Pages

Search

A Problem using yum on Linux Fedora Core 4 - Forums Linux

A Problem using yum on Linux Fedora Core 4 - Forums Linux


A Problem using yum on Linux Fedora Core 4

Posted: 13 Oct 2005 08:19 AM PDT

Enrique, I located and installed the proper python rpm's, and now yum
works like a charm. Thanks very much for your help!

Best,
WMD

--
Wayne Delia, net
Delta Iota Chapter Advisor, Phi Kappa Sigma at Marist College
"I'm beginning to sober up, and you're scaring me!" (Tom Servo, MST3K)

Why is Firefox printing in Linux so slow?

Posted: 13 Oct 2005 07:50 AM PDT

On Thu, 13 Oct 2005 07:50:45 -0700, usenet.1.massysett wrote:
 

I would first check that ALL your printer, CUPS and Firefox config files
are correct. I would access CUPS through its native HTML interface
(http://localhost:631) instead of KDE's printer tools, just to be sure.
Also, check that Firefox's Postscript output isn't being sent to
Ghostscript or some printer filters. The output should go directly to the
printer, unchanged.

As a last resort, uninstall your printer and, then, reinstall it using
CUPS' interface. Maybe, KDE's got a bug.

Stefan

FC4, Serial ATA, and RAID

Posted: 12 Oct 2005 09:48 PM PDT

On Thu, 13 Oct 2005 06:48:41 +0200, Screech <three> wrote: 

The FC4 iso had kernel-2.6.11-1.1369_FC4.i?86.rpm
The current kernel is 2.6.13-1.1526_FC4
 

If the numbers are right, they translate to

status=0x51 DriveReady SeekComplete Error
error=0x40 Uncorrectable ECC error

I believe the latter means data corruption on the disk or at a low level.
 

Sounds a bit disturbing, I should perhaps check a diff between the
2.6.11 and 2.6.13 kernels to see if this is justified (but I don't
have the time this week). It may well be as already the atapi
interface implements a packet interface modeled on the scsi protocol,
and I would guess that serial ata takes this evolution even further.
 

I don't really know anythig about what is failing, but I wonder why you
are placing the two disks on the same ide controller, and leaving the
other controller to run just the DVD player?

If no-one else know anything better to try, you could try to regroup
the disks, placing, say, the hdd on the other controller, making it hdb.
If there is something flakey in the ide controller hardware, perhaps
duplicating all disk write requests on the same controller is begging
to trigger the failure potentials.

(But I should also say that I don't know hwo much physical reality there
is behind the interface offered by the chipset that perhaps makes two
"virtual controllers" out of one, or even out of four.)

-Enrique

Can't install Debian 3.1 on HP Pavilion 7955 computer

Posted: 12 Oct 2005 10:47 AM PDT

First of all, thanks to those who have provided assistance. Others have
been less than helpful.

The reason I think Debian should have a tutorial on this is because
most PC users use Windows, and Windows XP is currently the newest
Windows version, so it could only help in getting more people to start
using Linux instead. Of course, the tutorial is not necessary, but I
didn't even find a notice explaining how burning a bootable CD is
different from burning the type of CD that most users commonly burn.

<i>Oh, come on, the issues here are so obvious only after you acquire
a number of concepts.</i>

That might be true, but only assuming that there is a guide
<i>explaining</i> these concepts.

<i>"Cease messing up".</i>

That's very helpful...

Bad Motherboard? Upgrade?

Posted: 12 Oct 2005 05:06 AM PDT

> What is that monitor program?

http://mbm.livewiredev.com/

The program in question is currently listed at the top of the page,
select a mirror to the right of it to download.

Cheers

bypassing hwclock?

Posted: 11 Oct 2005 10:44 PM PDT

On 2005-10-13, John Hasler <gt.org> wrote:
 
 

I suppose, unless you have one of those Dallas chips with the clock
incorporated into the battery pack.

