Linux-Misc Digest #168, Volume #21 Mon, 26 Jul 99 04:13:12 EDT
Contents:
Re: Numbering Xterms? (Warren Bell)
Listing Problem? ("Alok R Saboo")
Re: Umount fails on /mnt/cdrom (Stefan van der Eijk)
Re: Position X windows through cli (Rob Brown-Bayliss)
Calerda 2.2 Vs Mandrake 5.3 (A Dumbass)
A question about Bochs and ext2fs? ("Groman")
Re: Karl Marx was fat and hairy chap (Richard Kulisz)
kernel panic: unable to mount root fs on 03:42 (Scorpio)
Re: Where is NULL/__null? (Konrad Hambrick)
Re: A question about Bochs and ext2fs? (Sam Holden)
"who" problem (Ilan Finci)
Re: Shortcomings of Linux? (Darren Winsper)
Re: Where is NULL/__null? (Mark Brown)
Re: Numbering Xterms? (Cameron Hutchison)
Re: Shortcomings of Linux? (Javier)
GPHIGS for Linux ? (Herve Le Cornec)
----------------------------------------------------------------------------
Date: Sun, 25 Jul 1999 21:07:27 -0700
From: Warren Bell <[EMAIL PROTECTED]>
Subject: Re: Numbering Xterms?
Patrick M. Geahan wrote:
>
> Well, couldn't you hold an environment variable? Call it XTERM_Number or
> something, put the number one or zero in it to start, then exec each xterm
> from a shell script that increments the variable by one. Use said
> environment variable in the title.
>
Could you give me a rough idea how to do this? How would I increment the
number back as I'm closing down xterms. This sounds like it would be a
little tough.
Thanks
> --
>
> -------Patrick M [EMAIL PROTECTED]:3784715----------
> USENET Quote of the Week: "I'm still pondering whether i should
> pre-emptively register 'I can't believe it's not Jesus' as a name for a
> low-calorie communion wafer" - Tanuki on alt.sysadmin.recovery
------------------------------
From: "Alok R Saboo" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Listing Problem?
Date: Mon, 26 Jul 1999 09:43:19 +0530
Hello,
Is there any command in Linux similar to "dir/s" option in DOS?
Suppose we want to list all files with extension .EXE we could give "dir
*.exe/s" in DOS, what is the substitute command for Linux?
Thanx,
>From Alok.
------------------------------
From: Stefan van der Eijk <[EMAIL PROTECTED]>
Subject: Re: Umount fails on /mnt/cdrom
Date: Mon, 26 Jul 1999 07:19:19 +0200
Hi Brad,
Did you try:
fuser -k /mnt/cdrom ??
This should free up the cdrom... anyway, also have a look
at "fuser" manpage.
Greetings,
Stefan van der Eijk
[EMAIL PROTECTED]
> Hello,
>
> I'm using Redhat 6.0 /w kernel 2.2.10. Mount/umount works
> fine on /mnt/cdrom until another unix machine nfs mounts
> and then umounts the export. At this point umount reports
> the device is busy. I am sure /mnt/cdrom is not still
> remotely mounted and that nothing locally is using
> the mount. Fuser reports that user as kernel even
> though fstab has the user option set and the mount
> was mounted by a user. The only way to remove the
> cd is to reboot. I know there are some patches out
> there for the nfs but I am not sure where the problem
> is. Does anyone have any thoughts on this problem?
>
> Thanks
------------------------------
From: Rob Brown-Bayliss <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Position X windows through cli
Date: Mon, 26 Jul 1999 17:04:38 +1200
Peter Englmaier wrote:
>
> Most X-programs accept the '-geometry' option for specifying
> size and position of windows. See the man page for X.
Alas it seems most wm dock apps don't accept it, I have found only one.
Is it possible to setup a window manager for the Xlogin screen?
--
[EMAIL PROTECTED]
Zoo Station
--===<|>===--
------------------------------
From: A Dumbass <[EMAIL PROTECTED]>
Subject: Calerda 2.2 Vs Mandrake 5.3
Date: Sun, 25 Jul 1999 22:33:32 -0700
Hi
I got a copy of Mandrake. I went to buy a book and I was hard up to find
a book for Mandrake. But I found many books that came wirth either Red
Hat 5 or Caldera OpenLinux 2.2. I dont want Red Hat because it is just a
stripped down Mandrake and Caldera comes with PartionMagic and KDE
(which I really wanted). But what I was wondering is which would be
better for me (I am a NEWBIE) and are there any good books anyone can
recommend?
