Linux-Setup Digest #203, Volume #20 Mon, 11 Dec 00 14:13:09 EST
Contents:
CGI Bins linked under virtual domains RH 6.0 ("DLS")
Mandrake 7.2 install hangs ([EMAIL PROTECTED])
howto make boot floppy after install (Anthony Ewell)
Re: adaptec 2940uw and install ("Alk")
Re: What is the command to . . . ? (Kaz Kylheku)
Re: Linux and NTFS (Thoni Bernhard)
Re: mount points ("Peter T. Breuer")
Re: CGI Bins linked under virtual domains RH 6.0 ("DLS")
Re: Using LILO to boot 2 Linux distributions ("Chris Kassopulo")
Re: howto make boot floppy after install (Bit Twister)
Re: Two cards on IRQ 9 ??? (Carlos)
Q: I2C and lm_sensors fail to compile (Dirk Engelmann)
Re: Kernel sees only part of memory
Re: set-up DSL? (Michael Perry)
Re: SiS 6326 (Michael Perry)
Re: Two cards on IRQ 9 ??? (Michael Perry)
----------------------------------------------------------------------------
From: "DLS" <[EMAIL PROTECTED]>
Subject: CGI Bins linked under virtual domains RH 6.0
Date: Mon, 11 Dec 2000 09:20:03 -0800
I'm running RH 6.0 as a web hosting server. I setup a free BBS under the
primary server domain. When I try to set up the same script under one of the
virtual domains it always calls up the primary domain script instead of the
one in the virtual domain's cgi-bin.
I figure I have made some very basic error in my initial setup of the
server. The problem is I'm a newbie with enough knowledge to be a danger to
myself and others.
Any ideas on wheere I have gone wrong?
Thanks in advance
Dann
------------------------------
From: [EMAIL PROTECTED]
Subject: Mandrake 7.2 install hangs
Date: Mon, 11 Dec 2000 17:24:48 GMT
I sure could use some help!! I have been trying to install Mandrake 7.2
on my system. I have an Abit BP6 with dual 500's, an ATI all in wonder
(pci), and plextor CDR. The cdrom is master on ide1, the main
harddrive is a 17gig Seagate on ide0. the second harddrive is on the
hpt360 channel as master. I can boot from the cdrom into the mandrake
install and everything proceeds normally until... about halfway through
the package installs ..the system will just stop. The transfers just
hang and I am forced to do a hard reboot. I have tried disconnecting
the drive on the hpt360 channel. I have tried both swapping and
removing one of the processors (in case one of them was bad). I have
tried removing each of ram chips in sequence. all of this of course to
try and isolate the problem as either ram, the dma66 channeled hd, and
the processors. Any help is greatly appreciate. The interesting thing
is that the hang doesn't happen with Mandrake 7.0 which doesn't
recognize the highpoint dma. I have also been sucessful in installing
Gentus's version of RH (6.2). However, I really would like to install
Mandrake 7.2. Thanks
ken
[EMAIL PROTECTED]
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
Date: Mon, 11 Dec 2000 09:43:38 -0800
From: Anthony Ewell <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.redhat
Subject: howto make boot floppy after install
Hi,
I just installed Red Hat 7.0. When it came time
to make the special boot floppy, I only had
"unformatted" floppies and had to bypass the
event.
Is there a way to make this floppy up without
having to go through the install process again?
Many thanks,
--Tony
[EMAIL PROTECTED]
p. s. my ISP is going through some NNTP growing
pains and is not saving more than two days of postings.
If replying, would you please respond to both my eMail
address and the newsgroup.
------------------------------
From: "Alk" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: adaptec 2940uw and install
Date: Mon, 11 Dec 2000 19:43:08 +0200
Reply-To: "Alk" <[EMAIL PROTECTED]>
Check if Ctrl-A settings are completely same in means of exact ( not auto)
termination setup and other strange features setup
If you get crc error right after loading boot image from lilo - lilo is
booting different boot image your it was made for
Boot anything else and recreate lilo
AG
"E Moore" <[EMAIL PROTECTED]> wrote in message
news:912qpc$4jp$[EMAIL PROTECTED]...
> I replaced an Adaptec 2940uw with another 2940uw and Red Hat 6.2
> started to have problems with the new card. Processes would not stay
> active and when I restarted Linux again I received a CRC error. I
> reinstalled
> Linux and the installation hangs at different places in the install. I
have
> low level formatted my SCSI drive and checked the drive for errors.