--

John (dhs.org)

Rar files

Posted: 11 Oct 2005 11:33 AM PDT

Manta a écrit : 

Hello,

- rar for linux from http://www.rarlab.com/download.htm to compile or
from http://rpm.pbone.net <= not last version (command-line but
necessary for the GUI)

- file-roller as GUI for extract (never succeed compress-split with GUI,
use command-line), file-roller is the gnome equivalent for ark for kde.

- unrar from http://rpm.pbone.net

Bye.
Rv
--
.~. - http://www.web-space.tv/faq_abcf
/V\
/( )\ - Une recette de bière « open source » !:
^^-^^ http://linuxfr.org/2005/07/21/19331.html

Grub Network Bootdisk - where from ?

Posted: 11 Oct 2005 03:26 AM PDT

On Wed, 12 Oct 2005 09:20:21 +0200, hermann <net> wrote:
 

In order not to leave that part of the question unanswered, Grub must
be on an rpm package, and I would guess that package is called
grub-0.95...something.rpm. It should be on your installation CD.
Almost certainly it is already installed.

Next, how to make a grub floppy. Doing the command "info grub", pressing
TAB tree times (places the cursor on "Installation"), then pressgin enter,
pressing TAB three more times (Creating a GRUB boot floppy) and Enter:


# cd /usr/share/grub/i386-pc
# dd if=stage1 of=/dev/fd0 bs=512 count=1
1+0 records in
1+0 records out
# dd if=stage2 of=/dev/fd0 bs=512 seek=1
153+1 records in
153+1 records out
#

This is still not the "network boot" thing you want, I just quote this
to get those questions out of the way.
 

Sounds a bit strange to me, is this so that you could have something
different happen when you don't insert the floppy? Or, are we dealing
with a computer where you are somehow prevented from installing in the
disk?

I have never tried the network support facilities in Grub, but I have seen
something about that in the info command output.

Yes, I just checked, and it says:

Although GRUB is a disk-based boot loader, it does provide network
support. To use the network support, you need to enable at least one
network driver in the GRUB build process.

That brings up the question if Suse has enabled any network drivers in their
grub rpm. If not, you will have to compile your own rpm.

The procedure is this: Download the grub *source* rpm from your distribution
(perhaps you have a distro CD with all the source rpms).

Install it the normal way, rpm -i grub-whatever.src.rpm. Since I use
Fedora, which is a child of Redhat, there is a "redhat" in the following
paths, yours are probably a little different.

Edit the file /usr/src/redhat/SPECS/grub.spec.
Search for the line

%configure --sbindir=/sbin --disable-auto-linux-mem-opt

(Well, that is the redhat/fedora version of that line, yours may
be different.) Now add one option to that line, e.g.

%configure --sbindir=/sbin --disable-auto-linux-mem-opt --enable-3c509

(how to see the possible options - below) Save the file and build
the rpm like this:

rpmbuild -bb /...../grub.spec

You will find the result in /usr/src/.../RPMS/i386/grub....rpm or something
similar.

This is how simple it could be in an almost perfect world. In reality you
will have to install a couple of other rpms to satisfy the "build-require"
statements in the grub.spec file.

Issue the rpmbuild command above, and you will be told. Or look inside the
spec file.

In order to find the possible options,

rpmbuild -bp /..../grub.spec

This will unpack the tarball(s) and apply all the patches your distro
applies. Then find /..../BUILD/grub-0.95/netboot/README.netboot.
Notice that rpmbuild usually starts every build by deleting the build
source tree, and repeating the steps from there. I mention this in case
you find you have to modify something, you must then create a patch
file and add it to the spec file to have it reapplied when the source
is prepared again.
 

I don't know SuSE, there are others here who do. I would guess that
the distribution has a ready-made solution for diskless workstations
or for bootp use, but I don't know much about it.

-Enrique

dual boot problem...

