Sun-Spots-Request@RICE.EDU.UUCP (04/17/87)
SUN-SPOTS DIGEST Thursday, 16 April 1987 Volume 5 : Issue 8
Today's Topics:
Re: Screen Saver for Sun-3's (1)
Re: Screen Saver for Sun-3's (2) (long)
Re: Screen Saver for Sun-3's (3)
Super Eagles on SUN's
Re: mouse problems on SUN-3 (1)
Re: mouse problems on SUN-3 (2)
Re: mouse problems on SUN-3 (3)
Sun 3/50 hardware problems and Franz Lisp
Summary of Best Lispm/WorkStation Responses
Re: Video cable extension on Sun-3/160
Used Sun 2/50's
Re: yp slave servers
FLAME SUN 3.3UPGRADE
Taming SUN multibus memory (4 meg on a 150u)
Bug in yppasswdd
SUN Common Lisp question?
Terminal Emulators for Sun?
B/W addon monitor - second source?
Pascal runtime structures?
------------------------------------------------------------------------
Date: Fri, 20 Mar 87 20:43:26 EST
From: sundar@hermes.ai.mit.edu (Sundar Narasimhan)
Subject: Re: Screen Saver for Sun-3's (1)
This is with respect to your message regarding screen saving:
(we have had something like this working on our machines for some time
now.. but I am not sure I can post it).
It is easy to add an entry to your gettytab describing a normal
terminal (make a copy of the entry for '2' and call it 'h', say).
Then edit /etc/ttys for the console to be this type. ie.
the first line looks like "1hconsole".
Then hack /usr/src/etc/getty/main.c to blank the video (look at
/dev/fb and the appropriate ioctls), if the termtype is 'h' and to
turn the video back on again when someone types a character on the
keyboard or moves the mouse.
Now you have a fancy screen saver, that kicks in as soon as the user
logs out.
Of course you could change/add-endless-frobs-to that routine that blanks the
video and sits around waiting for a keystroke!
-Sundar
------------------------------
Date: 21 Mar 87 23:58:11 GMT
From: swatsun!babylon!garth@seismo.css.gov (Garth Snyder)
Subject: Re: Screen Saver for Sun-3's (2) (long)
> Date: Wed, 25 Feb 87 12:45:13 CST
> From: Charles Hampton Curley <wucs!chc@seismo.CSS.GOV>
> Subject: Screen Saver for Sun-3's?
>
> Is there a screen saver for sun3 consoles? I know there
> is screen saver that can be cranked up from the root menu in suntools
> but is there one that will blank the console after someone has logged
> out (maybe some other process could grab the console and release it
> if a key is typed). Currently, we turn the console off when ever
> we leave for an extended period of time, but some of the users seem
> to have a rough time remembering to do this.
>
> Thanks,
> Charles
I am enclosing a shar file of a screen saver program that we have used
with great success on our local Sun 3 network. The program should be
started in one of the /etc/rc files and will hang around until the machine
crashes or is stopped.
The program reads the /etc/utmp file every minute, checking to see if there
is someone logged in on the console. If there is not, it starts up the
lockscreen_default program and stats the /dev/console every second until
it detects activity (return), when it will white out the screen and print
the login prompt. Both of these frequencies can be easily modified in
the source code.
This methodology may seem a little kludgy, but it works great. It only takes
about 5 minutes of cpu per day, and that is with a large portion of time
spent in the stat phase. Reading /etc/utmp is a safe way to check for logins
on the console even though Sun 3's have problems maintaining this file
correctly because the problems only occur on pseudo-terminals.
This program is in the public domain. I would like to hear from anyone who
finds/fixes bugs in it or who has written a more sophisticated program for
dealing with this problem.
Good luck!