> The Adaptec 2940uw works with Windows (NT/95/98) but it is having problems
> in Linux. The card I am having trouble with is running Adaptec BIOS
> 1.34.3. My other Adaptec 2940uw is running OK on this PC and has Adaptec
> BIOS 1.32. I am wondering if there is a problem with versions of the
Adaptec
> 2940uw BIOS and Linux. Any help would be greatly appreciated !
>
> E. Moore
> [EMAIL PROTECTED]
> (remove NOSPAM from email address)
>
>
>
>
------------------------------
From: [EMAIL PROTECTED] (Kaz Kylheku)
Crossposted-To:
alt.os.linux.mandrake,alt.os.linux.slackware,alt.uu.comp.os.linux.questions,comp.os.linux.development.system,comp.os.linux.misc,comp.os.linux.networking,comp.os.linux.hardware
Subject: Re: What is the command to . . . ?
Reply-To: [EMAIL PROTECTED]
Date: Mon, 11 Dec 2000 17:58:50 GMT
On Mon, 11 Dec 2000 08:15:34 +0100, Josef Moellers
<[EMAIL PROTECTED]> wrote:
>Allen Wong wrote:
>>
>> In alt.os.linux.slackware Markus Amersdorfer <[EMAIL PROTECTED]> wrote:
>> > [EMAIL PROTECTED] wrote:
>
>> > find . -name '*.txt' -exec grep "Hello World" {} \;
>>
>> This works, but it's alot slower than "find . -type f -name '*.txt' -print |
>> xargs grep "Hello World".
>
>These solutions won't tell where they found the match.
>Markus' solution can be enhanced to do that:
> find . -name '*.txt' -exec grep "Hello World" {} \; -print
So can the other one:
... xargs grep --with-filename "Hello World"
or, more portably:
... xargs grep /dev/null "Hello World"
Note that exec'ing grep for each file may be a lot slower compared to using
xargs to build up a long argument list to pass to one grep instance.
------------------------------
From: Thoni Bernhard <[EMAIL PROTECTED]>
Crossposted-To:
alt.os.linux.caldera,alt.os.linux.mandrake,alt.uu.comp.os.linux.questions,comp.os.linux.help,comp.os.linux.networking
Subject: Re: Linux and NTFS
Date: Mon, 11 Dec 2000 17:23:33 +0000
hi,
linux r/w access to ntfs partition (win2000) works fine for me (suse 7.0)=
;
the only thing when booting win2000 the next time is:=20
give the /F flag for filesystemcheck;
i use kernel 2.2.16
greetings,
bernie
"Woody" <[EMAIL PROTECTED]> wrote:
> Can Linux read NTFS files on the same drive and or via Samba?
> If so, how.
> TIA
> Mike
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: mount points
Date: Mon, 11 Dec 2000 18:35:00 GMT
Gary Sandine <[EMAIL PROTECTED]> wrote:
> In case any students read this, I have to make a few comments:
>> Err, a banach space is a normed space with the completeness
>> axiom, isn't it? Sorry - my functional analysis is 20 years in the
>> past. But what you're asking is for every bounded sequence to have
>> a limit
> Not quite - the sequence (-1,1,-1,1,...) is bounded, but doesn't
> have a limit. Want every Cauchy sequence to converge.
Yes it does have *a* limit. It's just not a unique one. Do you want
me to say "limit point" instead of "limit"? Maybe that's the problem
for you. I meant that there is a point that is approached arbitrarily
closely in norm arbitrarily often as you proceed along the sequence.
>> In a linear space you
>> get a lot of points for free just by doing linear displacements and
>> recombinations.
> I have no idea what that means.
I meant that because you can scale by the field elements, you at least
get an image of the field embedded there, and that's already a lot of
points since the field is (presumably) real here. You can also form
linear combinations of vectors, and by the triangle inequality and
the fact that the norm scales linearly, the unit ball is convex, so
the unit ball has all convex combinations of all points within it.
That's a lot. Enough to make it likely that it has all limit points.
I think I can see that it has all limits if the space is finite
dimensional, since you just have to consider the way the components
of the basis behave. Hmmm ... yes, aiX+biY has to tend to aX + bY
if ai->a and bi->b by the triangle inequality and the linear norm
thing. That's encouraging. It leaves the infinite dimensional case as
the only one to worry about.
