Linux-Misc Digest #948, Volume #25 Thu, 5 Oct 00 06:13:04 EDT
Contents:
Re: "w" and "who" do not list the same users logged in (Robert Kiesling)
Re: MS Netmeeting under Linux? (Bill Gates)
Re: Problem with samba (Miguel Lastra Leidinger)
Linux install (Ray)
irq problems? pt II (asage)
Re: any hope? (win98 upgrade, lost mbr, no boot disk) (Thomas Regner)
Re: any hope? (win98 upgrade, lost mbr, no boot disk) (Eric)
Logger for Hardware-Traps? (Stefan Meister)
Re: Missing err msgs in errno.h (Floyd Davidson)
malloc() ("Ludwig Stroobant")
Recompiling Kernel ([EMAIL PROTECTED])
Help !! Why does changing the hostname cause so many problems ? ("Phil Hedley")
Re: Printing man pages (Floyd Davidson)
Re: slow? (Jeph Herrin)
Re: Linux as server: recommendations system & Linux (Neil)
Re: loading graphical interface, linux 6.2 (Neil)
Re: Installing Redhat 7.0 on a Dell Inspiron 3200 Notebook (Neil)
Re: tar but no dump - help? ([EMAIL PROTECTED])
Re: Linux install ([EMAIL PROTECTED])
Re: can't log in as root (Giacomo Amabile Catenazzi)
Re: any hope? (win98 upgrade, lost mbr, no boot disk) (Daniel Haude)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Robert Kiesling)
Subject: Re: "w" and "who" do not list the same users logged in
Reply-To: [EMAIL PROTECTED]
Date: 5 Oct 2000 02:43:03 -0400
Send a SIGHUP to the shell on that xterm to log the user out. At
least, that works for the bash shell, which doesn't respond to SIGTERM
or SIGQUIT. That might be more difficult, because then you
have to correlate the shell with the xterm's pty, and in ps or
top, tat may not be one of the choices that is offered.
In article <[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
>I'm not sure, but killing the xterms more gently with a kill -HUP may
>have logged out the users. Kill -9 is really quite drastic. No signal
>handler is usually invoked for signal 9---in fact, it may be impossible
>for a program to register a handler for signal 9---so no program that
>is killed that way has any opportunity to clean up after itself.
>
>Dave
>
>On Tue, 29 Aug 2000 20:04:28 GMT, Stewart Honsberger
><[EMAIL PROTECTED]> wrote:
>>On Sun, 27 Aug 2000 17:49:32 +0000 (UTC), David Efflandt wrote:
>>>If you kill sam, he is not properly logged out, so of course the logs will
>>>be wrong. Try logging him out instead (Crtl-D or logout command).
>>
>>But how does one go about killing several (dozen) logged-in users? I had,
>>well, an Xterm incident, and long story short I ended up with about 109
>>Xterms loaded. I killall -9 xterm'ed the whole she-bang and my system
>>continues to churn away, but there are still most of these users logged
>>on. The solution I had before was to create dozens of Xterms and hit
>>"exit" in all of them. That worked, but I missed the last 24, and I don't
>>relish the experience of creating 109 Xterms and exit'ing them all.
>>
>
>
>--
>``I always thought that anybody who told me I couldn't live in the
>past was trying to get me to forget something that if I remembered
>it would get them in serious trouble.'' --Utah Philips
--
Robert Kiesling
Linux FAQ Maintainer
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Bill Gates)
Crossposted-To: comp.os.linux.help,comp.os.linux.setup
Subject: Re: MS Netmeeting under Linux?
Date: Thu, 05 Oct 2000 06:58:28 GMT
On Wed, 04 Oct 2000 15:01:34 GMT, Bill Pringlemeir
<[EMAIL PROTECTED]> wrote:
The MBone tools are multiplatform tools for videoconferencing (video,
audio, whiteboarding, chat etc.).
this is a place where you can get it:
http://www-mice.cs.ucl.ac.uk/multimedia/software/index.html
BIll.
>>>>>> "John" == John Beardmore <[EMAIL PROTECTED]> writes:
>
> John> An interesting looking thing, but is there anything that will
> John> do cross platform telephony or audio video conferencing ?
