Linux-Misc Digest #95, Volume #27 Mon, 12 Feb 01 12:13:02 EST
Contents:
Re: MS to Enforce Registration - or Else ([EMAIL PROTECTED])
Bash arrays ("Chris West")
Kernel security (2.2.16)? (Stephen Cornell)
Re: Linux Sucks... well not really ("Nils O. Sel�sdal")
Re: Unable to handle kernel NULL pointer (Michael Heiming)
Re: Measuring user satisfaction with Linux OS ("Nils O. Sel�sdal")
Re: Unable to handle kernel NULL pointer ("Nils O. Sel�sdal")
Re: Newbie Question ("Mike Greenwood")
Re: stupid mistake (olliecat)
Re: dlopen problem ([EMAIL PROTECTED])
Re: trouble ppp 2.3.11 & Kernel 2.2.16 (Clifford Kite)
Mail for dummies. (Thaddeus L Olczyk)
GNOME/Enlightenment windowing crashed on me. (Jean-David Beyer)
Re: Unable to handle kernel NULL pointer (Jean-David Beyer)
Re: Difference between "su" and "su -" ? (Floyd Davidson)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.advocacy
Subject: Re: MS to Enforce Registration - or Else
Date: Mon, 12 Feb 2001 14:59:04 GMT
Only partially true. People seem to forget that Linux (while great) is
only a small piece of the picture. Most of "Linux" is really GNU which
has been around for quite awhile, I won't pretent to remember the exact
date but that was the brain child of RMS.
Carson
In article <[EMAIL PROTECTED]>,
jtnews <[EMAIL PROTECTED]> wrote:
> Who cares about politics or government?
> As far as I'm concerned most of the world
> governments impose high taxes and do much
> more than they should do "for the people".
> Everything except law enforcement and the military
> should be privatized and out of government
> hands. Then maybe things will get better.
>
> Science and technology really doesn't "belong"
> to one country or another, to suggest otherwise
> is rather silly. The perceived inequities
> of today can be really fleeting. All it takes
> is a few key employees to move from one country
> to another and to setup a corporation and, bang,
> the balance of power has changed.
>
> Steve Withers wrote:
> >
> > G'day
> >
> > I've been following developments closely on this.
> >
> > It appears that sometime this year all new Microsot software will need
> > to be registered....or it will cease to function.
> >
> > This may offer OS alternatives like Linux a huge toe in the desktop
> > door. With more and more homes being networked and having multiple PCs,
> > how many home and small business users are going to be forced to pay up
> > and stop using the "One CD fits all" approach they use today?
> >
> > In my own case.....I would have to upgrade 7 home PCs every year for
> > both Windows and MS Office.....to the tune of lots of dosh per annum. As
> > it is, I now have 3 of those systems on Linux...and quite happily.
> >
> > But some of my family members are reluctant to give up Windows. I may
> > have to suggest to them that they pay for it in future....as I will only
> > be paying for Linux software from 2001 onward.
> >
> > It is intersting that the US produced Windows......the country with
> > one-party (two faction) politics has also given us no choice on the
> > desktop. While politically diverse Europe with multi-party, proportional
> > systems as the politcal norm, has given us Open Software and Linux....
> >
> > Sort of the illusion of freedom (US politics) vs the reality of freedom
> > (European politics - outside Britain).
> >
> > There is an underlying cultural thing operating here somewhere...... :-)
> >
> > Steve
>
--
Carson R. Wilcox
Senior Architect
DMR Consulting Group
[EMAIL PROTECTED]
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Chris West" <[EMAIL PROTECTED]>
Crossposted-To: gnu.bash
Subject: Bash arrays
Date: Mon, 12 Feb 2001 15:14:55 -0000
I'm trying to create an array from a string using bash.
I've tried to do the following:
$ cat array_test
#!/bin/bash
declare -a myarr
echo "hello world" | read -a myarr
echo "myarr[0]=${myarr[0]}"
echo "myarr[1]=${myarr[1]}"
$ ./array_test
myarr[0]=
myarr[1]=
Any ideas how I can split the string into an array of words with bash?
