Sun-Spots-Request@RICE.EDU (William LeFebvre) (03/21/88)
SUN-SPOTS DIGEST Friday, 18 March 1988 Volume 6 : Issue 32
Today's Topics:
Re: Sun 4/280 power
Re: ping doesn't work with subnets and 3.4.2
A Suntools Background Rasterfile
CDC 9720-500 troubles on a 3/180
Problem: "tip" hangs
Mac simulation on Sun -- request for info
Detailed SUNos info wanted
rex question
What does "Restricted" mean?
Suns for general research computing?
Clues about "tty grp reset" message?
Experiences with modula-2 compilers for Suns?
Adding a new (non-Sun) display to a Sun?
Mods for lengthy usernames?
troff or Postscript previewer?
Experience with Sun 3/60 memory from clearpoint?
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 stored on "titan.rice.edu". For volume X, issue Y,
"get sun-spots/vXnY". They are also accessible through the archive
server: mail the word "help" to "archive-server@rice.edu".
----------------------------------------------------------------------
Date: Fri, 4 Mar 88 14:38:42 PST
From: weiser.pa@xerox.com
Subject: Re: Sun 4/280 power
Reference: v6n24
One person's opinion:
The rated MIPS of a RISC architecture are much more dependent on
optimizing compiler technology than the rated MIPS of a 68020. Here is
what this means in terms of the Sun-4:
If you are using the Sun-4 to crunch, it performs as promised. With
optimization level 3 or 4 turned on the code quality is very very good.
If you are using the Sun-4 in a heavy compile/edit/run loop, your will not
see that "10 MIP" performance, for the simple reason that you won't be
compiling your code with the optimizer on. (You won't, because (a) you
can't use dbx on optimized code, and (b) the compiler runs 3 times as long
at the high optimization levels.) The difference in dhrystone
performance on a sun-4 compiled with and without optimization is a factor
of 2. That still makes a Sun-4 faster than a top-end Sun-3 even without
optimization, but not by so much.
However, thinking long term, it makes sense now, other things being equal,
to buy Sun-4's if you believe the "SPARC story": that this architecture is
going to be getting faster and faster. Switch now and be ready later.
-mark
------------------------------
Date: Tue, 08 Mar 88 10:26:22 -0500
From: Mike Minnich <mminnich@udel.edu>
Subject: Re: ping doesn't work with subnets and 3.4.2
Reference: v6n25
> Since upgrading to 3.4(.2) and subnetting a class B subnet into a class C
> subnet, ping doesnt seem to work through our gateways.
The problem is that ping uses raw IP sockets for I/O. In the 4.2
networking code (upon which some portions of 3.4 are based), raw sockets
(SOCK_RAW) does not correctly route outgoing datagrams; instead it routes
them "to an interface". The result is that an IP datagram maybe goes out
the correct interface, but it ARP's for the destination IP address on the
local cable instead of forwarding the IP datagram to the appropriate
gateway on the local cable on the way to that subnet.
While pinging, look at your arp cache with arp -a and see if this is the
case. If you have source I can supply you with patches to correct the
problem.
mike
------------------------------
Date: Fri, 4 Mar 88 02:05:06 est
From: mike@stars.umd.edu (Mike Briley)
Subject: A Suntools Background Rasterfile
Uuencoded and compressed below is a sun rasterfile which so many people
around here enjoy using as a suntools background, that I thought it might
be nice to post. I'm sure quite a few other people out there also have
backgrounds more exciting than "desktop grey", and I would be interested
in seeing others. At any rate, the file needs to be first "uudecoded",
then "uncompressed". After that, just start "suntools" with both the "-i"
and "-background enterprise" options. Note that this is an inverted
image. This is largely so that the cursor will be visible. Also, for
reasons too complicated to explain, the image has been set up to allow for
west/east icon gravities. A non-inverted image, as well as other
backgrounds are available if there is enough interest. Needless to say,
the name of the file just about sums up its contents.