Garth Snyder
Swarthmore College
Swarthmore, PA 19081
...seismo!bpa!swatsun!garth
----------- Cut Here ------------
#! /bin/sh
# This is a shell archive, meaning:
# 1. Remove everything above the #! /bin/sh line.
# 2. Save the resulting text in a file.
# 3. Execute the file with /bin/sh (not csh) to create the files:
# wiped.c
# This archive created: Sat Mar 21 18:57:09 1987
export PATH; PATH=/bin:$PATH
if test -f 'wiped.c'
then
echo shar: will not over-write existing file "'wiped.c'"
else
cat << \SHAR_EOF > 'wiped.c'
/*
** File: wiped.c
**
** A program to blank the console screen if no one is logged in.
**
** This program reads /etc/utmp every minute to see if there is someone
** logged in on the console. If there is not, it blacks out the screen
** and stats /dev/console every second until it detects activity, then
** whites in the screen and goes back to reading /etc/utmp.
**
** This source file is formatted for four-space tabs.
**
** Garth Snyder initial coding
** Scott E. Schwartz added sleep() before main loop, so rebooting is nicer
** Scott E. Schwartz substituted pixrect ops for mmap.
**
** cc -O -o wiped wiped.c -lpixrect
*/
#include <utmp.h>
#include <sgtty.h>
#include <stdio.h>
#include <sys/file.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <sys/wait.h>
#include <pixrect/pixrect_hs.h>
#define BLACK 0xff
#define WHITE 0x00
#define UTMP_FREQUENCY 60 /* How often to check /etc/utmp */
#define STAT_FREQUENCY 1 /* How often to stat /dev/console */
struct pixrect *screen;
char *index();
main() /* No command line arguments used */
{
int fd; /* File desc for ctl terminal */
if(fork()) /* Run in background */
{
exit(0);
}
/* Close the control terminal */
if((fd = open("/dev/tty",O_RDWR,0777)) != -1)
{
if(ioctl(fd,TIOCNOTTY,0) != -1)
{
close(fd);
}
}
/* open frame buffer in pixrect */
screen = pr_open("/dev/fb");
/* sleep for a while before getting started */
/* necessary if this is to be run from rc.local */
sleep(60*5); /* 5 minutes should be ok */
/* Loop infinitely */
while(1)
{
wait_idle();
wipe_screen(BLACK);
wait_login();
wipe_screen(WHITE);
write_cons();
sleep(100); /* Allow 100 secs for login */
}
}
/*
** Wait until the console is not in use.
*/
wait_idle()
{
struct utmp person; /* An entry in /etc/utmp */
int fd; /* Fd for /etc/utmp */
u_char busyflag; /* Is console busy? */
while(1)
{
if ((fd = open("/etc/utmp",O_RDONLY,0777)) == -1)
{
died("could not open /etc/utmp");
}
busyflag = 0;
while(read(fd,&person,sizeof(person)) == sizeof(person))
{
if (nonuser(person) || (!strlen(person.ut_name)) ||
(strncmp(person.ut_line,"console",7)))
{
continue;
}
else
{
busyflag = 1;
break;
}
}
close(fd);
/* If not busy, escape, otherwise keep watching */
if (!busyflag)
{
return(1);
}
else
{
sleep(UTMP_FREQUENCY);
}
}
}
/*
** Wipe the screen.
*/
wipe_screen(pattern)
u_char pattern;
{
switch (pattern)
{
case BLACK:
pr_rop(screen, 0, 0, screen->pr_size.x, screen->pr_size.y,
PIX_SET, screen, 0, 0);
break;
case WHITE:
pr_rop(screen, 0, 0, screen->pr_size.x, screen->pr_size.y,
PIX_CLR, screen, 0, 0);
break;
default:
break;
}
}
/*
** Wait for someone to hit return on the console
*/
wait_login()
{
struct stat mortician;
long lastmod;
int pid;
union wait junk;
if ((pid = fork()) == 0)
{
execl("/usr/bin/lockscreen_default","lockscreen_default",0);
}
stat("/dev/console",&mortician); /* Get initial modtime */
lastmod = mortician.st_atime;
while(1)
{
sleep(STAT_FREQUENCY);
stat("/dev/console",&mortician);
if (mortician.st_atime != lastmod)
{
kill(pid,9);
wait(&junk);
break;