>
>Speak Freely. "http://www.speakfreely.org/"
>
>Bill
>
>--
>genetic class struggle Honduras Khaddafi Semtex munitions strategic
>Ortega Treasury plutonium Panama colonel Albanian smuggle Noriega
------------------------------
From: Miguel Lastra Leidinger <[EMAIL PROTECTED]>
Subject: Re: Problem with samba
Date: Thu, 05 Oct 2000 09:50:58 +0000
Reply-To: [EMAIL PROTECTED]
Natanael wrote:
> It could be that the win98 sends the passwords encrypted, while your
> samba expects the passwords as plain text. Look in the samba
> documentation for encrypted passwords.
I have already turned off password encryption in W98 but nothing happened.
Any other ideas ?
thanks
Miguel
------------------------------
From: Ray <[EMAIL PROTECTED]>
Subject: Linux install
Date: Thu, 05 Oct 2000 07:29:44 -0000
Hey all...I just bought Suse Linux version 7 and tried to install on my
computer....which is a p3 733 coppermine with a Creative labs Annilator2
video card. The problem starts when Im installing and asks me to configure
my monitor and video card...when I go to test it...my computer freezes and
I cant access anything! not even mouse or keyboard...I wind up hitting the
restart button.... ;o( please if anyone can shed some light...I would
appreciate it alot! Take Care!
Ray
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: asage <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: irq problems? pt II
Date: Thu, 05 Oct 2000 03:24:37 -0400
On Tue, 03 Oct 2000 14:47:34 -0400, asage wrote:
>>Sorry if this has been covered before (ad nauseum even).
>>I finally got my 3com modem working - yay. Now my Wacom tablet won't
>>work. Previously, when I installed the Wacom, I lost the modem. So it
>>must be an irq conflict, right?
[snipp]
(DancesWithCrows replied...)
>Plenty of info, but where is the Wacom tablet? Do you have 2 serial
>ports on this machine? Is this an internal or external modem?
[snipp]
>If you do have 2 serial ports and you have the Wacom plugged in to the
>second one, the device is /dev/ttyS1 (not S3, even if DOS thinks it's on
>COM4) and it uses IRQ 3 by default. No conflicts there!
Okay... I've figured out that I have 2 serial ports. It's an internal
modem.
In Linux, the modem just won't work on any other port than ttyS0. DOS
does think
that the Wacom is on COM4 (ttyS3), but I have tried to configure it for
ttyS1, and
ttyS3.
Nothing at all happened when it was configured for ttyS3.
When it was configured for ttyS1, when X was starting, it tried to
initialize the Wacom,
and then there was the error message:
Wacom tcgetattr error
Threshold = -80
New Threshold = -80
I'm guessing that that means it still can't find it on ttyS1?
Would it help if I just removed the Wacom and re-installed it?
TIA
Allison Sage
------------------------------
From: Thomas Regner <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: any hope? (win98 upgrade, lost mbr, no boot disk)
Date: Thu, 5 Oct 2000 10:02:14 +0200
Reply-To: [EMAIL PROTECTED]
Aslak Johansen wrote:
[...]
> There is, however, still a problem: How will Win98 react to your LILO?
> Microsoft has been pretty successfull in making their OS'es incompatibl=
e
> with other boot-managers. The safe way is to either install 'loadlin' o=
r
> to create a boot-disk.
[...]
As far as I know, this is no problem at all...
Win98 is bootable from lilo, and this is no microsoft issue...
regards
tom
--=20
________________________________
| thomas regner |
| noltestr.22 |
| 30451 hannover |
|--------------------------------|
| tel +49 (0) 511 21349817 |
| tel +49 (0) 172 1605452 |
| eml [EMAIL PROTECTED] |
| eml [EMAIL PROTECTED] |
| icq 87162930 |
|________________________________|
------------------------------
From: Eric <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: any hope? (win98 upgrade, lost mbr, no boot disk)
Date: Thu, 05 Oct 2000 10:21:35 +0200
Reply-To: [EMAIL PROTECTED]
RCD wrote:
>
> hi,
>
> thanks for the useful info,
>
> I'll try booting from the cd, then loadlin.
More trouble than LILO if you ask me, but that's a matter of opinion.
>
> Yeah, I think you're right about installing LILO. I've read that
> re-installing it may make
> win98 unbootable. I think I'll have to stick to a boot disk until this
That's just bullshit, LILO is very wel capable of booting win98. I've
done this so many times, and never encountered any problem whatsoever.
Just boot linux, either from CD or floppy and rerun LILO. All you need
is an entry in lilo.conf that points to the partition where win98 is
installed. There's nothing more to it to get win98 booted through LILO.
