SUN-Spots Digest, v4n22
Vicky Riffle
Sun-Spots-Request at Rice.edu
Sat Jul 19 02:24:52 AEST 1986
SUN-SPOTS DIGEST Friday, 18 July 1986 Volume 4 : Issue 22
Today's Topics:
Reply to: problems with arp
More Problems with SunView
Thanks for the SMD responses
ND on a Vax?
Paint program for technical publications?
2.2 listen() problems?
Make bug?
DDCMP?
Function keys in Suntools?
Workstation comparison?
RCS on Suns?
vt100 emulator?
Pac Man?
------------------------------------------------------------------------
From: dredge at zotz.STANFORD.EDU (Michael Eldredge)
To: Sun-Spots at Rice.edu
Subject: For Sun-Spots, a reply to someone's problem....
> Date: Wed, 9 Jul 86 14:28:33 edt
> >From: seismo!allegra!phri!roy at SALLY.UTEXAS.EDU (Roy Smith)
> Subject: Problems with arp?
>
> If I try to reboot alanine, it hangs with "tftp timeout" until I
...
> do an "arp -d alanine" on goober and that allows the boot to continue.
> There definately seems to be something wrong with arp. For example,
> looking at goober's arp table right now:
>
> goober% /etc/arp -a
> glutamine (192.9.200.16) at 8:0:20:1:ff:f5
> wombat (192.9.200.1) at 8:0:20:1:4f:2b temporary
> glycine (192.9.200.17) at 8:0:20:1:ff:f6
> histidine (192.9.200.18) at 8:0:20:1:ff:f3
> uracil (192.9.200.6) at 8:0:20:1:5e:30 temporary
> clone (192.9.200.11) at 8:0:20:1:47:ed
> alanine (192.9.200.12) at a:0:0:0:0:0 <====== this is not kosher!
> arginine (192.9.200.13) at a:0:0:0:0:0 <====== neither is this!
> asparagine (192.9.200.14) at 8:0:20:1:5e:96
> cysteine (192.9.200.15) at 8:0:20:1:ff:f4
>
> So, what did I do wrong?
I just finished fooling with this for sometime. We have several suns:
gloworm(server), pioneer, zotz, talbots, goose, bbc
Everytime bbc and only bbc would reboot, it would hang with TFTP timeout errors
and as soon as we logged on to the server and did 'arp -d bbc', BINGO- it
fininshed the boot.
The problem is in 'ifconfig'. Sun's version seems to have a really bad
heuristic for deciding if a word is an ether address, an ip address or
a hostname. The three are very different, and very easy to tell apart:
ether: 8:0:20:1:ff:f5
ip: 36.2.0.81
host: bbc
Certainly if you look at our hostnames, it is obvious where the problem
lies. The only host with problems is 'bbc'. The first character is
(looks like) a hex digit.
ifconfig is usually used to set the ip address, but if given an ether
address, it will override the kernel's idea of what the ether address
is. So in our case: 'bbc' was taken to be 'bbc:0:0:0:0:0'. It
would send out TFTP request with a return ether of 'bc:0:0:0:0:0'
(0xbbc gets truncated to the low order byte of 0xbc'), the server
sends it back, but ofcourse the hardware is listening on its
real ether address and never sees the responce.
The fix, until sun fixes it is (well a fix is, or my fix is) to make sure
that 'ifconfig' never gets confused. This can be done by always giving
it the ip address. To automate this, I have changed '/etc/rc.boot' to:
#
# @(#)rc.boot 1.1 85/12/18 SMI
#
# Executed once at boot time
#
hostname=bbc
add: ipadder=`ypmatch $hostname hosts | sed 's/[ ][ ]*.*$//'
> /etc/mtab
/etc/mount -f /
/etc/mount /pub
/bin/hostname $hostname
mod: /etc/ifconfig ec0 $ipadder -trailers up
mod: /etc/ifconfig ie0 $ipadder -trailers up
mod: /etc/ifconfig le0 $ipadder -trailers up
/etc/umount -at nfs
sync
Michael Eldredge
dredge at zotz.stanford.edu
-----------------------------
Date: Fri, 11 Jul 86 18:20:03 -0500
From: Christopher Byrnes <cb at mitre-bedford.ARPA>
Subject: More Problems with SunView
I've noticed a few more "minor" problems when trying to program in
SunView 3.0 on our Sun 3's. I'd like to find out if anyone else has
had these problems and if they have come up with better solutions than
I did. These problems have made me increasingly suspicious about
what's in the SunView documentation.
I've noticed that the cursor position in a text window is not always
updated when the cursor position is moved in response to an event
generated from another window. These events result in
textsw_possibly_normalize() being called, with the insertion point
(cursor) being placed somewhere in the normalized window. If it turns
out that textsw_possibly_normalize() doesn't have to do any scrolling,
then the movement of the insertion point does not take effect until
the mouse is moved (from where ever it currently is outside the text
window) into the text window. This is awkwards if you're trying to
use the text window's cursor as a pointer to text that is somehow
related to some activity you're performing in another window.