PS What is a good modem for linux (my current is a soft modem) ??
Thanx in advance,
Jeremiah
--
"Awww cram it churchy!!"
-Kirk Van Houten
------------------------------
From: "Groman" <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,alt.os.assembly,alt.os.development,comp.os.linux.help
Subject: A question about Bochs and ext2fs?
Date: Sun, 25 Jul 1999 23:00:02 -0700
hello. I just installed Bochs x86 emulator
for windows
and I can't make an image of my ext2fs bootable
floppy with dskimg.exe
it says unknow diskette type detected...
any suggestions?
------------------------------
From: [EMAIL PROTECTED] (Richard Kulisz)
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.advocacy,gnu.misc.discuss
Subject: Re: Karl Marx was fat and hairy chap
Date: 22 Jul 1999 06:30:18 GMT
In article <[EMAIL PROTECTED]>,
Jay Maynard <[EMAIL PROTECTED]> wrote:
>No, the question is *who* and how much. You want to screw everyone who has
>more than you do. Your writings smack more of jealousy and envy than
>anything else.
No. I want to screw anyone who has *power* over me. This is simple self-
preservation. I also recognize that *humans* want to screw anyone who
has more than they do.
>"Hello, Mr. Kettle? I'm Mr. Pot. You're black."
I never had a problem calling the kettle black.
>>And why the bloody hell shouldn't workers take the *wealth* /they/
>>created in the first place?!
>
>Because they *didn't* create all of it, yet you would have them take it all
>anyway?
They *did* create -*ALL*- of it. And even if they didn't, it would be a far,
*FAR* lesser injustice for workers (most of the population) to take all wealth
than be left with the current less than 10%.
>I suspect you wouldn't sully a copy of a Tom Clancy book with your hands,
>but, just in case you would, I'd suggest you read pages 365-366 of Executive
>Orders. He debunks the "working man" argument quite thoroughly.
Tom Clancy's work is pure militaristic propaganda. His message is "Trust
the CIA and the Army, it's made up of Human Beings, *Just Like You*". IOW,
Tom Clancy is a fascist or neo-fascist.
------------------------------
From: Scorpio <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.setup
Subject: kernel panic: unable to mount root fs on 03:42
Date: 26 Jul 1999 05:31:15 GMT
Hi all.
I've been runnuing 2.2.10 kernel for a while, and had to replace a
motherboard in my machine. After that, the kernel would not boot and
produces teh message:
Partition check:
hda: hda1
hdb: hdb1 hdb2 hdb3 <> hdb4
hdc: hdc1 hdc2 <hdc5>
[MS-DOS FS Rel. 12, FAT 16, check=n, conv=b, uid=0, gid=0,umask=022]
[me=0x1b, cs=6649, #f=221 , fs=23683, fl=1714243, ds=379592282, de=37899,
data=37959, se=38384, tf=-570602308, ls=56158, rc=0,fc=4294967295]
Transaction block size = 512
[MS-DOS FS Rel. 12, FAT 16, check=n, conv=b, uid=0, gid=0,umask=022]
[me=0x1b, cs=6649, #f=221 , fs=23683, fl=1714243, ds=379592282, de=37899,
data=37959, se=38384, tf=-570602308, ls=56158, rc=0,fc=4294967295]
Transaction block size = 512
Kernel panic: VFS: unable to mount root fs on 03:42
What is interesting is that 2.0.36 kernel boots perfectly.
I've noticed that htere is a difference in partitions being detected,
because 2.0.36 detects these:
Partition check:
hda: hda1
hdb: hdb1 hdb2 hdb3 <hdb5> hdb4
hdc: hdc1 hdc2 <hdc5>
but 2.2.10 misses teh hdb5.
I've recompiled the kernel couple of times. Included all IDE support htere
was, checked the BIOS for dangerous settings...nothing.
Any leads on what I should check and pay attention to in order to get the
kernel booted?
