Linux-Misc Digest #517, Volume #21 Mon, 23 Aug 99 14:13:13 EDT
Contents:
[ `id -gn` = `id -un` ] and "[: Command not found." (Habibi4me)
Re: Does hotmole still work with MS HoTMaiL (Jayan M)
linear raid trouble (nils grimsmo)
Re: anyone certified? (toby)
Re: Random Lockups and Freezes (John Girash)
Re: Is there a website for minimalist Linux users? (Jayan M)
Re: WTF is the difference between Linux and FreeBSD??? (John Girash)
Re: ISDN or 2-56k Modems and RH 6.0 ([EMAIL PROTECTED])
Re: Change LILO from MBR to a local partition? (Leonard Evens)
Re: Default mkdir attributes ("Christopher W. Aiken")
Re: No core file (Paul D. Smith)
win95 backup on Linux ([EMAIL PROTECTED])
Re: Can OS/2 users grow up and think like Linux users? (was: Can I switch from OS/2
to Linux and be happy?) (John L. Daschbach)
Re: Default mkdir attributes (Mapu)
----------------------------------------------------------------------------
From: Habibi4me <[EMAIL PROTECTED]>
Subject: [ `id -gn` = `id -un` ] and "[: Command not found."
Date: Mon, 23 Aug 1999 11:15:41 -0400
I have the following expression:
[ `id -gn` = `id -un` ]
and the tcsh from SuSE-6.1 distro complained "[: Command not found."
when executed. However, it just works fine under tcsh from the
RedHat-6.0 distro. What can go wrong with the tcsh from SuSE-6.1
distro?
Please kindly help me as I posted similar question a month ago and got
no response from this NG. TIA.
--
[EMAIL PROTECTED]
PS. Remove "4m" from e-mail address to enable reply.
------------------------------
From: Jayan M <[EMAIL PROTECTED]>
Subject: Re: Does hotmole still work with MS HoTMaiL
Date: Mon, 23 Aug 1999 14:47:40 GMT
Sorry, my mistake I did not read it properly..
Jayan
Richard Thomas wrote:
> On Thu, 19 Aug 1999 13:55:16 GMT, Jayan M
> <[EMAIL PROTECTED]> wrote:
>
> >"Pete G." wrote:
> >
> >> Does anyone know if hotmole 1.0 still works with Microsoft's Hotmail?
> >>
> >> I know that Microsoft has made some security enhancements over the past
> >> month or two and the version of hotmole I have is from April. At any rate
> >> ... I can't seem to get hotmole working. (see below) And the author's
> >> supportURL is DOA; which leades me to belive hotmole is no more.
> >>
> >> Any leads?
> >>
> >> (FYI: Hotmole was a GREAT program for reading/forwarding mail from Microsoft
> >> Hotmail. Like a simple version of fetchmail for HTTP)
>
> >Hey buddy,
> >
> >This is a linux newsgroup, not the MICROS~! WINDOW~1
> >
> >shoo shoo...
> >
> >Jayan
> >
>
> Hey buddy,
>
> It was a Linux question.
------------------------------
From: nils grimsmo <[EMAIL PROTECTED]>
Subject: linear raid trouble
Date: Mon, 23 Aug 1999 13:21:56 +0200
running debian 2.1 and linux 2.0.36 :)
i am playing with some disks, and trying to make linear raid work, but
can't. i do this:
# modprobe linear
# mdadd /dev/md0 /dev/hdb1 /dev/hdc1
# mdrun -pl /dev/md0
# mount /dev/md0 /test
i get no error messages, but when i run a 'df', md0 is no bigger than
hdb1 itself. i can't copy more than the size of the first disk to md0
either, and one if there was something on the second disk before i
linked them, it is not deleted as the hdb1 part of md0 is filled
completely. do i have to format the disks in some special way?
thanks
--
nils grimsmo
[EMAIL PROTECTED]
"du skal ikke plage andre, du skal v�re grei og snill,
og for�vring kan du gj�re hva du vil"
------------------------------
From: toby <[EMAIL PROTECTED]>
Subject: Re: anyone certified?