Just one thing you might want to consider, put lilo in a partitition and
not in the MBR, that way reinstalling win98 won't mess-up LILO.
Eric
> apparent problem is solved.
>
> (I just hope I can get in a make it)
>
> thanks again,
>
> ryan
>
> >
------------------------------
From: Stefan Meister <[EMAIL PROTECTED]>
Subject: Logger for Hardware-Traps?
Date: Thu, 05 Oct 2000 07:39:22 GMT
Hello,
is there anything (already included to Linux) what can log=20
Hardware-Errors?
Ive heared from then VMS-OS that there is a showinfo-command, which=20
displays info about the Hardware and the number of Errors, which occured=
.=20
I found procinfo, klogd and sylogd on linux, but they seem not to suppor=
t=20
something what i call: catching HW-traps and log them into a log-file.
Thank you
regards
stefan
------------------------------
From: Floyd Davidson <[EMAIL PROTECTED]>
Subject: Re: Missing err msgs in errno.h
Date: 04 Oct 2000 23:36:02 -0800
marvin greenberg <[EMAIL PROTECTED]> wrote:
>It's been a long time since I've done development on unix, but isn't errno.h
>or something similar supposed to have the various error defines?
>
>I'm trying to compile sajber jukebox, and something in the thread code is
>complaining about an undefined symbol EBUSY. I did a
>find /usr/include -exec grep EBUSY {} \; and found nothing. Out of curiosity
>I also looked for ENOMEM, ENOENT, etc... and found only some comments.
>errno.h and other similar headers don't define any of the familiar EXXXXX
>definitions.
>
>What gives?
Your use of find isn't likely to find anything which is system specific.
The "-follow" option to find will allow it to follow symbolic links, which
in this case is useful because the /usr/include/asm directory is a
symlink to /usr/src/linux/include/asm, and your usage will not look there.
But, even if you use the -follow option, it only tells you that somewhere
there is a file (or two) with EBUSY in it. You won't have a clue as to
which file it is though. However, here is an alias I have in my .bashrc
file to do just that,
alias include='find /usr/include -follow -name "*.h" | xargs grep'
It will let you know that EBUSY is found in both /usr/include/asm/errno.h
and /usr/include/scsi/sg.h, though clearly the former is what you are
interested in.
--
Floyd L. Davidson [EMAIL PROTECTED]
Ukpeagvik (Barrow, Alaska)
------------------------------
From: "Ludwig Stroobant" <[EMAIL PROTECTED]>
Subject: malloc()
Date: Thu, 05 Oct 2000 08:35:01 GMT
Hello,
I have a problem with malloc(). At some time in my program I do a malloc of
two pointers, and now it seems that, if I do a free on the second pointer
first, and then a free of the first pointer, I get a segmentation fault. If
I skip the freeing of the second pointer, the freeing of the first one is
ok.
If I look at the memory space of the pointers, it seems that, on freeing one
or the other pointer, one byte, that they have in common, is changed. So, if
I free the second pointer first, some bytes are changed, and if I skip this
one, the free operation on the other pointer changes some other bytes,
except for this one byte that they both change on their free operation. (I
don't know if I'm explaining it good).
What can I do about this problem?
How is the information about the size of pointer kept in Linux. If you do a
malloc on, lets say, 20 bytes, where is the size of that pointer being kept.
And is there any other info about that pointer being stored somewhere.
Ludwig
------------------------------
From: [EMAIL PROTECTED]
Subject: Recompiling Kernel
Date: Thu, 05 Oct 2000 08:38:45 GMT
I need to recompile my RedHat 6.2 kernel to disable CPU_ID support. Can
anyone give me some pointers as to how to do this?
Thank you
--
Steven Coutts
[EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Phil Hedley" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.admin
Subject: Help !! Why does changing the hostname cause so many problems ?
Date: Thu, 05 Oct 2000 08:51:35 GMT
Hi all,
I have a Redhat 6.2 workstation setup.
If I change my Hostname using Linuxconf or Netcfg and re-boot, I find that I
cannot start the Gnome desktop and the system seems to be working extremely
slowly.
I fixed this by editing /etc/sysconfig/network and putting in the original
host name.
1. Can anyone tell me what is going on ?
2. Is this the correct way to change the hostname ?
Thanks,
Phil Hedley
------------------------------
From: Floyd Davidson <[EMAIL PROTECTED]>
Subject: Re: Printing man pages
Date: 04 Oct 2000 23:55:55 -0800
James Silverton <[EMAIL PROTECTED]> wrote:
>Floyd Davidson wrote:
...