}
}
}
/*
** Print a diagnostic and die.
*/
died(message)
char *message;
{
int fd;
char buffer[100];
fd = open("/dev/console",O_WRONLY,0777);
sprintf(buffer,"Wipe daemon: %s\n",message);
write(fd,buffer,strlen(buffer));
close(fd);
exit(1);
}
/*
** Write the login message on the console
*/
write_cons()
{
char buffer[100];
char *temp;
int fd;
buffer[0] = '\015';
gethostname(&buffer[1],sizeof(buffer) - 1);
temp = index(buffer,'\0');
sprintf(temp," login: ");
fd = open("/dev/console",O_WRONLY,0777);
write(fd,buffer,strlen(buffer));
close(fd);
/* Be sure we don't pick up the console as the control terminal */
fd = open("/dev/tty",O_RDWR,0777);
if(ioctl(fd,TIOCNOTTY,0) != -1)
{
close(fd);
}
}
SHAR_EOF
fi # end of overwriting check
# End of shell archive
exit 0
------------------------------
Date: Sun, 29 Mar 87 12:36:47 EST
From: hope%gatech.gatech.edu@relay.cs.net (Theodore Hope)
Subject: Re: Screen Saver for Sun-3's (3)
We set up special "lock" user like this:
lock::998:10:LockScreen:/whatever/lock:/whatever/lock/myshell
Its login shell looks sort of like this:
stty -echo -tostop intr '^c'
exec /usr/bin/lockscreen_default
To "save the screen," log in as "lock"; a control-C will
kill lockscreen_default and you're back to getty.
---
Theodore Hope
School of Information & Computer Science, Georgia Tech, Atlanta GA 30332
CSNet: Hope @ gatech ARPA: Hope@Gatech.EDU
uucp: ...!{akgua,decvax,hplabs,ihnp4,linus,seismo,ulysses}!gatech!hope
------------------------------
Date: Sun, 22 Mar 87 16:05:13 EST
From: mo@seismo.css.gov (Mike O'Dell)
Subject: Super Eagles on SUN's
If SUN has had only 5 failures, then there must only be three
others besides me, because I have had 2 fail. I have a 280 with
2 Supers. One failed about 3 weeks after it arrived and was
replaced. The replacement failed this last week. When I contacted
Fujitsu, they admitted there were engineering problems with the drive
and were very up front about it. They acknowledged that their biggest
headache is the incredible history of the 474's legendary reliability.
The also said they had solved the engineering problems, they believed.
-Mike
------------------------------
Date: Mon, 23 Mar 87 07:36:20 PST
From: jvh%janus.berkeley.edu
Subject: Re: mouse problems on SUN-3 (1)
We have often seen the mouse problem where the mouse becomes unresponsive.
Buttons become ineffective when the mouse is held dead still, but still work
if the mouse is moved when the buttons are pushed. This makes the buttons
appear to have become erratic. The mouse movement is impaired, especially
when crossing windows, it seems to be "repelled" by them. We, too, have found
that either rebooting or unplugging and replugging the mouse fixes the problem.
Theories are that both these actions reset the microchips in the mouse.
We first saw this problem after 3.0 was installed, and we have seen it on
Sun 2/50s & 2/120s as well as the Sun 3 line. Don't know about damage, but
none of the mice has later died. Wish we knew what was going on.
------------------------------
Date: Sat, 21 Mar 87 17:59:43 EST
From: cmcl2!phri!roy@seismo.css.gov (Roy Smith)
Subject: Re: mouse problems on SUN-3 (2)
> Movement of the mouse is not tracked accurately (almost as if the
> movement was against some type of "resistance")
We've seen the same thing (we call it Drunken Mouse Syndrome). I
have no idea why it happens, but when it does we just reboot and don't
worry about it since it doesn't happen very often. One 3/50 had this on a
regular basis and we sent the mouse back for replacement, which seems to
have fixed the problem.
--
Roy Smith, {allegra,cmcl2,philabs}!phri!roy
System Administrator, Public Health Research Institute
455 First Avenue, New York, NY 10016
"you can't spell deoxyribonucleic without unix!"
------------------------------
Date: Wed, 25 Mar 87 23:11:24 PST
From: walker%spica.usc.edu@usc-oberon.arpa (Michael D. Walker)
Subject: Re: mouse problems on SUN-3 (3)
In reply to the person who complained about the 3/50 mouse getting
confused.
Yes, I have experienced the same problem. The mouse fails to track
properly and the buttons either don't work or take several seconds to
respond. I have had to unplug and replug the mouse to get it to
function properly again. I haven't had any problems with doing this
but it is annoying.