Date: Sat, 21 Aug 1999 16:26:16 -0400
Reply-To: [EMAIL PROTECTED]
Lye Nooks wrote:
> I'm very interested in studying for Linux certificate exam. I know there is
> a certificate for Redhat, and also a separate certificate shown at the
> website www.linuxcertificate.com
>
> [snip about MCSE and books].
>
> Thanks
Certificates are worthless. Unix got along just fine for years without
certifying anybody. You just asked a co-worker about a problem and a possible
solution, or egads, you read the fucking manual. Most MCSE's couldn't fuck their
way out of a wet paper bag, so why would we want to infect the Linux world with
people who crammed for a cursory exam and couldn't solve a realtime production
problem on their feet? It is a racket, at best, and at worst it keeps
perpetuating the need for vendors to create "installation wizards" and "IDEs".
Give me 'insmod' anyday. I'd rather have a tricky driver that does its job for
months on end than a simple to install piece of shit that could have been better
if the software designers didn't have to design a "easy-to-use installation
interface." The operating system is the IDE! People who buy CodeWarrior and shit
like that are only going to infect :Linux with Bloatware that is over coded and
doesn't work. we should make Linux harder to use so stupid people will stay away
from it and keep corporations from screwing it up. Give me my VI or give me
Emacs.
Toby
------------------------------
From: John Girash <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.x
Subject: Re: Random Lockups and Freezes
Date: 23 Aug 1999 11:51:01 -0500
In comp.os.linux.misc Mike Brown <[EMAIL PROTECTED]> wrote:
: I have been having an interrmitent lockup problem with my Linux
: system. Randomly the whole machine will lock. Nothing I can do will
: free it. There is no response from the keyboard at all so I can't get
: to another console or kill X. The mouse completely stops. The only
: thing to get it back is a complete hard system reset.
Hi Mike,
I've had similar problems before, when my X server was pushing my video card
too hard. Try looking in your XF86Config file and choosing more conservative
values for the Options under the video chipset Device; in my case (a CL card)
it was stuff like 'Option "fifo_conservative"' and 'Option "no_imageblt"'.
Yours will have different options I'm sure but there might be some that help.
good luck
jg
--
"don't listen when you're told about the best days in your life Spirit of
a useless old expression, it means passing time until you die." the West
/\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\/
-- John Girash -- girash @ cfa.harvard.edu - http://skyron.harvard.edu/ --
------------------------------
From: Jayan M <[EMAIL PROTECTED]>
Subject: Re: Is there a website for minimalist Linux users?
Date: Mon, 23 Aug 1999 14:46:34 GMT
486 25 MHz, 24MB RAM,
running debian 2.0, newsserver, fvwm for X
160 days uptime.
Another 486 25 MHz, 24MB RAM, 500MB HD, running
XFCE..
Jayan
andy wrote:
> Hey all you hardcore commandline users! Keeping that 486 alive, eh? I'm
> just tired of all these apps intended for the wicked blazing smoking
> kick ass 450Mhz-toting user.
> So how much fun can I have? YOU tell ME. I want to make my pitiful 486
> sweat without making it crawl! No KDE, No GNOME. So what do YOU run on
> YOUR box? Did you make it a game server? Gateway router? What?
> --
> apurugga AT mindspring DOT com
> annandy AT dc DOT seflin DOT org
------------------------------
From: John Girash <[EMAIL PROTECTED]>
Subject: Re: WTF is the difference between Linux and FreeBSD???
Date: 23 Aug 1999 11:45:53 -0500
Jordan <[EMAIL PROTECTED]> wrote:
: Do you know if Minix was a free system or what kind of computers it worked on?
What's this "was" business? From http://www.cs.vu.nl/~ast/minix.html :
WHAT IS MINIX 2.0
MINIX is a free UNIX clone that is available with all the source code.
Due to its small size, microkernel-based design, and ample documentation,
it is well suited to people who want to run a UNIX-like system on their
personal computer and learn about how such systems work inside. It is
quite feasible for a person unfamiliar with operating system internals to
understand nearly the entire system with a few months of use and study.
