Sun-Spots-Request@Rice.edu (William LeFebvre) (09/22/88)
SUN-SPOTS DIGEST Wednesday, 21 September 1988 Volume 6 : Issue 232
Today's Topics:
Re: compiler error: out of string space
Re: Information on TAAC board
Re: NULLs in 4.0 file
Re: Clock losing 30 days
Tooltool patch
Closing the architecture: Ciprico gets the last laugh?
SunOS 3.2 vs. SunOS 4.0 on Sun-4/280
Disk Drives and Controllers for SUN-4
Spurious "Your have new mail." msgs
Problem: UUCP and detached processes
RS232 handshaking on the Sun386i?
386i DOS window and color?
Timings for Sun3?
Suppliers/info re: sun3/50 phillips monitor power supply?
Video and mouse extenders?
Send contributions to: sun-spots@rice.edu
Send subscription add/delete requests to: sun-spots-request@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@rice.edu" or mail the word "help" to the same address
for more information.
----------------------------------------------------------------------
Date: Fri, 16 Sep 88 09:27:52 EDT
From: Bennett Todd <bent!bet@mcnc.org>
Subject: Re: compiler error: out of string space
Hmm. Assuming you have bounteous plenty of swap space allocated, I
wouldn't sweat bugs/deficiencies/whetever with the provided cc at all; I
right now am running with CC=gcc in all my makefiles, and if things keep
looking as beautiful as they have of late, I am liable to completely
replace cc with gcc as the main compiler (and maybe keep it around for a
little while as occ or something).
-Bennett
bet@orion.mc.duke.edu
------------------------------
Date: 16 Sep 88 06:21:56 GMT
From: hedrick@aramis.rutgers.edu (Charles Hedrick)
Subject: Re: Information on TAAC board
Reference: v6n228
Someone asked for information from people who have actually used a
TAAC-1. I conjecture that this will be hard to find, since as far as
we can determine, the TAAC is not yet available. We ordered one when
it was first announced by Sun. We haven't seen it yet. At SIGGRAPH,
one of our faculty was told that it had recently started shipping, and
that the backlog would be worked off by the end of Sept. However when
our salesman attempted to investigate, he was told that the production
version is not yet shipping, although there are pre-production
versions. He was not sure what difference, if any, there would be
between the versions, except that they are being made in different
plants.
------------------------------
Date: 16 Sep 88 00:30:41 GMT
From: jeff@tc.fluke.com (Jeff Stearns)
Subject: Re: NULLs in 4.0 file
abe@mace.cc.purdue.edu writes:
> Has anyone seen any problems with characters of a file being NULLed under
> SunOS 4.0? I have a situation in which a printer log file occasionally
> has a legitimate entry converted to NULLs for no apparent reason. The
> NULLs never seem to appear at the end of the file, but occur prior to the
> last entry.
We have seen this phenomenon on our systems running SunOS 4.0. It appears
to have been introduced with release 4.0. It isn't present on comparable
systems running release 3.5.2.
It shows up for us in one especially painful circumstance -- mail files.
I've seen several cases where mailboxes in the /usr/spool/mail directory
have have been corrupted when a piece of mail was delivered as NULLS.
There is evidence that the bug is provoked by programs such as tail -f or
little mailwatcher programs that periodically stat the file to see if it's
grown.
Perhaps you have some process periodically watching the log file from an
NFS client?
--
Jeff Stearns John Fluke Mfg. Co, Inc. (206) 356-5064
jeff@tc.fluke.COM {uw-beaver,microsoft,sun}!fluke!jeff
PS - Calling all users of the Vitalink TransLAN IV Ethernet bridge! Please
drop me a line.
------------------------------
Date: Fri, 16 Sep 88 09:09:29 EDT
From: Bennett Todd <bent!bet@mcnc.org>
Subject: Re: Clock losing 30 days
Just to add another data point, every time I reboot my diskless Sun 3/50
running SunOS 3.5 off a Sun 3/280 server (who also is running 3.5) I get
the same "WARNING: clock lost 30 days -- CHECK AND RESET THE DATE!"
stuff. I finally added an rdate servername to the end of rc.local and said
to heck with it. I have had the clock patch in on every machine around
here for a while now.
