Re: [newbie] Problems setting up paths

2000-01-07 Thread Matt Stegman

First, what you did does not do it- it will set the second path
mentioned (after the double pipe: ||) only if "sbin" is not already in the
path.  To ensure this is in your path, add another line (below this first
one) like so:
  export PATH=$PATH:/usr/local/netscape
Second, put this line at the bottom of /etc/profile to set it for all 
users.

And by the way, a brand-new-shell IS started just by opening a new xterm.
There's no need to restart X.

-Matt Stegman
<[EMAIL PROTECTED]>

On Mon, 3 Jan 2000, Mike Ladwig wrote:

> On Mon, 03 Jan 2000, you wrote:
> > How do you set the path? I am using the bash shell.
> > I go into my .bashrc file and there is a line that looks like this:
> > 
> > echo $PATH|grep -q /sbin || export PATH=$PATH:/usr/sbin/:/sbin/
> > 
> > then I add at the end of this line :/usr/local/netscape (I want netscape in
> > my local path)
> > 
> > I imagine this statement is making sure sbin is in my path, and it if isnt,
> > then it adds it ( I am not sure what the || means.. I imagine thats an or
> > command). But this doesnt work. I save the file, exit my xterm session,
> > restart another and /usr/local/netscape does not show up in my path.
> > 
> > Also this doesnt setup the global path so everyone will have this in their
> > path settings. Where do you set the global path at?
> > 
> > Thanks
> > Scott
> -- 
>  Scott,
>Are you also exiting X and logging back in ?  If all you do is edit the
> .bashrc and quit the xterm session.  Then open another xterm session your
> .bashrc doesn't get run.  Try going back to a login: then test your path. 
> After you log back in type "echo $PATH" (no quotes) to make sure it worked. 
> Hope this helps 
> 
>  -[ Mike ]-
> [EMAIL PROTECTED]




Re: [newbie] Distribution

1999-12-29 Thread Matt Stegman

Mein Deutsch ist nicht so gut, und ich weiss nicht "nennen."  Ich kann
gutes Englisch (und nur ein bisschen Deutsch) sprechen.

I didn't catch your last sentence, but other than that, it sounds like
you're asking for reasons to use Mandrake. If you're happy with SuSE, then
don't!  There should be only minor differences between the two, like setup
tools (Yast vs. linuxconf, for example). 

Your problem with X11amp is probably due a bug in one of two things: the
kernel module for your soundcard, or X11amp.  To fix it, try upgrading
your kernel and getting the latest version of XMMS (new name for X11amp).

-Matt Stegman 
<[EMAIL PROTECTED]>

On Tue, 28 Dec 1999, cb3hn wrote:
> Sehr geerte Damen und Herren,
> 
> ich habe zufällig Ihre Homepage gesehen und benutze seit einige Zeit
> schon SuSE Linux 6.0. Und habe vor die neue ver. 6.3 zu erwerben. Nennen
> sie mir gute Gründe Ihre Distribution zu überzeugen.
> 
> Meine momentane Hard/Softwarekonfiguration ist:
> 
> PC:1
> 
> Pentim III 500 / 128 MB SD-RAM (Internetgateway-Server)
> - Matrox Mystique 4MB
> - Creativ Labs Sondblaster AWE64
> - NE 2000 PCI Ethenetkarte 10 MBit/s (TCP/IP)
> - Elsa Mikrolink 56k pro
> 
> PC:2 (Client)
> 
> Notebook Pentim 133 / 40 MB EDO-RAM
> - Chips & Technoligies 65548 PCI 1MB
> - OPTI 930 Soundkarte / WSS und Soundblaster kompatible (probleme mit X11AMP)
> - Accton PCMCIA Ethenetkarte 10 MBit/s (TCP/IP)
> 
> ---
> 
> OS:SuSE Linux
> X-Server: KDE 1.0
> Browser: Netscape Navigator 4.5
> Editor/Dateimanager: Midnight Commander (sehr gut)
> FTP-Sever: WarFTPD
> Konfiguratonsmanager: YasT
> 
> 
> MfG: Guenter Hiller



Re: [newbie] Shutdown Rights

1999-12-27 Thread Matt Stegman

I believe that adding the appropriate usernames to /etc/shutdown.allow
will do the trick.  If that dowsn't work, or if you want to know more,
check out the man page for "shutdown.allow".

-Matt Stegman
<[EMAIL PROTECTED]>

On Sun, 26 Dec 1999, Rick Bonczek wrote:

> When I log in as root, I can issue the command "shutdown -h now", but
> when I log in as a normal user, I cannot.
> 
> How can I give a normal user the rights to shutdown a Linux workstation.
> 
> Rick Bonczek
> LAN Administrator (although still a newbie user)



RE: [newbie] Updating netscape ver. 4.7

1999-12-22 Thread Matt Stegman

Can't say as I ever had any dependancy problems with Netscape, but I
probably installed more packages than you did.  Does `rpm -q compat-libs`
come up with nothing?  If so, go get it from a Mandrake FTP mirror.  

-Matt Stegman
<[EMAIL PROTECTED]>

On Wed, 22 Dec 1999, Howard Lee wrote:

> Thanks Matt. I've tried it, and it still asks "compat-libs" is needed by
> netscape-common. Have you had this problem? Thanks.



Re: [newbie] Updating netscape ver. 4.7

1999-12-22 Thread Matt Stegman

It's already installed.  The command "rpm" uses the RedHat Package Manager
to install and uninstall packages that you download (filenames will end
with .rpm).  Some simple commands:

rpm -ivh filename.rpm#installs a package
rpm -Uvh filename.rpm#uninstalls a package
rpm -q package   #tells you if a package is installed,
 # and the package version
rpm -ql package  #lists files in an installed package
rpm -qi package  #gives a description of a package
rpm -e package   #uninstalls (erases) a package

For an example, run `rpm -q kernel`.  This will tell you what version of
the Linux kernel you have installed.  `rpm -qi kernel` gives you a
description of the "kernel" package.  

For Netscape, you can download the new version from
ftp://ftp.tux.org/pub/distributions/mandrake/updates/6.1/RPMS

You don't need all three.  You do need "netscape-common," and one of the
other two, depending on if you want the whole Communicator package with
mail and news and composer, or just the web browser (navigator).  Once
you've downloaded the appropriate packages, run `rpm -e netscape-common
netscape-communicator` to erase the current packages, and `rpm -ivh
netscape*` to install the new ones. 

-Matt Stegman
<[EMAIL PROTECTED]>

On Wed, 22 Dec 1999, Howard Lee wrote:

> [EMAIL PROTECTED] wrote:
> Thanks for the help. Where can I get this REDHAT package manager? Sorry I'm
> new to Linux. I'm using Mandrake Linux by the way. Thanks!



Re: [newbie] Monitor Configuration

1999-12-22 Thread Matt Stegman

... maybe.  Xconfigurator does an okay job, but the last time I used 
it (Mandrake 6.0, I think) it would not set the DefaultColorDepth line 
even if you did not pick a resolution for a color depth other than 8-bit.

if you have XF86Setup installed, that may work better.  If not, then
you'll have to (gasp!) edit an ASCII config file by hand.

Depending on what monitor settings you want to change, though, xvidtune
may the utility for you.  It's hard to say; you weren't very verbose in
your e-mail.

-Matt Stegman
<[EMAIL PROTECTED]>

On Tue, 21 Dec 1999, Joe Brault wrote:
>   I have a question for you all.  I recently did a dual boot on my Gateway 
> Solo laptop with Win98 and Linux-Mandrake 6.5.  I had an easy time 
> installiing the operating system, however, I need to change my monitor 
> settings.  Is there an easy command to use for this?  I can see my screen, 
> but it is quite large as is now... Thanks in advance as always!



Re: [newbie] /usr/local versus /usr/local

1999-12-22 Thread Matt Stegman

Did you not create the /usr/local partition on /dev/hdc when you installed
Mandrake?  If you did, it should already be mounted correctly.  Check the
output of `mount`.  What, if anything, does it list as mounted on
/usr/local?

If /dev/hdc1 (or whatever partition is to be /usr/local) is not already
mounted on /usr/local, you can sync the two by mounting the hdc partition
temporarily (on, say, /mnt/floppy unless you're already using a floppy
disk), copying the files over, then `rm -rf /usr/local` (this will
permanantly delete anything under /usr/local, so be sure the files are
properly copied or have a backup ready) and finally, umount the hdc
partition and remount it at /usr/local.

`cp -a` will copy while preserving directory structure, symlinks,
permissions and ownership.

Now you're set until you reboot.  You next need to add a line to
/etc/fstab like so:
/dev/hdc1/usr/localext2defaults  1 2

This tells your computer to mount /dev/hdc1 on /usr/local at next boot.
It also tells your computer things about the filesystem: it's of type
ext2, mount with default options, whether or not to dump(8) the
filesystem, and it's fsck(8) order.

A quick bit of UNIX... uh... trivia: the numbers in parentheses give man
sections for these words (commands, in this case).  You can run `man 8
dumpe2fs` (while the man page for fstab(5) refers to dump(8), the actual 
command on Linux is dumpe2fs(8)) to read about the dump command, and `man
8 fsck` to read about fsck.  The command names are just `dump` and `fsck`,
but the parentheses are used to indicate which specific man page to look
at.  In this case, there's only one for each, so the parenthetical numbers
are extraneous.  You could just as well run `man dumpe2fs`, but the
parenthetical numbers just make things clear.

While you may need to change the partition number when you copy this
line into /etc/fstab, don't change the dump or fsck numbers in the above
example.  Those are the correct values.

-Matt Stegman
<[EMAIL PROTECTED]>

On Tue, 21 Dec 1999, WH Bouterse wrote:

> /dev/hda; The 1-gig has M$/Linux-Mandrake 6.1 with 
> System Commander as boot manager.
> 
> /dev/hdc The second 400mg HD, I made a ext2 mounted as
> /usr/local. 
> 
> /dev/hda3   /   
> /dev/hdc1   /usr/local
> 
> How does one get Linux to always tell the difference
> between the two /usr/locals? I initially intended
> to have all new programs install in /dev/hdc /usr/local
> in fact I would have preferred to have all original
> programs which get put into /usr/local go into the "extra"
> /usr/local, i.e. the whole install spread out over 
> the two drives. I have "fumbled" with this same issue 
> many times over the last couple years. All the reading and
> misconfigurations have yet to shed real "light" on this,
> for me anyway.
> 
> So is the obvious to cp everything in one /usr/local
> to the other /usr/local maintaining all permissions,
> sym-links etc.? Or a cleaner "do it right the first time"
> kind of routine which I am missing?



Re: [newbie] Uninstall? Please Help a Newbie!

1999-12-22 Thread Matt Stegman

Whoa... a little bit of disinformation in here.  Seems people are confused
about what NT does and doesn't do on install.  I hope this clears some
things up:

> Okay...first...in order to install another OS...you must FIRST
> install the other OS, THEN linux.

Nah, that's just the easier way to do it.  Other operating systems don't
kill Linux, they just sometimes overwrite the MBR (Master Boot
Record; it tells your computer how to boot).  Most at least warn
you before overwriting it, except for (notably) Windows 95 and 98.  They
figure you won't ever use more than one operating system - as a friend of
mine said, "a monopolistic OS from a monopolistic company."

If you are going to overwrite your MBR, be sure your new bootloader
accepts multiple entries, or else have boot disks handy for all your
other operating systems.

> If your disk is smaller than 4 GB you probably don't have to worry about
> them, but remember that the Linux boot partition (well... any boot
> program)  needs to be below cylinder # ??? (127?)  Look this up if you
> run into trouble.

The limit you're referring to here is the 1024 cylinder limit; this is
built into PC BIOS.  When the computer boots, the BIOS cannot see beyond
1024 cylinders (this almost always translates to 8GB), so any hard disk
space you want to see without loading an OS needs to be below this.
Specifically, OS kernels and boot loaders must be below the 1024 cylinder
barrier.

> In fact I am pretty certain that the fdisk /mbr command alone will let
> you boot into NT   

ONLY if the NT partition is marked "Active" (sometimes referred to as
"Bootable").  Any partitioning program worth it's salt will tell you if
it's marked active.  

Why must it be "Active?"  Because DOS boot record does nothing but look
for the "Active" partition and load the boot record from that partition.
This is why DOS fdisk will not let you mark multiple partitions as
"Active," because DOS MBR cannot handle more than one "Active" partition.

> Actually, if you want to boot NT, I don't think you can boot with a DOS
> disk and do fdisk /mbr.  It'll write the DOS boot record.  NT over
> writes the DOS boot record on install...  If you have NT, boot with your system
> startup disks, then use the rescue disk that you made...

Not true.  "fdisk /mbr" will work (using DOS fdisk), but Windows
NT DOES NOT overwrite the boot record on install, that's the whole reason
he's having this problem.  NT's "bootloader" is contained entirely on it's
own partition.

> If I try to format the HDD's and install NT, when I boot it comes up
> with "LI" and then freezes. 

That would be because NT installation didn't overwrite the MBR, like
Win9x does.  When you cleared out Linux, you neglected to clear the MBR.
LILO is still there, but can't find your Linux kernel.  I _think_ the
"LI" means that LILO sees that the partition record has changed (I could
be wrong; I'm too lazy to look it up right now).

Like several people posted, booting off a DOS disk and running "fdisk
/mbr" will clear LILO out; running fdisk normally will allow you to mark
the NT partition active (if it's not already).  This will let you boot NT.
Alternatively, you could re-install LILO and point it to the NT partition.
In this case, it would not have to be marked active (NT doesn't really
care; the whole "Active" bit is for the benefit of the DOS MBR.)

-Matt Stegman
<[EMAIL PROTECTED]>



Re: [newbie] libc5 or glibc

1999-12-20 Thread Matt Stegman

Try `ls -l /lib/libc*`

I got (amongst other libs):  

-rwxr-xr-x 1 root root 1286208 Sep 9 13:48 /lib/libc-2.1.1.so* 
lrwxrwxrwx 1 root root 13 Nov 10 01:14 /lib/libc.so.6 -> libc-2.1.1.so*

The top one is GNU libc (glibc) 2.1.1.  This should be installed by
default on a Helios system.

If you need to upgrade, I'm not sure what I'd suggest.  A lot of things
depend on a certain version of libc, and if you upgrade it without
upgrading those other programs, you may break things.
It may just be easier to upgrade your distribution version, if you don't
alreay have the correct version of glibc.

-Matt Stegman 
<[EMAIL PROTECTED]>

On Fri, 17 Dec 1999, Hidong Kim wrote:
> How can I tell if my Linux installation is libc5 or glibc?  Is upgrading
> a library as simple as installing the rpm?  Thanks,



Re: [newbie] LILO problem need urgent help

1999-12-15 Thread Matt Stegman

Mount your Caldera filesystem and copy the portion of /etc/lilo.conf that
pertains to the Caldera kernel, and paste it into RedHat's lilo.conf.  Be
sure to specify the correct root partition (like you do for RedHat).  Then
re-run lilo, and you ought to be set.

-Matt Stegman
<[EMAIL PROTECTED]>

On Tue, 14 Dec 1999, NEW STAR wrote:
> I have 2 harddrive. one for Linux(caldera,slackware) and another for
> Dos(win98/NT.).
> Today i install Redhat6.1 to remove slackware. Installation is fine.when i
> reboot
> my computer LILO prompt shows only Redhat and Dos(used to open win98/NT)
> but not Caldera and also I'm not able to run Dos.Lilo show me that no Dos
> image
> loaded. Redhat run by default.
> caldera kernel is 2.0.3x, how can i keep redhat kernel for redhat
> and caldera kernel for caldera.