-mike@stars.umd.edu
[[ The compressed and uuencoded files has been placed in the archives
under "sun-icons/enterprise.uu" and is 16170 bytes in length. As the name
implies, it is a digitized black-and-white image of the Starship Enterprise
flying through the galaxy. It can be retrieved via anonymous FTP from the
host "titan.rice.edu" or via the archive server with the request "send
sun-icons enterprise.uu". For more information about the archive server,
send a mail message containing the word "help" to the address
"archive-server@rice.edu". --wnl ]]
------------------------------
Date: Thu, 03 Mar 88 23:08:13 -0800
From: Bob Brown <rlb@riacs.edu>
Subject: CDC 9720-500 troubles on a 3/180
We have four CDC 9720-500 380 MB (formatted) disk drives on our 3/180
server, run off a pair of Xylogics 451 controllers. Last week, one
started reporting hard errors. After some diagnositic work, we decided to
try reformatting it, and that failed. We tried it on another controller
and that failed, too. We decided that the disk itself had gone bad. No
problem. Call our vendor and get service.
Here, one week later, we are still sitting on a bad drive; at least three
phone calls daily are exchanged between RIACS and Group III (our vendor)
or CDC. No success yet. Now, just a few days back, another drive has
started reporting the same type of error. Curiously, it runs off a
different controller than the first bad drive.
Thoughts of lemons come to mind. CDC refuses to tell us if they have a
recent history of trouble with this drive. All of our serial numbers are
in the 300s.
Question for the net: does anyone out there use these CDC 9720-500s, and
has anyone experienced losses like this? I'd like to hear from Sun users
who have these units, and whether or not you've had troubles.
The lighter side is that Group III promised a MB-for-MB swap with double
eagles within one year if we decided we wanted it.
Reply direct. Thanks in advance,
Bob Brown
RIACS/NASA Ames
------------------------------
Date: Fri, 4 Mar 88 15:16:13 EST
From: darkstar!brian@uc.msc.umn.edu (Brian Utterback)
Subject: Problem: "tip" hangs
I use tip through a serial port and modem to connect with our network.
Sometimes (about once per day) spurious line noise causes tip to hang.
I'll be in a session, and I'll see some line noise, and the tip will no
longer respond. Examination of the modem indicates that the key strokes
are not being sent at all. However, tip will still respond to the tilde
escapes, so it is not completely dead. The only thing I seem to be able
to do, is to disconnect, and reinitiate the connection. This is a big
pain since I am usually going through 2 to 3 computers to get where I
want. (NH->DC->MN or even NH->CT->DC->MN) Does anyone know of a way to
fix the problem?
Brian Utterback |UUCP:{ihnp4!cray,sun!tundra}!hall!blu
Cray Research Inc. |ARPA:blu%hall.cray.com@uc.msc.umn.edu
One Tara Blvd. #301 |
Nashua NH. 03062 |Tele:(603) 888-3083
------------------------------
Date: Sat, 5 Mar 88 16:12:45 PST
From: crocker@tis-w.arpa (Stephen D. Crocker)
Subject: Mac simulation on Sun -- request for info
Does anyone know of Sun software that can emulate a Mac so that Mac
software can run on a Sun? Note that this is related but different from
Thomas Kurfurst's query in v6n18. He wants to run Sun applications and
see the results on his Mac II. I want to run Mac applications on the Sun.
(Why? There's a LOT of CHEAP, USEFUL Mac software out there...)
Reply to crocker@tis-w.arpa.
Steve Crocker
Trusted Information Systems
11340 Olympic Blvd. #265
Los Angeles, CA 90064
213-477-5828
------------------------------
Date: 06 Mar 88 16:03:24 EST (Sun)
From: franklin@csv.rpi.edu
Subject: Detailed SUNos info wanted
Where would I find more info about SUNos (or 4.3 in general) than is given
in the UNIX Internal Ref Man? For instance, where are the fields listed
by vmstat -s defined?