Running 2.0.7 libc Debian system.
The board is 430VX chipset.
TIA
Andrew
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: [EMAIL PROTECTED] (Konrad Hambrick)
Crossposted-To: uk.comp.os.linux
Subject: Re: Where is NULL/__null?
Date: 26 Jul 1999 06:21:47 GMT
In article <7nfgka$8r2$[EMAIL PROTECTED]>,
John Winters <[EMAIL PROTECTED]> wrote:
>In article <7nfc1c$[EMAIL PROTECTED]>,
>Konrad Hambrick <[EMAIL PROTECTED]> wrote:
>[snip]
>>Here is a handy trick to find out which ones you might need
>>from the standard /usr/include/ heirarchy ( it is one way to
>>do it, that is ;-):
>>
>>[tim@tim source]$ srcdir=`pwd` # optional -- save your place
>>[tim@tim source]$ export srcdir # optional
>>[tim@tim source]$ cd /usr/include
>>[tim@tim include]$ egrep '^ *#define *NULL ' `find . -name "*.h" -print` |less
>>[tim@tim include]$ cd $srcdir # optional -- restore your place
>>
>>replace the '^ *#define *NULL ' with whatever defined constant, etc
>>you want to know about.
>>
>>Here is the output on my RH 5.2 box:
>>
>>../libio.h:#define NULL (__null)
>>../libio.h:#define NULL ((void*)0)
>>../libio.h:#define NULL (0)
>>../malloc.h:#define NULL 0
>>../malloc.h:#define NULL ((__malloc_ptr_t) 0)
>>../stdio.h:#define NULL 0
>>../stdio.h:#define NULL (void*)0
>>../g++/streambuf.h:#define NULL (__null)
>>../g++/streambuf.h:#define NULL (0)
>>../pgsql/c.h:#define NULL ((void *) 0)
>>../pgsql/c.h:#define NULL 0
>>../python1.5/mymalloc.h:#define NULL ((ANY *)0)
>>../python1.5/object.h:#define NULL 0
>>../tcl.h:#define NULL 0
>
>The trouble with this sort of empirical method of working out where
>something should be is that it will often yield the wrong answer.
>A better method is simply to check the language specification.
>
>For C, the macro NULL is defined in stdio.h and stddef.h
>
>Quite a few of those headers above are arguably illegal for attempting
>to define the NULL macro. If they need the NULL macro they should include
>the relevant headers. More importantly, if you pick one of them and
>include it in order to get the NULL macro, your code may well break
>on a different or upgraded installation and it'll be your problem. If
>you include stddef.h to get the NULL macro and it later disappears from
>there you can with justification complain to the implementation provider.
>
>John
John --
I agree 100%.
OTOH, RH 5.2 has no stddef.h under the /usr/include/
directory tree.
At times, I have had to revert to this trick in order to
find the necessary include files when porting c-code to
some flavors of unix ( not to mention grepping nm output
to find out which non-std libraries I needed to link to
end up with a runable binary )
Anyhow. You are correct. Check the language spec first
but when that approach fails, tricks like grep *.h and
nm /lib/lib* can produce results while the language lawyers
work out the gory details.
-- kjh ( still a hacker after all these years ;-)
--
============================================================
Konrad J. Hambrick | email: [EMAIL PROTECTED] |
1111 Seacoast Dr. Unit 41 | home: (619) 423-4451 |
Imperial Beach, CA 91932 | |
------------------------------
From: [EMAIL PROTECTED] (Sam Holden)
Crossposted-To: alt.os.assembly,alt.os.development,comp.os.linux.help
Subject: Re: A question about Bochs and ext2fs?
Date: 26 Jul 1999 06:22:06 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 25 Jul 1999 23:00:02 -0700, Groman <[EMAIL PROTECTED]> wrote:
>hello. I just installed Bochs x86 emulator
>for windows
>and I can't make an image of my ext2fs bootable
>floppy with dskimg.exe
>
>it says unknow diskette type detected...