Re: [newbie] hdc1 win98 partition lost

1999-12-14 Thread Matt Stegman

I believe the "Server" option under install will wipe all your hard drives
of all data.  I don't think there's any way to get your data back now.
Unfortunately, you'll have to re-install Windows now.

-Matt Stegman
<[EMAIL PROTECTED]>

On Mon, 13 Dec 1999, Bernard Pidoux wrote:
> I have to hard disks on a Pentium pro PC system. Booting Win98 or Linux
> from different partitions using lilo did not work on both drives. So I
> decided to have one disk with Win98 and the other with Linux.
> I was happy I could boot from one disk with Win98 and the other with
> Linux until I made a new installation of Linux mandrake 6.1 with server
> class option on /dev/hda. Thes installation was a succes.
> But while doing this the installation process apparently erased the
> win98 FATand boot sector on hdc1 hard disk.
> I am pretty sure the disk has not been formated but I don't know why the
> installation removed the boot record and the FAT on Win98 drive.
> 
> Before I re-install Win98 on /dev/hdc, does somebody have any solution
> to recover the data on the win98 drive ?
> 
> Thanks,
> 
> Dr Bernard Pidoux, MD, PhD
> 



Re: [newbie] StarOffice install

1999-12-14 Thread Matt Stegman

A .tar file is an archive, similar to a .zip file in Windows... but
without compression.  To extract files from it, you'd use the command 
`tar -xf filename.tar`
If you want to see the filenames that it is extracting, put a "v"
inbetween the x and f.

Sometimes you'll get tar files that are compressed.  These will end with
.tar.gz or maybe just .tgz .   

If you like using GUI, you can use the program called "Archiver", which
will show up on the right-click menu of .tar.gz or .tgz files.  For some
reason, it may not show up for .tar files.  In this case, choose "Open
With" and Archiver will be under "Utilities."

After extracting the tar file, read the included READMEs.  Before posting
any questions about StarOffice, please check the archives of the mailing
list.  Someone set up a separate list for StarOffice questions, but I
don't know the address to subscribe.

Don't worry about showing how little you know about UNIX/Linux.  That's
why there is a "newbie" list in place- to help people learn.

-Matt Stegman
<[EMAIL PROTECTED]>

On Mon, 13 Dec 1999, Jeffrey L Sawyer Jr wrote:
> I didn't want to show how ignorant of linux or unix I am but I have to. I 
> downloaded StarOffice and it is in my home DIR as so51a_lnx_01.tar . I can 
> click on it and it opens up to show many other .tar folders. I am 
> using  KDE  and can't figure out how to install or setup StarOffice. Please 
> don't laugh to hard when replying.
> Jeffrey Sawyer



Re: [newbie] (Off Topic Sorta) SCO<->Linux

1999-12-10 Thread Matt Stegman

You may need to recompile your kernel, or at least compile a module for
the SCO filesystem (System V?).  Also, your kernel will need to support
iBCS (Intel Binary Compatibility Standard); I don't know if Mandrake does
this by default.  

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 10 Dec 1999 [EMAIL PROTECTED] wrote:
> This is a stretch. But...
> Speaking of SCO.  Is it possible for Linux to read SCO disks?  I have dBase IV
> for SCO, (really, it is a licensed copy), and would like to try to install on
> Linux, but I can't mount the floppy.
> I get an error saying wrong filesystem, bad superblock, etc.. no matter which
> filesystem I try to mount as..
> 
> Bad disks or bad user?



Re: [newbie] partition trouble

1999-12-10 Thread Matt Stegman

You really have eight swap partitions?  This isn't going to help stability
at all.  First of all, I seriously doubt it was a lack of memory that made
your computer lock up.  I can say for sure that while Netscape is not the
most stable of beasts, it is useable on my meager 64MB RAM + 80MB swap.  A
gigabyte of swap is way too much overkill.

Next time netscape locks up, see if you can kill it individually before
you have to reset your computer.  Try Ctrl+Alt+Bkspc to kill X; this may
help, even if X doesn't seem to be responding. 

-Matt Stegman
<[EMAIL PROTECTED]>

On Thu, 9 Dec 1999, raymond borowiak wrote:
> While downloading Netscape 4.7 for linux, my comp
> froze
> and I had to reinstall. Even though I tried fsck.ext2
> on the first partition and recovered but my second
> partition was not recoverable. Hence when I
> reinstalled
> from my mandrake CD-Rom ver 6.0, used one partition
> for
> native and 8 partitions for swap. As follows on a 2
> gig
> HD 1gig to / and the rest divided to the 8 swap parti-
> tions. My Question is, is this sufficient so that
> down-
> loads will not lock up my computer, which is linux
> only?



Re: [newbie] KDM and Gnome

1999-12-10 Thread Matt Stegman

make /etc/X11/prefdm (a symlink) point to /usr/bin/gdm

-Matt Stegman
<[EMAIL PROTECTED]>

On Thu, 9 Dec 1999, Joshua Hoffman wrote:
> How can I switch from KDM to GDM (or whatever the login manager for Gnome
> is called)?



Re: [newbie] GNOME to KDE

1999-12-10 Thread Matt Stegman

Run `switchdesk` from a terminal.  If you get something like "command not
found" try `desktopcfg`.

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 10 Dec 1999, David Loke wrote:

> How can I switch from GNOME to KDE full environment?



Re: [newbie] LILO

1999-12-09 Thread Matt Stegman

Check the list archives.  Some time ago I posted a short description of
how to do this.  I haven't tried it myself (I prefer LILO) but it should
work.   

-Matt Stegman
<[EMAIL PROTECTED]>

On Thu, 9 Dec 1999, Robert W. Oliver wrote:

> I don't like the NT bootloader myself, but there is a way to get it to
> load a Linux partition if anyone would be interested in doing this :)



Re: [newbie] X window remote access?

1999-12-09 Thread Matt Stegman

There's two parts to X: the server (the display unit, the machine you're
sitting at) and the client (the programs that connect to the X server to
display their output).  Telnet opens a session so you can run programs,
but you're not necessarily running an X Server which can accept display
output.

If it's on another UNIX machine, you just need to run `xhost
+remote.hostname` to add the remote host to the display list.  Then, on
the remote host (through your telnet session) you'll need to set the
DISPLAY environment variable approprately.  in bash, you can `export
DISPLAY=remote.hostname:0.0`.  Then try `xterm &`.  This should open an
xterm, displaying it on the machine you're sitting at.

The window manager is usually handled locally, unless you're using XDMCP
(at least, this has been true in my experience).  So while kfm and even
kpanel might work, kwm (the window manager component, handles window
borders, shading, focus, etc.) will not, unless you can terminate the
local window manager.

You might also look for secure shell (ssh); it automatically forwards X
connections.  Much easier, and more secure, than telnet & xhost.  In this
situation, ssh will even (by default) let you log in as root (no need to
su). 

-Matt Stegman
<[EMAIL PROTECTED]>

On Wed, 8 Dec 1999, Eric Mings wrote:

> I have read some information about x window being a server and client 
> program. I was under the impression the server and client could exist on 
> seperate computers. Is it possible to access a machine across a network 
> as root running K desktop and x windows? I clearly do not fully 
> understand how this works. I have used telnet some but that is pretty 
> limited for what I want. I would like to be able to colocate one of my 
> machines (running Mandrake 6.1 and K desktop) at an ISP and need to know 
> if I can access it for administration across the net with X windows/K 
> desktop. Could someone please clarify this for me. Thanks.
> 
> 
> Regards,
> 
> Eric Mings Ph.D.
> 



Re: [newbie] Installing Downloaded Files

1999-12-07 Thread Matt Stegman

First, you'll need to be sure you downloaded everything under the
/Mandrake directory.  This, and the right boot image, is all you need to
install Mandrake Linux.  You can get a boot image from the /images
subdirectory.  Be sure to take a look at the README file; it explains how
to use the boot images, and which one is right for you.

Once you've created a boot disk from the appropriate image, you can boot
from this, and it will step you through the installation.  Be sure you
have some unpartitioned space on your drive.

-Matt Stegman
<[EMAIL PROTECTED]>

On Mon, 6 Dec 1999, Jonathan Jackson wrote:

> This probably sounds a little weird, but I'm just starting Linux...
> 
>  Q: How do I install files I downloaded??



Re: [newbie] Getting rid of the penuin on bootup

1999-12-07 Thread Matt Stegman

Commenting it out just stops rc.local from overwriting it at boot time.
The penguin is still in there.  Try

echo -n > /etc/issue

This will clear out /etc/issue.  Next time you logout, you should get only
a login prompt.  Then, when you reboot next, you'll not get the penguin.

I don't think this is the solution you're looking for, though.  Before
showing the login prompt, something (mingetty?) clears the screen.  This
will wipe out the last of the boot messages in any case, whether or not
the penguin is displayed.

-Matt Stegman
<[EMAIL PROTECTED]>

On Sun, 5 Dec 1999, Benjamin Sher wrote:

> Dear friends:
> 
> I forgot the instructions on how to get rid of the penguin that greets
> you at bootup. Could someone please refresh my memory. I know that
> commenting out the /etc/issue section in /etc/rc.d/rc.local doesn't do
> it. Personally, I like the penguin. It's just that I would like to be
> able to scroll back and see the full boot sequence with its OK's after
> booting up.
> 
> Thanks so much.
> 
> Benjamin
> -- 
> Benjamin and Anna Sher
> [EMAIL PROTECTED]
> Sher's Russian Web
> http://www.websher.net
> 



Re: [newbie] Desktopcfg

1999-12-07 Thread Matt Stegman

If you've already established that you want to use WindowMaker as a user,
the default does not hold.  Try creating a new user.  This one should
default to XFCE.  The default holds only for those who haven't specified a
window manager in the drop-down box sometime previously ... or you might
try choosing the "default" option in the drop-down box.

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 3 Dec 1999, Josh McCaffrey wrote:
> I ran desktopcfg as root and chose to set the system wide default as
> Xfce, but when I log in as user, Window Maker starts; if I log in as
> root, Xfce starts.
> 



Re: [newbie] Window Managers and xdm

1999-12-01 Thread Matt Stegman

You'll also need to use KDE's "kdmconfig" tool to add it it to the
drop-down box.

-Matt Stegman
<[EMAIL PROTECTED]>

On Tue, 30 Nov 1999, Richard Yevchak wrote:

> I download blackbox and I was wondering how to add it to the list of wm's I am
> presented with when booting into run level 5.  From what I can tell xdm is run
> at start up.  I tried adding it to the the Xsessions file in /etc/X11/xdm but
> that didn't work.



Re: [newbie] Can't start KDE

1999-11-13 Thread Matt Stegman

Unless you have a lot of desktop icons you added, I'd suggest logging in
at a console prompt and running `rm -rf .kde` which will delete all your
personal KDE settings.

If you boot into runlevel 5 (where X starts automatically), don't login
there.  Press Ctrl-Alt-F2 after your computer finishes booting, and login
at the "penguin screen."  After you run the rm command above, logout and
press Alt-F7 to go back to X.  You can now login, and KDE will be
unmarred.

-Matt Stegman
<[EMAIL PROTECTED]>

On Sat, 13 Nov 1999, Gregory D. Fox wrote:

> Hi All,
> 
> I had KDE running ok until I choose a KDE desktop theme. I choose the theme
> and press ok to restart and since then I have not been able to get back into
> KDE. Sometime I get only the background and other times I get the black/gray
> background.
> 
> Any help will do.
> 
> TIA...
> 
> Gregory D. Fox
> [EMAIL PROTECTED]
> <>< <>< <>< <>< <>< ><> ><> ><> ><> ><>
> Liberty Computer Solutions/Liberty Marketing
> Web Site Hosting and Design
> <>< <>< <>< <>< <>< ><> ><> ><> ><> ><>
> (401) 823-5252/Fax: (401) 828-7057
> <>< <>< <>< <>< <>< ><> ><> ><> ><> ><>
> Visit our website at: www.libertymarketing.com
> 
> 



Re: [newbie] second HD partition

1999-11-08 Thread Matt Stegman

On  6 Nov, M L Cates wrote:
> I have a second backup hard drive on my system . I made a second
> partition on it using FIPS.
> 
> How do I (1) get linux to recognize it and (2) how do I reformat it for
> linux.

Probably.  A good way to find out is to run `fdisk -l /dev/hdd` (must
be run as root) which will list all the partitions on hdd (the secondary
slave hard drive).

To change things, try running `cfdisk /dev/hdd`.  This partition editor
is a little easier to use for newbies than fdisk.  You can't do quite
as much, but I'm sure it will suffice for your purposes.

I'll bet FIPS created a FAT16 or FAT32 partition.  Linux can see this
fine, but if you want to preserve file permissions and ownership,
you'll want e2fs on that partition.  The quick way to do this is change
it's type to Linux in cfdisk, then run `mkfs /dev/hdd2` ... replacing
hdd2 with the real partition name.
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] Choose Pentium? OR 586?

1999-11-08 Thread Matt Stegman

On  5 Nov, John Aldrich wrote:
> On Fri, 05 Nov 1999, you wrote:
>> Hi,
>> 
>> In menuconfig, (I have a feeling this is a dumb question) I
>> automatically have 386 with an "X" next to it.  I have a Pentium 233
>> mHz.  Is that technically a 586?  I'm not sure whether to tell it that
>> or Pentium.  My Linux book says to check 586 but I'm not sure why.
>> 
> Yes! A Pentium is a 586-class processor. I think Technically a
> Pentium is a 586, a Pentium Pro is a 686, a PII would be a 786, and a
> PIII would be an 886 

Not quite.  P-II and P-III are both 686 processors.  The first (or so I
hear) truly 7th generation x86 processor on the market is AMD's Athlon
processor.  The P-III is just a hyped up P-II, which is just a P-Pro
with MMX.  Of course, I'm over-simplfying, but I believe I'm mostly
correct.  I do know for sure that neither the P-II nor P-III qualify as
anything more than a "686" processor.

-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] Arbitrarily recognizes CD ROM drive

1999-11-08 Thread Matt Stegman

On  5 Nov, Karen Heiby wrote:
> Most of the time Linux can't mount my cdrom drive, but it recognized it just
> 10 minutes ago for the first time.
> 
> Why does it recognize it and sometimes it won't?

My first thought would be to check the IDE cable (I presume it's an IDE
drive) to be sure it's plugged in all the way.  That would be the
"ribbon" cable attached to the back of the drive.

-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] kernel boot problem

1999-11-03 Thread Matt Stegman

On  3 Nov, cyberclay wrote:
>   I tried to upgrade my kernel (using rpm -Uvh kernel*.rpm instead of
> kernel -i kernel*.rpm).  Then I forgot to run lilo before rebooting, and
> now linux can't boot.  I have no other boot media besides the Linux
> Mandrake CD (which does not have a rescue option).  I have Windows 98 on
> this system, and I'm able to do everything fine in there.  How can I boot
> linux?

Go to http://www.toms.net/rb download the DOS install, read the README,
and boot with the root boot disk you'll make.  Once you're up, mount
your root partition on /mnt, and run
# chroot /mnt /sbin/lilo

This should install lilo, and you'll be set to reboot!
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] KDE is unreadable - weird transparent windows

1999-11-03 Thread Matt Stegman

> Wow! Someone who seems to have seen this before! I have a Jaton
> Video-87AGP which is a Trident 3D Image 975 chip. Supported, by
> all accounts, by the X Server. 

Ooh, I'm sorry.  I've had all sorts of problems with Trident chipsets. 
I stay very far away from them now.  Also, everybody (well, almost) I
talk to says every Trident card they've worked with has been horrible. 
Not a very good reputation.

> I have scrolled through the resoulutions using ctl-alt-[num+/num-]
> to no avail.

That should go through different screen sizes (i.e. 640x480, 800x600),
but not change your color depth.  If you find that one of these
resolutions will work for you while a different screen size doesn't (at
the same color depth) then you'd do good to move that resolution to the
front of the line in XF86Config and delete the ones that don't work.

