Re: fvwm question..

[bing yu]

On Tue, Oct 05, 2004 at 11:12:24AM +1000, Cameron Hutchison wrote:
> I wasn't referring to anything XFS specific. XFS implements standard
> unix permissions, so you want to look at the man pages for chmod and
> chown.
Thank you, Problem solved by simply chmod of the mounting point.
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 

-- 
Cold hands, no gloves.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Software RAID 0 and linux 2.6.8.1

[Niels]

I shall explain my choises. I really want to buy a RAID 0 (striping) 
controller, because my current harddisk is the bottleneck in my current 
system. I have a dual XP-M 2800+ with 1Gbit NIC and 1GB memory. I have 
read enough about RAID 0. I really want it! I know that it is not safe, 
but I can backup my data.

I choose for a software RAID controller because it is cheap. An hardware 
controller would cost me six times more, but would be at most 10% 
faster. That's not worth it. Installing this RAID card is easy under Win 
XP, because I can just put a disk in my pc with the drivers at setup 
time and it will work.

I know this will be another story for debian. I've read stories of 
people who got the Promise drivers working under 2.6.8, but those 
drivers are not fast. I use 2.6.8 because I have a SMP setup. I want no 
other drives than the two of my RAID setup in my pc because I hate noise.

I buy a software RAID controller because the price difference between a 
software RAID SATA controller and a SATA controller is < €10,-

What are the troubles I will run into? I can load things in my BIOS, 
because I have 2MB BIOS memory (MPX2).

The adress you gave me was wrong, 
http://www.1U-Raid5.net/HowTO/SW-Raid-HOWTO.txt this one is correct.

Alvin Oga wrote:
On Mon, 4 Oct 2004, Niels wrote:
 

I want to buy a Promise Fasttrack TX2 plus RAID 0 controller with two 
SATA hard disks. The controller is a software controller. I want no 
other hard disks than these two in my PC. Can I boot of this setup with 
the 2.6 kernel?
   

yes...
 

Is it possible to put Windows XP and Debian Linux on the 
same RAID 0 array?
   

yes
 

Is it difficult to install?
   

no
-- trick is to install windoze on the 1st partition so it thinks its C:
--- bigger problem ...
- why use 2 disks in raid0 ( aks stripping ) which might allow you to read
 faster but also makes 2 disks look like 1 bigger disk
- if one disk dies... both disk are useless
- i think you want to use (sw or hw ) raid1 ( mirror )
- if one disk dies ... all your data is still intact on the other
disk
- gazillion raid howtos urls
http://1u_raid5.net/HowTo/SW-Raid-HOWTO.txt
c ya
alvin
 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: How to patch..

[condor_rl]

Eric,
you give a big amount of informations and a clear view to me about lots
of thinks i did not know.
Now I'm sure I have to study...a lot... :)
Also the link are very good!

..now I go to open a new topic on the list .. :)

Thanks again
Lorenzo

* Eric Gaumer <[EMAIL PROTECTED]> [031004, 09:25]:
> 
> Uncompress these first.
> 
> Apply the large patch first (mm) because chances are that the rtc-patch
> will fail here. If the rtc patch isn't against 2.6.8.1-mm4 then you may
> see hunks that fail. You'll need to fix these failures by hand.
> 
> If you are not familiar with C code and/or kernel code, then this may be
> difficult for you to fix. It may be easier to have two source trees.
> Apply the mm patch to one and the rtc patch to the other. Then you can
> create your own patch that has mm and rtc merged together. This gives
> you a little flexibility as what the patches were applied to. Bitkeeper
> is good for this because you can clone trees easily, using hardlinks
> (see below).
> 
> > 
> > Or should I use a dfferent way?
> > 
> > If I forget what patches I have installed how can can I see the list of
> > installed paches? :)
> 
> Yes and no... Just using patch and diff you wont be able to see without
> keeping a directory of patches you applied to the tree.
> 
> A utility called "Quilt" exists that allows you to push and pop patches.
> With this you can see what patches have been applied. Quilt is available
> via apt-get. This may sound really cool but be warned... Quilt is not
> something I would consider "for newbies". You should have a firm
> understanding of kernel patching first.
> 
> http://packages.debian.org/unstable/devel/quilt
> 
> There is also another utility called ketchup that may be of some help to
> you. This is an easy way to keep your tree up to date. It also supports
> the mm tree as well as a few others.
> 
> http://kerneltrap.org/node/view/2976
> 
> You could also check out Bitkeeper. It's free of charge (providing you
> keep your source open) and it makes merging changes easier. It's also
> GUI driven. Here is an article on kernel source control that makes for a
> decent primer.
> 
> http://linuxjournal.com/article.php?sid=6183
> 
> > 
> > 
> > Thanks and sorry for the log list of questions...I'm a Newb...
> > Lorenzo
> 
> 
> -- 
> Eric Gaumer <[EMAIL PROTECTED]>
> -- 
> Eric Gaumer <[EMAIL PROTECTED]>



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: using scp to move files

[martin f krafft]

also sprach Kamaraju Kusumanchi <[EMAIL PROTECTED]> [2004.10.04.1741 +0200]:
>I need to move some files from a server to client. I am planning to
> use scp for this. Using scp, I can only copy the files. But I want to
> move the files instead of copying them. Any ideas? The manual of scp
> does not have any option for this.

scp source server:dest && rm source

-- 
Please do not CC me when replying to lists; I read them!
 
