Sun-Spots-Request@Rice.edu (William LeFebvre) (10/18/88)
SUN-SPOTS DIGEST Sunday, 16 October 1988 Volume 6 : Issue 262
Today's Topics:
Re: Project Planning Software
Re: Force periodic password changes
Using Return-Path: for replies
pr_tex_* bug under 4.0
C/A/T Troff to ditroff?
System V shared memory "zombies"?
Lockdowns for 3/60s?
Registered Domain Address?
386i vs ksh-i?
monitoring logins, forcing password chg?
Large screen display for SUN video output?
FAX equipment/software?
M*cAPP file open tool?
SNA3270?
Discrete modelling tools?
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: 12 Oct 88 19:47:24 GMT
From: Bill Randle <billr@tekred.cna.tek.com>
Subject: Re: Project Planning Software
I am using the SunTrac program from Sun to do project planning with. It
does PERT and critical path analysis, lets you do resource leveling,
assign overtime and generates graphical and tabular reports. The user
interface is fairly straight forward and easy to use (it does help to go
through the tutorials). I would definately recommend it. As an aside, for
the PC world I've heard that a program called Project Manager's Workbench
is one of the best.
-Bill Randle
Tektronix, Inc.
billr@saab.CNA.TEK.COM
------------------------------
Date: Wed, 12 Oct 88 19:04:13 PDT
From: walker%skat.usc.edu@oberon.usc.edu (Michael D. Walker)
Subject: Re: Force periodic password changes
Reference: v6n235
Peter Ho: <ho@hac2arpa.hac.com>:
>Does anyone out there have software to force users to change password
>every so often on a SUN?
>
>[[ Is this even possible without rewriting "/bin/login"? --wnl ]]
No, just mv their .cshrc to something else, and mv a script to .cshrc.
This script will loop running passwd until it exits correctly and then
mv's their .cshrc back onto the script. Another method would be to change
their login shell to such a script and the script could do a chsh.
We setup new user accounts with their SSN as the password and a .cshrc
script that loops until they change it. This script then mv's a default
.cshrc onto itself.
Mike Walker :-] arpanet: walker%wei.usc.edu@oberon.usc.edu
USC University Computing Services uucp: sdcrdcf!oberon!wei!walker
Phone: (213) 743-2957 bitnet: walker@kylara
------------------------------
Date: Wed, 12 Oct 1988 15:00:21 EDT
From: The Pirate King <chet@cwjcc.cwru.edu>
Subject: Using Return-Path: for replies
In Sun-Spots Digest, v6n255, lai@vedge.UUCP (David Lai) asks:
> How do I get mail(1) to use Return-Path for replys?
>
>...It seems as if mail
>prefers to use the path in From: header instead of the Return-Path:
>header, which seems to always contain the proper return path.
>
>Can I tell mail to use Return-Path?
No, and you shouldn't be able to. [[ Quite right! --wnl ]]
User agents should only use the Reply-To: header field, if it exists, or
the From: field. The Return-Path: header is there for the benefit of Mail
Transport Agents, usually to send errors back to the originator. It
should contain a valid path back to whoever sent the mail, true, but
that's not always where you want the replies to go.
(Case in point: I send mail from lots of machines, most of them hooked
together with NFS so I have the same home directory but different spool
directories on all. I want replies to go to cwjcc.CWRU.EDU, no matter
where the mail was sent from. If you use the Return-Path: header to
reply, my efforts will be defeated.)
[[ Another case in point: Sun-Spots digests get sent from the address
"sun-spots-request@Rice.edu". That's where the return-path point to. But
if someone replies to a digest, the mail should be sent to
"sun-spots@rice.edu". Thus the "Reply-to" points there (although the
"From" does not). But I understand the original poster's frustration:
not all uucp mailers change the "From:" line in the mail header to reflect
passage through the machine. So the address there ends up being useless.
This is indicative of poor system/network management. --wnl ]]
Chet Ramey
Network Operations Group
A.R. Jennings Computing Center, Case Western Reserve University
chet@cwjcc.CWRU.EDU
------------------------------
Date: 12 Oct 88 13:28 -0500
From: Roland Schneider <sch%ee.uofm.cdn@relay.ubc.ca>
Subject: pr_tex_* bug under 4.0
( this has also been sent to hotline@sun.com )
I have run into a problem with the following:
/* program t.c */
#include <stdio.h>
extern short pr_tex_dotted[]; /* used for pw_line() textures */
main()
{
short *b;
b = pr_tex_dotted;
printf("%d\n", b[0]);
}
compile with "cc t.c -lsuntool -lsunwindow -lpixrect" under SunOS 3.5
On a SUN 3/260 (SunOS 3.5) the program gives the correct result (1)
On a SUN 3/160 (SunOS 4.0) the program gives the correct result (1)
Now recompile under SunOS 4.0:
On a SUN 3/160 (SunOS 4.0) the program gives an incorrect result (25087)
On a SUN 4/280 (SunOS 4.0) the program gives an incorrect result (-25117)
The problem is that the program gets the wrong address for pr_tex_dotted.