--
:-]
Arpa: walker%castor.usc.edu@usc-oberon.arpa
Uucp: ihnp4!sdcrdcf!usc-oberon!castor!walker
------------------------------
Date: Mon, 23 Mar 87 12:12:10 PST
From: franz!fray!cox@ucbarpa.berkeley.edu (Charles A. Cox)
Subject: Sun 3/50 hardware problems and Franz Lisp
I just finished talking to Alan Kostinsky from Sun Microsystems about the
problem many franz users have discovered running lisp on 3/50 clients.
The problem was that between one third to one half of the time, lisp will
get a memory error. This error is not entirely reproducible since doing
exactly the same thing a second time may or may not result in the same
memory reference error. If it doesn't get the error right away, things run
to completion. Also, running the exact same binary in exactly the same way
on a non sun3/50 (eg a server) never has the problem.
I found out that Sun has recently diagnosed this hardware problem with the
3/50. They didn't say when the fix would be out, but I've been told this
problem is being given a high priority.
Charley Cox
------------------------------
Date: Mon, 23 Mar 87 15:05:58 PST
From: taylor%plu@ames-io.arpa (Will Taylor)
Subject: Summary of Best Lispm/WorkStation Responses
This is a summary of responses I received to a request for opinions
and experience on the best Lisp Machine (Lispm) or AI workstation.
I received techical summaries and internal reports from marketing
reps of the following companies:
Symbolics, TI and Integrated Solutions (AI workstations).
Except for the Symbolics vs. Explorer and Symbolics vs. Xerox comparisons
which appeared on the net last year, I received no extensive comparisons
of two or more Lispms/WorkStations.
I did get responses (positive & negative) from users with opinions (op)
and/or experience (ex) on a particular machine. First a short summary,
then some detailed comments.
machine configuration type positive negative
----------------------- ---- -------- --------
Apple Macintosh II op 1 0
Hewlett-Packard 350 workstation op 2 0
HP-UX, integrated Lisp/Prolog
Intel 3086 op 1 0
Golden Common Lisp
LMI - 0 0
Sun 3/160, 2/160 diskless ex 3 0
Sun 3/280 server, 16-20 MB memory
Symbolics - 0 0
Tektronix 4400 AI Work Station op 1 0
VAX AI Work Station ex 2 1/2 1 1/2
Xerox 1186 0 2
As you can see the response was not great.
Now some detailed comments:
-------------------------------------------------------
From: spar!malcolm@decwrl.dec.com (Malcolm Slaney)
Organization: Schlumberger Palo Alto Research
In article <8703010658.AA21849@ucbvax.Berkeley.EDU> you write:
>11. A SUN without disks is useless.
No, No, No, No, No.
But you must have enough memory so you don't swap. I have a Sun3/160 on
my desk with 16M of memory.....I NEVER PAGE or SWAP!!!!!
If you do start paging then things lose real fast. Franz Lisp image are
small and you can probably get by with less.
I think the reason that memory is so critical to the current generation
of Sun Lisp's is because of their swapping garbage collection. Every
few minutes it must touch every page of your dynamic area. If you have
to go to disk then chances are you will flush one of the pages you
are currently using (isn't least recently used wonderful???).
I have seen Lucid and Franz Common Lisp running anywhere between .5 and
4 times a Symbolics machine. Things are even faster with a Sun3/260.
It is safe to say that Sun Lisps have caught up with Symbolics machines
on speed.
Now if they just had the environment.... [for program development? - wmt]
P.S. I keep a Sun on my desk because that is my religion of choice...but
whenever I have a real hairy problem debugging my lisp I run to the
Symbolics machine.
--------------------------------------------------------
From: in%"beane%bartok.dec@decwrl.dec.com" 13-FEB-1987 07:06
I suppose somebody working at Digital can be expected to have a very
positive opinion of the AI VAXstation, but I do, even so.
I especially like the ability to run lots of completely independent processes,
especially VMS ones doing mail, file transfer, access to other resources in
the network without any LISP overhead.
The editor is especially good, compared to other editors on the VAX. I much
prefer it over EMACS, which I have stopped using. I've written several
editor extensions (eg., menus for common commands) which I'll be glad to
send you in hardcopy (to get the screen images).