I'd like to use space or tab as field separators.
------------------------------
From: Stephen Cornell <[EMAIL PROTECTED]>
Subject: Kernel security (2.2.16)?
Date: 12 Feb 2001 15:40:01 +0000
Red Hat have announced a kernel update for versions 6.x; their current
version stood at 2.2.16-something, and they recommend to upgrade to
2.2.17-something. They state that this is in response to three local
root exploits that have been found. However, I haven't seen any
reference to these exploits on any security sites (securityteam or
securityfocus). Accordingly, I don't know whether these affect
generic kernels, or just Red Hat ones (I know they do a fair bit of
tweaking, and their kernels differ from the standard ones on
kernel.org).
I run a self-compiled kernel 2.2.16 on one of my machines - do I need
to upgrade it or not?
--
Stephen Cornell [EMAIL PROTECTED] Tel/fax +44-1223-336644
University of Cambridge, Zoology Department, Downing Street, CAMBRIDGE CB2 3EJ
------------------------------
Reply-To: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
From: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
Subject: Re: Linux Sucks... well not really
Date: Mon, 12 Feb 2001 16:40:14 +0100
"Anurodh Pokharel" <[EMAIL PROTECTED]> wrote in message
news:967sgt$5dg$[EMAIL PROTECTED]...
> well, not really and sort of..
> I was recently having adiscussion with one of my windows using friends
> trying to show him the light.
> In the course of the discussion we came to a point where we had to
> discuss net capabilities. I told him the usual about servers and security
> compared to windows.. then we camp upon browsers.. linux has no good
> browser anymore.. IE unfortunately is faster, more stable and renders
Not faster...
> better than Netscape 4.xx .. mozilla is just big and slow... when it
> shoudl be small and fast... so i find myself with asking why is our OS
> handicapped when it comes to browsers... when it does everything better..
Tried konqueror?
------------------------------
From: Michael Heiming <[EMAIL PROTECTED]>
Subject: Re: Unable to handle kernel NULL pointer
Date: Mon, 12 Feb 2001 16:52:34 +0100
Carsten Huettl wrote:
> Hello,
>
> from time to time my rh6.2 maschine hangs While trying to login it
> says:
> Unable to handle kernel NULL pointer dereference at virtual adress
> 00000762
> current -> tss.cr3 = 00101000, %cr3 = 00101000
> ...
> What is wrong here?
> How do I fix this problem?
>
> TIA
> C.
Hello,
as Jean-David Beyer allready posted, it sounds like a memory problem,
try compiling something big (ie. kernel) to stress your CPU/RAM, if it
stops with the message:
gcc caught fatal Signal 11
You can be ~sure some of you RAM is broken, try to change some RAM, to
find out which is broken.
Good luck
Michael Heiming
------------------------------
Reply-To: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
From: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
Subject: Re: Measuring user satisfaction with Linux OS
Date: Mon, 12 Feb 2001 16:56:30 +0100
<[EMAIL PROTECTED]> wrote in message
news:960hq2$k4o$[EMAIL PROTECTED]...
> Hello to all Linux users. Some of you guys may have read my plea for
> help some of you may not. I am currently completing my master degree in
> Information System and I need respondents from all Linus users.
> I have adopted a EUCS instrument to aid me in my research. The End User
> Computing Satisfaction instrument was formulated by Doll and Torkdazeh
> in 1988 and validated by confirmatory factor analysis. The
> questionnaire may appear abit confusing but it has been proven an
> effective tool in measuring end user satisfaction. Please take the time
> to fill out the short 12 question instrument at
> http://www.gonzalo.net/satisfaction.htm. All opinions are welcome.
> Your help is critical to the completion of my degree.
> Assistance will be greatly appreciated. To all those who have filled
> out the questionnaire...Thank You!
> Mahalo Nui Loa (Thank you very much in Hawaiian)
I cannot really understand how you can "measure user satisfaction" with
that scheme you put up....