Posted: 10 Oct 2005 12:47 AM PDT

com wrote:
 

Good, I'm glad you fixed the problem and your welcome.

 

Start a new thread and let us know about the modem many are supported.

Vist; http://linmodems.org/ and download/use the scanModem tool.


--
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety." -- B. Franklin, 1759

Wireless networking/Mepis/OS question

Posted: 09 Oct 2005 08:05 PM PDT

Bill Marcum wrote:
 

Thanks. I'll keep that in mind if I ever run into it again. (See below)

 

Okay. I did a google search yesterday for "0000:00:07.1" which gave me
the hint to use lspci like you said. And it turns out to be the IDE
interface on my machine as well. And the hd works, BUT it makes me
wonder since I just had to replace the hd because the last one failed.
Hmmh...

 

To tell you the truth, I didn't really like Mepis anyway -- can't tell
you why other than I have some familiarity with Fedora and so I know
where the config files are and just generally how to do stuff with it.
So I loaded FC4 on it.

For whatever reason I don't get the "video mode" problem with grub. I
don't recall if Mepis was using lilo...

Still struggling with the wireless, though. Dl'd the ndiswrapper pkg
plus a modified kernel (16K vs. 4K stacks) with another machine and
loaded them up via sneakernet. Followed the instructions in the
ndiswrapper wiki and have gotten as far as being able to scan and find
the AP correctly. Wireless tools is the wrong release (17 vs 18) so I
hope I can manually get this thing working well enough to do a system
update. The odyssey continues...

Anyway, thanks for the reply,

Rod

Kernel 2.6 mouse dead

Posted: 09 Oct 2005 05:22 PM PDT



J.O. Aho <net> says... 

From http://www.slackware.com/ page is

"Slackware 10.2 includes the Linux 2.4.31 kernel, with Linux
2.6.13 available in the /testing directory. For the first time,
a 2.6 kernel with support for SCSI, RAID, and SATA is offered
as a boot option in the installer (called "test26.s")."

test26.s is on Slackware 10.2 distribution disk. Slackware
10.2 users are told to select test26.s to have Slackware 10.2
with 2.6.13 Kernel. Is not test26.s on Slackware 10.2
distribution disk same mouse enablings as bare.i on same disk?
I installed exactly same including format two times except
kernel different. NO other changes!

Please to look at
http://www.tux.org/pub/distributions/slackware/slackware-10.2/kernels/test26.s/config
has
CONFIG_INPUT_MOUSEDEV=y
CONFIG_INPUT_MOUSEDEV_PSAUX=y
CONFIG_INPUT_MOUSE=y
CONFIG_MOUSE_PS2=m

All same except CONFIG_MOUSE_PS2=m is not CONFIG_MOUSE_PS2=y
but also has
CONFIG_MOUSE_SERIAL=m
CONFIG_MOUSE_INPORT=m
CONFIG_MOUSE_ATIXL=y
CONFIG_MOUSE_LOGIBM=m
CONFIG_MOUSE_PC110PAD=m
which I think are menu choices. I menu choose PS/2 and also
menu choose Intellimouse once for testing. Both menu choices
both mice work with bare.i not with test26.s My mice port is
normal PS/2, not USB.

Other Slackware 10.2 users using test26.s with no mouse problem
so I think mouse enablings are on.

Sorry if my English is bad.

test26.s mouse problem

Posted: 09 Oct 2005 03:47 PM PDT

On 10/10/05 14:07, reclusive monkey wrote: 

PS/2
--
A computer is like an air conditioner,
it stops working when you open Windows.
Registered Linux user #337974 <http://counter.li.org/>

no option to boot FC4 after install

Posted: 09 Oct 2005 10:43 AM PDT

On Sun, 9 Oct 2005 18:43:58 +0100 in comp.os.linux.setup, Richard O
Hora favored us with... 

Did you install GRUB or another dual-boot manager? If so, check your
disk to find where it was installed, and set that partition active.
(You can do this from an Administrator account in Wondows XP by
running diskmgmt.msc.)