Oh, yeah, I've never used any other machine, so no comparison, just
praise.
--------------------------------------------------
From: in%"@charon.mit.edu:meltsner@athena.mit.edu" 25-FEB-1987 19:09
We use Vaxstation II's + Vax LISP (Ultrix) here, and I'm fairly
happy. But --
1) Lucid's LISP is twice as fast on the same machine, given enough
memory.
2) Ultrix LISP's don't yet support the window system, although the
VMS ones do.
3) DEC memory and disks are notoriously over-priced. Consider buying
a minimal system in a BA123 box, and getting an Emulex disk ($3000
MIT price for a 140 meg drive+ controller, installed) and a third-part
memory board.
Personally, I like the Vaxstation. The machine feels fairly solid,
the software doesn't crash and everything install very easily. DEC
service is expensive, but ubiquitous. In general, the Symbolics stuff
is wonderful if you are willing to devote the guru time to keeping it
running. DEC has an acceptable product which seems much easier to support.
I have never managed a cluster of LISPM's, but I do manage our 5 machines
in not much more than 3-4hours/week (I don't do backups....).
--------------------------------------------------
From: cmh@purdue.edu (Christoph M Hoffman)
We used the 1180 and 1186 Xerox dandilions at Cornell for a year or so
in our research on solid modeling. We wern't too thrilled about them,
because of the poor floating point handling and because they were very
hard to learn. The trouble with floating point was software related:
The compiler boxed everything, so it made a lot of work for the garbage
collector. Also the fp precision turned out troublesome. The net effect
was that we couldn't work with surfaces of (algebraic) degree 4 or higher,
which excluded for example the torus. Here at Purdue we now use Symbolics
machines, and so does Cornell.
--------------------------------------------------
Well, after all that, what are we going to do?
We now have a configuration of a Symbolics file server, serving:
4 - Symbolics 3640's
2 - Symbolics 3620's
1 - Symbolics 3670
2 - LMI 2+2 (3 lisp, 1 unix)
1 - Xerox 1186
1 - TI Explorer
We also have a beta-test machine from Integrated Inference Machines
and a VAX 780/VMS with Franz, Interlisp, OPS-5, MRS, ITP, etc.
We are tentatively thinking of expanding our facility by adding:
3 - Xerox 1186's. Reasons : ease of learning, superior windowing,
Common Loops, NoteCards, Xerox PARC innovation,
inexpensive
1 - TI Explorers. Reasons : ease of learning, different yet similar to
Symbolics, integrated Lisp/Prolog, source code
2 - Sun 3/260 diskless and Sun 3/260 server. Reasons: fast, NeWS, numeric
speed, work station versatility, easy access to
other languages
It is worth noting that Symbolics, TI Explorer & Sun 3/260 are all in the
same price ballpark. Xerox is considerably less, however it is not designed
for the development of large systems.
Hope that this is of benefit - Will
P.S. Please send me your comments - thanks
--------------------------------------------------------------------------
Will Taylor - Sterling Software, MS 244-17,
AI Research & Applications Branch
NASA-Ames Research Center, Moffett Field, CA 94035
arpanet: taylor@ames-pluto.ARPA
usenet: ..!ames!plu.decnet!taylor
phone : (415)694-6525
------------------------------
Date: Tue, 24 Mar 87 02:50:55 PST
From: hoptoad.uucp!gnu@cgl.ucsf.edu (John Gilmore)
Subject: Re: Video cable extension on Sun-3/160
It would be nice if Sun sold a console extender kit, since it took me
two days plus about $200 of wire, connectors, etc to get mine all set.
Then again, it would have taken three months to get one from Sun, even
if it was on the price list! :-)
I just extended my color video, keyboard, and mouse to 70 feet. The video
had no trouble -- I just used four 1/4'' 75-ohm coaxial cables (Belden #8241)
and didn't use the thin Sun cables at all. It's messy, but the picture
is great. You can get spiral wrap stuff that will keep the 4 cables
together, but that just makes a THICK cable which is hard to bend. Less
mess, though. As I recall, the color boards were tested with 100'
cables when the board was first designed, so there's a good chance
(no guarantees of course) that it'll work OK to that distance at least.