Name: Wm. Randolph Franklin
Telephone: (518) 276-6077
Papermail: ECSE Dept., 6026 JEC, Rensselaer Polytechnic Inst, Troy, NY, 12180
[[ Well, the kernel structure named "sum" is a struct vmmeter, and that is
defined in "/usr/include/sys/vmmeter.h". The comments in the file contain
very brief descriptions of each field's meaning. If you want more
information than that, I'm not sure where you might find it. --wnl ]]
------------------------------
Date: 06 Mar 88 16:00:32 EST (Sun)
From: franklin@csv.rpi.edu
Subject: rex question
I have a fileserver romeo and diskless node juliet that mounts its
filesystems from romeo. Therefore romeo's /etc/exports lists juliet for
each partition. (juliet's /etc/exports also makes all its filesystems
exportable back to romeo just in case romeo should want them.) Now, from
juliet I try to say
on romeo date
This fails thus:
on romeo: rexd mount: not in export list for romeo:/usa
unless I modify romeo's /etc/exports to export its own partitions to
itself also! Is this usual?
Name: Wm. Randolph Franklin
Telephone: (518) 276-6077
Papermail: ECSE Dept., 6026 JEC, Rensselaer Polytechnic Inst, Troy, NY, 12180
Telex: 6716050 RPI TROU (General RPI Telex Number)
Fax: (518) 276-6003 (General RPI Fax Number)
------------------------------
Date: 4 Mar 88 11:21:56 GMT
From: shapiro@blueberry.inria.fr (Marc Shapiro)
Subject: What does "Restricted" mean?
What the @#!!%%$$## does "Restricted" mean? It seems I get this message
whenever I try to compile something as root on my 3/280 server, ever since
we installed SunOS 3.4. The compilation then runs into some strange state
and aborts. Some other programs also seem to exhibit this wonderful
behaviour.
It never happens on a client. This message is not listed in the error
list of intro(2).
Is there any rationale for this annoyance?
Marc Shapiro
INRIA, B.P. 105, 78153 Le Chesnay Cedex, France. Tel.: +33 (1) 39-63-53-25
e-mail: shapiro@inria.inria.fr or: ...!mcvax!inria!shapiro
------------------------------
Date: 04 Mar 88 13:47:00 EDT
From: BRUCE WINER <BWINER@CARLETON.BITNET>
Subject: Suns for general research computing?
We at Carleton University are investigating the potential for use of Suns
in support of general research computing, with a significant component of
social science/statistical applications. The following characteristics of
the computing environment are anticipated:
(1) a variety of Sun hardware, including server(s) and diskless workstations.
(2) provision of "timesharing" access for dumb terminals.
(3) interconnections to Macs, PCs, non-Sun LANS.
(4) a mixture of user types, from naive to experienced. Some would want
to have as little to do with Unix as possible.
(5) definite interest in statistical software: SPSS-X, BMDP, TSP, etc.
We would very much appreciate hearing from anyone who has experience with
Suns in this sort of academic/research context, as to critical
hardware/software choices, general viability as an alternative to
mainframe support, and anything else you think might be relevant. We'll,
of course, summarize replies and post to the net. Thanks.
Bruce Winer, BWINER@CARLETON.BITNET
Computing Services,
Carleton University,
Ottawa, Ontario Canada K1S 5B6
(613) 564-7141
------------------------------
Date: Fri, 4 Mar 88 13:50:45 EST
From: darkstar!brian@uc.msc.umn.edu (Brian Utterback)
Subject: Clues about "tty grp reset" message?
Two or three times a day, I open a cmdtool window from sunview, and when I
go to use it, a message flashes in it that says something like this:
"tty grp reset"... and the the window disappears. I don't know exactly
what the error is because it only appears for a second when I least expect
it. The console prints a message when this happens, which says: " A
command window has exited because its child exited. Its child's process
id was 27805 and it died due to signal 1." This is exact because this one
stays around. Does any one have any clues?
Brian Utterback |UUCP:{ihnp4!cray,sun!tundra}!hall!blu
Cray Research Inc. |ARPA:blu%hall.cray.com@uc.msc.umn.edu
One Tara Blvd. #301 |
Nashua NH. 03062 |Tele:(603) 888-3083
------------------------------
Date: Fri, 4 Mar 88 22:22:48 PST
From: Jim Hudgens <hudgens@ray.met.fsu.edu>
Subject: Experiences with modula-2 compilers for Suns?