>> What I would do is see if there is a ghostscript driver for your
>> printer, and use that to print a PostScript file generated with
>> the -t option to man. ...
>
>Hi Floyd!
>
>What I am saying is that I have a Laserjet 4L which is not a postscript
>printer. In order to use it with Linux, I followed the SuSE instructions
>and installed apsfilter for lpr. This worked very well and ordinarily, I
>can forget that I don't have a real postscript printer. For instance, a
>multipage Abiword document is printed perfectly.
>
>The command "man find | lpr", for example, produces a multipage
>printout but the pages are not ejected at the man headers. This is not
>particularly bothersome tho' it would look at bit neater if the page
>ejects occurred. I just wondered if it was possible to get man pages to
>eject properly but I don't regard it as a major problem nor one
>requiring a change to a different printing program.
>
>Incidentally, the bold headers in the document print correctly but if I
>use "man -t find | lpr", the printout is much the same but the
>previously bold characters become doubled, normal weight and separated
>by by about a half letter space; not satisfactory at all!
Hi Jim,
Hmmm... I think you might want to take this to one of the gnu
groups, or where ever you can get help with Ghostscript. It
sounds to me like a problem with font generation/selection in
the conversion from Postscript to HPCL for the HP backend
to ghostscript.
What do the files look like viewed with Ghostview?
I've never used an HP 4L, so I can't really help much. I have
used several different HPCL printers, both color and monochrome
and both DeskJet and LaserJet, and did not experience that with
any of them. There usually are some slight differences, but
nothing that ugly.
--
Floyd L. Davidson [EMAIL PROTECTED]
Ukpeagvik (Barrow, Alaska)
------------------------------
From: Jeph Herrin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: slow?
Date: Thu, 05 Oct 2000 11:09:47 +0200
Hal Burgiss wrote:
> The error message still worries me.
my stupid mistake the error message ("FAILED") was for the prior boot down
item (sound card, which i haven't configured yet). swap dismounts just
fine; i was seeing it wrong as it scrolled past.
>
>
> >Well, this is an open question. the xfree86 website had details for
> >my card (AT24 chipset, Alliance Promotion card), but they didn't
> >work, so i had to finagle a bit to get it to go. i wrote xfree86 about
> >this, but no reply. maybe go back to generic 600x400? i could try to
> >see if it improves performance, but then i can hardly use the thing
> >(and i may never luck into the right configuration again).
>
> I would suspect the answer is here somewhere. You might try different
> color depths. Are you 24 or 32 now? If so drop to 16. Are you running
> the latest X server? Not meaning 4.x, but would at least be 3.3.6.
>
I'm running 3.3.6, and 8 bit color depth; i deleted the 24 & 32 bit
entries. The problem i had was that my card has 1502kb of ram,
which xfree86 would find and reject. so i coded it as 1024kb -
it works, but i have no idea what is happening at the hardware level.
jeph
--
[EMAIL PROTECTED]
------------------------------
From: Neil <[EMAIL PROTECTED]>
Subject: Re: Linux as server: recommendations system & Linux
Reply-To: [EMAIL PROTECTED]
Date: Thu, 05 Oct 2000 11:28:48 +0100
On 04 Oct 2000 21:13:24 GMT, [EMAIL PROTECTED] (Ldrpdx) wrote:
>I was wondering if anyone had any opinions or recommendations for systems
>(hardware configurations) and Linux "brand".
Most distros are about the same. If ifs security you are looking for then go for
FreeBSD over linux.
------------------------------
From: Neil <[EMAIL PROTECTED]>
Subject: Re: loading graphical interface, linux 6.2
Reply-To: [EMAIL PROTECTED]
Date: Thu, 05 Oct 2000 11:31:31 +0100
On Wed, 04 Oct 2000 19:26:07 GMT, [EMAIL PROTECTED] wrote:
> While I'm at it,
>what are the differences between X Window, GNOME,
>and KDE, aren't they all inter related? Sorry,
>I'm a brand new user.
>Thanks
type startx
GNOME and KDE are window managers/desktop environment - X windows is the
underlying architecture which enables all X based applications.