> Locally compact normed linear space (locally compact in the norm
> induced topology, that is) = finite dimensional. But be careful,
Yes, I think I definitely know that, though I don't see the trick that
forces it just now... oh, I suppose one considers a sequence of
unit base vectors. What does it have as a limit point?
> there are two non-equivalent definitions of locally compact which
> are standard.
Really? Surely it's that every point has a compact neighbourhood (or
possibly a base of compact neighbourhoods).
The first seems to me to be equivalent to the unit ball being compact,
because a compact neighbourhood must be bounded. Conversely if
the unit ball is compact, then by translation it is a neighbourhood
of any point you choose.
> I'm shocked. I see that you used to know what many of these things
> mean. Quite well, I'm sure, if you still remember them well enough
> (after 20 years, as you say) to string related items together as you
> did. So much for my clever example. :)
Well, I was a research (pure) mathematician for some years of my life.
But analysis is not my area. Logic and algebra is/was. But I probably
taught the third year functional analysis course at least once, despite
not being particularly wonderful at it(pure) . One needs a very visual
imagination for analysis, and I am better with symbolic reasoning than
spatial reasoning. Come to think, I don't recall ever teaching
functional analysis, though I do remember teaching measure theory,
complex analysis, number theory (I had Roger Taylor as a student
when he was in his first year, and I was in my fourth/first of
teaching, and I remember distinctly being battered by his insight.
Took me two weeks once to prove a statement he questioned. No
surprise that later he took up Wile's proof and completed the Fermat
demonstration).
Back to On-Topic stuff ...
>>> are questions such as, "OK - so with multiple partitions, I'll
>>> c:, d:, e:, etc. drives, right?" Questions which I think are
>>> reasonable.
>> ???
> I have heard that question so many times, I can't help but consider
> it reasonable, coming from a user with only MS Win experience. I
> usually offer the "DOS-Win-to-Linux-HOWTO" to begin with, explaining
Yes, this is a useful start.
>>> Place 20 meg /boot at front of drive. Then (usually) comes a
>> Fair enough, but not necessary, and completely impossible for the
>> L-user who already has a win partition there and can't figure what to
>> do about it.
> Agreed. Good point. But when we build from scratch, we *always*
> do this.
Not necesarily :-). I don't, for example, because I put a 2GB dos
partition, and then start with swap and root, wich are only a couple
hundred MB, so I don't have trouble with lilo even if it were the sort
of lilo one might have trouble with, of, under, ...
>> Again, I agree. Except that /usr and /var are very good ideas!
>> /usr because it keeps / small (and hence you can keep an extra copy
>> around) and /var becuase it keeps / practically readonly.
>> And mixing /var and / allows you to wipe your linux distro in other and
>> more unforeseen ways!
> Perhaps I should read the partitioning how to.
It is quite good, I assure you!
> Really, we advise as we do because having a big / partition is bad
> for anyone who actually intends to use Linux. But keeping it
> relatively simple is good, too (to begin with), for folks who are
> completely unfamiliar with Linux. When they change and wipe out
> their MS Win partition, they can change the partition scheme. :)
Absolutely. But (as I said, I hasten to remind you :-), these are
exactly the sort of arguments that apply to your house, your desk, or
any sort of organisation. If you divide it into separate areas by
function or other criteria, then you increase "safety" and "stability"
at the expense of "flexibility". How is that different from organising
your disk? If you let the kids play anywhere, then you get all kinds
of interesting outcomes.
>>> An answer to the stupid question above is that every absolutely
>>> convergent series converges. Are there others?
>> Oh .. is that all?
> Other than the definition of Banach space, it's the only one I can
> think of. I can think of many sufficient conditions, but not so
> many that are necessary.
Thanks. Hmm .. I have a copy of Dieudonne on the shelf. Let's see.
5.3.2 is the theorem that in a banach space, absolutely convergent
implies convergent. I see that a locally compact normed space is finite
dimensional (for him, locally compact means that every point has a
compact neighbourhood). Apparantly it's Riesz's theorem (one of).