------------------------------
Reply-To: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
From: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
Subject: Re: Unable to handle kernel NULL pointer
Date: Mon, 12 Feb 2001 16:59:06 +0100
"Carsten Huettl" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hello,
>
> from time to time my rh6.2 maschine hangs While trying to login it
> says:
> Unable to handle kernel NULL pointer dereference at virtual adress
> 00000762
> current -> tss.cr3 = 00101000, %cr3 = 00101000
> ...
> What is wrong here?
> How do I fix this problem?
When i got that one it was some bad RAM.. so you better find out which it is
and replace it...
Compiling some C++ progs should now pretty much fail, try that.. remove some
memory(if you have more than one)
start up, compile and see if it still fails...
------------------------------
From: "Mike Greenwood" <[EMAIL PROTECTED]>
Subject: Re: Newbie Question
Date: Mon, 12 Feb 2001 15:54:30 -0000
Hi,
I did this recently to backup and restore the win98 partition on my
win98/Linux box.
dd if=/dev/hda1 of=/home/myimage
This creates a file called /home/myimage which is an image of the partition
/dev/hda1.
Note that if you're doing the transfer between separtate parts of the same
physical disk, you'll want to play with the blocksize option. From memory,
append bs=4M to set a 4megabyte block size. This will minimise disk
thrashing.
Hope this helps
Mike
<[EMAIL PROTECTED]> wrote in message news:9670jl$kba$[EMAIL PROTECTED]...
> What is the best way to create an exact copy of a hard drive partition.
> Both the source and destination partitions are ext2fs formatted.
>
> Thanks in advance
> Kevin
>
>
> Sent via Deja.com
> http://www.deja.com/
------------------------------
From: olliecat <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux
Subject: Re: stupid mistake
Date: Mon, 12 Feb 2001 16:08:51 GMT
Sandy,
Booting from the RH 7.0 cd I have two mount points /mnt/runtime and /mnt/source
- no system. mount reports:
/tmp/cdrom /mnt/source
/tmp/loop /mnt/runtime
There's got to be way to do this without necessitating a re-install, I just
don't know enough about the internals to do it. But there is a good side to all
of this.
As I was typing your reply I realized this was the only machine on my network
that was still running RH 6.1. I popped in the 6.1 install cd and did an
upgrade thereby recovering my libc file. It seems to be working now...
Thanks for everybodies comments and help.
sandy wrote:
> Yes, I think you use F5 after the install screen appears.
>
> Linux boots in a RAM disk and mounts your system
> root partition as something like /mnt/system (do a "df").
> After you straiten some things out you can test by changing
> the root to the proper place (e.g. "chroot /mnt/system")
> and try a couple of test programs.
>
> Oh. I just re-read your post. You are on the correct track but
> I just think you're a little confused. The rescue filesystem is
> in RAM ... this isn't your "real" disk. Your root partition
> on your "real" disk is mounted as indicated above, so you
> need to "cd" down to it.
>
> Your disk partition device files are not under /dev they
> would be under something like /mnt/system/dev (see above).
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: dlopen problem
Date: Mon, 12 Feb 2001 15:57:00 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
> [EMAIL PROTECTED] wrote:
> >
> > Does anyone know why I get a segmentation fault
> > on this code?
> > void* handle = NULL;
> > char name[64]="/home/data/build/ilImagedata.so\0....garbage";
> > if ((handle = dlopen(name, RTLD_LAZY)) ==NULL)
> >
> > name isn't really defined like that, but for the purpose of this
problem
> > it's addequate.
> > If anyone could make a suggestion, I'd really really appreciate it.
> >
> > Sent via Deja.com
> > http://www.deja.com/
>
> There isn't enough code here to know. However, let's say for the
moment
> that in case of failure, you run dlerror, or in the case of success,
> something else. If you have success/failure reversed in your test, it
> will do this. Running dlerror after success is a problem, as well as
> running some other related functions if it really failed. I think the
> key is in what happens immediately after the "if((handle..." stuff,
> within the branches of the "if".