------------------------------
From: Neil <[EMAIL PROTECTED]>
Subject: Re: Installing Redhat 7.0 on a Dell Inspiron 3200 Notebook
Reply-To: [EMAIL PROTECTED]
Date: Thu, 05 Oct 2000 11:32:10 +0100
On Wed, 04 Oct 2000 16:40:35 GMT, [EMAIL PROTECTED] wrote:
>I toke me over a year to get X working right and now I am thinking
>about installed Redhat 7.0?
try the linux-dell-laptops mailing list at www.egroups.com
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.admin,comp.os.linux.hardware,linux.redhat.misc,linux.scsi
Subject: Re: tar but no dump - help?
Date: 5 Oct 2000 09:35:29 GMT
In comp.os.linux.admin ekkis <[EMAIL PROTECTED]> wrote:
> hi all,
You need to tell dump how large the tape is (-s switch) default is
a very small tape (in your output you can see that dump
estimates that it will need 99.97 tapes for your backup !)
Be aware that switches for dump is a little "unnatural" i guess
you should do :
dump -0usf 1000000 /dev/st0 /home/ftp/mp3z
but do check your manpage first !
> I'm having a problem backing up a file system to tape using "dump". I can
> tar to the device (read and write) so I know there's nothing wrong
> with the hardware or drivers... can anyone help point me ine right
> direction?
> here's my command and it's output:
> root@beowulf:/root # dump -0u -f /dev/st0 /home/ftp/mp3z
> DUMP: Date of this level 0 dump: Mon Oct 2 13:32:48 2000
> DUMP: Date of last level 0 dump: the epoch
> DUMP: Dumping /dev/hdb1 (/ (dir home/ftp/mp3z)) to /dev/st0
> DUMP: Label: none
> DUMP: mapping (Pass I) [regular files]
> DUMP: mapping (Pass II) [directories]
> DUMP: estimated 3913078 tape blocks on 99.97 tape(s).
> DUMP: Volume 1 started at: Mon Oct 2 13:32:49 2000
> DUMP: dumping (Pass III) [directories]
> DUMP: dumping (Pass IV) [regular files]
> DUMP: Closing /dev/st0
> DUMP: Volume 1 completed at: Mon Oct 2 13:34:44 2000
> DUMP: Volume 1 took 0:01:55
> DUMP: Volume 1 transfer rate: 333 KB/s
> DUMP: Change Volumes: Mount volume #2
> DUMP: Is the new volume mounted and ready to go?: ("yes" or "no") y
> DUMP: Cannot open output "/dev/st0".
> DUMP: Do you want to retry the open?: ("yes" or "no") n
> DUMP: The ENTIRE dump is aborted.
> the drive has 4 tapes. When dump first asks
> whether the "new volume" is mounted, I switch to the next tape by issuing:
> root@beowulf:/var/log # mtx -f /dev/sg1 next
> Unloading Data Transfer Element into Storage Element 1...done
> and then respond "y" (which fails).
> Some useful info:
> root@beowulf:/root # uname -a
> Linux beowulf 2.2.14-5.0 #1 Tue Mar 7 20:53:41 EST 2000 i586 unknown
> root@beowulf:/root # tapeinfo -f /dev/sg1
> Product Type: Tape Drive
> Vendor ID: 'ARCHIVE '
> Product ID: 'Python 28849-XXX'
> Revision: '4.CM'
> Attached Changer: No
> DataCompEnabled: yes
> DataCompCapable: yes
> DataDeCompEnabled: yes
> CompType: 0x20
> DeCompType: 0x20
> ActivePartition: 0
> EarlyWarningSize: 0
> NumPartitions:0
> MaxPartitions:1
> MinBlock:1
> MaxBlock:16777215
> root@beowulf:/root # mtx -f /dev/sg1 status
> Storage Changer /dev/sg1:1 Drives, 4 Slots ( 0 Import/Export )
> Data Transfer Element 0:Full (Storage Element 2 Loaded)
> Storage Element 1:Full
> Storage Element 2:Empty
> Storage Element 3:Full
> Storage Element 4:Full
> Also, something weird, the directory I'm trying to dump is a little over
> 3GB:
> root@beowulf:/root # ls /home/ftp/mp3z |head -1
> total 3885396
> and my tapes should fit 4GB (uncompressed), however if you'll notice in the
> dump output:
> DUMP: estimated 3913078 tape blocks on 99.97 tape(s).
> the number of tapes calculated is BIG! What's up with this?
> any help greatly appreciated. - erick
> p.s. please e-mail so I don't have to keep coming back to the newsgroup,
> thx!