I'd suggest just playing around with it.  Set up several different
resolutions under different color depths.  `startx -- -bpp 16` I think
will start X in 16-bit color.  You'll have to exit and run `startx --
bpp 24` to use 24-bit color.  Anyway, just run through a bunch of
resolutions in several color depths and see what works and what doesn't.

On the other hand, the second of Pablo's links looked pretty good.  I
didn't go through it too much- I don't have one of these cards.  I'd
guess, though, that it'll work pretty well for you.

-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] executing programs

1999-10-31 Thread Matt Stegman

> I am trying to learn how to do some C programming and my question is this: How
> do I execute the program after it is compiled?  I have tried typing the name
> of the program in at the prompt and it says "command not found".  I have tried
> to type "sh" before the filename and am told "cannot execute binary file".

First of all, make sure the file has executable permissions.  If it
doesn't, you'll need to add them with `chmod +x a.out` before trying to
run the program.

Second, your current working directory is not in your path.  You'll need
to run the program like `./a.out` to specify that the program is in the
current directory.

> Any help would be appreciated, also, what does the "sh" before a filename do?

That would start up a new instance of the Bourne Shell and execute the
program inside that new shell.

-Matt Stegman
<[EMAIL PROTECTED]>




Re: [newbie] Windows 98

1999-10-28 Thread Matt Stegman

How do you mean, you set up Windows 98 to login to the Linux box?  Are you
authenticating your Windows login through Linux?   Or do you mean you're
telnet-ing into the Linux box?

Or is it something else entirely?

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 27 Oct 2000, jeff wrote:

> Hello all.
> 
> I have setup a linux server with Mandrake 6.1. Boy was that easy. Then I
> connected my windows box and setup all the params to log onto my linux
> box. It has logged on and everything seems fine except it asked for a
> password. No matter what password I send it, it says that it is
> incorrect. I do know that the linux box has only two passwords. Root
> account and my personal account. Tried both. What password is it asking
> for and if it's a bug in windows or linux, how do I fix it?
> 
> Jeff
> 



Re: [newbie] What's the best way to autostart programs upon bootup/logon?

1999-10-28 Thread Matt Stegman

Notice, also, that your ~/.bash_profile will run ~/.bashrc, so everytime
you start a new shell (like opening a Konsole window) anything in
~/.bashrc will also be run.

-Matt Stegman
<[EMAIL PROTECTED]>

On Thu, 28 Oct 1999, John Aldrich wrote:

> On Wed, 27 Oct 1999, you wrote:
> ...you could put it in rc.d/rc.local. If you just want it to
> happen on logon, you can put it in the .bashrc file. If you
> want it to happen anytime you open a Konsole window, you'd
> put it in .bash_profile.
> Any of those should work and are equally valid.
>   John
> 



Re: [newbie] Partition setup help

1999-10-28 Thread Matt Stegman

>   | > The solution, then is to create the partitions using the DOS fdisk utility,
>   | > including the ones you want for Linux. Then restart the 'puter, and run fdisk
>   | > again, but this time remove the partitions you want for Linux. Install Win9x,
>   | > then install Linux, recreating the partitions with disk druid. I think this 
>will
>   | > work.
>   | > 
>   | The problem with this, is that it makes the Linux
>   | partitions visible to Windows, which, I've heard, can cause
>   | corruption, as Windows tries to "fix" the "grunged" file
>   | system and can't.

... what?  This should definitely not happen!  Windows ignores partition
types that aren't FAT or FAT32.  Windows can indeed see the partitions,
but fdisk for DOS is so limited that it doesn't display them correctly.
If Windows does not recognize the partition type, it ignores the
partition. I've never seen Windows try to "fix" a Linux filesystem.  When
did this happen to you?

You should not create partitions for Linux under Windows.  Well, you can,
but you'll have to edit them later.  Windows will automatically set the
partition type to FAT.  In order to create an ext2 filesystem on the
partition, you first must change the partition type to Linux.  It's better
to simply not create partitions from Windows' fdisk unless they are
partitions to be used by Windows.

> Even if yoy create the partitions in DOS, restart in DOS, remove the partitions
> for Linux, restart in DOS, Install Win, restart with Linux installer and create
> the partitions again with disk druid, DOS will still see the Linux partitions,
> and try to "fix" them? Hmmm... I didn't know that. I never tried it before for
> Linux, but it sounded right. I just thought that when the partitions were
> removed in DOS, they would just be empty space on the disk, or extended
> partition. Then you could do anything you wanted to with the space. Guess
> I gave the Redmond boys too much credit this time. Thanks for seting me
> straight.

You sure are going about this the long way.  Why create partitions when
you're going to delete right away?  It's not like you have to fill up
the whole disk for DOS or anything.  Just create the partitions you need
for DOS in DOS, then format them and install your OS, then install Linux.
Disk druid or fdisk will let you create the Linux partitions during the
installation.  DOS/Windows should not try to touch your partitions at all.
The only problem might be if they have a FAT label in the partition table,
but a non-FAT filesystem on them.  That may cause Windows to "fix" the
partitions.  But it shouldn't happen, as long as you stick to creating an
OS's partitions with the right tool- DOS fdisk for DOS partitions, Linux
fdisk (or disk druid) for Linux partitions.

-Matt Stegman
<[EMAIL PROTECTED]>




Re: [newbie] Problem with swap partition

1999-10-18 Thread Matt Stegman

Swap problems aren't usually good, but I think you may have enough RAM to
make up for them... :)

Double-check the output from `fdisk -l /dev/hda` to make _sure_ the
partition is there.  If it is, try `ls -l /dev/hda*` to make sure the
files corresponding to the partition are there, and that permissions are
set correctly (should be 644 or 600... 644 would show up as -rw-r--r--
in ls).

All the above assumes that Linux is installed on the primary master hard
drive.

> Note, I boot Linux from floppy because I am a CIS student who can not afford to
> break winnt so am afraid to mess with mbr. 

Don't worry- NT doesn't care about the MBR.  It's only it's own
partition's boot sector that has to be intact.  You can safely put LILO in
your MBR.



-Matt Stegman
<[EMAIL PROTECTED]>




Re: [Re: [newbie] boot disk & LILO installation crash down]

1999-10-17 Thread Matt Stegman

> BTW this is NOT being sent in HTML

This one doesn't seem to be, but...

Today, I'm using Pine (ver 3.96L) to read my mail.  Depending on where I
am and what my computer is booted to, I may use Pine ver 4.10 or TkRat
(ver 1.2) to read my mail.  Depemnding on how things are sent, it either
takes me a lot of trouble to read HTML mail, or it's as easy to read as
plain text mail.

With Pine 3.96L, the HTML either shows up as an attachment beneath the
plain text version, ot both the plain text and HTML show up as attachments
to an empty mail.  When both are attachments, I won't read it, unless the
subject line really interests me.  When it takes no extra effort to read
plain text, I'll read it and suggest that the poster turn off HTML in the
future, as a courtesy.

When using Pine 4.10, HTML is interpreted and displayed fine.  No
additional attachments are shown.  Thus, unless the sender throws in a
link or something, I almost can't tell that HTML is being used.

TkRat works a lot like the older Pine, except it's easier to read
attachments.  By default, the "source code" of the attchment is shown, so
when it'a plain text there's no effort required.  HTML is a little more
difficult to read, but I can usually parse it okay.  Still, if someone
uses HTML, I'll rebuke them (hopefully in a polite manner).

The post I was responding to had an attachment of type "text/HTML".  I
assumed this was a duplicate version of the mail, only formatted in HTML
for those mailers that can read it.  This prompted me to suggest turning
off HTML in the mailer.

> About the guy who flamed getting flamed re: being TOLD to turn off HTML...I
> can agree with some of his opinion.

I suppose so, but from what I hear, mail clients are in use that are even
tougher than my old Pine on HTML mail- as in, it takes a LOT of extra
effort to read it.  So, it's easier for everyone if plain text is used.
Not to mention that ASCII is the universal format.

> > First tip: turn off HTML in your mail client when posting to this list.
> > Plenty of us use mail clients that don't handle HTML too well, and tend to
> > disregard posts using HTML- or flame those who use it.
> 
> So either quit whining about the use of HTML, or get a better Mail reader

I like my mail readers just fine.  Frankly, I see no reason to switch just
because someone can't take a little time to be courteous.  I try to stay
away from flame wars, but as has been said plenty of time before, HTML
eats up bandwith and offers little in the way of necessary features.  Most
of the time, what few extra features it offers are raely used.
Punctuation can emphasize just as well as boldness or italics.  I see no
reason to use HTML when plain text suffices just as well, while retaining
more compatibility and making less fuss.  If you insist, I can quit
whining about the use of HTML; from here on out I'll just ignore it.  Less
effort for me, certainly.

Not too long ago, Civileme posted that he forsees this list degenerating
into two lists- those who post in and respond to HTML, and those who
don't.  I hope that doesn't happen; I'd hate for anyone to be restricted
to helping only a certain group of people.

-Matt Stegman
<[EMAIL PROTECTED]>





Re: [newbie] sound card install

1999-10-17 Thread Matt Stegman

A good place to start would be running `sndconfig` from a command line (as
root, not a normal user).  Does this work for your card?

-Matt Stegman
<[EMAIL PROTECTED]>

On Sun, 17 Oct 1999, M. L. Cates wrote:
> I need some basic information on how to go about getting linux to
> recognize my sound card. I am running Mandrake 6.0
> 
> My sound card is a BTC 1817  ( opti 931 ). It is an inexpensive
> 16 bit generic type sound card.



Re: [newbie] boot disk & LILO installation crash down

1999-10-17 Thread Matt Stegman

First tip: turn off HTML in your mail client when posting to this list.
Plenty of us use mail clients that don't handle HTML too well, and tend to
disregard posts using HTML- or flame those who use it.

How are you booting into the installation?  Off the CD, or do you use a
boot floppy?  Also, it would be nice to quote exactly any error text you
get.  You're right - vague desciptions of errors aren't too often useful.

In my experience, when you change the default settings for LILO in the
installation, you usually get problems.  You might do well to take the
default and change it after you get your system running.

Regarding errors, you'll find them on the third and fourth consoles, I
think (been too long since I last re-installed).  Too see them, press
Alt-F3 or Alt-F4 at any time.  To get back to the install (which runs on
virtual console 1), press Alt-F1.

Can you tell us anything more?

-Matt Stegman
<[EMAIL PROTECTED]>

On Sat, 16 Oct 1999, granduke wrote:

> Hi everyone,
> 
> this is the first time I get on list. 
> I have tried to install Mandrake 6.0 on my Pentium, 2Gb, computer
> without success. Two main problems occcur me:
> 
> first - when the installation process gets to the rescue disk step it
> doesn't seem to find a diskette driver (the light doesn't turn on at any
> moment) and then obliges me to ommit this step.
> 
> Second - After the diskette driver problem has occured LILO is not able
> to install itself, I have tryed all possible options given in the
> installation menus. My hard disk is recognized by LBA mode.
> 
> So, after all these things I have all inside but I'm not able to reboot
> because I have no LILO installed nor a system disk to boot from.
> 
> Can anyone help me??
> 
> I guess I will have tu put on some more information in order to bring
> you a consistent clue, but at the moment my knowledge about Linux is
> limited so when you were so kind to do so ask me whatever you find
> necessary.
> 
> Thank you very much in advance
> 



Re: [newbie] Re: LICQ personal user info

1999-10-11 Thread Matt Stegman

On 11 Oct, Jaguar wrote:
> I d/l'ed LICQ in RPM from the net...and installed it.  How does a person pass
> your personal info (read UIN) to the program.  Is it a parameter/flag during
> LICQ starting, or an OPTION somewhere

Run the program from a command-line first.  Run a Konsole or Xterm or
similar program.  The first time Licq is run (or if it doesn't find a
personal info file) it will ask you for it- but not in a window; You
must be at a command line.
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] How to use ns-install that comes with v47 of Netscape for Linux 2.0

1999-10-11 Thread Matt Stegman

On 11 Oct, Dan Brown wrote:
> From: Richard Salts <[EMAIL PROTECTED]>
> 
>> Another question:  How does one unpack and install an *rpm program?
> 
> # man rpm

Good Lord, that's a lot of information!  Man pages say way too much for
most normal purposes, _especially_ the one for `rpm`.  Hell, even 
`rpm --help` outputs several screens of info.  I'd just recap the
basics:

rpm -ivh filename.rpm  # Installs a package
rpm -q package-name# Do I have package-name installed?
rpm -qa# What packages do I have installed?
rpm -Uvh filename.rpm  # Upgrade to new version of package
rpm -e package-name# Un-installs package-name

Take note of how "package-name" used in the query modes and uninstall
mode does NOT end in ".rpm".  If you don't know the name of a package,
try

rpm -qa | grep word

Where "word" is something that should be in the package.  For instance,
say you're wondering what GNOME packages are installed.  You run

[user@localhost user]$ rpm -qa | grep gnome
gnome-audio-1.0.0-7mdk
gnome-audio-extra-1.0.0-7mdk
gnome-core-1.0.9-9mdk
gnome-core-devel-1.0.9-9mdk
gnome-games-1.0.2-13mdk
gnome-games-devel-1.0.2-13mdk
gnome-libs-1.0.40-1mdk
gnome-libs-devel-1.0.40-1mdk
gnome-media-1.0.40-1mdk
gnome-objc-1.0.2-5mdk
gnome-objc-devel-1.0.2-5mdk
gnome-pim-1.0.9-1mdk
gnome-pim-devel-1.0.9-1mdk
gnome-print-0.8-1mdk
gnome-users-guide-1.0.5-7mdk
gnome-utils-1.0.13-1mdk
switchdesk-gnome-1.7.0-3mdk

There you go!  Let's now say you want to get rid of the GNOME utilities
(the second-to-last package on the list).  You'd run

rpm -e gnome-utils

No version numbers or file extension.  Does this make sense?  Hopefully,
it's more readable than a man page.
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] What now?

1999-10-09 Thread Matt Stegman

On  9 Oct, Joe Brault wrote:
> Hello,
> 
>   I have finished all of what the mini howto tells you to do to install a
> zip drive, but I do not know how to actually access my drive.  What is the
> command to actually mount the drive?  Or am I forgetting some steps in
> between.  Thanks in advance!
> 
> 
> Joe :)

Use the `mount` command.  Hopefully, somewhere in the HOW-TO there was
a step about putting a line in /etc/fstab.  Just mention either the
device name or the mount point (directory) to mount, and you'll be able
to access it.  Let's say that your zip drive is /dev/hdd (did you
create a '/dev/zip'?) and to be mounted at /mnt/zip.  You could give
either of two commands:

mount /dev/hdd
mount /mnt/zip

Both would work; mount just looks in /etc/fstab for the corresponding
line.  Be sure it's user-mountable ("user" being one of the
mount options) or else only root will be able to use the drive.
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] Widgets?

1999-10-09 Thread Matt Stegman

On  7 Oct, Kaplan, Paul wrote:
> I created a color scheme in kde 1.1.2 that has mostly dark gray backgrounds.
> However, in dialog boxes the check boxes use a black "x" to indicate an item
> is selected.  Since black is hard to see on dark gray, I was if it is
> possible to change the color of the "x"s to something that provides a higher
> color contrast.

Sure- on the "Style" menu, change uncheck the Windows 95 style
widgets.  The resulting "checkboxes" show up well on dark backgrounds.

-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>





Re: [newbie] Fwd: licq & libstdc++-libc6

1999-10-09 Thread Matt Stegman

On  6 Oct, David M. Kufta wrote:
>  I was attempting to instal licq-0.70 and found failed deps for
> libstdc++-libc6. Has anyone managed to get licq running on 6.1 and if so
> I would certainly appreciate some advice as to what I need to do to find
> and install libstdc++-libc6 without wrecking other deps on my 6.1 system.