 .''`. martin f. krafft <[EMAIL PROTECTED]>
: :'  :proud Debian developer, admin, and user
`. `'`
  `-  Debian - when you have better things to do than fixing a system
 
Invalid/expired PGP subkeys? Use subkeys.pgp.net as keyserver!


signature.asc
Description: Digital signature

Linux 2.6 kernel modules links...

[condor_rl]

Hi,

I have just upgraded to linux kernel 2.6.
But I have a lot of problems with kernel modules.
Now I think the time is arrived, and I would like to understand how kernel modules 
"work" and fit with kernel.
For example what is /etc/modules.conf, what is modprobe?
And what is new in kernel 2.6 concerning modules?

Looking for "2.6 kernel modules" with google on Internet I can find lots
of informations but I did not found an overall view about kernel modules
and related config files for 2.6 kernel.

I'm sure I'm missing something, but I will be glad if you could help me
with some links to documents you judge good start poins..

THK to all
Lorenzo


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: using scp to move files

[Cameron Hutchison]

Once upon a time martin f krafft said...
> also sprach Kamaraju Kusumanchi <[EMAIL PROTECTED]> [2004.10.04.1741 +0200]:
> >I need to move some files from a server to client. I am planning to
> > use scp for this. Using scp, I can only copy the files. But I want to
> > move the files instead of copying them. Any ideas? The manual of scp
> > does not have any option for this.
> 
> scp source server:dest && rm source

and what about the other way?

scp server:source dest && rm ???


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: using scp to move files

[martin f krafft]

also sprach Cameron Hutchison <[EMAIL PROTECTED]> [2004.10.05.1024 +0200]:
> and what about the other way?
> 
> scp server:source dest && rm ???

scp server:source dest && ssh server rm source

-- 
Please do not CC me when replying to lists; I read them!
 
 .''`. martin f. krafft <[EMAIL PROTECTED]>
: :'  :proud Debian developer, admin, and user
`. `'`
  `-  Debian - when you have better things to do than fixing a system
 
Invalid/expired PGP subkeys? Use subkeys.pgp.net as keyserver!


signature.asc
Description: Digital signature

Re: Linux 2.6 kernel modules links...

[Andrea Vettorello]

On Tue, 5 Oct 2004 10:18:08 +0200, [EMAIL PROTECTED]
<[EMAIL PROTECTED]> wrote:
> Hi,
> 
> I have just upgraded to linux kernel 2.6.
> But I have a lot of problems with kernel modules.
> Now I think the time is arrived, and I would like to understand how kernel modules 
> "work" and fit with kernel.
> For example what is /etc/modules.conf, what is modprobe?
> And what is new in kernel 2.6 concerning modules?
> 
> Looking for "2.6 kernel modules" with google on Internet I can find lots
> of informations but I did not found an overall view about kernel modules
> and related config files for 2.6 kernel.
> 
> I'm sure I'm missing something, but I will be glad if you could help me
> with some links to documents you judge good start poins..
> 

I hope you have installed the "module-init-tools" package. 

In short modules.conf is a configuration file for loading kernel
modules, modprobe is a program to add and remove modules from the
running kernel. The difference between insmod is that modprobe try to
resolve modules dependencies.

To have more details "man modules.conf" and "man modprobe" or you can
look to the kernel-howto (but don't know how is updated)...


Andrea


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Samba issues

[Gabriel Granger]

Top man!
That sorted it out straight away, thanks very much I cant believe 
it was something so simple

Gabe
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Hi, About Kernel 2.6.6 with NPTL on Debian.

[NetSnake]

Hi, all
I compile kernel-source 2.6.6 on Debian, but it seems non NPTL support.
I compile kernel-source-2.6.6, glibc6 2.3.2-ds1-16, glibc-i686
2.3.2-ds1-16 glibc6-dev-2.3.2-ds1-16
On a machine the NPTL works fine!
But on another machine, the same kernel(I tried same kernel config and
same binary kernel iamge) and libc, the NPTL NOT work!
I can not find any diff between two machine!
How can I make sure one system already support NPTL? 3ks!
I use j2sdk 1.4.2, run rmiregistry to test this feature, one machine
have 1 process and 8 threads, one machine have 8 process.
I also use OpenLDAP compiled with thread support, have some issue on
these two machine.
Any suggestion?