--
Peter H�kanson
IPSec Sverige (At the Riverside of Gothenburg, home of Volvo)
Sorry about my e-mail address, but i'm trying to keep spam out.
Remove "icke-reklam"and "invalid" and it works.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Linux install
Date: Thu, 05 Oct 2000 09:29:44 GMT
In article <[EMAIL PROTECTED]>,
Ray <[EMAIL PROTECTED]> wrote:
> Hey all...I just bought Suse Linux version 7 and tried to install on
my
> computer....which is a p3 733 coppermine with a Creative labs
Annilator2
> video card. The problem starts when Im installing and asks me to
configure
> my monitor and video card...when I go to test it...my computer freezes
and
> I cant access anything! not even mouse or keyboard...I wind up hitting
the
> restart button.... ;o( please if anyone can shed some light...I
would
> appreciate it alot! Take Care!
This could be a problem of the BIOS configuration:
check the following in BIOS config, chipset features:
video memory cache mode : uc (=uncacheable)
the best way ;-))) is, of course, to use xf86config instead of sax.
Volker
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
Date: Thu, 05 Oct 2000 11:40:18 +0200
From: Giacomo Amabile Catenazzi <[EMAIL PROTECTED]>
Subject: Re: can't log in as root
Steve Gage wrote:
>
> Hey group,
>
> In 3 years of using linux, here's something I've never encountered:
>
> I can't log in at the console as root! I can login as any user and su to
> root, but can't log in as root. If I type "root" at the login prompt, it
> doesn't even get around to asking for a password - after a few seconds
> it just says "Login incorrect".
>
> This seems to have come out of a blue sky following a reboot today.
> System is Debian 2.2. Anyone have any clue what could be going
> on here?
>
> TIA,
>
> Steve
you can login as root only on the terminal listed in /etc/secure* (I
don't remebmer the true name!), but you can 'su' anywhere!
Thus I think that there is a problem in your /etc/secure*.
Try to login as root in virtual terminal 1 (CTRL-ALT-F1 if you are in X,
ALT-F7 or CTRL-ALT-F7, to return in X). It this works it is definetely a
/etc/secure* problem.
giacomo
PS. Debian has a nice support team on irc: irc.debian.org channel
#debian.
------------------------------
From: [EMAIL PROTECTED] (Daniel Haude)
Crossposted-To: comp.os.linux.setup
Subject: Re: any hope? (win98 upgrade, lost mbr, no boot disk)
Date: 5 Oct 2000 09:47:12 GMT
Reply-To: [EMAIL PROTECTED]
On Wed, 04 Oct 2000 15:39:13 -0700,
RCD <[EMAIL PROTECTED]> wrote
in Msg. <[EMAIL PROTECTED]>
| if anyone knows of a solution I'd like to hear it.
|
| I have (had) a dual boot (with lilo) redhat, win95 and thought I'd
| finally upgrade to win98.
|
| I stupidly forgot to make a bootdisk.
|
| I thought I lost the original MBR, but when I scan the disk with NAV
| 2000 it says I have 2 MBRs.
|
| Can I recover and edit the MBR from windows to restore the lilo
| information?
|
| If I'm screwed, I'll just do another clean install of redhat, but I'd
| really like to know if there's any hope of recovery first.
Other people have already posted useful information, let me just point out
a thing or two:
The MBR is not part of any of the two OSs. Your Linux system is still
perfectly intact. The only problem is to get it to boot. To boot, any
Linux system needs only two things:
1) a kernel and
2) a root filesystem.
At the moment, there is no way to boot the kernel that's on your Linux
partition because you do not have a boot loader that could access it (you
trashed that with win98 -- it always happens). So you have to boot some
other kernel -- from a bootdisk or, easier, with loadlin from the DOS
prompt. That kernel needs to be told where the Linux root partition is, it
will then mount that under / and go on with the normal boot process.
So what do you do?
Copy a Linux kernel and loadlin.exe onto your W98 disk. Go into DOS mode.
Say:
loadlin linux root=/dev/hda2
(replace 'linux 'with the name of your kernel image, and 'dev/hda2' with
the name of your Linux root partition)
The system will boot up (possibly with some warnings due to a different
kernel version). Once the system is up, login as root and type
lilo
and all is like before, except now you have animated menus in Windows.
--Daniel
--
"The obvious mathematical breakthrough would be development of an easy
way to factor large prime numbers." -- Bill Gates, "The Road Ahead"
------------------------------
** 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
******************************