-Bennett
[[ This was sent in later... --wnl ]]
Well color me embarassed! Why can't I double check these danged things
before I go and shoot off my mouth in public.... I must have dreamed about
installing the TOD patch on all our Suns and /usr/sys/OBJ files; I just
checked and the 3/50 I posted about earlier today didn't have the fix in.
I applied it and the problem vanished.
Very sorry.
[[ S'okay. We recently (in the past week) had a problem with a machine in
a different department. We called Sun support about it, and they said
"are you sure the TOD patch is installed? That's one of the TOD bug's
symptoms". We said "Of course it's installed. It's installed on all our
machines." Well, guess what? It wasn't. The lesson? Don't just assume
that you've installed that patch---double check it. --wnl ]]
-Bennett
bet@orion.mc.duke.edu
------------------------------
Date: Thu, 15 Sep 88 14:31:24 EDT
From: Chuck Musciano <chuck@trantor.harris-atd.com>
Subject: Tooltool patch
For those who have tried to move tooltool to SunOS 4.0, you may have
encountered difficulties during compilation. The attached patch will fix
the problem.
On a more interesting note, this was the bug. I had a routine like this:
bad_decl(b, c, d, e)
int a;
int b;
int c;
int d;
int e;
Note the erroneous declaration of a. Under 3.x, this was not flagged, and
things compiled correctly. Under 4.0, this is caught.
Thanks to Paul Emerson, who found this for me. I haven't gone to 4.0 yet.
Chuck Musciano
Advanced Technology Department
Harris Corporation
(407) 727-6131
ARPA: chuck@trantor.harris-atd.com
#! /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 shell archive."
# Contents: patch
# Wrapped by chuck@melmac on Thu Sep 15 14:27:04 1988
PATH=/bin:/usr/bin:/usr/ucb ; export PATH
if test -f 'patch' -a "${1}" != "-c" ; then
echo shar: Will not clobber existing file \"'patch'\"
else
echo shar: Extracting \"'patch'\" \(599 characters\)
sed "s/^X//" >'patch' <<'END_OF_FILE'
X*** tooltool.c.orig Thu Sep 15 14:24:27 1988
X--- tooltool.c Thu Sep 15 14:24:57 1988
X***************
X*** 285,291 ****
X /************************************************************************/
X static struct pixrect *better_menu_image(text, chars, pixels, font)
X
X- Panel panel;
X char *text;
X int chars;
X int pixels;
X--- 285,290 ----
X***************
X*** 313,319 ****
X /************************************************************************/
X static struct pixrect *better_button_image(text, chars, pixels, font)
X
X- Panel panel;
X char *text;
X int chars;
X int pixels;
X--- 312,317 ----
END_OF_FILE
if test 599 -ne `wc -c <'patch'`; then
echo shar: \"'patch'\" unpacked with wrong size!
fi
# end of 'patch'
fi
echo shar: End of shell archive.
exit 0
------------------------------
Date: 16 Sep 88 06:28:35 GMT
From: hedrick@aramis.rutgers.edu (Charles Hedrick)
Subject: Closing the architecture: Ciprico gets the last laugh?
I just got a "news release" from Ciprico. In November, they are going to
start shipping a version of their Rimfire disk controller that emulates
enough of a Xylogics controller so that it can boot with standard Sun boot
ROM's and boot programs. Of course you will still need a Ciprico device
driver in the kernel, since the emulation will only be done during
booting. (This is for performance reasons.) The board uses the Sun VME
form factor, so it doesn't need a VME to VME adapter board.
------------------------------
Date: Thu, 15 Sep 88 17:36:23 EDT
From: Christopher Rabson <albion@opus.cs.mcgill.ca>
Subject: SunOS 3.2 vs. SunOS 4.0 on Sun-4/280
We have acquired a pair of Sun-4/280s over the past six months, to replace
our two aging VAX 11/780s. One of our Sun-4/280s is running under SunOS
3.2, and the other is currently running SunOS 4.0 on a "trial" basis. We
also have several Sun-3/50s and a couple of Sun-3/280s running SunOS 3.5.