Has anyone gotten menus containing MENU_GENERATE_PULLRIGHT_ITEM
items to work like they're suppose to? No matter how much I tried, I
could not get an action/notify procedure to be called after the user
makes a choice from one of the items in the generated pullright menu.
I tried associating action and/or notify procedures to the main
generated pullright menu and/or the individual items within that menu.
For some reason the procedure I'd like to have called never was. I
got around this problem by placing the action/notify procedure inside
the menu generator procedure when the "operation" is MENU_DESTROY. I
assume that MENU_DESTROY is done after the user makes a choice, so the
MENU_SELECTED_ITEM seems to be still valid. Am I overlooking
something, or am I making dangerous assumptions about how generated
pullright menus work.
I've also noticed that the menu is generated and displayed
immediately when the parent menu (the one containing the
MENU_GENERATE_PULLRIGHT_ITEM) is displayed. You'd think that the menu
would not be generated and displayed until the user moves the mouse
to the end of the ITEM (where the right arrow is). I've tried to set
MENU_INITIAL_SELECTION_SELECTED to FALSE without any result. This
doesn't seem to happen with other (non-generated) pullright menus.
I wish the Notifier was more generous when clients try to make
reentrant calls to notify_start(). My problem came when I tried to
have one top level frame create a new frame (and its associated
windows), passing control to the notifier through window_main_loop().
To allow both frames to run concurrently, a UNIX fork() called is
issued to create a new UNIX process that the second window can run in.
The problem comes when the second (new) frame calls
window_main_loop(). When notify_start() is called within
window_main_loop(), the Notifier thinks that another frame is still
running because this new UNIX process inherits all the data (including
whatever flag is set to prevent notify_start() from being called
twice). No amount of trickery such as calling notify_stop() to break
out of the original call to notify_start() or trying to avoid
notify_start() by calling notify_dispatch() will work. The SUN User
Support people inform me that you cannot create new frames/windows
this way, you have to use wmgr_forktool() (like "suntools" itself does).
This restriction makes programming in SunView awkwards when you are
making heavy use of asynchronously created and terminated UNIX
processes. For example, wmgr_forktool() doesn't report the PID of the
UNIX process which is forked() to run, making control of you processes
more difficult. Is there anyone else out there who thinks that these
restrictions are a problem?
-----------------------------
Date: Tue, 15 Jul 86 07:43:53 EDT
From: scott at mitre-gateway.arpa (John A. Scott at vlsi-a)
Subject: Thanks for the SMD responses
Shortly after I sent out my plea for help I discovered my problem.
It seems that the same controller is unable to use an eagle disk drive
and the CDC (9766) disk drive at the same time (the boot program complains
about not being able to find the eagle disk label). This steams from
the fact that the disk drives use their own "standard" interface (I think).
For those of you that already knew this, my humble apologies.
--John
-----------------------------
Date: Sun, 13 Jul 86 12:28:21 edt
From: Ken Mandelberg <km%emory.csnet at CSNET-RELAY.ARPA>
Subject: ND on a Vax
Can anyone give me a pointer to where I can get ND for a Vax?
It needs to run on 4.3BSD Vax, and support diskless Sun 3/50s.
Thanks
Ken Mandelberg
Emory University
Dept of Math and CS
Atlanta, Ga 30322
{akgua,sb1,gatech,decvax}!emory!km USENET
km at emory CSNET
km.emory at csnet-relay ARPANET
-----------------------------
Date: Mon, 14 Jul 86 12:03:41 pdt
From: bake at ee.UCLA.EDU (Dr Rich Baker)
Subject: Paint program for technical publications?
Last month I asked for pointers to a good program for creating technical
drawings for conference papers, technical journals, etc., that runs
on the Sun and creates ditroff or postscript for laserprinting. It should
be able to do block diagrams, schematics, simple charts, etc., with
typeset legends.
I received many responses, but unfortunately they all were requests for
what I found out. So lots of us are interested in such a program . . .
Could I ask a second time that those of you who have a satisfactory
program to take a minute to comment on it?
I'll post results.
R. Baker
bake at ee.ucla.edu
{ucbvax!} ucla-cs!bake
-----------------------------
Date: Mon, 14 Jul 86 14:16:27 PDT
From: hight%sri-claire at sri-tsc
Subject: 2.2 listen() problems
Has anyone had problems with listen(s, 5) not maintaining a backlog
of tcp connections? It seems that if my process is busy while
new connections come in, my process never sees the new conection
when it gets back around to doing another accept(). In fact, the
client process thinks it has made the connection, and successfully
writes data to the server process. But the server never sees the
connection if it was not waiting at the accept() call. I've even tried
backlog = 0 and 1, to workaround the problem of the "missing connections",
but no luck.
John Hight
SRI International
(415)859-4279
-----------------------------
Date: Wed, 16 Jul 86 08:38:41 edt
From: ramsdell%linus at mitre-bedford.ARPA
Subject: Make bug?