I presume you're trying the one from Cooker?  I'd bet that the best way
to install it would be to install Cooker first, then Licq.  You could
always use the --force option with rpm, but like you say, that _may_
break things.

Again, I'll recommend a different that you go get a different licq rpm
from http://rpmfind.net/linux/RPM/licq.html
There's a .70f RPM at the top of that page that works fine on any
Mandrake 6.0+ system (maybe even 5.x, but I haven't tried it) _without_
(this is the important part for me) any required qt2 or libstdc++
library upgrades!

The only downside is that you have to specify the option on the command
line- run `licq -p qt-gui` instead of just `licq`.  Bero says that the
Cooker package gets around this, but I'll have to take his word for it,
as I'd rather not jump through hoops upgrading my libraries just to
install licq.

Not to knock the Mandrake distribution or anything; it's a fine
product that I recommend to anyone who wants to use Linux.  With licq,
however, I'm more comfortable running this other RPM than the one
offered in Cooker.  Now, they may have uploaded a new one that doesn't
require QT2 & new libstdc++, or perhaps made qt2 and/or libstdc++
easier to install (i.e. no dependency issues with 6.x), in which case
I'd try it again.  As it stands, though, I'll stick with what works
right now.
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] Dual Boot Win98/Linux

1999-10-07 Thread Matt Stegman

On  7 Oct, Jesse Royall wrote:
>  I have been trying to install Linux on my system. Tried several
> different things and I either get Win98 or Linux. I downloaded the
> Mandrake .iso and burned my CD and made my boot disk/setup disk. I can
> get it to install and run Linux. But I can't get it to dual boot. What am
> I missing here?

Check the archives, or read the lilo.conf man page.

[user@localhost user]$ man lilo.conf

to read the man page.  The mailing list archive is at

http://www.mail-archive.com/newbie@linux-mandrake.com/

There you can search for 'dual boot and lilo' or something similar.
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] Download Mandrake

1999-10-07 Thread Matt Stegman

On  7 Oct, Mike Easter wrote:
> you need it it contains important system files

No, you don't need it. 
/RedHat is just a symlink to the /Mandrake directory.  This is, I would
guess, to maintain compatibility with RedHat install disks.  I can't
think of any other reason, anyway.
-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] Aol Instant Messenger

1999-10-07 Thread Matt Stegman

On  7 Oct, Payne Stanifer wrote:
> No, it is a shell script. So what do I do now?

You're not telling us what we need to know.  I'd suggest running the
shell script to start with.

Which AIM client are you using?  What's the web site?  I'd help, but I
simply don't know what the problem is.  Okay, you can't install it...
well, there are many, many methods of program installation under Linux-
RPM, DEB are two packging systems that jump to mind.  Something might
come in a tarball of source code.  Or they might just tar up binaries. 
ICQ for java is distributed as compiled binaries, but require you to
run a shell script to determine environment variables.  Might this be a
situation similar to yours?

-- 
 -Matt Stegman
 <[EMAIL PROTECTED]>



Re: [newbie] the programming language

1999-09-19 Thread Matt Stegman

The kernel (which is what gives the name "Linux" to the operatinf system)
was written in C. You'll find that most applications for Linux are written
in C, as it's a very vesatile language (or so I'm given to understand,
I've yet to master it).

However, it's not the only language used.  Many programs are written in
C++ (a successor to C, I think mostly to add graphics capabilities), and
you may even encounter FORTRAN or Pascal.  Java is becoming a little more
widly used, and UNIX systems are full of shell scripts, usually written
for the Bourne Shell (/bin/sh).  Speaking of scripting, you'll also
probably see programs written in Tcl/Tk, Perl (i.e. DiskDrake), and
Python.

Of course, there are many more out there- this is just what I can think of
right off the top of my head.  I suppose, though, the answer you were
looking for was just that the Linux kernel, and most applications, were
written in C.

-Matt Stegman
<[EMAIL PROTECTED]>

On Sun, 19 Sep 1999, Joe Brault wrote:

> Sorry for the overly simple question... but can anyone tell me what
> language Linux was written in?  Thanks in advance and don't laugh too long
> and hard!
> 
> Joe :)
> 



Re: [newbie] losing lilo

1999-09-08 Thread Matt Stegman

You'll need to point lilo to the right config file: try

bash# /mnt/sbin/lilo -C /mnt/etc/lilo.conf

This tells lilo to use the config file /mnt/etc/lilo.conf.

You may also want to symbolically link /mnt/boot as /boot, because your
old config file points to /boot for all it's info.  If you already have a
/boot directory on the rescue disk (I don't know, never used it) then:

bash# mv /boot /old-boot && ln -s /mnt/boot /boot

Perhaps this will work for you?

-Matt Stegman
<[EMAIL PROTECTED]>



Re: [newbie] screen size

1999-09-07 Thread Matt Stegman

I agree- that whole desktop-bigger-than-the-screen thing bugs me, too.
You can get around it by making sure you either don't have a "Viewport"
section or commenting said section out (by starting the line with a pound
sign: Shift-3).

The other problem is that it sounds like you have 2 resolutions defined in
XF86Config:

> I use the Xconfigurator RIVIA TNT driver.
> I choose the 800 X 600 and the 1024 X 786 at 24bit
> I can switch between the two by clt alt + with no problems

When X starts up, it reads the section applying to the chosen color depth,
and then sets the desktop size to the largest resolution present.  Thus,
if you have listed "640x480" "800x600" "1024x768" "1280x1024" then your
desktop size will be 1280 pixels by 1024 pixels.  When you change
resolutions by Ctrl-Alt-+ (or minus) your desktop size will not change.
As far as I can tell, there is no way around this.  This will happen,
regardless of your "Viewport" settings.  

-Matt Stegman
<[EMAIL PROTECTED]>

On Tue, 7 Sep 1999, Ed Santiago wrote:

> I am running Linux Mandrake 6.0 with the 2.2.9-27 kernel. Everything
> seems to work fine but I have this question. The same that many others
> have had.
> I want to run my video at 800X 600 but it doesnt fit the screen. I have
> to use the mouse to move the screen around to see all of the desktop. I
> have to run at 1024X 786 in order to view the whole desktop. I have
> rerun the Xconfigurator and I still have the same thing.
> I use KDE almost exclusively.
> I have a 17" monitor.
> I have a Creative Labs Graphics Blaster RIVIA TNT 16mb (PCI) video card.
> 
> I have no problem getting into console mode and running Xconfigurator.
> But what do I need to do in order to make the 800 X 600 fit the screen?
> 
> Thanks in advance,
> Ed Santiago
> 
> 





Re: [newbie] grep and replace

1999-09-03 Thread Matt Stegman

Maybe this would do it (as a shell script):

#! /bin/sh
for $i in `find .`; do
cp $i $i.old
cat $i.old | sed -e "s/MarketingWorks\/Collard
Associates/MarketingWorks" > $i
next;

I am NOT a shell script guru- you can tell because my for loop syntax is
wrong, and I'm not sure what the proper way to write it is.  I hope I
get my idea across, though.  You might run it as `script 2> /dev/null`
because it will generate an error message for every directory under ./
as I couldn't figure out a way to get `find` to list only files, or 
`ls -R` to not print out directory names.  
With this pseudo-script, you also get backups of your HTML files
(with a .old extension) in case it screws up.  If it works, then you can 
rm `find . |grep .old`
to recusively remove all old files.  I think.

-Matt Stegman
<[EMAIL PROTECTED]>

On Thu, 2 Sep 1999, James Stewart wrote:

> Been investigating the problem I posted yesterday a bit more, and have a
> slightly better way of expressing it now ;)
> 
> What I want to do is like issuing:
> 
> grep -r MarketingWorks\/Collard Associates /home/httpd/marketingworks.co.uk
> 
> but instead of listing the results I want to replace it with simply
> MarketingWorks
> 
> Another way of looking at it is issuing perl's
> s/MarketingWorks\/Collard Associates/MarketingWorks/;
> 
> but recursively through the directory structure.
> 
> Can anyone suggest how to do this?
> 
> James.
> -- 
> James Stewart |  Britlinks |  The Phantom Tollbooth
> [EMAIL PROTECTED] | http://www.britlinks.co.uk | http://www.tollbooth.org
> 
>  Sixpence None The Richer UK -- http://www.britlinks.co.uk/sixpence/
> 




Re: [newbie] Display problems and LILO question

1999-09-03 Thread Matt Stegman

On Fri, 3 Sep 1999, Hershel S Robinson wrote:
> 1 How do I change the screen resolution within Linux?  It is too high
> and everything is hard to read.

If your XF86Config file has multiple resolutions defined in it, you can
switch by pressing Ctrl-Alt-plus or Ctrl-Alt-minus (use the + and - keys
on your numpad, not beside the backspace).  You can change the default by
editing /etc/X11/XF86Config.  Look for a section like this:

Section "Screen"
Driver  "svga"
Device  "My Video Card"
Monitor "My Monitor"
DefaultColorDepth 24
Subsection "Display"
Depth   24
Modes  "1024x768"
ViewPort0 0
EndSubsection
EndSection

See where it says "1024x768"?  Change that to whatever resolution you
like- as long as there's a modeline for it above this section.  Which
shouldn't be a problem.  If changing this doesn't help, change the section
that begins with the line 
Driver "accel"
This is the accelerated X server, and your card may use that instead.

> 2 When I run Gimp, my screen colors change--they become so dark I can
> not read the buttons on the title bar of Gimp itself.  When I activate
any other window, the colors return to normal again.  What's the problem?

Don't know about this- I suppose it might be a problem with your video?
What color depth are you running in?  Try changing to a different color
depth (GIMP requires at least 16-bit).

> 3 I tried to change the default system of LILO from somewhere inside
> the linuxconf, but when I activated the changes, it generated an error.
> I think I saw somewhere that LILO does not run so well with Windows NT.
> Is this true?  Even if it is true, is there a way to change the default
> system?  (I am expecting to get another hard drive in a few months, but
> in the meantime, I only have one)

I've never used LILO with Windows NT, so I can't say.  What I did use was
a rather awesome boot manager called OS-BS (see
http://www.prz.tu-berlin.de/~wolf/os-bs.html).  The beta version is free,
and works perfectly well.  You could also get he new one, with gads of
features, for some cash (don't remember how much).  I can say from
experience that this will let Windows NT & Linux & Windows 95 co-exist
very well.

If you do decide to use this other boot manager, keep in mind that LILO
will still have to be installed- installed on the Linux partition, instead
of the MBR.  OS-BS should go in the MBR.  And as always, read README's
first.

-Matt Stegman
<[EMAIL PROTECTED]>




Re: [newbie] how to dual boot

1999-09-02 Thread Matt Stegman

Whoops- I lied.  The command for step 4 should be:

dd if=/dev/hda1 of=/root/BOOTSECT.LIN bs=512 count=1

Of course, you still replace hda1 with your Linux partition.

-Matt Stegman
<[EMAIL PROTECTED]>

From: Matt Stegman <[EMAIL PROTECTED]>
Subject: Re: [newbie] how to dual boot

Here's what I know (I haven't tested this, but I think it will work okay):
1: Boot into Linux.
2: Edit /etc/lilo.conf such that the line "boot=" is set to
   "boot=/dev/hda1" (change hda1 to your linux partition).
3: Run /sbin/lilo.
4: Run the command 
dd if=/dev/hda1 of=/root/BOOTSECT.LIN
   Where you substitute your linux partition for hda1.
5: Put BOOTSECT.LIN somewhere where NT can see it- on a FAT16
   partition or a floppy disk.
6: Boot to NT.
7: Copy the BOOTSECT.LIN file to C:\.
8: Edit C:\BOOT.INI (you'll have to `attrib -r -s -h c:\boot.ini` 
   first.)
9: Add a section like so:
C:\BOOTSECT.LIN="Linux Mandrake 6.0"
10: Save & Exit.  Run `attrib +s +r +h c:\boot.ini`

You should be done.  If things don't work out, you can "rescue" your NT
install using the rescue disk set (run rdisk if you haven't got one).

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 3 Sep 1999, peterwt wrote:

> i install both winnt4 and mandrake 6.0 on my computer.
> i want to use nt'boot loader to boot mandrake.
> how to do that? now i use a floppy disk to boot mandrake.
> i have tried to do so as how-tos , unfortunately it does not work.
> thanks in advance.
> 




Re: [newbie] how to dual boot

1999-09-02 Thread Matt Stegman

Here's what I know (I haven't tested this, but I think it will work okay):
1: Boot into Linux.
2: Edit /etc/lilo.conf such that the line "boot=" is set to
   "boot=/dev/hda1" (change hda1 to your linux partition).
3: Run /sbin/lilo.
4: Run the command 
dd if=/dev/hda1 of=/root/BOOTSECT.LIN
   Where you substitute your linux partition for hda1.
5: Put BOOTSECT.LIN somewhere where NT can see it- on a FAT16
   partition or a floppy disk.
6: Boot to NT.
7: Copy the BOOTSECT.LIN file to C:\.
8: Edit C:\BOOT.INI (you'll have to `attrib -r -s -h c:\boot.ini` 
   first.)
9: Add a section like so:
C:\BOOTSECT.LIN="Linux Mandrake 6.0"
10: Save & Exit.  Run `attrib +s +r +h c:\boot.ini`

You should be done.  If things don't work out, you can "rescue" your NT
install using the rescue disk set (run rdisk if you haven't got one).

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 3 Sep 1999, peterwt wrote:

> i install both winnt4 and mandrake 6.0 on my computer.
> i want to use nt'boot loader to boot mandrake.
> how to do that? now i use a floppy disk to boot mandrake.
> i have tried to do so as how-tos , unfortunately it does not work.
> thanks in advance.
> 



Re: [newbie] How daunting is a kernel upgrade?

1999-09-02 Thread Matt Stegman

On Thu, 2 Sep 1999, Alan  Schussman wrote:
> My case is that I'm running Mandrake 6.0 and would like to upgrade the
> kernel to get rid of some of those mount/unmount errors that have been
> described here already. Is the best upgrade for me the one at cooker
> (kernel-2.2.11-2mdk.src.rpm), as opposed to the multitude of files I find
> at kernel.org? Is it preferable for me to stick to a Mandrake release of
> any given kernel?

Given that Cooker is under development, not exactly kosher Mandrake 6.0,
I'd say that it's safest to go with the 2.2.9 RPMs from the Mandrake
Update icon on your KDE or GNOME desktop.  Now, the 2.2 kernel should be
stable and quite usable in all it's incarnations, and generally the bigger
numbers are better than the smaller ones (bugfixes, etc.).  There probably
wouldn't be any problem with using the 2.2.11 kernel from Cooker, though-
the choice is up to you.

You should know, however, that the .src.rpm file will NOT install the
kernel.  It will install the kernel source, from which you may compile
your own.  If you feel comfortable doing so, then go for it.  If not,
well... you can always keep your old kernel, and boot to that if you have
problems.  Or you could install a kernel binary RPM from Cooker
(don't know which one, I haven't looked.  I'd image it's called
kernel-2.2.11-2mdk.rpm, though).  Given the profuseness of documentation
on the subject, you should be able to find out almost anything you want to
know (about downloading source, compiling & installing, not about
Mandrake kernel RPMs).

-Matt Stegman
<[EMAIL PROTECTED]>





Re: [newbie] Doc

1999-09-02 Thread Matt Stegman

Look in the "docs" directory on the CD.  As for installation, look for
files called README or README.html or README.txt.  Also, if you see HTML
files, look for index.html as it will usually have useful information too.

-Matt Stegman
<[EMAIL PROTECTED]>

On Thu, 2 Sep 1999, Iulian Ungureanu wrote:
> Hi I'm planning to install Mandrake 6.0 and I wander if there is any
> printable documentation 
> on step by step installation and HOWTO's.