>
> D. Stimits, [EMAIL PROTECTED]
>
I tried getting rid of the if statement altogether
and just running dlopen. I get the same
Segmentation fault error.
This is the code executed if the handle ==NULL
fprintf(stderr, "Couldn't load %s: %s\n",
direntp->d_name,dlerror());
continue;
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Clifford Kite <[EMAIL PROTECTED]>
Crossposted-To: comp.protocols.ppp,comp.os.linux.questions
Subject: Re: trouble ppp 2.3.11 & Kernel 2.2.16
Date: Mon, 12 Feb 2001 09:27:24 -0600
Rao Garimella <[EMAIL PROTECTED]> wrote:
> I am having problems using pppd (ppp 2.3.11 [and ppp 2.4.0]) with linux
> kernel 2.2.16 (both from the slackware 7.1 distribution.).
> When I try to fire up a ppp connection (using the script ppp-go generated by
> the pppsetup program), pppd dies and the file /var/log/syslog has a message
> from the "chat" program saying that "Can't get terminal parameters -
> Input/output error".
There are various reason for this message. It means that chat can't
get parameters that it needs from the device file, /dev/ttyS0 in your
case. Check to make sure that it is configured correctly and that
no other device is using it or it's IRQ.
> I tried to do it manually using minicom and manual starting of pppd. I
> started minicom and dialed into my ISP which initialized PPP. Then I quit
> _without_ resetting minicom and started pppd with the command "pppd -detach
> /dev/ttyS0 38400 &". I get the following error:
> ioctl(SIOCSIFMTU): No such device 19
This means that an attempt to set the PPP interface MTU failed because
there wasn't a PPP interface (19 is the "No such device" message number).
> tcflush failed: Input/Output error
This means that the ttyS0 buffer couldn't be cleared before pppd
terminated.
> I've setup PPP several times before with multiple machines and modems with
> Kernel 2.0.33 and an older version of PPP. So clearly something has changed
> in the kernel or PPP that is messing it up. Just don't know what!
It's not the kernel or pppd, unless one has been modified from the
standard editions. I have pppd 2.3.11 and the 2.2.16 kernel on a box
here and they work fine together.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* A salute to Inspector Baynes, of the Surry Constabulary, the only
police Inspector to ever best Mr. Sherlock Holmes at his own game.
"The Adventure of Wisteria Lodge", by Sir Arthur Conan Doyle. */
------------------------------
From: [EMAIL PROTECTED] (Thaddeus L Olczyk)
Crossposted-To: comp.os.linux.networking
Subject: Mail for dummies.
Date: Mon, 12 Feb 2001 16:19:11 GMT
Reply-To: [EMAIL PROTECTED]
Two things:
1) I would like to change the way I handle email, but
I have a ton of email in a proprietary email client
( Forte' Agent ) that I do not believe I can convert.
I have one idea of a way of converting, forward the
mail to myself. The problem is that then the email
becomes useless because it does not "display" the
original headers ( they probably appear in the message
but not in the From, and Subject when I browse my email ).
Is there someway I can read them and strip the "wrapping"
headers?
2) I am generally not that familiar with email ( I know the basics,
there are protocols pop3, smtp etc. but that's about it ). Is
there some link that describes the components of an email system
(MTA, MDA etc. ) well enough for a guy like me to figure it out
well enough to setupmy own one/two person system?
TIA
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: GNOME/Enlightenment windowing crashed on me.
Date: Mon, 12 Feb 2001 11:46:11 -0500
I was working in Netscape and an Xterm when my monitor clicked off and
tried to restart. No success. /var/log/messages had this to say:
Feb 12 11:27:57 valinux gnome-name-server[32582]: input condition is:
0x10, exiting
Feb 12 11:27:57 valinux PAM_pwdb[31319]: (gdm) session closed for user
jdbeyer
Feb 12 11:27:57 valinux gdm[31319]: gdm_auth_user_remove: /home/jdbeyer
is not owned by uid 0.