I am not happy with the SunOS 4.0 flavour of UNIX--I find it difficult to
accept what Sun has done to the UNIX directory hierarchy, and the larger
kernel and more "sluggish" response (compared to the Sun-4 under 3.2) is
disturbing. Since I do not get the final word on whether we go with 4.0
or not, does anyone havve any real horror-stories on 4.0, particularly in
a multiuser environment? (We plan to run 32+ users simultaneously per
machine, and we have the 3.2 sub-netting fix.) Is there any hope for
SunOS 4.1, or can we expect the same bloated kernel and associated
performance?
Any e-mail responses would be appreciated.
Christopher Rabson McGill University
Senior Technician School of Computer Science
e-mail: albion@oliver.cs.mcgill.ca Montreal, Canada
------------------------------
Date: 16 Sep 88 13:25:07 GMT
From: gt-eedsp!jensen@gatech.edu (P. Allen Jensen)
Subject: Disk Drives and Controllers for SUN-4
This is a request for information on vendors of disk drives and
controllers for SUN-4 systems (all models including SUN-4/110). I would
also appretiate comments on anyones experience with drives and
controllers. I will post a summary of the responses I get.
Thanks,
P. Allen Jensen
Georgia Tech, School of Electrical Engineering, Atlanta, GA 30332-0250
USENET: ...!{allegra,hplabs,ihnp4,ulysses}!gatech!gt-eedsp!jensen
INTERNET: jensen@gteedsp.gatech.edu
------------------------------
Date: Fri, 16 Sep 88 10:09:17 EDT
From: csrobe@work1.icase.edu (Charles S. [Chip] Roberson)
Subject: Spurious "Your have new mail." msgs
Several newer users to our system (Sun 3/160 + 19 3/50 + 1 3/60) are
seeing (or have seen) notification of new mail without actually receiving
any mail. As far as we can tell, they are not losing any mail; so, it
seems to be only an annoyance. Older users on the system don't appear to
have this problem.
Does a anybody have a suggestion on how we can stop these spurious messages?
many thanks,
-chip
Charles S. Roberson ARPANET: csrobe@icase.edu
ICASE csrobe@[128.239.1.30] (cs.wm.edu)
MS 132C BITNET: $csrobe@wmmvs.bitnet
NASA Langley Rsch. Ctr. UUCP: ...!uunet!pyrdc!gmu90x!wmcs!csrobe
Hampton, VA 23665-5225 Phone: (804) 865-4090
------------------------------
Date: Wed, 14 Sep 88 14:53:37 EDT
From: cognos!geovision!curry!graham@uunet.uu.net (Graham Ashby)
Subject: Problem: UUCP and detached processes
Scene is this: User logs in over the modem, starts up a process in the
background, then exits. Uucp tries to dial out, and gets a NO DEVICE
error.
The trouble, as far as I can see, is that the process group that the user
started is still active, so that the tty driver blocks any read attempts
(returns -1 status).
Now, I am sure that this process is not going to do any reads or writes to
the terminal, so I want UUCP to get in. Is there any easy way of doing
this? Does one have to force the process to detach itself from the
terminal (whatever ioctl it is)?
BTW, this is on a sun4 running 3.2.
Graham Ashby, Geovision Corp, Ottawa, Canada. {smart-mailer}!geovision!graham
------------------------------
Date: Thu, 15 Sep 88 14:47:26 MST
From: ron%catuc.UUCP@arizona.edu
Subject: RS232 handshaking on the Sun386i?
I have never sent anything to the net before, let alone a moderated,
digested group, so I hope you'll pardon any errors on my part.