Be sure you cut the cables to the exact same length (within an inch or
two), since an extra 8 inches or so of cable cause 1ns delay at the
speed of electricity in coax. There's only 10ns per pixel, so you could
end up with the red image visibly skewed to the right, or something...
If you encounter that, you might lengthen the short cables with little
barrel connectors or something, rather than tear up the long cable(s) and
remake the connector(s).
For the mouse and keyboard I had to construct a cable too. I used wire
with 4 pairs (8 wires - Belden #9504) and wired them up to pins 1-8 of
a DB-15 (one male, one female). This cable goes between the back of
the Sun and the connector on the keyboard curly cord. See the back of
the 160 hardware installation manual for the pinouts. You can't run
the 5V (VCC) through 70' of cable though, by the time it gets there it
is too low a voltage to run the kbd/mouse. So you need to put a little
transformer near your console, and feed power to the kbd/mouse
connector with it. I got a little "battery eliminator" type thing at
the local electronics store and set it for 4.5 volts and it works fine.
You can probably get by with a 4-wire cable (xmit to kbd, rcv from kbd,
rcv from mouse, ground) if you need to. Hook up both the transformer
ground (-) and the ground wire leading back to the Sun to the "GND"
pins on the connector that plugs into the keyboard, again check the
pinout to see which ones these are. I think I supplied VCC from the
transformer on pin 9 and ground on pin 8.
Make sure that your transformer doesn't swing to too high a voltage
when running your keyboard and mouse -- often they assume they are
driving a cassette player or something which draws more juice. If it
does, you can probably put a resistor in series with it or something --
I'm no EE about things like that. If the voltage is too high (say
above 6 volts) you might damage the keyboard/mouse. Mine runs at about
5.5 volts. 5.0 would be ideal.
It has been running for two whole days now and seems to be fine! :-)
John Gilmore
PS: This should work fine on a 2/160 too. You can probably do similar
things with the monochrome monitor, too. Then again, you could get
an extra 3/50 and run cheapernet instead...just one cable to deal with.
------------------------------
Date: Mon, 23 Mar 87 13:41:41 EST
From: det@hel-ace.arpa (Doug Tyrol)
Subject: Used Sun 2/50's
The U.S. Army Human Engineering Laboratory is interested in
adding to its current network of Sun 2/50's, 2/120's and 2/170's. We
would be interested in hearing from anyone willing to sell us 3-4
working Sun 2/50's at a reasonable price. You can write to me at the
arpanet address listed below or call (301) 278-5890 or autovon 298-5890.
thanks
Doug Tyrol
det@hel-ace.arpa
------------------------------
Date: Tue, 24 Mar 87 15:16:22 EST
From: mcgrew@topaz.rutgers.edu (Charles)
Subject: Re: yp slave servers
My suspicion is that your are not running ypserv on the slave server
(so that when the master goes down, there's no ypserv running for the
domain). If you're not, put in the ypserv line in your slave server's
rc.local again (with or without -i depending on what you need).
Charles
------------------------------
Date: 24 Mar 87 23:18:00 GMT
From: hi!cyrus@hc.dspo.gov (Tait Cyrus)
Subject: FLAME SUN 3.3UPGRADE
FLAME ON
This is about the second or third time that I have gotten a new
release of SunOs from SUN and have found the upgrade shellscripts
full of bugs (features :^) ). I am getting tired of having to run
SUN's shellscripts over and over, fixing bugs each time.
When I went from SUN 3 SunOs 3.2 to 3.3 I tried running their upgrade
script about 12 times (yes 12) before I got all of the bugs worked out.