This appears to have something to do with the 4.0 shared libraries, as dbx
changes its mind about what pr_tex_dotted is when the program is executed.
Roland Schneider
Dept. Electrical Engineering
University of Manitoba
Winnipeg, Canada
------------------------------
Date: Wed, 12 Oct 88 11:22:14 -0700
From: greg@aerospace.aero.org
Subject: C/A/T Troff to ditroff?
Is it possible to get ditroff output on a Sun with SunOS 4.0? If not (and
I think not) is there a free C/A/T troff-2-ditroff out there? All replies
welcome....
greg@aerospace.aero.org
[[ Am I missing something here? You said that you think it is not likely
that one may produce ditroff output under SunOS 4.0. Then you ask for
something to convert from C/A/T to ditroff? Did you get that backwards?
To answer your first question: it depends on the output device. A
ditroff converter exists for most every reasonable ouptut device. One may
install it via printcap such that "lpr -n" will use the filter at the
appropriate time. Does that help? --wnl ]]
------------------------------
Date: 11 Oct 88 19:24:21 GMT
From: harvard!ll-xn!munsell!jwf@gatech.edu (Jim Franklin)
Subject: System V shared memory "zombies"?
Has anyone using System V shared memory found a clean way of dealing with
the potential "zombie" shared memory segments? The problem is that that
shared memory segments created by calls to shmget() live forever (until
the next reboot), unless they are destroyed using shmctl(). Once a
process writes to a shared segment, swap space is allocated for the
modified pages, and it can't be released until the segment is destroyed.
This is a useful feature but a potentially dangerous one. A process must
do careful error-handling to avoid clogging up the system with shared
memory segments allocated by it if the process dies unexpectedly.
There are two utilities ipcs(1) and ipcrm(1) that can be used as a brute
force way of finding and destroying these zombie segments. ipcs(1) will
write to stdout a list of all shared memory segments, including the
reference count and time of last attach. ipcrm(1) will destroy a
specified shared memory segment. These two utilities could conceivably be
pasted together in a shell or awk script and run via cron to kill off
zombie segments.
But what constitutes a zombie segment is clearly application specific: the
fact that a segment is unattached doesn't make it a zombie, nor does the
length of time it has been unattached. There is nothing to say that a
process can't allocate a shared memory segment, write data to it, detach
it, and expect the segment to still be around days later. (Except the
crontab entry which destroys them -- Might Makes Right ;-).
It is possible for a process to create a shared memory segment that will
be automatically destroyed when the process exits. Doing a shmget(), then
a shmmat(), then a shmctl(IPC_RMID) will make the segment disappear as
soon as the last process mapping the segment exits. This is similar to
doing an open() in /tmp immediately followed by an unlink() to create a
true temporary file. The only problem with this is that *all processes*
that intend to share the memory segment must attach it before the
shmctl(IPC_RMID) call is made, otherwise their shmat() calls will fail.
But this isn't really a workable solution either. Suppose you have a
collection of RPC-based server processes that will perform some
computations on shared memory segments. Suppose that there is a control
process that allocates some shared memory segments, makes RPC calls to
server processes as required, detaches the shared memory segments, and
then repeats. The shmctl(IPC_RMID) solution doesn't work in this model,
unless the control process sends an RPC to every server process, asking it
to attach to all the shared memory segments, even if the server process
will never be used! This is ugly and slow.
Has anyone found a clean solution to this problem?
"System V Shared Memory -- The Memory that Wouldn't Die."
Thanks ...
-----
{harvard!adelie,{decvax,allegra,talcott}!encore}!munsell!jwf
Jim Franklin, EPPS Inc., 23 Crosby Drive, Bedford, MA 01730
Phone: (617) 663-2115 x4015
------------------------------
Date: 12 Oct 88 17:02:00 EST
From: "Dave Anderer" <dave@vax.oit.udel.edu>
Subject: Lockdowns for 3/60s?
We're putting a number of 3/60's in a public area of the campus this fall.
The area will be supervised most of the time, but there may be periods we
miss. I can easily see someone wanting to walk out with a whole system or
parts of one (8 mb. of SIMMs or a 16" monitor.)
Does anyone make commercial lockdowns we can use to secure the bases and
monitors (both 16" and 19")?
I could also use ideas for securing the keyboards and mice, though I'm
willing to just lock those in a cabinet if need be.
Thanks.
------------------------------
Date: Wed, 12 Oct 88 19:09:00 PDT
From: versatc!tran@sun.com (Tony Tran 982-4317)
Subject: Registered Domain Address?
I've just recently received a new network number of 192.42.146 from SRI.
How do I go about modifying our system and network so that we can be
considered a registered domain?
We previously used default IP address from SUN (192.9.200.xxx). What is
the advantage/disadvantage of switching to 192.42.146 and having a domain
defined as versatc.COM in terms of uucp map, and inter-domain
communication?