MINIX has been written from scratch, and therefore does not contain any
AT&T code--not in the kernel, the compiler, the utilities, or the
libraries. For this reason the complete source can be made available (by
FTP or via the WWW).
MINIX has evolved over the years, so several versions exist. Two of these
are still current. The rest are obsolete. The current versions are:
MINIX 2.0 (Intel CPUs from 8088 to Pentium)
MINIX 1.5 (Intel, Macintosh, Amiga, Atari, SPARC)
Note that the last-modification date of that webpage is in 1999. It's not GPL
but is free (both speech- and beer-wise) for education and research purposes.
jg
ps. I'm really glad to see the "corrections" posts here weren't done as flames.
--
"don't listen when you're told about the best days in your life Spirit of
a useless old expression, it means passing time until you die." the West
/\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\//\\/
-- John Girash -- girash @ cfa.harvard.edu - http://skyron.harvard.edu/ --
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: ISDN or 2-56k Modems and RH 6.0
Date: Mon, 23 Aug 1999 15:52:41 GMT
Mark,
I have a USR V.Everything Courier flashed to V.90 and I always get a
45,333 bps connection and occasionally 46,666. My downloads are
typically 4.0-5.5kbps on compressed files. I have had really good
experiencies with my ISP and USR <not 3com> modems (I once ran a 4 line
BBS a few years back).
David.
In article <[EMAIL PROTECTED]>,
Mark Presley <[EMAIL PROTECTED]> wrote:
> What's easier under Linux I can't help with. But I would caution you
that
> if you are to far from the Telco for DSL then you are probably too far
for
> a relaible v.90 connect. It will connect slow or drop you. It may
> sometimes connect at an ok rate (45333) but then give you SH*T for
> throughput.
>
> In your kind of environment if you choose modems stay away from the
Lucent
> chip. It is more likely to fail the connect near the max distance. On
the
> otherhanda Rockwell will connect at a higher speed but then throughput
will
> suck. If you do anything try a USR/3COM.
>
> I hate 56k technology it really sucks!
>
> Another to far from the telco user...
>
> [EMAIL PROTECTED] wrote:
>
> > I am deciding on ISDN vs. 2-56k modems for faster internet access. I
am
> > too far from the telco co so these are my only choices using todays
> > technologies.
> >
> > Linux- What I need to know is which is easier to setup under RH
Linux
> > 6.0 ? <as I don't have alot of spare time (who does?)>.
> >
> > Thanks,
> > David.
> > --
> > David Morris / CNA
> > NetWare / IntranetWare / NT Integration Specialist
> > New Redhat Linux 6.0 user ;)
> >
> > Sent via Deja.com http://www.deja.com/
> > Share what you know. Learn what you don't.
>
>
--
David Morris / CNA
NetWare / IntranetWare / NT Integration Specialist
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Leonard Evens <[EMAIL PROTECTED]>
Subject: Re: Change LILO from MBR to a local partition?
Date: Mon, 23 Aug 1999 10:58:43 -0500
[EMAIL PROTECTED] wrote:
>
> Hi, I have win95+RH6.0 with LILO installed in the MBR of /dev/hda, recently
> I added a second hard drive and decided to install NT in C:\(/dev/hda1, linux
> was in /dev/hda3), so there I went, everthing seemed to be fine, I partitioned
> the second Hard drive to be NTFS, and NT system was installed in C:\, then
> for some unfortunate reason my linux crashed several times and when I finally
> got it back, I found when I'm in NT, the NTFS parttion was gone, it's still
> visible under linux, and since my LILO is in MBR, the NT emergency disk doesn't
> work, all I can think of is reinstall NT4.0, however this time I'm thinking of
> putting LILO in the boot sector of /dev/hda3 instead of MBR of /dev/hda, my
> question is, since my /dev/hda3 is the linux native partition with a lot of
> data in it(I use linux most of the time), will uninstall LILO(shouldn't be
> a problem) and reinstall it "locally" damage the file system that's already
> in /dev/hda3? I've checked the LILO HOWTO but couldn't seem to find the
> proper answer, can somebody give me some hint about this? Thanks in advance.