FLAME OFF
This letter is for those of you who are about to upgrade your SUN's
to 3.3. Specifically those of you who have the following configuration:
a diskfull 3/(160,180) server, MINUS tape drive, running YP. The 3.3UPGRADE
has a feature (bug) that when it tries to find the name of the remote
tape drive host, using ypmatch, it fails BECAUSE /usr/etc/ypserv is
NEVER started. The script starts /etc/portmap and /etc/ypbind, BUT
NOT /usr/etc/ypserv. The diff is:
*** /etc/3.3UPGRADE.org Tue Mar 24 15:29:40 1987
--- /etc/3.3UPGRADE Tue Mar 24 15:29:46 1987
***************
*** 277,282 ****
--- 277,283 ----
grep -w $TAPEHOST /etc/hosts > /dev/null 2>&1
else
/bin/domainname ${DOMAIN}
+ /usr/etc/ypserv
/etc/portmap
/etc/ypbind
ypmatch $TAPEHOST hosts > /dev/null 2>&1
Another feature of the 3.3UPGRADE script is found when it upgrades
clients. Being a neat system manager (ha if there is such a thing :^) )
I edited the /etc/nd.local file to be all lined up, column wise, and neat.
This causes a problem with the script because it searches for the following
"user CLIENT_MACHINE_NAME 0" where all white space is ' ' (a space).
Being 'neat', I had edited /etc/nd.local and had replaced all of the
white spaces with tabs (to line everything up nice and neat).
I you are a 'neat' system manager, then you also might have this problem.
As I said the script searches for lines containing "user CLIENT_MACHINE_NAME 0"
to find out what partion, i.e. /dev/ndl?, to mount.
After spending about 2 hours getting the upgrade to run I said to myself
that the next 3.2 system I upgrade, we have two systems, will go real
fast because all I have to do is copy the modified script over to the
second system and run it. Much to my dismay, but expected now that I
think about it, the last thing the script did was to remove itself
(being 'neat' :^) ). Oh well..........
These are the only things that I ran into when upgrading my SUN 3 3.2 system
to 3.3.
NOTE:
I helped someone upgrade their SUN 2's to 3.3 and the only problem
was in /etc/rc.boot. The first if statement was wrong. If you
tried to boot single user (i.e. b -s) it never stopped at single
user. It keep on going, that is doing the fsck's and starting of
daemons. If you have SUN 2's (slooowwwwwwww 8-) ), check the first
if. It should look something like:
if [ $1x = singleuserx ]
Disclaimer:
I am a strong supporter of SUN computers. I like them. It is that
I just get tired of 'hacking' at things to get them to work.
--
@__________@ W. Tait Cyrus (505) 277-0806
/| /| University of New Mexico
/ | / | Dept of EECE - Hypercube Project
@__|_______@ | Albuquerque, New Mexico 87131
| | | |
| | hc | | e-mail:
| @.......|..@ cyrus@hc.dspo.gov or
| / | / {gatech|ucbvax|convex}!unmvax!hi!cyrus
@/_________@/
------------------------------
Date: Thu, 26 Mar 87 15:43:40 CST
From: jjhnsn@ngp.utexas.edu (James Lee Johnson)
Subject: Taming SUN multibus memory (4 meg on a 150u)
A colleague has persuaded me to document in an appropriate place,
i.e. sun-spots, my experiences with SUN multibus memory. I hope that
someone finds this information useful.
We have a SUN 150u file server that has been running with two 1-megabyte
memory boards. We planned to bring it up to 4 megabytes by adding two
more 1-megabyte boards salvaged from a 100u.
The obvious approach was to place the two additional boards in multibus
slots 4 and 5. This produced too many memory errors. SUN suggested that
we would need to have the P2 bus modified and/or add a P2 terminator
board. They weren't very encouraging.
After studying the hardware installation manuals for the 100u, 150u, and
120, I hypothesized that the 150u would work reliably if the cpu board
was placed in the middle of the memory boards. This approach required
that the cpu board be placed in a multibus slot other than slot 1, a
non-standard configuration for a 150u.
I placed the cpu in slot 4, the first meg to the right of the cpu, the
second meg to the left of the cpu, the third meg to the right of the
first meg, and the fourth meg to the left of the second meg.
The 150u has been running reliably with this configuration for 9 days.
It had no errors in 310 passes of the sysdiag pmem test.
Supposedly, newer multibus SUNs do not have this problem. However, our
CS department told me that they had similar problems when trying to
run four 1-megabyte memory boards in their 120s.
If your multibus SUN is having memory problems, you might try moving the
cpu board out of slot 1, and moving some of the memory boards (and ONLY
memory boards) above it. The strategy is to get the memory boards as
close to the cpu board as possible. The goal is to reduce noise on the
"a" partition of the P2 bus.