[newbie] Bad ISO Image?

1999-08-21 Thread Matt Stegman

I downloaded the mandrake60-2.iso image off a mirror, and while snooping
around, fount several broken links under Mandrake/instimage.  Is this
particularly bad, or can I go ahead & burn this without worrying about it?

-Matt Stegman
<[EMAIL PROTECTED]>



Re: [newbie] failing to unmount

1999-08-21 Thread Matt Stegman

On Sat, 21 Aug 1999, you wrote:
> Hi Matt. I can connect to the internet with Netscape and KFM (that's all
> I've tried so far), but when I try that updates icon I get a can't connect
> error. Do you have to configure it to connect? I have a DSL connection
> through a proxy server.

The "updates" icon on the desktop does not have to be configured separately to
connect to the Internet.  However, your error message may mean that it cannot
connect to a mirror site.  Try using a different mirror.

 -Matt Stegman
<[EMAIL PROTECTED]>



Re: [newbie] Changing the Window Manager

1999-08-20 Thread Matt Stegman


On Fri, 20 Aug 1999, Axalon Bloodstone wrote:

> Only a little, ;) try the enlightenment from cooker 0.16 will implement /
> does implement, the ability to drag apps around in the pager from desktop
> to desktop, aswell as around the virtual desktop. Just don't drop them
> between pagers, they haven't fixed it to snap them back where they were so
> they wind up disapearing

I downloaded it a few days ago, actually.  I'm getting ready to move
tomorrow, though, so I haven't installed it yet, and probably won't until
next week.  All the same, AfterStep _will_ let you drag windows between
desktops.  It's something I wouldn't want to be without- except that I
like the alt+{left|mid|right} button in E that lets you drag windows,
resize them, and get that nifty window function menu without having to go
all the way up to the title bar.

The only reason I'd rather use AfterStep is the cool pager, and that I
notice a speed difference (AfterStep is much faster on my machine).  If
this pager works well, I may just have to buy some RAM to make up the
difference.  Too bad it's going up in price again.

-Matt Stegman
<[EMAIL PROTECTED]>




Re: [newbie] Does Linux use the Bios for Harddrives?

1999-08-20 Thread Matt Stegman


> I have a 520 meg harddrive on my NEC laptop.. 
> 
> I am going to run Slackware 4 on it.. because I can do a Floppy install
> with it.
> Can you do a FTP install of Mandrake?? 

Yes, you can.  Get the bootnet.img file from /updates/6.0/images off your
local mirror

> Anyway.. how do you think I should partition my harddrive for a basic X
> internet laptop.. nothing too complicated.. just web email and IRC.. 

I'd say either: 
1) / and swap space
2) / and /home and swap space

The only reason you'd want to separate out /home is a) if your root
partition becomes corrupted, you can still preserve your personal files
and b) if you need to upgrade, you don't lose everything when you format
the root partition.

> I am getting a 1.4 gig for the laptop.. so I will be using that in future
> for Linux I think.

That sounds better.  With the way distributions are these days, less than
1GB of hard drive space seems like way too little.

-Matt Stegman
<[EMAIL PROTECTED]>




Re: [newbie] RealPlayer G2 and MP# problems

1999-08-20 Thread Matt Stegman

> My sound configuration is fine as far as I can tell.  Audio CD's and wav
> files play clearly, it's just the .rm, .ram, .mp2, and .mp3 formats that are
> garbled to the point of incomprehension.

In that case, it sounds like the software decoder you're using has bugs
in it.  What software do you use to decode the MPEG audio files?

-Matt Stegman
<[EMAIL PROTECTED]>




Re: [newbie] failing to unmount

1999-08-20 Thread Matt Stegman

Really?  Out of curiosity, are you shutting down from KDE (runlevel 5) or
the console (runlevel 3)?  Also, what is the output of 
`rpm -q initscripts`?

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 20 Aug 1999, Wayne Boaz wrote:

> FWIW. I've done that and I still get umount - Failed when I shut down...
> 
> Wayne



Re: [newbie] Changing the Window Manager

1999-08-20 Thread Matt Stegman

> Steve wants to know how to start a different window manager from the command
> line, not from within KDE (or booting to initlevel 5).

Right, that's what `desktopcfg` does.  Well, kind of.  It is a console
utility that allows you to change your default desktop- your choices are
GNOME, KDE, AfterStep, or "Plain X11" which means fvwm2, I think.  If you
run desktopcfg as root you may change the system default.

And since you asked...
Another way to change the window manager that starts when you run `startx`
is to put a command in ~/.xinitrc.  This file is a shell script run
after X has started (it is usually what loads a window manager).
`desktopcfg` uses this file to start your window manager (WM). You may
safely delete this, and specify your own script.  Mine follows:

#/home/matthead/.xinitrc
exec kde

I think that I'm going to replace "exec kde" with "exec enlightenment" as
I'm tired of KDE, and would like to use Enlightenment for a while.  The
only thing I don't like is that you can't move windows around in the
pager (a point for AfterStep, 'cause you can in there!)...  but that's
off-topic.

Yet another way to specify a window manager to run with X is by appending
it to `xinit` (`startx` is a shell script that runs `xinit` with some
default arguments).  Try, for instance, `xinit /usr/bin/enlightenment`
This will start X and then run the Enlightenment WM.  Because of the way
`xinit` interprets your arguments, you have to specify the full path to
the program you wish to run in X (it doesn't _have_ to be a WM).  It
doesn't even matter if the path to the program is in $PATH. If you're
interested in the why, try `man xinit`.  

Which window managers are included with Mandrake?  Here's a few (I
installed the extra window managers package, so all these may not be on
your system): 
/usr/bin/enlightenment
/usr/bin/gnome-session 
/usr/bin/kde
/usr/X11R6/bin/AnotherLevel
/usr/X11R6/bin/afterstep
/usr/X11R6/bin/fvwm
/usr/X11R6/bin/fvwm2
/usr/X11R6/bin/wmaker

I'm sure there are more.  Also, gnome-session starts GNOME, which is not a
window manager, but rather a "desktop environment."  This includes
the panel, the right-click menu on the desktop, the icons, and more.  
GNOME uses Enlightenment as it's WM by default, but can be configured to
use WindowMaker, or some other WM I can't remember right now.  The same
goes for KDE, execpt that I don't think you can successfully use any
WM other than kwm, which is KDE's default.

This is probably much more information than you ever cared to know, but
it's still not all!  However, I'll bet `desktopcfg` is plenty for most
people.  The thing is, if you want more control over what happens when you
start X, well, you've got it. This is one of many advantages that Linux
has over Windows: the freedom of choice.  For any given task, there are
usually several different ways to accomplish it, each with it's own unique
advantages & disadvantages.  This means you can find the best way to get a
job done- you're not locked into someone else's idea of "the best way,"
like you are with proprietary systems (Microsoft being only the most
prominent).

-Matt Stegman
<[EMAIL PROTECTED]>





Re: [newbie] Changing the Window Manager

1999-08-20 Thread Matt Stegman

You can run `desktopcfg` from the command-line, or (from KDE) click on the
K Menu, then Utilities> Desktop Switching Tool.  This will let you choose
bewteen Gnome, KDE, and AnotherLevel.  The command-line tool gives you a
few more options.

For even more control, you could just put the window manager directly in
your .xinitrc file.  If you're interested, I can explain it, but otherwise
I don't want to take up bandwidth with a bunch of stuf nobody wants to
read.

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 20 Aug 1999, Steve Spiller wrote:

> This may seem like a dumb question, but I have only run Linux-Mandrake
> for 1 week. When I run startx, it automatically boots into KDE. This is
> fine, but I would like to test the other Window Managers that I
> installed. How do I use startx to boot into WindowMaker (for example)?
> 



Re: [newbie] failing to unmount

1999-08-20 Thread Matt Stegman

I believe that if you download the initscripts update RPM, the umount
problem will be fixed.  If you can connect to the internet from Mandrake,
click the "updates" icon on the KDE desktop.  Choose a nearby mirror, and
away you go!

If you can't get online from Linux, you can surf to
http://www.linux-mandrake.com/en/fdownload.php3 and choose a mirror near
you.  You'll find update RPMs in the updates/6.0/RPMS directory.

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 20 Aug 1999, Joe Brault wrote:

> HELP!!! my linux system is failing to unmount the file system.  How can I
> have the system unmount correctly???  Help!!!
> 
> Thanks
> 
> 



Re: [newbie] Dual Boot

1999-08-20 Thread Matt Stegman

I'm guessing you want the choice to boot between Linux & Windows?  Well,
you just need to setup lilo a little differently.

First, you need to know which partition your Windows C: drive is.  If
you're not sure, run `fdisk -l /dev/hda` (assuming Windows is on hda) and
look for a FAT partition.  Once you know the Windows drive (we'll say it's
/dev/hda1) you need to edit lilo.conf (in any editor you like).

There will be a section that looks like:

image=/boot/vmlinuz-2.2.9-19mdk
label=linux
root=/dev/hda5
read-only

You need to add a section either just below these lines or just above
them.  Whichever one is first will be loaded by default.

other=/dev/hda1
label=win
table=/dev/hda

Above is an example section for Windows.  Adding this to lilo.conf (and
running `/sbin/lilo` after saving it) will let you boot Windows from
/dev/hda1 by typing "win" at the LILO prompt.

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 20 Aug 1999, Jason Cunningham wrote:

> In some of the messages I have been reading there is mention
> of determining which OS to boot with.  Mine will only boot
> on Linux.  There is no choice.  I did not partition the whole
> hard drive to Linux.  I believe windows is still there.
> 
> Jason
> 



RE: [newbie] Not getting anywhere!!

1999-08-20 Thread Matt Stegman

I think that this is due a video card driver limitation in XFree86.  Bad
settings for the monitor would produce different results.  I know that my
card (Diamond Stealth 3D 4000) works fine at 32bpp in Windows, but in
Linux, I can't use 32bpp without running into what you're describing.  You
might try a lower color depth, but the same resolution- does that help?

-Matt Stegman
<[EMAIL PROTECTED]>

On Fri, 20 Aug 1999, Rick Fry wrote:

> I'll be the first to admit that the monitor that I use is a $200 special at 
> Sears. And, I too am getting the black on black menus and background until I 
> put a graphic back there. Then the labels are black on black and so are the 
> menus until I highlight them. Right now, I'm using Windows 2000 release 
> candidate 1 with the monitor set at 1280x1024 at 24 bit color with NO 
> problems like I'm getting in Linux. Even Windows 98 second edition uses 
> these parameters.
> 
> 
> Original Message Follows
> I had a similar video configuration problem.  The default setting
> Xconfigurator was choosing was giving a black screen with blackish icons (I
> think that means that the refresh rates don't synch).  My problem was solved
> by manually choosing my display settings in Xconfigurator.  Another thing I
> tried is reducing the type of monitor to one not quite so good (in custom
> monitor section).  My monitor, being on the cheaper end of the scale does
> not work well at its maximum settings.  By reducing the setting the picture
> quality improved considerably.
> 
> I suffered similar problems using my hardware with Windows 95.  My monitor
> does not perform well at its maximum refresh rates etc and I have terrible
> problems under Win95 with my AWE64 soundcard.  If the soundcard is enabled
> in the device manager, Windows will not boot.  I have to boot with the
> soundcard disabled and then manually turn on the soundcard using Device tab
> - a real pain in the arse if you forget to turn it off at the end of the
> session (Windows crashes on startup, reboot into safe mode, turn soundcard
> off, reboot again).
> 
> I might have problems with sound permissions with my Linux box, but at least
> the card makes noise!
> 
>  > -Original Message-
>  > From:  Rick Fry [SMTP:[EMAIL PROTECTED]]
>  > Sent:  Friday, 20 August 1999 9:43
>  > To:[EMAIL PROTECTED]
>  > Subject:   Re: [newbie] Not getting anywhere!!
>  >
>  > My sentiments exactly. I finally got my video configuration to go 
> 1024x768
>  >
>  > but the menus [for the most part] are black on black. This makes it just 
> a
>  >
>  > tad difficult to read.
>  >
>  >
>  > Original Message Follows
>  >
>  > Well I think I am going to give up on my install of Mandrake.
>  >
>  > I cannot get it to recognize my modem or sound card.
>  >
>  > I have found out the my modem is on Com3 IRQ11   but have found no way..
>  > even with everyones help of finding it or configuring it.
>  >
>  > As for my sound card it is found by linux when it boots.. but it aborts
>  > putting it in memory because it is sharing an IRQ with another item on 
> the
>  > PCI bus.. which is fine in windows but not in linux I guess..
>  >
>  > So thanks all for your help.. but this is just not hardware friendly
>  > enough
>  > for my current system.
>  >
>  > I guess if I want to run Linux I am going to have to build a specific box
>  > with all the specific hardware able to be detected and run.
>  >
>  > James
>  >
>  >
>  >
>  > ___
>  > Get Free Email and Do More On The Web. Visit http://www.msn.com
> 
> 
> ___
> Get Free Email and Do More On The Web. Visit http://www.msn.com
> 



Re: [RE: [[newbie] realy dumb and revisited Q]]

1999-08-17 Thread Matt Stegman

Um, actually, neither.  COMMAND.COM is the DOS shell- i.e. bash
equivalent.  COMMAND.COM is, however, much more limited.  The equivalent
to autoexec.bat & config.sys would be startup scripts- the /etc/rc.d
directory tree.  LILO is a boot manager- there's no real equivalent to it
for DOS, because DOS is, well, lame.  LILO chooses which of the kernels
you have on your system to boot.  You can't have, much less boot,  
multiple kernels in DOS.  LILO also has the ability to load boot sectors
from any partition off any hard drive.  DOS's MBR loads only the boot
sector off the active partition; it cannot handle multiple PRIMARY
partitions, much less multiple bootable (active) partitions.  LILO, on the
other hand, can even boot from logical drives inside extended partitions.  

That, as little and incorrect as it may be, is my two cents...

-Matt Stegman
<[EMAIL PROTECTED]>

On Tue, 17 Aug 1999, Theo Brinkman wrote:

> Actually it's closer to the linux equivalent of DOS's 'COMMAND.COM'.
> 
>   - Theo
> 
> Joseph Gardner wrote:
> > 
> > Ok, ok I get the point.  It's the linux equivalent of DOS's config.sys and 
>autoexec.bat, right.
> > Regards,
> > 
> > Joe
> > 
> > -Original Message-
> > From:   Michael Scottaline [SMTP:[EMAIL PROTECTED]]
> > Sent:   Tuesday, August 17, 1999 2:31 PM
> > To: [EMAIL PROTECTED]
> > Subject:Re: [RE: [[newbie] realy dumb and revisited Q]]
> > 
> > Joseph Gardner <[EMAIL PROTECTED]> wrote:
> > What does one mean rerun lilo (I know RTFM, but I don't have my books with me
> > 8-))
> > 
> > Regards,
> > 
> > Joseph Gardner
> > Senior Designer / Technical Support
> > Kirby Company
> > Cleveland, OH
> > 
> > >From a command line just type:  lilo
> > 
> >   
> > 
> >Part 1.2Type: application/ms-tnef
> >Encoding: base64
> 




Re: [newbie] Can I install Linux like this?

1999-08-17 Thread Matt Stegman

> Okay, I have a computer onto which I've downloaded al of Mandrake 6.0 (both
> the distribution and the CD-ready ISO version). I have another computer
> onto which I'd like to load Linux. Right now the computers are not
> networked, but that's the plan. I don't have a Linux CD-ROM (or a burner to
> make my own), and out of stubborness I am refusing to get one -I have all
> the files, I oughta be able to do this.