I had this problem with FC3 -- the install went fine but it didn't
set the active partition to the one where the boot loader is. I did
that manually, then rebooted and the GRUB loader came up.

--
Stan Brown, Oak Road Systems, Tompkins County, New York, USA
http://OakRoadSystems.com/
"You find yourself amusing, Blackadder."
"I try not to fly in the face of public opinion."

How do I configure drivers in Debian

Posted: 09 Oct 2005 05:30 AM PDT

From experience (and I've used various versions of Unix since ca. 1978
and Windows versions since 2.x in about 1982--as a result, personally,
I'd rather use Linux):

I just tried to install Windows XP (of an unknown vintage) and Linux
(RedHat version 9, from the same source, i.e. these were the versions
used at my workplace) on an IBM a30p laptop. Not the latest model,
but not ancient either. The laptop came to me running w2k. I did not
upgrade the w2k to xp, but instead installed xp from scratch.

The Windows XP version installed "correctly" first time and it wasn't
orginally configured for that version. The only difficulty was the
laptop defaulted to a 640x800 resolution until I went to the IBM
website and downloaded a new driver.

The Linux version installed and had the same 640x800 resolution
problem. Of course, going to the IBM website didn't resulting in
finding a new Linux driver. With some help, I did find the
appropriate XFree86 config file and inserted the "1600x1200" line in
the correct place and that was good. Next, I had to resolve a mouse
cproblem, which required downloading a whole new version of Linux,
CentOS 4.1 was recommended to me. That took a considerable amount of
time and effort. Of course, the new version of Linux broke some of
the shell scripts I use to invoke emacs, so that the shell windows
inside of emacs didn't show prompts.

I don't yet have the built in wireless talking to Linux. It worked
with XP, the first time I booted it. I haven't started trying to
tackle that problem because I can plug the laptop into the wired net
at home, so the wireless was a lower priority item. Moreover, even
when I have the wireless working under Linux, I won't be able to use
it that way at work, because the wireless VPN software I need to get
into the corporate internet is a Windows only program.

So, don't use me as your expert if you want the Linux installation to
be graded as easier. This wasn't the first that I've done. I have
one desktop that has run Redhat 5.0, 5.1, 6.something, 7.2, and 7.3.
I will upgrade it to CentOS 4.1 also, after I have my laptop running
as I like.

However, if you want something that works out-of-the-box with minimal
intervention, my experience is that one has better luck installing
Windows. I think there is a simple reason for that, every hardware
vendor ships drivers that work with Windows. Thus, if your brand x
hardware device doesn't work, one simply gets the brand x Windows
driver for it.

You're only in trouble if the vendor stops selling that hardware
device and a new version of Windows comes out and the old driver isn't
compatible with the new version. Then, you are truly out of luck, you
either have to stay with the old version of Windows forever or buy a
new hardware device. That's actually the way they want it too, it
simply guarantees obsolescence. That's the price one actually pays
for using Windows.

However, I'm not sure the Linux story is much better. I can't count
the number of times I've had some part of my Unix-Linux/Emacs
installation switch to something new and incompatible, sh to csh to
ksh to bash, twm to ctwm to vtwm to fvwm to kde to gnome, lilo to
grub, sccs to rcs to cvs to subversion, gnu emacs to xemacs, rnews to
gnus, rmail to vmail. I know some incarnations of the scripts I used
to run on Ultrix and SunOS still are runnning on my system, but only
because I've spent hours over the years dealing with the
incompatibilities and foibles of the versions they've been ported
over. Hmm, the latest version of Linux has changed something about
the way it writes ext2 file systems and my re-partitioning (Partition
Magic v 7) software will no longer change partitions on the
disk--hopefully I can get qtparted to work on them. Moreover, a lot
of the hardware I scrapped over the years, *never* worked with any
Linux version I had. Maybe if it had, it would still work with Linux
today, maybe....