Feb 12 11:27:57 valinux gdm[31319]: gdm_auth_user_remove: Ignoring
suspiciously looking cookie file /home/jdbeyer/.Xauthority
Feb 12 11:28:02 valinux gdm[1498]:
gdm_slave_windows_kill_ioerror_handler: Fatal X error - Restarting :0
Feb 12 11:28:06 valinux gdm[1504]:
gdm_slave_windows_kill_ioerror_handler: Fatal X error - Restarting :0
Feb 12 11:28:10 valinux gdm[1510]:
gdm_slave_windows_kill_ioerror_handler: Fatal X error - Restarting :0
[snip]
This went on for a while, and I turned off my monitor, since it did not
seem to like it. I did not succeed in getting to a console with
Control-Alt-PF1, but I may have messed up in annoyance. I ssh'ed in from
my other machine, put this one in run level 3. I then turned on this
monitor and got a console. I did init 5, and everything is OK again.
Any idea what "gnome-name-server[32582]: input condition is: 0x10,
exiting" is all about. I.e., what is an input condition, and what is the
significance of 0x10?
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 11:35am up 14 days, 20:03, 5 users, load average: 2.12, 2.09, 2.08
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: Unable to handle kernel NULL pointer
Date: Mon, 12 Feb 2001 11:50:48 -0500
Jean-David Beyer wrote:
>
> Carsten Huettl wrote:
> >
> > Hello,
> >
> > from time to time my rh6.2 maschine hangs While trying to login it
> > says:
> > Unable to handle kernel NULL pointer dereference at virtual adress
> > 00000762
> > current -> tss.cr3 = 00101000, %cr3 = 00101000
> > ...
> > What is wrong here?
> > How do I fix this problem?
> >
> > TIA
> > C.
>
> I would guess memory problems, at least as the first thing to check. If
> that is not it, possibly hard drive or hard drive controller problems.
> You are sure you have not been cracked, right?
>
P.S.: there is a memory test program, called memtest86 that seems pretty
good. Since I have 512 Megabytes of memory, it takes over 2 hours to
run. If your machine is up enough, you might wish to download it.
http://reality.sgi.com/cbrady_denver/memtest86/
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 11:45am up 14 days, 20:13, 5 users, load average: 2.07, 2.09, 2.09
------------------------------
From: Floyd Davidson <[EMAIL PROTECTED]>
Subject: Re: Difference between "su" and "su -" ?
Date: 12 Feb 2001 06:54:03 -0900
Arctic Storm <[EMAIL PROTECTED]> wrote:
>When I change to root super user, I usually issue the command
>su -
>And this has worked for me well.
>What's the difference between "su -" and "su" for root super user?
You are getting a number of responses, all of which are either
somewhat wrong or incomplete.
Read the man page! It says,
The optional argument - may be used to provide an
environment similiar to what the user would expect
had the user logged in directly.
Specifically that means su will retain the environment of the
shell which invokes it, except when the - option is used and
then the environment will be the same as a login shell for the
user su'd to.
So when does it make a difference? Well... maybe or maybe not in
several different situations.
Consider a large machine, where several different individuals
are responsible for various systems admin tasks. Each of them
wants to work in their own customized environment of course, and
obviously the root login shell cannot be made suitable for all of
them. Hence that root login shell might be a very bare bones
environment, and none of the admin users will ever use "su -" to
become root. Instead they all use just "su", and retain their
own environment (which with a few tricks in the init files can
be changed somewhat when the user is root).
On a workstation with only one person accessing root, root can be
customized in one specific way, and "su -" would be very
convenient to switch between a normal userid and root.
Another example might be when a system admin is asked to trouble
shoot a problem for a user. There is clearly no point in su'ing
to that user's account and running an errant program with the
environment being used by root, so using the - option will
provide the systems admin with an environment identical to the
user with a problem.
--
Floyd L. Davidson <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska) [EMAIL PROTECTED]
------------------------------
** 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 by posting to comp.os.linux.misc.
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
******************************