Peter
------------------------------
From: "DLS" <[EMAIL PROTECTED]>
Subject: Re: CGI Bins linked under virtual domains RH 6.0
Date: Mon, 11 Dec 2000 10:40:18 -0800
BTW - my httpd.conf file virtual hosts is setup as follows for these
domains.
<VirtualHost 209.166.87.76>
ServerAdmin [EMAIL PROTECTED]
ServerName www.tek-1.com
ServerAlias tek-1.com
DocumentRoot /vhome/tek-1.com/html
ErrorLog logs/error_log
ScriptAlias /vhome/tek-1.com/cgi-bin/ cgi-bin/
Options ExecCgi Includes Indexes
</VirtualHost>
Hope this helps. Anything else you need to know about how the httpd.conf
file is setup as just let me know and I'll provide a txt file for your
review of my actual conf file.
Dann
"DLS" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I'm running RH 6.0 as a web hosting server. I setup a free BBS under the
> primary server domain. When I try to set up the same script under one of
the
> virtual domains it always calls up the primary domain script instead of
the
> one in the virtual domain's cgi-bin.
>
> I figure I have made some very basic error in my initial setup of the
> server. The problem is I'm a newbie with enough knowledge to be a danger
to
> myself and others.
>
> Any ideas on wheere I have gone wrong?
>
> Thanks in advance
>
> Dann
>
>
------------------------------
From: "Chris Kassopulo" <[EMAIL PROTECTED]>
Subject: Re: Using LILO to boot 2 Linux distributions
Date: Mon, 11 Dec 2000 13:38:22 -0500
I had the same difficulty. Lilo was on the mbr and Grub on hdc.
To boot hdc, Lilo pointed to Grub which loaded the kernel from
hdc. I wanted to have the choice of booting to hdc from Lilo.
The partition that contains the kernel you want to boot must be
mounted from the partition that contains Lilo. The path to the
kernel is the path in the Lilo partition. For example:
image=/mnt/edrive/boot/vmlinuz-pc97-2.2.14-modular
Hope this helps.
[EMAIL PROTECTED]
"Dennis Lee" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Does anyone know if this is possible? Some pointers to documentation
> would be appreciated. Here is what I am trying to do.
>
>
> Disk distribution:
>
> /dev/hda1 that other OS
> /dev/hda5 /
> /dev/hda6 /usr
> /dev/hda7 /home
> /dev/hda8 /var
> /dev/hda9 swap
> /dev/hda10 /
> /dev/hda11 /usr
> /dev/hda12 /home
> /dev/hda13 /var
>
> What goes where:
> /dev/hda5 to /dev/hda9 are for Red Hat
> /dev/hda9 to /dev/hda13 are for Slackware.
> They share swap since only one distro is up at any one time.
>
> What I did:
> 1. Installed that other OS.
> 2. Installed Red Hat with LILO in the MBR.
> 3. Installed Slackware with LILO in the root partition.
> 4. Boot into Red Hat and edit /etc/lilo.conf to include
> description for Slackware.
> 5. Ran /sbin/lilo...
>
> When I ran /sbin/lilo I got a message saying the image
> /vmlinuz (Slackware's) does not exists although I specified
> that root=/dev/hda10.
>
> What gives?
>
> Thanks and regards,
>
> dennis lee
------------------------------
From: [EMAIL PROTECTED] (Bit Twister)
Crossposted-To: comp.os.linux.redhat
Subject: Re: howto make boot floppy after install
Reply-To: [EMAIL PROTECTED]
Date: Mon, 11 Dec 2000 18:37:22 GMT
Try
mkbootdisk --device /dev/fd0 --version $(uname -r)
On Mon, 11 Dec 2000 09:43:38 -0800, Anthony Ewell <[EMAIL PROTECTED]> wrote:
>Hi,
>
> I just installed Red Hat 7.0. When it came time
>to make the special boot floppy, I only had
>"unformatted" floppies and had to bypass the
>event.
>
> Is there a way to make this floppy up without
>having to go through the install process again?
>
--
The warranty and liability expired as you read this message.
If the above breaks your system, it's yours and you keep both pieces.
Practice safe computing. Backup the file before you change it.
Do a, man command_here or cat command_here, before using it.
------------------------------
From: [EMAIL PROTECTED] (Carlos)
Crossposted-To:
comp.os.linux.networking,alt.linux.slakware,alt.os.linux.slackware,alt.os.linux.slackware
Subject: Re: Two cards on IRQ 9 ???