What I am wondering about is RS232 hardware handshaking with the Sun386i
running under SunOS 4.0. We have a device that we want the Sun to
communicate with over RS232 at 9600 baud or better. The problem is that
the device can send characters to the Sun faster than the Sun can buffer
them and get them out. Note that the baud rate is not the problem, rather
it is the character rate is too high for the Sun (we don't lose bits, we
lose characters). The device does not have the capability to use XON/XOFF
but can use hardware handshaking to stop sending characters when its CTS
line goes low.
However, the Sun, according to the documentation, only supports hardware
handshaking in one direction, that is when it is sending data the
receiving device can drop the CTS line, as seen by the Sun, and the Sun
will stop transmission until the CTS line is re-asserted. We need this in
the opposite direction, where the Sun can de-assert the device's CTS line
when the Sun's buffer is full.
If anyone knows of an undocumented feature on the Sun which does this, a
workaround for this, a simple hardware device to somehow allow
handshaking, or any other solution to this problem, I would be eternally
indebted to him or her (well, at least till next week).
Thank you very much,
Ron Johnson
arizona!catuc!ron
or
(602) 881-8600
------------------------------
Date: Thu, 15 Sep 88 13:39:20 EDT
From: fed!m1edb00@uunet.uu.net (Eric D. Boutilier)
Subject: 386i DOS window and color?
When I open a DOS window on the 386i it always comes up in Monochrome
mode. My beta test manual doesn't tell me how to fix this... it says "If
you have a color monitor, color graphics appear in a color DOS window."
There is a file called setup.pc that seems to hold this kind of
information but the manual doesn't tell me anything about specifying CGA
color in this file.
If I run the PC program 'setup' which asks questions about the system and
(on a PC) stores the settings in non-volitale memory, I can change from
monochrome to CGA color... but only for that one DOS window. The worse
thing about this is that 'setup' is an incredibly stupid program that
takes you through all these yes/no questions before you can change the
monitor type.
Thank you.
--
Eric Boutilier
UUCP: uunet!fed!m1edb00
(202) 452-2734
------------------------------
Date: Thu, 15 Sep 88 17:46:29 pdt
From: kcooley@teknowledge-vaxc.arpa (Keith Cooley)
Subject: Timings for Sun3?
I am looking for timings of fundamental operations in SunOs 3.x and 4.0,
on say, a sun 3/60. In particular I am interested in timings for low level
operations such as process context switching, system call overhead, fork,
exec, read/write system calls. Also pointers to articles about the
performance of Sun 3 would also be useful.
Keith Cooley
internet: kcooley@teknowledge.arpa
uucp: { uunet|sun|ucbvax|decwrl|uw-beaver}!kcooley%teknowledge.arpa
------------------------------
Date: Thu, 15 Sep 88 19:09:53 CST
From: Robert Coleman <COLEMAN@UREGINA1.BITNET>
Subject: Suppliers/info re: sun3/50 phillips monitor power supply?
I have a problem with the power supply board of a sun3/50 phillips monitor
power supply board. The voltage generated is not sufficiently high to
produce a raster.
Recommendations of suppliers for these boards (at reasonable cost of
course) would be appreciated.
Two items on the board seem suspect.
1) Q4 on the power supply board
2N6396 (SCR)
This seems to be a standard number and although a local
supplier has not been found, it should be possible to get
one eventially. Still advice is wellcome.
2) OCI (opto-isolator)
Mexico 1
K8902 8607
This number appears to be a sun in-house number. Does someone
know what it really is?
Other suggestions regarding possible sources of trouble would be most
appreciated.
Please send responses to
coleman@uregina2.bitnet
Thank you
------------------------------
Date: Thu, 15 Sep 88 16:47:07 PDT
From: Darrell long <darrell@midgard.ucsc.edu>
Subject: Video and mouse extenders?
I have a 4/110 in my office and it makes a lot of noise. I would like to
move it to the machine room, which is about 50 feet away. I expect that I
will need to get some sort of signal booster in order to do this. Does
anyone have any experience doing this?
Thanks, DL
Darrell D.E. Long
CIS Board of Studies
University of California
Santa Cruz, CA 95064
------------------------------
End of SUN-Spots Digest
***********************