If I could have paid L100, i.e. $200 US and have had them install a
complete working Linux distribution on my laptop where the mouse,
screen, and wireless all worked (and so did the scripts I've used for
years so that I wouldn't have to "fix" anything), you bet I would have
done it.

-Chris

Fedora setup - modem not detected

Posted: 08 Oct 2005 10:53 PM PDT

example.tld (Moe Trin) wrote in
news:phx.az.us:
 

.....snip...

 

Once again many thanks for your help

Alan (another 'old guy' )

After slack upgrade fonts too small when print webpage

Posted: 08 Oct 2005 02:10 PM PDT

09 Oct 2005 03:29 UTC, tekla typed: 

One to subscribe to for Slack questions - alt.os.linux.slackware

Although you'll find Slackware people everywhere :-)

Debian bootable on external USB-harddisk

Posted: 08 Oct 2005 07:39 AM PDT

Andreas Rittershofer schrieb:
 
hi,
you can try to boot it with help of a RUNT-Linux Boot-Floppy
whith it, you can boot from any USB-Device (USB-Stick, ...)

RAID chipset SiI680

Posted: 08 Oct 2005 03:28 AM PDT

Giampiero Gabbiani <it> did eloquently scribble: 

Don't know...
There are some disk throughput tests, like bonnie++ to test access, read,
write and copy speeds but you'll just be testing the kernel's RAID code
rather than the controller (which is being used by the kernel as a simple
IDE). Only windows supports the raid part of the card fully, because the
manufacturers supplied the driver.

Testing windows running on the raid card vs linux using kernel RAID wouldn't
be a valid comparison because the test would be mainly windows v linux,
rather than card vs kernel in that situation...
Suppose a way to handle the comparison would be to install windows on the
RAID, run the tests, install it on the card in standard none-raid mode and
do the tests again, then do the same with linux using standard IDE and
kernel RAID... see the comparisons side by side...
--
__________________________________________________ ____________________________
| co.uk | "Are you pondering what I'm pondering Pinky?" |
|Andrew Halliwell BSc(hons)| |
| in | "I think so brain, but this time, you control |
| Computer Science | the Encounter suit, and I'll do the voice..." |
------------------------------------------------------------------------------

Problem with Redhat 9 & XP Pro

Posted: 07 Oct 2005 02:53 PM PDT

> I'm having trouble with dual-booting RH 9 & Microsoft XP Pro. I 

First, I would avoid reinstalling XP as long as I could, unless you
have a "ghost" copy of your previous installation, or it was simply a
vanilla instalation you haven't customized. If you reinstall XP, you
may lose your customizations. Attempting to "repair" it, is less
likely to have that problem. However, I would still wait on doing
that until I exhausted other options.

As to your specific problem, it sounds like you may have changed the
partitions as you installed RH. That can cause XP problems,
specifically if you insert or delete partitions before the XP
partition. For example, it may not have been creating the RH
partition that gave you problems, but deleting the DOS partition that
is the cause of your woes. Did XP used to boot through the DOS (FAT I
presume) partition?

Next, I see you are keeping XP in an extended/logical partition? Did
you have it there before you installed RH? Or, did you move it from a
primary partition to an extended partition when you installed RH? If
you changed it from primary to extended, then I suspect this is the
cause of your problem. As far as I know, XP (or any windows variant
for that matter) likes to have its boot code in a primary partition.
You can get XP (and the other NT, OS/2 variants) to boot in an
extended partition if you also have a FAT primary partition to keep
their boot info in, but you have to set them up that way in the
beginning.

What I suspect is happening, is that when you tell XP to boot, it
looks for an appropriate primary partition, finds HD1 which is
currently empty, but is formatted FAT ans says ah here is my primary
partition (C: drive) to find my boot info in. It then, looks for the
XP installation there and doesn't find it, because the info is in HD5,
and complains.