Date: 11 Dec 2000 12:40:30 -0600
In article <[EMAIL PROTECTED]>,
Lodewijk Otto <[EMAIL PROTECTED]> wrote:
>On Monday December 11 2000 at 09:15 [EMAIL PROTECTED] (M. Buchenrieder)
>wrote:
>
>> [EMAIL PROTECTED] (Lodewijk Otto) writes:
>>
>>> On Friday December 08 2000 at 22:02 Thomas SMETS <[EMAIL PROTECTED]>
>>> wrote:
>>
>>>> Well on slack 7.0,
>>>> I got the following problem ...
>>>> I've two 3COM cards modules 3c59x.o (I think)
>
>>> Sharing IRQ's is not a good idea under Linux, not good at all.
>>
>> These is a grossly misleading statement. IRQ sharing works - if
>> the drivers and the hardware both support it.
How? I could never make two different PCI cards share a IRQ, they simply
pick one and end-up blocking IRQ's that I would like my ISA sound card to
use... I haven't seen in the BIOS any setting allowing me to force a PCI
card to take a particular IRQ. Could anyonew explain how to do it?
>> A parallel port doesn't need any IRQ - as the standard printer driver
>> is using polling mode.
I use (and recommend) tunelp to make the driver use IRQ and thus reduce
CPU overhead.
Carlos
------------------------------
From: Dirk Engelmann <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.install,linux.redhat
Subject: Q: I2C and lm_sensors fail to compile
Date: Mon, 11 Dec 2000 19:55:14 +0100
Hello,
I have a problem with the I2C and the lm_sensors packages
I recently installed the Kernel 2.4-test11, and it is running fine.
Now I wanted to add some hardware monitoring(fan rotation, temperatures
etc.)
but the packages fail to compile.
I get :
[root@McGyver i2c-2.5.4]# make
grep: /usr/src/linux/include/linux/autoconf.h: Datei oder Verzeichnis
nicht gefunden
make: *** Keine Regel vorhanden, um das Target
�/usr/src/linux/include/linux/rhconfig.h�,
ben�tigt von �kernel/i2c-elektor.d�, zu erstellen. Schluss.
Which means in english:
[root@McGyver i2c-2.5.4]# make
grep: /usr/src/linux/include/linux/autoconf.h: File or Directory not
found.
make: *** No rule to make �/usr/src/linux/include/linux/rhconfig.h�,
required by �kernel/i2c-elektor.d�. Stop.
The link /usr/src/linux is pointing to the actual(2.4.-test11) kernel' s
source tree.
As far as I can see, all that's missing is the autoconf.h file.
But I found this file in the old (2.2.16) Kernel source tree.
Of course, I won't use that one.
How do I produce this file?
Any hints would be helpful.
Thanks
Dirk.
------------------------------
From: <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware
Subject: Re: Kernel sees only part of memory
Date: Mon, 11 Dec 2000 13:55:48 -0500
Usually, specifying the amount of memory at boot time should fix it.
However, I must tell you that when I increased my computer's memory from 64
Mb to 128 Mb a while back, the kernel detected all of it.
I seem to recall something about finicky bioses, for which the work around
was to tell the kernel howmuch memory you have, at boot time.
This is done by typing the following at boot time.
linux mem=256M
| |
| ----------- This is a boot time parameter. Lilo passes this
verbatim to the kernel.
|
-------------------- This is the name of the kernel image ( You may
have another kernel image file )
To see what all are the bootables, hit TAB at the LILO prompt.
HTH
Peter Linde <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Dear all,
>
> I am running RH Linux 7.0 on an HP Kayak, which has
> 2x733 Mhz processors and 256 Mb of memory. I also run Windows 2000
> from a separate disk. While booting Linux, the kernel reports and uses
> only 64 Mb of
> memory. Under Windows the entire memory is seen and used.
>
> Any hints?
>
> Peter Linde
> Lund Observatory
> Sweden
>
------------------------------
From: [EMAIL PROTECTED] (Michael Perry)
Subject: Re: set-up DSL?