>any suggestions?
cp /dev/fd0 /mnt/dosC/wherever_you_want.img
(after using said disk to boot linux of course)
--
Sam
Just don't create a file called -rf. :-)
--Larry Wall
------------------------------
From: Ilan Finci <[EMAIL PROTECTED]>
Subject: "who" problem
Date: Mon, 26 Jul 1999 05:49:00 +0000
Hi,
I've installed RH 6.0 on a new computer, and for some reason, the
command "who" shows that no one is logged in (although I'm working on
the computer). We have few other computers with RH 6.0 and there "who"
do work.
I also noticed that when someone is rlogged in to the machine - who
shows him (but only him). I've tried to compare few files permissions to
the computers that do work, and I see no difference.
What can cause such problem?
Thanks in advance,
Ilan
--
MobilEye Vision Technologies LTD
24 Mishol Hadkalim st. Jerusalem, 97278, Israel
Tel: 972-2-5866989 Fax: 972-2-5867720 Personal Fax: 1-801-912-3251
Email: mailto:[EMAIL PROTECTED] http://www.mobileye.com
------------------------------
From: [EMAIL PROTECTED] (Darren Winsper)
Crossposted-To: comp.sys.amiga.misc,comp.os.linux.advocacy
Subject: Re: Shortcomings of Linux?
Date: 26 Jul 1999 06:01:43 GMT
On Sun, 25 Jul 1999 20:34:42 GMT, Casper <[EMAIL PROTECTED]> wrote:
> Give it up DUDE! You know NOTHING about Miami, and I've seen from
> your previous post that you are completely CLUELESS, and you continue
> to make an ass out of yourself.
Just out of interest, what are the advantages of Miami over pppd?
--
Darren Winsper - http://easyweb.easynet.co.uk/darren.winsper
"God - error 38522: God does not exist - Warning 71154: the following
characters have been ignored: God." - Some mainframe with a crisis of faith.
------------------------------
From: Mark Brown <[EMAIL PROTECTED]>
Crossposted-To: uk.comp.os.linux
Subject: Re: Where is NULL/__null?
Date: 26 Jul 1999 07:49:34 +0100
[EMAIL PROTECTED] (Konrad Hambrick) writes:
> OTOH, RH 5.2 has no stddef.h under the /usr/include/
> directory tree.
You'll find it in /usr/lib/<gcc-version>/include/stddef.h - it's
provided by gcc rather than the libraries. There should be no problem
with ANSI C support on Linux.
--
Mark Brown mailto:[EMAIL PROTECTED] (Trying to avoid grumpiness)
http://www.tardis.ed.ac.uk/~broonie/
EUFS http://www.eusa.ed.ac.uk/societies/filmsoc/
------------------------------
From: Cameron Hutchison <[EMAIL PROTECTED]>
Subject: Re: Numbering Xterms?
Date: 26 Jul 1999 07:08:22 GMT
Warren Bell <[EMAIL PROTECTED]> writes:
>Patrick M. Geahan wrote:
>> Well, couldn't you hold an environment variable? Call it XTERM_Number or
>> something, put the number one or zero in it to start, then exec each xterm
>> from a shell script that increments the variable by one. Use said
>> environment variable in the title.
>Could you give me a rough idea how to do this? How would I increment the
>number back as I'm closing down xterms. This sounds like it would be a
>little tough.
An environment variable isn't going to work very well, unless you lanuch
each xterm from the same shell. This is because each process has its own
environment and a change in one processes' environment will not be noticed
in another's.
I've just quickly knocked up a script that should do what you want. The
script assumes that you want yout xterms named "xterm n" where n is a
number not starting with 0. I've included 2 versions of the function
"next_num" in the script - one selects the next xterm number by taking the
largest value already on an xterm title and adding one - the other finds
the first free number.
The script uses the xlsclients to get the names of all the windows open
on your X server and bases its decisions on that. This has the advantage
that your xterms will be correctly numbered even if you run xterms from
multiple machines. An envvar or local file wont be able to do that.
Watch out for programs that change the title of the xterm. For instance,
vim changes the title bar when started and then back again when done. If
you start an xterm while the title has changed, then you might end up
with duplicated xterm numbers.
Just cut the following out and stick it in a script, make the script
executable and run this instead of "xterm". You can call the script
"xterm", but if you do make sure you stick the full path of the xterm
binary on the last line (otherwise the script will keep calling itself).
If you want to know more about how this works and cant figure it all
out, mail me and we can discuss it.