[[ Well, for one, you are guaranteed that your Internet numbers are
unique. This might not be a big deal unles you have a gateway onto some
part of the Internet. I strongly recommend using the new ones. Even if
you aren't connected now, you might be someday, and it could become a big
problem at that time if you don't make the switch now. --wnl ]]
Tony Tran
{sun|pyramid|mips|vsi1}!versatc!tran
------------------------------
Date: Wed, 12 Oct 88 08:51:21 edt
From: chemabs!chemabs!lwv27@tut.cis.ohio-state.edu (Larry W. Virden)
Subject: 386i vs ksh-i?
Can anyone give me a set of changes for ksh-i so that it will work under
SunOS 4.0 on a 386i? Currently, tilde processing, vi editing mode, and
emacs mode all have various problems.
Larry W. Virden UUCP: osu-cis!chemabs!larryV
674 Falls Place, BITNET: lwv27@cas.bitnet
Reynoldsburg, OH 43068-1614 CIS: [75046,606]
------------------------------
Date: Wed, 12 Oct 88 11:50:18 EDT
From: edsews!mancer!wintermute!dharvey@uunet.uu.net (Doug Harvey)
Subject: monitoring logins, forcing password chg?
Does anyone know how to easily force someone to change their password
after a certain period of time?
[[ Someone just asked that. Check recent Sun-Spots digests for ideas.
--wnl ]]
Also, how about maintaining a log of failed login attempts?
Thanks,
Douglas Harvey UUCP: {uunet.uu.net}!edsews!mancer!dharvey
Electronic Data Systems ARPA: edsews!mancer!dharvey@uunet.uu.net
(313) 556-0791
------------------------------
Date: Wed, 12 Oct 88 10:31 DST
From: DOONG@reston.unisys.com
Subject: Large screen display for SUN video output?
I think this matter has already been discussed here, but I don't remember
which issue, and I need the info.
Does anyone have any info on taking the video output of a Sun 3/xx (color
and/or mono) and feeding it to a large screen projection system (6ft.)
I know that the 66Hz refresh and the very high scan rate (60+KHz I think)
would cause some problems. Any info on capable system would be
appreciated.
e-mail: doong@reston.unisys.com or
doong@eneevax.umd.edu or
doong%reston.unisys.com@relay.cs.net
------------------------------
Date: Wed, 12 Oct 88 22:38:21 -0400
From: Henry B.J. Krempel <krempel@pacrat.npac.syr.edu>
Subject: FAX equipment/software?
Has anyone heard of a way to hook up a FAX machine to a SUN and view the
result in a SunView (or NeWS) window? I thought I heard of an X product
once, but I'm looking for either of the other two.
Thanks,
Henry B. J. Krempel <krempel@pacrat.npac.syr.edu>
Northeast Parallel Architectures Center (NPAC)
Syracuse University
250 Machinery Hall
Syracuse, N.Y. 13244
------------------------------
Date: Wed, 12 Oct 88 22:44:30 -0400
From: Henry B.J. Krempel <krempel@pacrat.npac.syr.edu>
Subject: M*cAPP file open tool?
I seem to recall some discussion in Sun-Spots a while ago concerning a
tool to bring up a little scrolling pop-up window for specifying a file to
open (like on the M*cintosh). Frame and Artisan have tools like this.
Through extensive grepping of my spot-archive, I've come up with nothing.
Does anyone know where I can get this code?
------------------------------
Date: Wed, 5 Oct 88 10:44:17 +0100
From: mcvax!phcisa!frans@uunet.uu.net (Francois Staes)
Subject: SNA3270?
Is there anybody on the net who knows of a tool running on top of Sunlink
SNA and which behaves like the WASP application on an IBM-PC.
This means that it would provide a simple command language which can be
used for constructing automatic login script and things like that.
I know of the DAI interface but that gives problems in writing login
scripts : once DAI starts talking at a LU, it is impossible to start the
te3278 on that same LU ( and vice-versa ).
Francois Staes ( ...!mcvax!phcisa!frans )
Philips Int.
C.ISA - AIT
The Netherlands
------------------------------
Date: Wed, 12 Oct 88 12:23:13 BST
From: mcvax!ritd.co.uk!mr@uunet.uu.net
Subject: Discrete modelling tools?
I'm looking for some simple discrete modelling tools (defining and
measuring queue based systems). Nothing very elaborate needed.
If you have any (or now of any) such tools could you please drop me a
line. Public domain programs would be expecially wonderful. Summaries can
be posted if interest is shown.
Martin Reed, Racal Imaging Systems Ltd
uucp: mr@ritd.co.uk,{uunet!ukc!ritd,sun!sunuk!brains}!mr
Global String: +44 252 622144
Paper: 309 Fleet Road, Fleet, Hants, England, GU13 8BU
------------------------------
End of SUN-Spots Digest
***********************