If I were in your shoes, and the XP partition used to be a primary
partition, then I would convert the XP back to a primary partition and
put HD4 (or HD3 & 4) in the extended partition. You will find Linux
much more cooperative about booting while the root directory is not a
primary partition.

Consider my setup:
HD1 - a FAT32 w2k partition
HD2 - a FAT32 w2k partition with also boot files for booting XP out of an extended partition
HD3 - an NTFS xp partition
HD4 - an extended partition with the following logicals inside
HD5 - /boot (ext2) where I keep grub
HD6 - / (ext2)
HD7 - /var (ext2)
HD8 - /tmp (ext2)
HD9 - linux swap
HD10 - /usr (ext2)
HD11 - /home (ext2)
HD12 - a FAT32 partition for sharing data between the various OS's
HD13 - another FAT32 partition for sharing data between the various OS's
HD14 - an NTFS xp partition which boots out of HD2
HD15 - the "system restore" partition

See, how I have most of the Windows OS's in primary partitions down at
the bottom--the only exception being HD14, and that boots out of HD2.
Believe me that makes them much happier about being booted into.

The other major difference between my setup and yours is that I use
"System Commander" as my boot menu (and that's what in the MBR).
However, I think either lilo or grub in the MBR will work too.
Especially, because if I boot the Linux partition (from System
Commander) and then tell it (grub) to boot one of the Windows
partitiions, that still works.

Hope this helps,
-Chris

SSH tunneling of the Oracle Installer

Posted: 07 Oct 2005 09:59 AM PDT

Figured out my issue.

1. I only needed the xdm and xauth rpms installed to tunnel the X
traffic via ssh
2. I modified /etc/hosts to include the line

127.0.0.1 localhost.domain localhost

sshd_config was setup properly (I.E. ForwardX11 yes) and once hosts was
modified every thing worked like a magic cookie! I hate simple
problems, as they usually take the most time to resolve. Thanks
everyone for your input!

FC4, md raid-1 not started at boot.

Posted: 07 Oct 2005 08:22 AM PDT

In comp.os.linux.setup Peter T. Breuer <it.uc3m.es>: 
 
 
 

Have no problems with it on a few systems, the only problem was
the initial setup in kickstart. Strange enough, once tried to
setup FC3 in full softraid, but alas the installer would only
want to setup an array on the same hd, which doesn't make sense.

This works with RHEL 3 just fine, perhaps their is a reason why
this feature is missing from Fedora?

Other distro couldn't boot from softraid at all after successfully
mirroring a running system. So this could explain your paranoia
about it? It seems to depend on the distro.
 
 

Lilo, at least the version from RHEL does write stuff on both
physical disks in software raid1, so you can boot from both, if
your mobo is capable enough. Think I posted an example recently
in this "theater".

--
Michael Heiming (X-PGP-Sig > GPG-Key ID: EDD27B94)
mail: echo qr | perl -pe 'y/a-z/n-za-m/'
#bofh excuse 342: HTTPD Error 4004 : very old Intel cpu -
insufficient processing power

problem reading loop mounted .iso file on disk, get "Input/output error"

Posted: 06 Oct 2005 10:04 AM PDT

I would just like to thank all the people who helpd me get my laptop
upgraded. It is now running CentOS 4.1 and that fixes my "mouse"
problem, which is how this all started. (Yes, I do have a few minor
glitches due to things changing (and some of those I have already
fixed), but all-in-all the process has come to a successful
conclusion.) I can now go back to being mostly a Linux user and not a
system installer. That makes me happy.

Again, thanks to all.
-Chris

BTW, anyone who contributed and happens to be in the Boston/Worcester
area and wants a beer, drop me a line, I will gladly oblidge.