Comments welcome.
========cut here==========
#!/bin/bash
#
# Start an xterm with a unique title. Titles are made unique by finding a
# unique number and adding it to the title.
#
# [EMAIL PROTECTED] - 26th June 1999
#
# Note: There are two next_num functions - comment out the one you dont
# want to use.
# This next_num function prints the number that is one larger than the
# largest number in the list. It assumes the list is supplied in sorted
# order (numerically, not lexographically).
function next_num()
{
if [ -z "$1" ] ; then
echo 1
else
# rely on sed matching the longest string it can.
echo $(( $(echo $1 | sed 's/^.* //') + 1))
fi
}
# This next_num function prints the lowest next available number, given a
# list of numbers in sorted order.
#function next_num()
#{
# local LAST_NUM=0;
# for NUM in $1 ; do
# if [ $NUM -ne $(($LAST_NUM + 1)) ] ; then
# break
# else
# LAST_NUM=$NUM
# fi
# done
# echo $(($LAST_NUM + 1))
#}
NEXT_NUM=$(next_num "$(xlsclients -l | grep ' Name: xterm [1-9][0-9]*' | sed 's/^.*
Name: xterm \([1-9][0-9]*\)/\1/' | sort -n)" )
exec xterm -title "xterm $NEXT_NUM" "$@"
========cut here==========
--
Cameron Hutchison ([EMAIL PROTECTED]) | Onward To Mars
GCS d--@ -p+ c++(++++) l++ u+ e+ m+(-) s n- h++ f? !g w+ t r+
------------------------------
From: Javier <[EMAIL PROTECTED]>
Crossposted-To: comp.sys.amiga.misc,comp.os.linux.advocacy
Subject: Re: Shortcomings of Linux?
Date: Mon, 26 Jul 1999 00:02:44 -0700
Chris Lee wrote:
>
> In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> >
> >
> >
> >Chris Lee wrote:
> >>
> >> In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> >> >
> >> >Chris Lee wrote:
> >> >>
> >> >> In article <7n442u$hiu$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> >> >> >
> >> >> >In comp.sys.amiga.misc Chris Lee <[EMAIL PROTECTED]> wrote:
> >> >> >> Funny, what you are claiming doesn't really seem to be the case.
> >> Follow
> >> >> the
> >> >> >> following newsgroups for a while and you'll discover that.
> >> >> >
> >> >> >*grin*. Oh, well...
> >> >> >
> >> >> >I get my information directly from ISPs, from PPP login traces,
> >> >> >from other developers and from IETF workgroups (PPPEXT among them),
> >> >> >and I can assure you that I would not have gone to the trouble of
> >> >> >implementing and field-testing MS-CHAP if it had not been necessary.
> >> >> >
> >> >> >But of course if you feel the random argumentative rambling on
> >> >> >Usenet is more informative and objective then that is your choice.
> >> >>
> >> >> And I get my information from the people who *ACTUALLY CONNECT TO
> ISPs*
> >> in
> >> >> the *REAL WORLD*
> >> >
> >> >Hey, obnoxious moron, you are talking to the guy who PROGRAMMED Miami,
> >> >and several other packages for the Amiga!!!
> >>
> >> Who cares?
> >
> >Who cares if you and people of your type care or not?!?!?
> >
> >If you don't care then why are you trolling here?!?!?!
> >If you don't care then why are you responding to him?!?!???
> >If you don't care why even have a computer?!!?!?
> >
> >> Just like you it doesn't mean he knows a damn thing about what
> >> the people who don't use the Amiga use to connect to the internet.
> >
> >So someone who shipped one of the best TCP/IP stacks on any platforms
> >doesn't know, and a jackass like you who claims real world experience
> >with no clue knows??!?!?!?
>
> Who says it's the best? Amiga users? Give me a break....
>
So you had to quote the entire fucking message to add that useless
comment of yours?!?!??
Yes it is the best, show me otherwise!!!
------------------------------
From: Herve Le Cornec <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: GPHIGS for Linux ?
Date: Mon, 26 Jul 1999 00:49:47 +0000
Hi,
Is the graphic library GPHIGS (from G5G) available
for Linux ?
Thanks.
HCl
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