I'm not sure I absorbed all that, but I think the answer is that
changing your first line of lilo.conf to
boot=/dev/hda3
and running lilo will not damage the file system. But if you
are concerned, first do
dd if=/dev/hda3 of=block.hda3 bs=512 count=1
while in some directory not on /dev/hda3. You could even use
of=/dev/fd0 and put it on a floppy. Then if somehow or other
the first sector (which is where the lilo boot loader goes)
gets damaged, you can restore it by reversing the if and of.
(The arguments are essential and you can get really messed up
if you leave them out.)
Make sure you also use fdisk to make /dev/hda3 the only active
partition on that drive. Otherwise it won't work.
You can't put the lilo boot loader in a logical partition.
In that case you use the enclosing extended partition.
I have a similar system, but my lilo actually goes in a
Linux swap partition, which is supposed not to be allowable.
If you want to learn about that do man mkswap.
Most people opt to use the Linux+NT-loader mini HOWTO approach
which lets the NT boot loader boot both NT and Linux.
--
Leonard Evens [EMAIL PROTECTED] 847-491-5537
Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
------------------------------
From: "Christopher W. Aiken" <[EMAIL PROTECTED]>
Crossposted-To: redhat.security.general
Subject: Re: Default mkdir attributes
Date: Mon, 23 Aug 1999 13:09:55 -0400
See man pages on "umask" ( w/o quotes).
...cwa
Michael Klatsky wrote:
> I'm curious- recently, when I make a new directory, the default attributes
>
> are as follows:
>
> drwxrwsr-x 2 mapu mapu 1024 Aug 19 18:39 testing
>
> I believe the dirs are being created with the SGID bit set. How can I turn
>
> this behavior off? Should I turn this off?
>
> Thanks in advance.
>
> Michael
>
> ------------------ Posted via CNET Linux Help ------------------
> http://www.searchlinux.com
--
===================================================================
Christopher W. Aiken
[EMAIL PROTECTED]
SuSE 6.1, Kernel 2.2.7
Mandrake 6.0, Kernel 2.2.9
The box said 'WIN95/98 or better.' so I installed LINUX!
------------------------------
From: [EMAIL PROTECTED] (Paul D. Smith)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: No core file
Date: 23 Aug 1999 13:15:44 -0400
Reply-To: [EMAIL PROTECTED]
%% [EMAIL PROTECTED] writes:
g> However, do you agree that commercial applications for Linux
g> should dump core if an unexpected, and not really manageable
g> error occurs?
It's a tough question. As a developer, yes, I think that The Right
Thing is to dump core whenever the program is in a state where recovery
is impossible, or which is completely unexpected and invalid. This is
particularly important for libraries. The old SunOS 4.1.x C runtimes,
for example, were _horrible_ at this: they'd try to catch all kinds of
things "for" you. For example, if you passed a NULL or invalid FILE* to
fprintf(), it wouldn't complain. If you passed a nul pointer to printf
for a %s format, it would print "<NULL>" in the output. The malloc
library was extremely forgiving (to be fair, many were) and you could
use memory after you'd freed it. It made any program developed on SunOS
take much, much longer to port. I wish the libs would just core dump on
illegal values.
Obviously the first goal is to handle all errors gracefully. In short,
the _only_ valid reason for a core dump is probably programmer
error... any other kind of error (from an external source) should be
handled--even if the handling is merely an informative error message and
a controlled exit.
However, it's bad mojo for commercial applications to dump core;
customers feel that programs that do are buggy and unstable. I guess if
you follow the "first goal" above, then they're right :).
--
===============================================================================
Paul D. Smith <[EMAIL PROTECTED]> Network Management Development
"Please remain calm...I may be mad, but I am a professional." --Mad Scientist
===============================================================================
These are my opinions---Nortel Networks takes no responsibility for them.
------------------------------
From: [EMAIL PROTECTED]
Subject: win95 backup on Linux
Date: Mon, 23 Aug 1999 17:07:17 GMT
Can I backup and restore entire win95 system (/dosc )on a linux
partition from within Linux. Thanks !