We are currently looking at the commercially available Modula-2 compilers
for Sun-2 and Sun-3 systems, and are interested in hearing about any
experiences others may have had with various compilers. Currently we are
investigating the SUN Modula-2, as well as products sold by Ana-Systems,
and Oasis.
Does anyone out there have any positive or negative comments about the
above compilers, or possibly know of any other sources of Modula-2
compilers? We are particularly concerned about the symbolic debugger
support each of the packages provides. Please respond directly to me.
Thanks.
Jim Hudgens
hudgens@ray.met.fsu.edu
------------------------------
Date: Fri, 4 Mar 88 21:12:10 +0100
From: Jean Huens <mcvax!prlb2!kulcs!jean@uunet.uu.net>
Subject: Adding a new (non-Sun) display to a Sun?
We are thinking about adding another (not sun) display and bit map
interface to a sun workstation. Did somebody already try to do this.
Does it involve (from software point of view) more than writing a new
pixrect driver? Does one need more sources than available in a normal sun
distribution? Thanks.
Jean Huens jean@kulcs.uucp ...mcvax!prlb2!kulcs!jean
Katholieke Universiteit Leuven, Dept. Computer Science
Leuven, Belgium
------------------------------
Date: Sat, 5 Mar 88 18:49:18 EST
From: erm@cs.williams.edu
Subject: Mods for lengthy usernames?
Does anyone out there know of a modification/hack/param setting/compile
option/spell or suitable incantation to make Suns recognize more than 8
characters ?? Since 3.2 (through 3.5), you can happily stick names longer
than 8 chars in /etc/passwd, but if you do, they can never login. Even
you only enter the first 8 chars, you're bounced.
Normally, this wouldn't bother me, but the Computer Center is trying to
set a uniform standard for usernames on all campus machines, and many of
the faculty already have VAX accounts with usernames beyond 8 chars. You
see the problem.
Anyway, I suspect nothing short of a hack to the getty source code (which
I don't have) could fix this. However, if I'm wrong, would someone please
let me know.
Evan R. Moore
Williams College Computer Center
Internet: erm@cs.williams.edu
BITNET: 91ERM@WILLIAM
CSnet: erm@williams.edu
[[ Sadly, fixing "login" and "getty" won't entirely take care of the
problem (yes, getty deals with usernames). Other programs that deal with
usernames tend to make the bad assumption that the name will never be
longer than 8 characters. All of those programs would have to be fixed as
well. It might help if there was a parameter defined in an include file
to be the maximum length of a username (similar to MAXNAMLEN for files)
but such a parameter has not traditionally existed in Unix systems. --wnl]]
------------------------------
Date: Fri, 4 Mar 88 14:15:15 EST
From: peloquin@qtp.ufl.edu
Subject: troff or Postscript previewer?
Does anyone know of a troff or Postscript previewer which would give a
good representation of all the special characters and variations in point
size?
Renee Peloquin
...ufcsv!ufqtp!peloquin
peloquin@orange.qtp.ufl.edu
peloquin@ufpine.bitnet
------------------------------
Date: 7 Mar 88 14:27 -0500
From: naren <naren%sce.carleton.cdn@ean.ubc.ca>
Subject: Experience with Sun 3/60 memory from clearpoint?
Does anyone have experience with sun 3/60 memory from Clearpoint? The
rep. was saying that the Sun buys the memory from Clearpoint and the
memory should work just fine. These are 1mb chips. However, I was also
told that he can supply 3/60 expansion memory using 256k chips at an
almost 20% price reduction. Any comments?
naren
Narendra Mehta,
Systems & Computer Eng. Department, Carleton University, Ottawa, Canada
UUCP: {allegra,decvax,ihnp4,linus,utzoo}!watmath!clan!naren
ARPA: naren%sce.carleton.cdn%ubc.csnet@csnet-relay.arpa
------------------------------
End of SUN-Spots Digest
***********************