Well, as soon as they're networked, you'll be ready to go.  I assume the
files are downloaded onto a Windows machine?


> What I'd like to do, is
> 
> 1) Make a boot floppy from my downloaded Mandrake files (which I'm sure is
> simple but I don't know how)

Pretty simple- in the /dosutils directory is a program called RAWRITE.EXE.
Copy this to C:\.  Now, copy /images/bootnet.img to C:\.  Now, from a DOS
prompt, run 

C:\>rawrite -f bootnet.img -d a

Be _sure_ you have a cleanly formatted floppy with no bad sectors on it in
the floppy drive.  This will produce your boot floppy.  Actually, you may
want to use the updated boot image: /updates/6.0/images/bootnet.img from a
Mandrake mirror.

> 2) Boot the other computer up with Linux, using the floppy

Hmmm... put the floppy in the drive, and boot... I guess you should make
sure that the floppy is first in the boot sequence.

> 3) Put network support on the other computer via floppies (I imagine the
> network shouldn't take more than a floppy, and I assume whatever kernel I
> boot with ill have network support).

Well, you do have to have a network card in there, too.  The floppy image
contains a multitude of drivers, I'm sure the list is in some READMEs
somewhere.

> 4) Do a network install

Follow the directions on the screen.  It's not that hard.  My first
install with 5.3 was an FTP install, so I know even a moron like me can
get through it okay.

> I don't really know how to do any of these, but I think that 2, 3, and 4
> are covered in the Mandrake docs so I'm not too worried. Mostly it's 1
> that's driving me nuts - I can't find anything explicitly telling me how to
> make the boot floppy! I'm sure that's my fault, but where oh where?
> 
> The other option would be to network the computers together, and do an FTP
> install. However, I have yet to see an FTP install explained (though I have
> seen it mentioned, both on this list and in the docs).

It's as much of a no-brainer as any other network install.

> Can anyone help? 

I hope I did.

> Thanks,
> Chris 

You're welcome, Chris.

-Matt Stegman
<[EMAIL PROTECTED]>





Re: [newbie] lnx4win

1999-08-17 Thread Matt Stegman

I'm afraid that Linux cannot reside on a DOS partition without some
special means to protect the parts of the filesystem that Linux relies on.
Certain security features in Linux/UNIX- file ownership & the permission
modes- are features inherent in the filesystem, not the OS.  FAT
filesystems do not have these features, and so any program that makes use
of them simply cannot run off a FAT partition.  UMSDOS is a way of
"protecting" Linux from the problems (well, some of them anyway) of FAT.

Using loadlin, the "root=" part should always point to the root partition.
The kernel image (which is usually present in /boot) should reside on the
DOS partition with the loadlin executable.  This is because loadlin cannot
read ext2 filesystems.  The kernel is loaded from DOS via loadlin, DOS
gets pushed out of memory, and Linux takes over.  The Linux kernel can
read ext2 (unlike DOS) and then gets the root partition- from the loadlin
parameter- and mounts it at /.

I suppose- although I'm not sure how this could be implemented- that it
would be possible to have an ext2 filesystem contained in a file on a FAT
partition.  I don't know if the kernel would accept a DOS path\filename
for the root partition (probably not), but if you can round up a team of
hackers, you might be able to use this to use Linux from a FAT partition
without usinng UMSDOS.

-Matt Stegman
<[EMAIL PROTECTED]>

On Mon, 16 Aug 1999, Sean Armstrong wrote:

> I have installed the new Mandrake 6.0.  However I can not seem to get
> loadlin running.  I set it to my new kernel and directed it towards the
> root partition.  I think the second part is my problem.  Should I have
> directed it towards my boot partition?  Anyways, it begins to load linux
> and then panics and stops.  Also I had the same problems when I tried to
> use Mandrake's lnx4win software to load linux on to a dos partition.
> The readme for this is vague at best and I was wondering if anyone knows
> of a better source to direct me how to load linux onto a dos partition
> without the use of umsdos.
>
> thanx,
> SA




Re: [newbie] arrgh vfat and msdos unknown?

1999-08-15 Thread Matt Stegman

On 15 Aug, Will wrote:  
> well I got those 2 emails telling me how to get into my windows drive
> but both of your approaches didn't work. I reinstalled linux somewhat 
> successfully (for the 3rd time!) but get a few warnings about the wrong
> kernel ver in system.map and an error with mounting somewhere when 
> booting. I don't know if loadlin is doing it since that was the first 
> time I've used it. 

Could you quote the exact text of the messages?  If they scroll off the
screen too quickly during boot, you should be able to use `dmesg` to view
boot messages from a command prompt. 

> How do I mount my FAT32X and FAT16B drives onto linux? Is vfat used for
> FAT32 and msdos used for FAT16?

Nope.  Both should accommodate FAT12, FAT16 and FAT32.  The difference
between vfat and msdos is that "vfat" preserves Windows 95 long file names
(and lowercase letters!) while "msdos" uses the old MS-DOS 8.3 format. 

> I know the commands are:  > mkdir /mnt/windows and /mnt/ddrive 
> mount -t vfat /dev/hda1 /mnt/windows and 
> mount -t msdos(i'm assuming for fat16) /dev/hda5 /mnt/ddrive 
> I've tried both in the console as root and it says unknown file system
> or something like that. 

Well, that would suggest that either a) you don't have these filesystems
compiled into the kernel or b) the module isn't loaded (although it should
be by default).  Again, it might help if you included the exact error
text.  Since you've already re-installed twice, and are experiencing other
problems, you really might try re-installing again.  A stock installation
of Mandrake should have full support for DOS/Windows filesystems. 

> of the FAT16 and FAT32 file systems, which is superior in file 
> read/write/transfer speed and efficiency? I purposely created another 2
> gig FAT16 partition (faster than a 13 gig FAT32 partition) to hold all 
> the stuff I want to host with linux. I'm under the impression that the 
> smaller a partition is the faster and more efficient it becomes, and I 
> feel this would be very important if I'm hosting and files are 
> constantly being created/written/removed. please correct me if I'm 
> wrong. 

I believe that FAT32 is faster than FAT16 for reading, writing, etc. 
FAT32 also has the advantage of a smaller cluster size.  What does this do
for you?  Well, let's say you have a small file you save on your drive-
only 1 or 2 kilobytes.  On a 2GB FAT16 partition, cluster size is 64kB, so
this file _occupies_ 64kB, even though it's less than 2kB in size.  With a
FAT filesystem, every file must occupy an integer number of clusters.
Cluster size is determined by partition size.  The smaller the partition,
the smaller the cluster size.  I don't have the conversion table for
cluster size, but I know it's available on the Internet. FAT32 has smaller
cluster sizes for larger partitons- my 7.2GB drive was once formatted as a
single FAT32 partition with 4kB clusters. 

-Matt Stegman 
<[EMAIL PROTECTED]> 




RE: [newbie] bash scripts

1999-08-14 Thread Matt Stegman

> Store the process id of the ghostview. You can send signal to the
> process for termination using kill. But this kills the process abruptly
> and many of cleanup operation by the process would not be done.

I believe you can send a termination signal rather than a kill signal.  I
don't know what the `kill` command sends by default (too lazy to read the
man page), but I think if you do `kill -TERM ` the process should 
terminate, after doing whatever cleanup it needs- assuming it recognizes
the TERM signal.

 -Matt Stegman 
 <[EMAIL PROTECTED]>



Re: [newbie] The shutdown and halt user

1999-07-31 Thread Matt Stegman

`shutdown` checks a file /etc/shutdown.allow (which is just a list of
usernames) for users allowed to run it.  Everyone that you want to have
permissions to shut down the machine should have their username listed in
/etc/shutdown.allow.

 -Matt

On Sun, 1 Aug 1999, Ikhlasul Amal wrote:

> I met on /etc/passwd that there exist shutdown and halt username. When I
> tried to use their feature (shutdown and halt as shell), I got an error
> message telling that I had no permission for using shutdown. How can I
> set shutdown so certain users/groups are allowed executing it without
> having to have root previledge? (I ever met that option on one
> configuration dialog -- but, ooppss, forget what it was!)
> 
> -- 
> @mal
> [EMAIL PROTECTED]
> 



RE: [newbie]Change the command line prompt

1999-07-30 Thread Matt Stegman

You might be interested in the Bash-Prompt-HOWTO.  It goes into much more
detail than I can/will here, and should be available in /usr/doc/HOWTO or
if you didn't install the documentation, at
http://metalab.unc.edu/LDP/HOWTO/Bash-Prompt-HOWTO.html

Basically, the prompt is stored in the PS1 environment variable.  This can
include environment variables, and special backslash-escaped variables
including the following (`man bash` for an exhaustive list):

\h the hostname up to the first `.'
\H the hostname
\u the username of the current user
\w the current working directory
\W the  basename  of the current working directory
\$ if the effective UID is 0, a #, otherwise a $

Thus, the default prompt is PS1="[\u@\h \W]\$ ".  you can see this by
typing `set`, which shows a list of all environment variables.  The way to
set the prompt on login is to put this command in one of the scripts that
bash runs when it starts.  According to the bash man page, bash will
execute /etc/profile first, then one of ~/.bash_profile,  ~/.bash_login,
or  ~/.profile, whichever comes first.  ~/.bash_profile should exist, and
should tell bash to run ~/.bashrc if it exists.  ~/.bashrc should be
telling bash to run /etc/bashrc.  Thus, if you put the PS1 command in
/etc/bashrc, it will be run for all users, and will overwrite the prompt
setting in /etc/profile, but each user still has the option to overwrite
your prompt by appending their own PS1 command to ~/.bashrc.

In the above paragraph, ~ stands for the user's home directory (same as
$HOME).  I hope this gives you something to start from.  Remember that the
HOWTO goes into much more depth on the subject, including colors and
running commands inside the prompt!

 -Matt




Re: [newbie] shuting down by ord. user..

1999-07-29 Thread Matt Stegman


> How about creating a quick script '/etc/threefingersalute', like the one
> below:
> - -
> if [ $USERNAME = "ROOT" ]; then

I'd suggest you use $LOGNAME instead of $USERNAME.  On my system, `echo
$USERNAME` returns nothing.  `echo $LOGNAME` on the other hand, returns
the username of whoever's logged in.  It works just fine with `su` too.
Also, "root" should be lowercase.

>  halt
> else
>  logout
> fi

Other than that, it looks good.

 -Matt



Re: [newbie] Removing LILO

1999-07-29 Thread Matt Stegman

Lilo resides in the Master Boot Record (MBR) which is, I think, the first
sector of the drive, not included in any partitions.  Thus, you can wipe
all the partitions without touching the MBR.  To clear it, run 
`fdisk /mbr` from DOS.  This will clear LILO out of the MBR.

 -Matt

On Thu, 29 Jul 1999, Dennis Podein wrote:

> Hello , here is my dilemma . I have to clean out this computer
> , remove linux and set it up with windows 98 for my daughter
> for college . Anyway , I MS Fdisked it , then formatted it .
> When it goes to boot , LILO kicks in . I thought that would be
> gone . MS Fdisk doesn't find it , neither does disk druid , or
> linux Fdisk . How do I get LILO out of here ?
> 
> 



Re: [newbie] 2 X-related questions

1999-07-28 Thread Matt Stegman

When you put the CD in a windows machine, what do you see on the cd?  Do
you see one file (mandrake60-1.iso) or do you see several directories
(Mandrake, RedHat, images, etc.)?

If you see the former, well, now you've got a fine shiny coaster.  You
made a mistake when you burned the CD.

The .iso file is actually an entire filesystem.  It's an "image" of a CD.
What you should do is select whatever option says something similar to
"burn the CD with an ISO image" (it may not say "ISO").  This will put the
*filesystem* (as opposed to the *file*) on the CD.

If you knew this already, then you probably should go to the /images
directory and select the appropriate floppy boot image (I believe the
README provides a good synopsis).   

 -Matt

On Wed, 28 Jul 1999, Amit K Khandelwal wrote:

> i recently d/l the 600 MB ISO file to burn on a CD-- i burned it on a CD
> using adaptec softaer, but now i'm confsued as to what i am suppposed to
> do next. the CD is not bootable and i don't know how to access the file..
> can anyone help me with this problem???
> 
> thank..
> 
> [EMAIL PROTECTED]
> or
> [EMAIL PROTECTED]
> 



Re: [newbie] 2 X-related questions

1999-07-28 Thread Matt Stegman

Whoops, I lied.  The list of resolutions is individually defined for each
color depth, so this doesn't make sense: how would you set the default
color depth after you've decided which depth to use?

You need to put that line right ABOVE the first "Display" subsection.  Be
sure it's in the "Screen" section for the server you're using, too.  It
won't do much good if it's in the wrong server section.

If you are not sure which server you're using, go with "svga" first.
That's probably the more common, unless you have asouped-up video card.
Then you might try the "accel" server (Accelerated X server) if SVGA
doesn't work for you.

 -Matt

On Wed, 28 Jul 1999, Matt Stegman wrote:

> The way to increase your color depth in X is to add a line to
> /etc/X11/XF86Config:
> 
>   DefaultColorDepth 
> 
> This line goes in the "Display" subsection of the "Screen" section for
> your X server (I think- just put it by the list of resolutions, wherever 
> that is).
> 
>  is the color depth you want in Bits Per Pixel (8, 16, 24, or
> 32).
>  -Matt
> 
> 



Re: [newbie] 2 X-related questions

1999-07-28 Thread Matt Stegman

The way to increase your color depth in X is to add a line to
/etc/X11/XF86Config:

DefaultColorDepth 

This line goes in the "Display" subsection of the "Screen" section for
your X server (I think- just put it by the list of resolutions, wherever 
that is).

 is the color depth you want in Bits Per Pixel (8, 16, 24, or
32).
 -Matt




Re: [newbie] shuting down by ord. user..

1999-07-28 Thread Matt Stegman

> > /etc/shutdown.allow is a file that contains the names of users aloowed to
> > shutdown the system.  It may not be present on your system yet; you'll
> > have to create it.
> 
> Is it a list of user names (or UID's)?  Is there a man page on it?  I'm
> away from Linux right now... I don't feel like starting up another
> computer.

No man page on shoudown.allow, I'm afraid.  There is, however, a man page
on shutdown which briefly mentions /etc/shutdown.allow.
/etc/shutdown.allow is a list of usernames indicating which users may
execute the shutdown command.

> Will Ctrl+Alt+Del reboot the system if it's typed at *any* terminal?
Yes.  Well, I'm not sure if it will work while in X, but from any of
/dev/tty* it will work (as opposed to /dev/pts/*).

 -Matt




Re: [newbie] Problems with updatedb ?

1999-07-27 Thread Matt Stegman

Do you have a FAT or FAT32 partition mounted anywhere?  If so, I think you
can fix this by 
First: killing the offending process (`killall slocate`)
Second: Telling updatedb to ignore that partition in the future
(edit /etc/updatedb.conf and add the mount point of the FAT
partition to the EXCLUDE section)

I had this same problem and telling updatedb/slocate to exclude the FAT
partition fixed it.

 -Matt

On Tue, 27 Jul 1999, Dominique Deleris wrote:

> Hi list.
> 
> I have launched updatedb as root, and now I have a process (slocate)
> that has been running for 25 minutes, consuming 80-90 % cpu-time, and
> that does not seem to come to an end.
> 
> I've cheked my /var/lib/slocate directory, and it shows :
> 
> -rw-r--r--   1 root root0 Jul 27 21:51 slocate.db
> -rw-r-   1 root slocate 49410 Jul 27 21:52 slocate.db.tmp
> 
> What's happening ?
> 
> Thanks for your help.
> 
> Dominique
> 



Re: [newbie] shuting down by ord. user..