If you want more technical details, please contact me by email.
jj
--
James Lee Johnson, UTexas Computation Center, Austin, Texas 78712
ARPA: jjhnsn@ngp.utexas.edu jjhnsn@ut-ngp.ARPA
UUCP: allegra!ut-ngp!jjhnsn gatech!ut-ngp!jjhnsn ihnp4!ut-ngp!jjhnsn
seismo!ut-sally!jjhnsn harvard!ut-sally!jjhnsn
------------------------------
Date: Fri, 27 Mar 87 15:32:27
From: craig@icsg.uci.edu (Craig Rolandelli)
Subject: Bug in yppasswdd
When yppasswd looks up an account name in the master yp password file, it
only matchs enough characters to find the pattern it is looking for. For
instance, if you have two accounts "foobar" and "foo", and "foobar" is first
in the passwd file, you will never be able to change the password of the
account "foo". You must place "foo" in the password file before the account
"foobar". Yppasswd will always find "foobar" when looking for "foo". SUN
said that they will fix this in the next release.
Craig Rolandelli (craig@icsa.uci.edu)
University of California, Irvine
Computer Science Department
(714) 856-4222
------------------------------
Date: Mon, 23 Mar 87 21:29:43 EST
From: bzs@bu-cs.bu.edu (Barry Shein)
Subject: SUN Common Lisp question?
Creating a somewhat large array just falls over:
(make-array '(512 512) :element-type 'single-float)
Error: Not Enough Storage After GC
It should use up around 1MB (not that excessive I don't think.)
Anyone know if there is some magic cookie that needs to be tweaked? I
am quite sure it's not swap space or user limits I am running up
against, at least I shouldn't be (pstat says ~50MB swap available,
limit sez 245MB datasize.)
-Barry Shein, Boston University
------------------------------
Date: Mon Mar 23 23:10:28 1987
From: decvax!mcnc!rti-sel!talos!peterson@seismo.css.gov (Steve Peterson)
Subject: Terminal Emulators for Sun?
Does anyone have information (implementations, experiences
source listings, etc.) about terminal emulators for Sun
Workstations. I particularly want emulators for Tektronix
4010-4014, 4105-4205, 4125, and DEC vt100-220-240. Thanks.
--
Steve Peterson (804) 274-5351
UUCP: ...{seismo, decvax}!mcnc!rti-sel!talos!peterson
Philip Morris Research Center; P.O. Box 26583; Richmond, Va 23261
------------------------------
Date: 28 Mar 1987 1821-MET (Saturday)
From: mcvax!ethz!prl@seismo.css.gov (Peter Lamb)
Subject: B/W addon monitor - second source?
Does anyone know of a second source for B/W monitors for
Sun3s? We can make the cables ourselves, but I don't want
to have to be hacking the hardware.
thanks,
Peter Lamb uucp: seismo!mcvax!ethz!prl
Tel: (01) 256 5241 (Switzerland) eunet: prl@ethz.uucp
+411 256 5241 (International) earn: PRL@CZHETH5A
Institut fuer Integrierte Systeme Institute for Integrated Systems
ETH-Zentrum ETH-Zentrum
8092 Zuerich 8092 Zurich
------------------------------
Date: Tue, 31 Mar 87 11:13:04 CST
From: rbbb@titan.rice.edu (David Chase)
Subject: Pascal runtime structures?
Does anyone have an accurate set of C delcarations describing some of the
Pascal runtime data structures (on a Sun)? I have this version
("h00vars.h" from the TeX distribution) which I know was wrong under 2.0.
/* Copyright (c) 1979 Regents of the University of California */
/* sccsid[] = "@(#)h00vars.h 1.11 5/8/84"; */
/* THIS VERSION COMES BACK FROM SUN AND ALLEGEDLY MATCH OUR CURRENT
COMPILER 8/4/84 */
The definite error is in the declaration for the "bool" enum type; "1" is
not "TRUE" to Sun Pascal 2.0. Given this (and the passage of time) I do
not trust other declarations in the file. Can anyone offer a more
up-to-date version of this file?
David
------------------------------
End of SUN-Spots Digest
***********************