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (John L. Daschbach)
Crossposted-To: comp.os.os2.misc,comp.os.os2.apps,alt.os.linux
Subject: Re: Can OS/2 users grow up and think like Linux users? (was: Can I switch
from OS/2 to Linux and be happy?)
Date: 21 Aug 1999 10:34:43 -0800
Background:
I've used OS/2 nearly daily since v2.0 (and still have all the
boxes to show for it), Unix for 15 years (Linux for 3 years or
so), and Mac's for 8 years or so (a late comer). I've never *used* a
Windows PC as my machine, but I have used one for a few minutes here
or there. I've been programming in C for 15+ years, Perl for 12+
years, Lisp and Scheme for 10+ years, and a smattering of other
languages in small amounts (assembler in the old days, matlab/octave,
rexx, sh, ...)
Perspective:
OS/2 (with Object Desktop) still has one of the best user
interfaces going. Much of this is because of the WPS and SOM.
Nothing I have used or read about approaches OS/2 in terms of desktop
power *and* configurability. However this comes at a price, and that
price is stability. I have had OS/2 up for weeks, but at times,
especially when using *real* applications, the WPS gets totally
wacked, windows don't repaint, objects partially freeze on the screen,
the system won't respond to keyboard or mouse. This is typically when
using the combination of Navigator and a Lotus product (WordPro, 123,
Approach, Freelance). The only solution is a hard reboot. Other
times programs have problems and there is no way to kill them, at
least using the tools I have. In other words, OS/2 when it works is
great, when it doesn't it's hell.
Solution:
The only way, I think, to achieve the power of OS/2 is to have
a file system with meta data (extended attributes, resource fork).
Linux at present has no file system with meta data. The BeOS does.
IBM should open source the WPS, SOM and HPFS, and let clever Linux
programers port them to Linux, perhaps mapping SOM to Corba. Already
all the normal GNU file tools work with the HPFS extended attributes
so that code could just be part of the GNU tools for Linux. I would
expect that Gtk+ will be powerfull enough that it can replace PM.
This could really be a killer desktop environment. For a server you
wouldn't run the Linux/WPS and you'd get the great Linux stability (I
have an old Pentium with Linux that takes data from 6 instruments and
has been up for over a year). On the desktop you'd still have the
underlying OS stability, but the UI might be less stable (anytime you
have the complexity of the WPS you invite problems).
The problem is that I don't see what's in it for IBM, so it
won't happen. For big business, the thin client solution will win
out, since it allows central controll. I think IBM is still shooting
for desktops with Java which keep their meta-data on an IBM server.
It's a good business model, and I don't think the power-user desktop
is something they have any interest in pursuing.
Thoughts?
-John
--
John L. Daschbach
[EMAIL PROTECTED]
------------------------------
From: Mapu <[EMAIL PROTECTED]>
Crossposted-To: redhat.security.general
Subject: Re: Default mkdir attributes
Date: Mon, 23 Aug 1999 13:53:41 -0400
I looked at man umask, and umask seems only to deal with read, write and
execute perms. What I am looking for is info regarding setting the UID/GID
bit in the file attributes...or perhaps I'm missing something>
Thanks for your response.
Michael
On Mon, 23 Aug 1999, Christopher W. Aiken wrote:
> See man pages on "umask" ( w/o quotes).
>
> ...cwa
>
>
> Michael Klatsky wrote:
>
> > I'm curious- recently, when I make a new directory, the default attributes
> >
> > are as follows:
> >
> > drwxrwsr-x 2 mapu mapu 1024 Aug 19 18:39 testing
> >
> > I believe the dirs are being created with the SGID bit set. How can I turn
> >
> > this behavior off? Should I turn this off?
> >
> > Thanks in advance.
> >
> > Michael
> >
> > ------------------ Posted via CNET Linux Help ------------------
> > http://www.searchlinux.com
>
> --
> -------------------------------------------------------------------
> Christopher W. Aiken
> [EMAIL PROTECTED]
> SuSE 6.1, Kernel 2.2.7
> Mandrake 6.0, Kernel 2.2.9
>
> The box said 'WIN95/98 or better.' so I installed LINUX!
>
>
>
------------------------------
** 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
******************************