1999-07-27 Thread Matt Stegman

/etc/inittab allows commands to be differed by runlevel (i.e. do this in
runlevel 3, do that in runlevel 5) but not by user.  What you can do
instead, however, is add the -a option to the shutdown command in
/etc/inittab:

In inittab, change the line that says
ca::ctrlaltdel:/sbin/shutdown -t3 -r now
to say 
ca::ctrlaltdel:/sbin/shutdown -a -t3 -r now

> are these specific to certain versions of linux, or are these facilities
> present on all versions?

Well, I suppose if you have a really, really, really old version of init, 
it might not have had the Ctrl-Alt-Del catch implemented yet, but If
you're using Mandrake, Debian, RH, SLackware, whatever, it will work just 
the same. As to shutdown, well, every distribution had damn well better
include it- you can't safely shut down the machine without it!

 -Matt

P.S. Is "differed" really a word?

On Tue, 27 Jul 1999, Theo Brinkman wrote:

> Can you control what Ctrl-Alt-Del does by user?  (i.e.: Let root reboot
> the system that way, but have it just log everyone else out?)
> 
>   - Theo
> 
> Matt Stegman wrote:
> > 
> > /etc/shutdown.allow is a file that contains the names of users aloowed to
> > shutdown the system.  It may not be present on your system yet; you'll
> > have to create it.
> > 
> > Also check /etc/inittab to see if Ctrl-Alt-Del is being caught.  If so,
> > anybody (not just those in shutdown.allow) may shutdown the system via the
> > "three finger salute."  What you might do is change the Ctrl-Alt-Del
> > command to "logout" or something else that's harmless.
> > 
> >  -Matt
> > 
> > On Tue, 27 Jul 1999, Bobby Raagas wrote:
> > 
> > > I want to give a special acces to a user in my linux box to shutdown it,
> > > how will i do it? I'm using MDK 6.0 (Venus)
> > >
> > > Thanks
> > >
> 




Re: [newbie] shuting down by ord. user..

1999-07-27 Thread Matt Stegman

/etc/shutdown.allow is a file that contains the names of users aloowed to
shutdown the system.  It may not be present on your system yet; you'll
have to create it.

Also check /etc/inittab to see if Ctrl-Alt-Del is being caught.  If so,
anybody (not just those in shutdown.allow) may shutdown the system via the
"three finger salute."  What you might do is change the Ctrl-Alt-Del
command to "logout" or something else that's harmless.

 -Matt

On Tue, 27 Jul 1999, Bobby Raagas wrote:

> I want to give a special acces to a user in my linux box to shutdown it,
> how will i do it? I'm using MDK 6.0 (Venus)
> 
> Thanks
> 



Re: [newbie] Oh, yeah

1999-07-26 Thread Matt Stegman

Sounds like bzip2.  Except that bzip2 uses symlinks, instead.  I believe
that "bunzip2" and "bzcat" are both _symlinks_ to "bzip2."

Although, I imagine that if you're paranoid about deleting files, you
might use hard links as a "backup."  As was stated earlier, all regular
files are hard links to the file data.  When all hard links are gone
("rm"ed) then the data is inaccessable, and may be overwritten.  Thus, you
might have a backup folder full of hard links to all your important files.

If, for some reason one of these gets deleted, you can re-link it, because
the data is still there- hard linked from the backup folder.  This way,
you keep backups to guard against deletion (but not corruption!) of files
without actually keeping a second copy of the file in question.
 
 -Matt

On Mon, 26 Jul 1999, Richard Myers wrote:

> 
> On Sun, 25 Jul 1999, Andy Goth wrote:
> > Then what are hard links good for?
> 
> There are two applications which do something similar. Call them xyzzy and
> plugh. xyzzy gives you a help menu, but plugh is for expert users who
> don't need (and don't want) a menu.
> 
> I write a better application-- better than both of them, but it
> incorporates the features of both.  Everyone starts using my application
> instead of the two old ones.
> 
> And I include a flag that you can set-- either menu, or no menu. But my
> users get tired of always having to turn that darn menu on or off.
> 
> Soo, I come up with this great idea. I create two names (make that two
> *file* names) for the new application, xyzzy, and plugh. These are hard
> links to the same file. They both do the same thing-- launch my program.
> 
> Well, they do one thing different. They place different values into one
> special variable made available to my program, which keeps track of what
> filename was used to launch my program. And those two different values
> are: xyzzy, and plugh.
> 
> In my application I test whether it was called with the command xyzzy. If
> so, I make it act like xyzzy, menu and all.
> 
> If my application was called by the name plugh, I know this must be an
> expert user-- she's calling it by the name of the old program she used to
> use-- and so for her my application automatically provides no menu to take
> up screen space and get in the way.
> 
> Hard links make both my user groups happy.
> 
> 
> (Honesty in posting-- I didn't come up with any of these ideas, but I
> think they are nifty. ;-)
> 
> Trivia-- xyzzy and plugh are actually magic words from the 70's computer
> game "advent".
> 
> 
> best wishes,
> 
> richard myers
> 




Re: [newbie] How do I wipe the MBR?

1999-07-23 Thread Matt Stegman

Can't you boot off the NT CD?  If not, you should boot to DOS from a
floppy, and run the NT setup program to make the NT boot floppies.  Then
boot from those. 

 -Matt

On Fri, 23 Jul 1999, Ty Mixon wrote:

> I'm trying to re-install WinNT on hda, but every time I try to boot, 
> Lilo jumps in and asks what I want to boot.  I need to get rid of that 
> temporarily.  How do I do it?
> 
> I've tried Linuxconf from KDE, but it doesn't stop it.
> 
> Thanks,
> 
> Ty
> 
> 
> 



Re: [newbie] Installing

1999-07-23 Thread Matt Stegman

> The Ontrack program--DDO--is there to make sure that the whole drive can
> be accessed without cylinder wrap.  The proprietary format is there to
> make sure that the drive cannot be read without DDO.  Guess what.  Linux
> read it with no trouble (and with no DDO loaded).  

This means you use a floppy to boot Linux, right?

>   DOS, on the other
> hand...  I'll have to load DDO for Mr. Bill's sake.  I don't want to
> lose my data if the system is booted into DOS with a floppy that doesn't
> load the DDO, so I guess I'm best off leaving the proprietary format on
> there.  In that case, I could use the floppy but not the big hard disk
> (only the smaller one).  I can boot from floppy *and* use the DDO if it
> let it boot from disk, press space at the right time, and insert a
> floppy (thanks to a feature of the DDO).

Another thing you could do is make a copy of the MBR that loads DDO and
then put that MBR on a floppy.  It may not work, but then again, it may.
You could then put LILO in the MBR of hda (since LILO can read your drive
without the aid of DDO) and use the floppy to boot DOS.  I think it
depends on what you want to use as your primary operating system.

> I have the option of doing a BIOS format.  The DDO will still work, but
> then I can access the drive without it and risk cylinder wrap.  I read
> that 1024 cylinders is around 528MB or so.  Hmm.  That's pretty
> limiting.

Well, I don't know how your BIOS addresses the disk, but the way I
understand it is that by CHS (Cylinders Heads Sectors) the BIOS can handle
1024 Cylinders, 256 heads, and 64 sectors per track.  As each sector is
512 bytes, this allows up to (1024*64*256*512)=8589934592 bytes, or
8192MB, or 8.0GB.  LBA simply addresses each sector sequentially, and I
believe the upper limit there would depend on how many bits you use for
addressing the disk.  If today we use 32-bit processors, then you can
address (512*2^32)=219902322 bytes, or 2.0TB- I know the G is close to
the T, but this isn't a typo either.  T is an abbreviation for "Tera" or
10^12.  This is, of course, assuming my logic is correct.

I talked to somebody about this, and things are a little clearer now.  It
sounds like your old BIOS can't handle drives larger than 512MB. The
Ontrack program gets around this.  Don't reformat until you get newer
BIOS.  Until then, continue to use the Ontrack program.  What you might do
(if you're currently using floppies to boot Linux) is just let your
computer boot to DOS, and use loadlin to load a copy of the Linux kernel
off the 125MB DOS partition.  I don't know too much about passing options
to loadlin, but I think the program's on the Mandrake CD and I know
there's documentation on the web (try
http://metalab.unc.edu/LDP/HOWTO/mini/Loadlin+Win95.html).

> I get it, except for cfdisk.  I only heard of fdisk.  Is that a typo?  I
> know the C key is really close to the F key...

No typo.  cfdisk is a "Curses-based fdisk," hence the "c."  It's a little
prettier than fdisk, and I'd think easier for a newbie.  Curses refers to
the libraries used to generate the text-mode menus.
 -Matt





Re: [newbie] Downloading Files

1999-07-22 Thread Matt Stegman

This is a fault of Netscape.  It foten thinks that binary files are
actually text files, and so loads the binary file into the browser, giving
you plenty of gibberish.  It has something to do with file extensions and
MIME types on the server, and I'm not sure if you can configure Netscape
to override them (i.e.  specify *.rpm, *.gz, *.bz2, etc. as binaries).
Anyway, what you CAN do is hold down  while you click the link.
That will tell Netscape that no matter what, you want to save the target
of this link, not display it  I've been trying to download the KxICQ file in Netscape 4.6. Previously
> this has not been a problem. Just in the last day, instead of asking
> where it should save the file to, it simply downloads the file into the
> browser, giving me all sorts of gibberish. Has anyone run into this
> problem? Any suggestions for how I can correct it?
> 
> TIA
> 
> Andy Etzler
> 



Re: [newbie] Installing

1999-07-22 Thread Matt Stegman

> How do I install LILO on a different disk?  How do I make sure it
> doesn't trash the Ontrack program?

Well, the "boot=..." line tells LILO where to put itself- in the MBR of
the specified drive, or in the boot sector of the specified partition.  As
for trashing the Ontrack program, there's one idea I have to get around
it.  I'll explain in detail at the end.

> LBA?

Has to do with the low-level format of the drive.  I believe LBA is the
(de facto) standard way to get around the 1024 cylinder limit.  Then
again, I often don't know as much as I think I do, so somebody help me if
I'm wrong.  At any rate, I'm sure that this wouldn't be a problem with
only 3GB.  I think you can address either 8 or 8.4GB with 1024 cylinders.

> I am dealing with a 1993 computer.

Honestly, I'd recommend you upgrade it.  Go out, look for a "bare bones
system" with CPU, M/B, RAM.  I found really good prices (and systems) at
http://www.shoppingplanet.com (where I got my computer).  Then put your
old cards in the new computer, and you can upgrade the rest of the
hardware at your leisure.

> What's the difference between an extended partition and the other kind?
> The kind that would be called hdb2?

An extended partition can contain multiple partitions (logical drives)
inside it. For a more detailed explanation see
http://www.harris-lp.k12.ia.us/hlp/~jws/~jws/comp/PCInfo/Boot/DEFAULT.HTM
(click on the "Partitions and Volumes" link first).  I was looking for a
better page, but couldn't find one.  Still, this one's not bad.

> The disk is formatted with the Ontrack Proprietary Format...

Which means that if you ever want to access it without their special
driver, you'll have to reformat it- WITHOUT USING THEIR FORMAT.

As long as you don't have any data you want to keep, you shoul dbe able to
do this.

> A problem presented by the DDO is "the operating system must not require
> special code in the MBR."  LILO?

Yup.  Of course, it [LILO] doesn't _have_ to go in the MBR.

> Perhaps I can put LILO into the primary partition, put DOS and Windows
> into that partition, and put Linux on another partition.

Not a bad idea- in fact, this is what I was thinking you might try (if
you still want to use this OnTrack thing).  In lilo.conf, set the boot= to
a primary partiton.  It doesn't particularly matter which one (if you
have multiple primary partitions, which I don't think DOS likes too much).
Run cfdisk and make sure that partition, and NO OTHER ONE, is "Bootable"
(meaning active in DOS terms).  Run /sbin/lilo and reboot.  LILO should
come up, prompting for the image to boot.  As long as boot= is set to a
partition, not a drive (/dev/hda1, not /dev/hda) LILO will stay off the
MBR.

> table=/dev/hda  #Pass this partition table table to the other OS
> 
> What does that last line mean?

I'm not entirely sure myself.  I think that by passing the partition table
on to the other OS, it lets that OS know which drive it's being booted
from (I guess some operatig systems aren't smart enough to figure this out
on their own- *snicker*). 

 -Matt





Re: [newbie] Oh, yeah

1999-07-22 Thread Matt Stegman

> I heard that I should have two Linux partitions.  Can anyone give me
> more information on this?

Sure.  You should have one partition for files, and another for swapping.
Of course, you can distribute your file system between several partitions- 
I have /home on a separate partition so that when I format / to install
Mandrake 6.1, I won't lose all my personal files.

On a side note, would it be possible to selectively install 6.1 RPMs on my
6.0 system?  So that I don't have to re-format / anyway?

>> mind.  If only I could map "\Program Files" to be on a different
>> partition...
>
>Of course, such things are trivially easy in Linux...  Wonder why
>Microsoft chose such a half-assed method of drive management...

Probably for the same reason DOS can only use 640kB RAM.  I can
hear it now... "Nobody will ever have a drive larger than 2GB!"

 -Matt




Re: [newbie] Sound -- probably revisited.

1999-07-21 Thread Matt Stegman

Have you run 'sndconfig' yet?

 -Matt

On Wed, 21 Jul 1999, Beo d'Wulfie wrote:

> The soundcard I'm using is a sb16.
> It doesn't work. I've gone through the help docs and followed their
> instructions but still no go.
> 
> Can anyone help with this???
> 
> A few things to note:
> 
> - Yes, the speakers are turned on.
> - Yes, the system bell does come through the internal speaker.
> - No, I don't know how to decompress a module or recompile a kernal, but
> I'm a quick study and will find out how if this is what it takes to
> correct the problem.
> 
> Beo
> the newly frustrated.
> 



Re: [newbie] Installing

1999-07-21 Thread Matt Stegman

I don't know about that "special driver" you mention.  It sounds like a
problem with your motherboard (or IDE controller not recognizing 2 disks).
If that's so, then what I'm about to say won't help, and may even hurt
(you'd have to re-format the MBR and put that "special driver" on again).

/etc/lilo.conf is the config file for lilo.  When you run /sbin/lilo, it
will read this config file (or another if you use the -C option).

*/etc/lilo.conf (on my machine):

boot=/dev/hda   #tells where to put LILO (in the MBA of
#the first IDE drive)
map=/boot/map   #Specifies location of map file
install=/boot/boot.b#Which file to install in boot sector?
prompt  #Tells LILO to prompt before booting
timeout=50  #Wait 50 tenths of a second before going to
#default selection
image=/boot/vmlinuz-2.2.9-19mdk #Kernel image to load
label=linux #What you type at the lilo prompt
root=/dev/hda1  #Which file system is / (root)
read-only   #Mount the root fs as read-only (in case
#it has errors)


Instead of an "image=..." entry for Linux kernels, you need an
"other=..." entry for another OS.  You need to add an entry like this (at
the bottom of the file): 
other=/dev/hda2 #Which partition's boot sector do I load?
label=dos   #What you ype at the lilo prompt
table=/dev/hda  #Pass this partition table table to the other OS

You might try this, to see if it works.  Be sure to save all your boot
disks, at least the first time, just in case.  Be sure to modify this as
is appropriate for your system.

 -Matt

On Wed, 21 Jul 1999, Andy Goth wrote:

> My question: How can I get LILO to prompt me on which OS to use on
> booting without the use of floppies?




Re: [newbie] DHCP problems (fwd)

1999-07-20 Thread Matt Stegman

You might try to use the "-h" option on pump (to request a specific
hostname).  Make sure you have the latest version of pump (0.6.7-2mdk I
believe).  If so, you might want to try a different DHCP client (i.e.
dhcpcd).