Does anyone know how to write Makefiles that:
1) Run on both Sun 2's and 3's.
2) Allow the specification of an environment variable (via a macro
declaration) to be used as part of a file name on the left hand
side of a rule.
The following obvious solution does not work:
# Link Makefile.
# This makefile works fine on Sun 2's,
# but fails on Sun 3's.
# On a Sun 3,
# make install
# gives:
# Make: Don't know how to make .y. Stop.
# CP is defined in the environment to
# be a directory (it ends with "/").
# The problem goes away if you remove the
# last rule. It also goes away if you
# define ROOT = $(CP) which fails when
# you are using Sun 2's.
ROOT = $${$(SYSTEM)}
SRCS = link.sh
SYSTEM = CP
all: $(ROOT)link
$(ROOT)link: $(SRCS)
cp $(SRCS) $@; chmod +x $@
-----------------------------
Date: 17 Jul 1986 09:39-EDT
From: Katsushi.Ikeuchi at n.sp.cs.cmu.edu
Subject: ddcmp
Does anyone have a software package to handle DEC's DDCMP protocol
on SUN? --- ki
-----------------------------
Date: Thu, 17 Jul 86 18:57:45 EDT
From: Steve M. Burinsky <smb at mimsy.umd.edu>
Subject: Function keys in Suntools
Not being a SunView programmer, I haven't been able to figure out how I would
do the following:
(1) Assign a function (e.g. "zoom window to full screen") to some
currently unused function key, and
(2) Turn off the special meanings of function keys like F7 (open/close
window) in A SPECIFIED WINDOW. I know I can mess around with the keyboard
tables or use setkeys, but this changes the keyboard interpretation for the
entire suntools environment (not just one window). I want to port my
favorite editor to the Sun, and would like to have F7 perform an editor
function. Currently, the editor never sees the escape sequence. It is
swallowed by suntools, and my window closes.
Any input would be appreciated. I have looked at /usr/src/sun/suntool/
{shell,sun}tool.c, but haven't found any clues. Granted, they could be there;
they just haven't jumped up and waved at me.
Thanks,
Steve
smb at mimsy
-----------------------------
Date: Fri, 18 Jul 86 11:55:07 +0200 (Central European Summertime)
From: XBR2D96D%DDATHD21.BITNET at rice-goliath.arpa (Knobi der Rechnerschrat)
Subject: Workstation comparison
Hallo,
we (see below) consider buying one or two high performance 32-Bit graphic
workstations.
During the process of selection and evaluation we found in the "Dataquest
Research Newsletter 1986-7" a comparison between Apollo-Domain (DN580),
Sun 160C and 3 other workstations. There is a table that says:
Relative performance (1 is best, 5 is worst)
Apollo DN580 xxx yyy zzz Sun 3/160C
Computing 3 4 5 2 1
Graphics 1 5 4 2 3
Networkng 1 3 5 4 2
Applic. 2 5 4 1 3
My questions are:
a.) Where does the 3 for Apollo Computing come from?
b.) Where does the 3 for Sun 3/160C graphics come from?
I would be most grateful if people who really know the answer (Dataquest)
could send me a short (or long) explanation.
PLEASE respond directly to <XBR2D96D at DDATHD21>, because I'm not a member
of this list.
Regards
Martin Knoblauch
TH-Darmstadt
Dept. Physical Chemistry 1
Petersenstrasse 20
D-6100 Darmstadt
West-Germany
BITNET: <XBR2D96D at DDATHD21>
-----------------------------
Date: Fri, 11 Jul 86 15:53:45 EDT
From: penny at CCA.CCA.COM (Penny Rheingans)
Subject: RCS on Suns
Has anyone out there been successful in geting rcs to run on a Sun?
We're using Sun 2's running Version 3.0 and would like to use rcs to
manage our source code. Unfortunately, the port seems to involve much
more than just moving the sources and remaking, as the rcs code seems
to be machine dependent.
Have any of you done this successfully?
Penny Rheingans
penny @ cca.cca.com
-----------------------------
Date: Mon, 14 Jul 86 22:40 EDT
From: HUDGENS%FSU.MFENET at LLL-MFE.ARPA
Subject: vt100 emulator
Are you aware of any information about the availability of a
'vt100-window' for a Sun workstation, which correctly emulates the
characteristics of a vt100? We currently do a lot of work on a Vax from
a Sun-2, and must edit in line mode, with the terminal set as 'unknown'.
We have two choices: one is to develop a 'vt100-window' for the Sun, and
the other is to find some software for the VAX that will emulate the Sun.
Any responses would be appreciated.
Jim Hudgens
-----------------------------
Date: Sat, 12 Jul 86 14:10 EDT
From: PACE%FSU.MFENET at LLL-MFE.ARPA
Subject: pac man
Someone on the net a few weeks ago said they had ported a pac-man
like game to the sun3. Could someone please tell me where I can get
this or who to contact.
thanks
Don Pace
Florida State University
-----------------------------
End of SUN-Spots Digest
***********************
More information about the Mod.computers.sun
mailing list