
Subject: printing on laserwriter connected to CMC terminal server box?

I have  a Sun-3/50 with both ttya and ttyb already in use. In order to get
more ports for dumb terminals I bought a 10 port Ethernet terminal server
box from (CMC) Communications Machinery Corp.  The CMC box has its own
network address.

I now want to connect a Sun laserwriter to one of the CMC ports and have
the Sun users be able to "lpr" to this printer. We use Sun's transcript
software.  Has anyone done this or have some software that will do this?

Thanks
  Jeff Gordon                        gordon at nprdc.arpa
  Navy Personnel R & D Center        !ucsd!nprdc!gordon
  San Diego, Calif. 92152

------------------------------

Date:    6 Oct 88 01:46:11 GMT
From:    heather at cs.ucla.edu
Subject: 2 questions: scsi tape drives and 3270 connections

I have two questions:

1) I am looking for a 9 track reel tape drive from my Sun 3/50 (which only
has a SCSI port).  I know of two drives--one from Cipher, with an add-on
SCSI converter and one from Kennedy with built-in SCSI capability.  Both
salespeople have claimed they will work with the Sun and have appropriate
drivers, etc.  Does anyone have experience with either of these drives on
a 3/50?

2) Someone else here would like to connect a serial port of their 3/50
directly to a coaxial cable attached to our IBM 3090 mainframe.  They
would like to access the mainframe using a 3270 interface.  I am familiar
with tn3270 but am not familiar with any products that run over a coax.
Can any of you fill me in?

Please reply for either question directly to me; I will summarize to the
newsgroup.

Thanks very much,

Heather Burris, UCLA

------------------------------

End of SUN-Spots Digest
***********************



More information about the Comp.sys.sun mailing list