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Hi, About Kernel 2.6.6 with NPTL on Debian.

[Andrea Vettorello]

On Tue, 05 Oct 2004 17:32:22 +0800, NetSnake <[EMAIL PROTECTED]> wrote:
>  Hi, all
>  I compile kernel-source 2.6.6 on Debian, but it seems non NPTL support.
>  I compile kernel-source-2.6.6, glibc6 2.3.2-ds1-16, glibc-i686 2.3.2-ds1-16
> glibc6-dev-2.3.2-ds1-16
>  On a machine the NPTL works fine!
>  But on another machine, the same kernel(I tried same kernel config and same
> binary kernel iamge) and libc, the NPTL NOT work!
>  I can not find any diff between two machine!
>  How can I make sure one system already support NPTL? 3ks!
>  I use j2sdk 1.4.2, run rmiregistry to test this feature, one machine have 1
> process and 8 threads, one machine have 8 process.
>  I also use OpenLDAP compiled with thread support, have some issue on these
> two machine.
>  Any suggestion?

Try look to the "libc6-i686" package, IIRC to use NPTL you need
support on your libc6 (someone correct me)...


Andrea


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Transport debian woody from laptop to PC

[Olle Eriksson]

On Tuesday 05 October 2004 06.06, s. keeling wrote:
> Incoming from Oliver Fuchs:
> > I am running a debian woody box for a long time now on my laptop.
> > I want to install it now with the (nearly) same configuration on my
> > PC. Can I transport the laptop system (for example via tar) to my PC
> > or won't this work.
> [...]
> The archives at lists.debian.org mention this often.  You should think
> carefully about this though.  Some apps run on a laptop may not be
> appropriate on a PC.  pcmcia, anachron, ...  It shouldn't cause any
> real problems you can't work around, however.

How about the other way around? To copy a partition from a desktop 
computer to a laptop via cp. Are there any questions asked during a 
Debian install about whether it is on a desktop or laptop computer and 
what things are differently on a laptop?

Regards
Olle Eriksson


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Installation recommendation

[Ritesh Raj Sarraf]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi,
I've just purchased a laptop and will install debian onto it. I just
wanted your generous suggestions regarding installation.
I am planning to use the following partition structure:
/boot -- 150mb
/home -- 10GB
/usr -- 15GB
/usr/local -- 5GB
I don't think having /tmp, /var etc on a laptop would be much
beneficial. Also since Debian already has most softwares packaged
/usr/local doesn't need be much big. And since I've a target to become a
DD soon, whatever I install into /usr/local will some time soon go into
/usr .
I'll be using Maildir format for emails. So no /var tension. /home will
do that.
Is there anything else I should be concerned for installation on a
laptop ? Any better suggestion ?
rrs
- -- 
Ritesh Raj Sarraf
RESEARCHUT -- http://www.researchut.com
Gnupg Key ID: 04F130BC
"Stealing logic from one person is plagiarism, stealing from many is research".
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (GNU/Linux)
Comment: Made with pgp4pine 1.76

iD8DBQFBYnu14Rhi6gTxMLwRAjMVAJ9l+if7xQDUlDaWN6P2giMW0ld5BQCcCdQ2
5sa4oJNVZseDlJ2P7pVWvGA=
=1o4p
-END PGP SIGNATURE-
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sendmail Install Dramas

[David Powell]

Andrea Vettorello wrote:
>http://bugs.debian.org/cgi-bin/pkgreport.cgi?which=pkg&data=sendmail-base&archive=no)
>are reported a couple of problems ugrading (one is solved =)...
Ah, thanks for the link to the bug tracking info - adding the group sasl 
worked, as mentioned at: 
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=273719

So all I did was :
1. apt-get install sendmail (which caused the errors as previously 
discussed)
2. addgroup sasl
3. apt-get -f install (this time all went smoothly)

Thanks again!
--
--
Moondrake Trust
e: [EMAIL PROTECTED]
m: 0415 341 821
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Adaptec 2100S: System freeze when trying to backup with tar

[Cristian Livadaru]

Hi, I recently installed a Adaptec 2100S SCSI HW Raid controler and two HD which
are mirroerd.
Kernel version is 2.4.22-ow1 and I compiled the i2o module to the kernel and I
thought that everyithing works fine, well actualz it does work perfect till I
start a tar.
After a short while the system just hangs and I have to reboot it.
Is there a problem with the Driver that's in the kernel ? alas adaptec has no
software for debian, just redhat and suse. I could use alien to install the rpm
packages but I'm not sure if this will work or solve my problem.
Should I try kernel 2.6 ?