Date: Mon, 11 Dec 2000 08:32:54 -0800
On Fri, 01 Dec 2000 01:15:49 GMT, Werner Puschitz
<[EMAIL PROTECTED]> wrote:
>If you have already set up your account with you ISP on Windows then
>download PPPoE for Linux from http://www.roaringpenguin.com/pppoe/
>
>Install the rpm package with rpm -ivh <package name> and run /usr/sbin/adsl-setup
>which
>will create DSL configuration files for you specific ISP. After that use adsl-start
>and
>adsl-stop to connect and to disconnect.
>
>Werner
>
>
>
>Dean Kwak wrote:
>
>> hello,
>>
>> i'm trying to set up my earthlink DSL to my linux box. is it even
>> possible to use DSL with my RED hat 6.2 linux box? i heard that it is
>> possible with @home cable but not sure if it'd be working with DSL. I'd
>> be appreciated if you could help me how to set-up if it's possible...
>>
>> thanks
>>
>> please reply to [EMAIL PROTECTED]
>
Not all dsl lines need the pppoe stuff. As an example, my pacbell dsl (not
PBI) uses the alcatel athome external dsl box. This is a simple ethernet
device which plugs into the phone line and has a cat5 ethernet connector
which I then feed to a firewall/NAT box. For me, I brought up two network
interfaces in debian potato, told it that one was for the external network
and one for internal, plugged the internal network card into a hub, did some
fuddling/fiddling with routes, and installed and modified the ipmasq deb
package.
I also have one static IP address but my ISP does not stop the use of NAT.
Its nice to have an enlightened (and BSD/Linux) ISP. :)
--
Michael Perry
[EMAIL PROTECTED]
==================
------------------------------
From: [EMAIL PROTECTED] (Michael Perry)
Subject: Re: SiS 6326
Date: Mon, 11 Dec 2000 08:36:56 -0800
On Sat, 2 Dec 2000 03:16:50 -0800, fciii <[EMAIL PROTECTED]> wrote:
>In order to take advantage of the sis 6326 chip acceleration, you need to
>run XFree86 >=4.0.
>Loaded beta 4.0 early this year, diaplay ran fine at 16 bpp depth but
>wouldn't support 24 bpp. Now am using stable release of 4.0.1, 24 bpp
>works fine.
>
Exactly what I did too. On a system I have with a SiS card integrated, the
video with redhat 6.2 seemed to go through stages. If I did nothing on it,
it seemed okay; but when I opened an xterm/eterm, horizonatal lines would
form on the screen where the window was. I found a driver on SiS website
which fixed some of it. My best alternative was to decide that SiS video
cards are problemmatic overall at least for me. I went to the Diamond Viper
770d cards running the Xi accelerated desktop server. Pretty good
performance, fast graphics!
The system with the SiS card ended up having win98se installed on it
anyways. Video works pretty well with the SiS "vga" drivers on it.
--
Michael Perry
[EMAIL PROTECTED]
==================
------------------------------
From: [EMAIL PROTECTED] (Michael Perry)
Crossposted-To:
comp.os.linux.networking,alt.linux.slakware,alt.os.linux.slackware,alt.os.linux.slackware
Subject: Re: Two cards on IRQ 9 ???
Date: Mon, 11 Dec 2000 08:57:22 -0800
On Fri, 08 Dec 2000 21:02:40 GMT, Thomas SMETS <[EMAIL PROTECTED]> wrote:
>Well on slack 7.0,
>I got the following problem ...
>I've two 3COM cards modules 3c59x.o (I think)
>First card is fine (ping telnet http) but the second is
>on the same IRQ (9). The two cards put them selves on the IRQ 9 when I
>look in dmesg or in ifconfig -a.
>
>Do i have to turn PnP off ?
>
>If yes, how
>
>Thomas,
>
>
I have two etherexpress 100 cards which do not share irqs in one box but in
my debian workstation, I had a sblive and a etherexpress which tried to
share irqs, but they did not play well together. In fact, the network would
grab the irq at boot time and not share at all even with irq sharing enabled
in the bios. So what I did in the interests of peace and prosperity for all
linux devices consisted of moving the sound card down a pci slot. This took
care of the insensitivities of certain linux devices :)
You can also use Becker's config utility at scyld, but I have never done
that. I tend to flash the bios using the 3com boot diskettes when I use a
3com card. The etherexpress cards are the ones I'm using now. Work great
every time!
--
Michael Perry
[EMAIL PROTECTED]
==================
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.setup.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Setup Digest
******************************