Can't use internal network after dialup modem is used -- get ping: sendto: Operation not permitted

Posted: 05 Oct 2005 05:36 PM PDT

On Fri, 07 Oct 2005 19:02:02 +0200, Jean-David Beyer <com> wrote:
 

If you had nameservers on one of the other "hardware sockets", too bad.
If your nameserver (named, bind) wants to talk to ns1.google.com, and it
already knows that ns1.google.com has address 216.239.32.10, then
it does as I wrote, and leaves to the kernel to figure out how to reach
216.239.32.10. The kernel uses the routing table, and finds that with
the last line in the routing table, the mask is zero, so

(216.239.32.10 & 0.0.0.0) == (0.0.0.0 & 0.0.0.0)

oh, yes, that makes zero on both sides of the equal sign, and yes,
0 == 0 that is true, this line in the routing table is the one we will
use. Now, the rest of the line is

0.0.0.0 46.99.491.1 0.0.0.0 UG 0 0 0 ppp0

so the interface is ppp0. Oh, wait, this line has 'G', so we must do
a little more processing, see if we have a line with 46.99.491.1
(the gateway) as the "destination" and an 'H' flag (a host route),
yes we have that, but then again, the interface on that line is "ppp0"
That settles the question, we have to use the IP of ppp0 as our
source IP. 

It "knows" because it knows what addresses it is trying to reach with
its questions. Those addresses come from the configuration or from
answers obtained earlier. If there is an excelent and knowledgeable
name server sitting 10 inches away on eth0, too bad its address was not
listed in the configuration, because it may instead be trying to get
an answer from the other side of the earth.

Please correct me anyone, Is it not possible to configure named to allways
ask a particular name server first, and only do the formally correct
search for authoritative answers if the first server does not know?
 

I had RHL7.3 a looong time ago and can't remember except through the rosy
glasses we tend to see the bad old days when we were kids.

Yet I guess it was not that whimsically, it was in its configuration and in
the routing tables. It would be addressing the messages to specific IP
addresses, there is no broadcasting around for answers in dns. And so, if it
used the wrong destination IP address, it would not have helped to have
the world's best nameserver connected to eth0, even if it were to send on that
interface.

Or, have I forgotten? Does somebody remember what confusion reigned
in the Linux routing code? Or did Linux then fall back to sending
duplicated datagrams on all interfaces if it had no matching route?

By the way, what you call "hardware ports" is usually called "interfaces",
even the term "(network) interface" is also use for things that don't
really exist in hardware. I have a ppp0 interface that emits its segments
on eth0, whereas I also have eth0 as a separate intgerface with an IP
address. Yet the segments emitted by ppp0 do not have IP addresses, only
ethernet addresses. (But then there are some IP addresses buried deeper
inside the segemnts.) In this way, ppp0 is a pure software interface in
my case, yet all data that enter my house or leave it, do so over "ppp0".
You find the word "interface" in abbreviated form in e.g., "ifconfig", in
"/etc/sysconfig/network-scripts/ifcfg-eth0" and many other.

-Enrique

Best partition techniques for Linux/Winxp

Posted: 04 Oct 2005 12:35 PM PDT

sherry wrote:
 

I've never experienced that. Keep the windope partition to FAT32
then you got access and resizability option open with all the major
liveCDs (http://www.livecdlist.com )
I generally install windope / linux in any order,
then using QtParted reparition if necessary, and then using
Mepis liveCD and its GUI grub installer install grub
to boot up the linux/windope install by configuring
the /boot/grub/menu.lst file.
Even if windopes illegally overwrite the boot sector akin
to malware to wipe out the linux install, it can still be brought
back to life with Mepis liveCD grub installer.

linux install on sony laptop

Posted: 04 Oct 2005 09:47 AM PDT

On 10/04/05 18:47, GIJoe wrote: 

I did several slackware installations via network mounting the cdrom via
nfs. It is not a standard setup and required some manual work but always
worked OK.

Ciao
Giovanni
--
A computer is like an air conditioner,
it stops working when you open Windows.
Registered Linux user #337974 <http://counter.li.org/>