I believe the pump commandline for the hostname thing  would be 
pump -i eth0 -h $HOSTNAME
I hear (by way of Axalon) that some DHCP servers require the client to
specify a hostname.  
"man dhcpcd" could tell you more, but it doesn't work on my system, and
the pump man page isn't very verbose on that subject (or any other).

dhcpcd is available for download from:
http://www.cps.msu.edu/~dunham/out/dhcpcd-1.3.17-pl2.tar.gz (for 
2.2 kernels)

 -Matt

On Mon, 19 Jul 1999, Theo Brinkman wrote:

> I'm having trouble getting hooked up to the network at work.  We're
> using DHCP, but my machine doesn't seem to be getting assigned an IP
> address.  
> --
> On boot I see:
> Bringing up eth0
> Delaying initialization of eth0   [FAILED]
> --
> When I run ifconfig, I see the following for eth0:
> 
> eth0  Link encap: Ethernet  HWaddr 00:00:86:35:95:EA
>   UP BROADCAST RUNNING  MTU:1500  Metric:1
>   RX packets:5 errors:0 dropped:0 overruns:0 frame:0
>   TX packets:15 errors:0 dropped:0 overruns:0 frame:0
>   collisions:0 txqueuelen:100
>   Interrupt:9 Base address:0x300
> --
> After wards, eth0 quietly vanishes, leaving just the 'lo' entry.
> 
>   - Theo
> 




Re: [newbie] RoadRunner Person

1999-07-20 Thread Matt Stegman

Kansas?  Where's that?  Oh, that flat piece of land in the middle of
nowhere... :)

I should start by saying that I got my Road Runner through Multimedia.  If
you're getting it through TCI (or someone else) you may have to use
rrlogin; I didn't.  If you're getting it through Multimedia, you shouldn't
have to.

First, verify that the service works in Windows.  If you already have the
Road Runner software installed and all your hardware works, great.  If
not, try this (WITHOUT installing RR software):
Under "Control Panel/Network," set the properties for "TCP/IP
Protocol."  Choose "Obtain an IP Address automatically."  The DNS and
Gateway tabs should be devoid of information- although "hostname" can be
whatever you like.  On the WINS tab, WINS _should_ be disabled, but
someone I know had to choose "DHCP for WINS resolution" before it would
work.  This is probably the safer choice; if WINS is unavailable, using
DHCP for it won't break a thing.  Leave the rest of the tabs alone.  Click
"OK", exit Network settings, and reboot.  Once you're up, try to ping...
oh, say 216.71.116.161.  If you can, you don't have to use rrlogin!  If
not, you should use rrlogin.

If you already use the Windows RR software to login, you might try to
disable it (assuming that it starts up with Windows), reboot, and see if
you can connect.

In the end, there wasn't much that I did to get hooked up with the cable
modem.  In 'netconf' you can turn on DHCP, set the hostname to whatever
you desire, delete gateway/DNS info, and exit.  If it works on exit,
congratulations!  If not...

What I did to get it working at this point was download dhcpcd from
http://www.cps.msu.edu/~dunham/out/dhcpcd-1.3.17-pl2.tar.gz and
tar -zxf dhcpcd-1.3.17-pl2.tar.gz
cd dhcpcd-1.3.17-pl2/
su -c "make; make install"
After that, pump worked for me.  Go figure.  The only thing I can think of
is that this installed an updated library or some other file that pump
also calls on.

If you need to use the rrlogin program, it is available here:
http://people.qualcomm.com/karn/rr/rrlogin.c
Compile it using 
gcc -o rrlogin rrlogin.c
Then, you might wish to copy the "rrlogin" executable to /sbin or
/usr/local/sbin or wherever you please.  Test it.  Does it work?  I sure
hope so, because if it doesn't, I have no idea how to help.  I didn't
write this program, and have never used it (like I said, I don't need to
with Multimedia).  To start the login program on boot, there's two things
you need to do: make a password file, and add the command to
/etc/rc.d/rc.local.  A good location for the password file is /root/, and
be sure to change the mode to 600 (so only user "root" can read and write
it).  Call it, say, /root/rr.password and it should contain two lines:
username
password
Now add the following line to rc.local:
/sbin/rrlogin < /root/rr.password

Please, let me know if this works.  Maybe then I actually WILL write up a
mini-HOWTO.

References:
http://metalab.unc.edu/LDP/HOWTO/mini/DHCP.html
http://people.qualcomm.com/karn/rr/
http://usmcug.usm.maine.edu/~kpesce/rr/

 -Matt Stegman
 <[EMAIL PROTECTED]>




Re: [newbie] DHCP and RoadRunner

1999-07-13 Thread Matt Stegman

I guess it's time to relate more info regarding my problem (which I should
have done initially- sorry).

> Maybe the Internet Police are going to come after me.

Isn't that what IP stands for...?
Each ISP independently determines how it will hand out IP addresses.  This
particular ISP just doesn't want to give out static IPs; they'd rather use
DHCP.  To quote from Road Runner's home page:

"Will having a Road Runner cable modem give the customer their own IP
address?
Not permanently. Customers will be assigned an IP address dynamically when
they open a Road Runner session by switching on their computer and clicking
on the Road Runner icon."

> Determining IP information for  ethX... failed <- that your error?

Yes, that's it exactly.  It takes a couple minutes to timeout, and watching
das blinkenlights on the hub, I notice that they only blink once every
15-30 seconds during this time.

> You will need to locate the linux
> rrlogin, hopefully someone has already customised it for your area (I
> haven't fiddled with it for a few months maybe it's not a compiletime
> option only anymore) They tend to get upset and cut you off when you
> refuse to run the rr program (even on windows). You should be able to
> obtain a dhcp lease in windows use winipcfg to findout ip gw dns and all
> that fun stuff, now reset the machine so it doesn't release its lease
> (windows does release this if you do shutdown doesn't it?) now setup your
> linux cfg with the ip info you retrieved and it should function (winipcfg
> should be able to tell you when the lease expires..)

I haven't been cut off in Windows (well, not yet anyway).  This leads me to
believe that the Road Runner login program isn't necessary.  If it's not for
Windows, why should it be for Linux?  On the other hand, maybe I'm just not
patient enough- they'll kick me off later.
In any case, I am still initially allowed to connect without using the
program in Windows.  Why would I be treated differently for using Linux?

As to the DHCP lease- it's only for a few hours.  Even so, I guess it's
worth a try to see if I can get it working with static IP, even if only for
a few hours.  I'll try this when I get home this afternoon.  Thank you, all,
for your suggestions.

-Matt Stegman
<[EMAIL PROTECTED]>



[newbie] DHCP and RoadRunner

1999-07-12 Thread Matt Stegman

Our community recently got cable modem access, via RoadRunner services.  My
family signed up right away, and we now have cable modem access to the
Internet (everybody applaud)!  However, I now have a problem.

My system is dual-boot, and I've set the Windows side up no problem.  All I
had to do was tell it to use DHCP ("Obtain an IP address automatically"),
and set the IRQ and IO port settings for my ISA NE2000 network card, and all
was well.  Linux, on the other hand...

Up 'till now, I've used Linux on a network with no problem.  This was with a
static IP, user-specified gateway, DNS, etc.; now that I think of it, I was
using Festen, though.  When I installed Venus, I was (temporarily) off the
network.  Now I'm back, and with problems.  My IRQ and IO settings are good
(in linuxconf), and I tell it to use DHCP, and leave all the other info
blank (no DNS, gateway, etc).  Then, after loading the DHCP module (I think,
is this what's happening?) , Linux comes back after timing out and says it
couldn't get an address from the DHCP server.

Given that Windows doesn't need to use the RoadRunner setup, Linux shouldn't
have to, either (right?).  Is there something I'm missing, i.e. that Windows
takes care of behind the scenes, but Linux doesn't?  I'd rather not have to
rely on Windows for internet access...

Thanks,
-Matt Stegman
<[EMAIL PROTECTED]>



Re: [newbie] starting xwin

1999-07-02 Thread Matt Stegman

Oops, sorry- I guess I assumed you'd want to boot to it.  There is another
way; you can run 'kdm' from the console.  It has the exact same effect, and
has to be run as root.  Also, GNOME has it's own desktop manager thingy...
called 'gdm'.   Just in case you're interested.

-Matt Stegman
<[EMAIL PROTECTED]>

- Original Message -----
From: hevnsnt <[EMAIL PROTECTED]>
To: Matt Stegman <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Friday, July 02, 1999 12:19 PM
Subject: Re: [newbie] starting xwin


> Hey btw, GO K-STATE! heh..
> Yeah, i knew all that, but I dont want it to boot directly to that.. is
> there any other way?
> -hevnsnt




Re: [newbie] X screen resolution

1999-07-02 Thread Matt Stegman

From: jsm <[EMAIL PROTECTED]>
Subject: [newbie] X screen resolution


> In Xwindows with Redhat you can change screen resolution with Ctrl+Alt
> + or - . It does not work with my Mandrake 6.0 & KDE . How can I change
> resolution ?
The reason it doesn't work with Mandrake is because the default for X is to
support only the resolution you picked during the installation.  You can
change this a couple different ways:
run xconf (it's a pretty self-explanatory program, but you'll have to
set up everything again, video card, mouse, monitor frequencies, etc) or,
edit /etc/X11/XF86Config by hand- which isn't very hard.
If you choose the second option, "pico" is a good editor to use.  Once
you've loaded /etc/X11/XF86Config, scroll down until you see  Section
"Screen."  Now look for the particular server you're using (probably svga).
In here, you need to edit the "Modes" line.  It should look something like:
Modes "640x480" "800x600" "1024x768"
The first resolution will be the one that the X server starts with.  You
should be able to get to the other ones using <+> or <-> (make
sure you use the keys on the Numpad!).


> Also , does Linux support USB ?

AFAIK, only experimentally.

> Thanks for the help
>
> Jeff



Re: [newbie] starting xwin

1999-07-02 Thread Matt Stegman

I think you're talking about kdm- in which case you might try 'init 5' (as
root, I believe).  This should start X on tty7, and give you a login prompt
with the drop-down box of window managers/desktops.  If you like it, you can
set up /etc/inittab to make this the default- change
id:3:initdefault:
to
id:5:initdefault:
and you're done!  The next time you boot, you'll go straight into runlevel
5, which will start X and kdm automatically.

Also, instead of using an alias for startx --bpp16, you might add
DefaultColorDepth 16
to /etc/XF86Config, under the "Display" section for your server.  That way,
it'll take hold for all users.

-Matt Stegman
<[EMAIL PROTECTED]>

- Original Message -
From: hevnsnt <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, July 02, 1999 12:34 AM
Subject: [newbie] starting xwin


> Ok guys/gals, I have another question..
>
> I would like to set up an alias to start xwindows in 16bit but I want it
> to go to the windowmanager choice screen.. I know how to setup an alias,
> know how to start x in 16bit, but I have no Idea on how to start it to go
> to that window choice screen without having it boot directly to x...
> Anyone have any Ideas?
>
> -Bill
>
>



Re: [newbie] wine

1999-06-30 Thread Matt Stegman

> ok, I want to run MS Office with Wine emulator, but I have to edit the
> configuration file of wine to do so, but I dont know how to do it! It ask
for
> the Unix path of my floppy, my C drive, my D drive (both from windows) and
my
> temp drive... anyone knows what could be my path giving that  linux is
resident
> in my D drive of windows?

Not sure what that last bit means- I didn't think there was a working
driver for reading ext2fs drives from windows, or that you could install
Linux on a FAT partition (apart from a UMSDOS install).
I can tell you that by the default install of Mandrake, the floppy path
is /mnt/floppy and CD-ROM is /mnt/cdrom.  Other than that, it depends on
where you have your Windows partition(s) mounted.  If you have only one
drive in Windows, you can comment out the D: drive part in wine.conf.  Then,
you'll need to mount your C: drive.  Wherever you mount it, that's what
you'll put in WINE's config file.  I can't tell you too much more, as I've
only used WINE once or twice before, and am not sitting at my computer right
now.  I'm sure, though, that if you poke around the Internet long enough,
you'll be able to find anything you might need to know about WINE.  Some
experimentation can work wonders, too, if you're not sure about something.
And the great part about Linux is, unless you're working as root (which you
should not be) you can't hose your entire system if something goes really
wrong(like what happens in Windows)- just that one user, at the worst!
Have fun, and good luck.
-Matt Stegman
<[EMAIL PROTECTED]>



Re: [newbie] 3COM Fast Etherlink

1999-06-28 Thread Matt Stegman

> Best bet is to re-install and set up your card during the process.

Or you could just run 'netcfg'

-Matt Stegman
<[EMAIL PROTECTED]>

- Original Message -
From: Daryl Walsh <[EMAIL PROTECTED]>
Subject: Re: [newbie] 3COM Fast Etherlink


> hmmm, I have that card (3COM 3c900b) on my dual-boot NT/Mandrake (Venus)
> box.  It's an excellent card.
> The Venus installer allows you to set up this card during the installation
> process--I believe it is the last 3COM card listed and it's nicknamed
> "Boomerang."  It's listed under 3c900, in any case.




Re: [newbie] Can't mount CDROM

1999-06-16 Thread Matt Stegman

Do you mean that when you click the CD-ROM and floppy icons in KDE, you get
an error?  If so, you might try editing the files that generate the icons,
i.e. the .kdelnk files in KDE's home directory (probably
/usr/share/applnk/Desktop or something similar, but I can't verify as I'm
not sitting at my home computer).

The virtual desktop can be fixed in /etc/X11/XF86Config by finding the
"Display" subsection of the "Screen" section for the server you're using,
and commenting out the "Virtual" line, or by running "xconf" or "XF86Setup"
and changing your settings through those programs.

-Matt Stegman
<[EMAIL PROTECTED]>

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, June 15, 1999 8:11 PM
Subject: [newbie] Can't mount CDROM


> When I try to mount the CDROM or floppy drive, it says "Must specify a
file
> system". Can anyone tell me how to do this? Also, I have a virtual desktop
> displayed, how can I change the settings for that? Thanks!
>
> Matt
>



Re: [newbie] Partitioning

1999-06-14 Thread Matt Stegman

Are you sure LILO is installed on that partition?  The default, when you
install Mandrake is to put LILO in the master boot record of hda, but if
you're using another boot loader, then that will wipe LILO off your MBR when
it's installed.  I believe that when you use another boot loader in your
MBR, LILO needs to be installed on the partition that that boot loader
points to.  Did you make a boot disk during the install?  If so, boot off
that, and check /etc/lilo.conf to see if the first line reads
boot=/dev/hda2
where /dev/hda2 is the partition that linux is installed on (not necessarily
hda2 on your machine).  If it says just "hda", you need to change it to the
partition that Linux is installed on (i.e. hda2, hda3, etc.).  After that,
run "/sbin/lilo" and reboot.  Assuming that you put in the correct
partition, it should work.  I hope I made the instructions clear.  If not,
rebuke me and make me try again.
Good luck.
-Matt Stegman
<[EMAIL PROTECTED]>

- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, June 14, 1999 10:56 AM
Subject: [newbie] Partitioning


> I have Mandrake 6.0 that came with Partition Magic, and after using it,
> everything installs fine, but when it's time to dual boot (it came with
Boot
> Magic too) , it seems like it doesn't see Linux on the new partition. It
just
> hangs when tryign to boot Linux. Anyone suggetions would be greatly
> appreciated!
>
> Matt Rose
>



[newbie] Re: XWindows Interface Other than KDE

1999-06-11 Thread Matt Stegman

> I have the power pack edition and having used the KDE for a while I would
> like to just use Xwindows instead as I would like to compare that with the > KDE. Is 
>it possible to disable the
> kde and use just a more conventional xwindows interface?

It sure is.  Try running "desktopcfg" fromthe command-line.  Choose your window 
manager, and restart X.

-Matt Stegman
<[EMAIL PROTECTED]>



  1   2   >