Cris


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Recomended dual head video card

[Paolo Alexis Falcone]

On Tue, 5 Oct 2004 14:44:49 +0900, Nick Hastings <[EMAIL PROTECTED]> wrote:
> 
> Hi all,
> 
>for reasons I won't bother going into I'm about turn a server
> machine into a workstation (and perhaps occasional game machine).
> 
> There are two nice 1280x1024 17 inch digital LCD displays available
> (Mitsubishi RTD179S) [1], so I thought I'd go out and get myself a nice
> dual head video card. But which to choose? My budget is fairly generous.
> 
> My machines motherboard has and AGP slot so I thought an AGP card would
> be best.
> 
> I've had a bit of a look at the ATI web site and it seems that the
> "FireGL Z1-128" [2] might suit me well.
> 
> I briefly looked into nvidia cards, but seem to remember there may be
> some issues with drivers for 2.6 kernels.
> 
> I saw the nvidia vs ati thread a few days ago, but there seemed to be
> little in the way of conclusions or specific reports of experiences
> with cards. Anyone have anything to add?
> 
> Any insights would be much appreciated. This is the first time that I've
> ever had to choose a video card.
> 
> Please note that I'm not scared of compiling kernels or running bleeding
> edge software: I've been using 2.6 since the "pre" days and Debian
> unstable for about 4 years.

Matrox cards are quite good for dual displays, although the price
comes at a premium too.


-- 
Paolo Alexis Falcone
[EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Recomended dual head video card

[Henrique de Moraes Holschuh]

On Tue, 05 Oct 2004, Nick Hastings wrote:
> I've had a bit of a look at the ATI web site and it seems that the
> "FireGL Z1-128" [2] might suit me well.

Either that, or any nVidia card with dual outputs. You can have it dual-DVI
or DVI+VGA (the DVI port has the VGA signals as well, and the cards come
with an adaptor plug).

> I briefly looked into nvidia cards, but seem to remember there may be
> some issues with drivers for 2.6 kernels.

Not anymore.  And they are the fastest ones you will find. As long as you
don't mind relying on non-free drivers, because if you do, the FireGL *may*
be a much better choice (I don't know if it uses free drivers or not).

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sendmail Install Dramas

[Andrea Vettorello]

On Tue, 05 Oct 2004 21:12:03 +1000, David Powell
<[EMAIL PROTECTED]> wrote:
> Andrea Vettorello wrote:
> 
>  
> >http://bugs.debian.org/cgi-bin/pkgreport.cgi?which=pkg&data=sendmail-base&archive=no)
>  >are reported a couple of problems ugrading (one is solved =)...
> 
> 
> Ah, thanks for the link to the bug tracking info - adding the group sasl
> worked, as mentioned at:
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=273719
> 
> So all I did was :
> 
> 1. apt-get install sendmail (which caused the errors as previously
> discussed)
> 2. addgroup sasl
> 3. apt-get -f install (this time all went smoothly)
> 
> Thanks again!
> 

No problem, happy to be able to help.


Andrea


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Adaptec 2100S: System freeze when trying to backup with tar

[Andrea Vettorello]

On Tue,  5 Oct 2004 13:31:17 +0200, Cristian Livadaru <[EMAIL PROTECTED]> wrote:
> 
> Hi, I recently installed a Adaptec 2100S SCSI HW Raid controler and two HD which
> are mirroerd.
> Kernel version is 2.4.22-ow1 and I compiled the i2o module to the kernel and I
> thought that everyithing works fine, well actualz it does work perfect till I
> start a tar.
> After a short while the system just hangs and I have to reboot it.
> Is there a problem with the Driver that's in the kernel ? alas adaptec has no
> software for debian, just redhat and suse. I could use alien to install the rpm
> packages but I'm not sure if this will work or solve my problem.
> Should I try kernel 2.6 ?
> 

I don't know what kind of kernel is 2.4.22-ow1, but i'll suggest you
to try with one of the latest 2.4.x series (2.4.27 IIRC) instead of
trying 2.6.x.

There are strange error messages on syslog?


Andrea


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Adaptec 2100S: System freeze when trying to backup with tar

[Cristian Livadaru]

Citat Andrea Vettorello <[EMAIL PROTECTED]>:

> On Tue,  5 Oct 2004 13:31:17 +0200, Cristian Livadaru <[EMAIL PROTECTED]>
> wrote:
> >
> > Hi, I recently installed a Adaptec 2100S SCSI HW Raid controler and two HD
> which
> > are mirroerd.
> > Kernel version is 2.4.22-ow1 and I compiled the i2o module to the kernel
> and I
> > thought that everyithing works fine, well actualz it does work perfect till
> I
> > start a tar.
> > After a short while the system just hangs and I have to reboot it.
> > Is there a problem with the Driver that's in the kernel ? alas adaptec has
> no
> > software for debian, just redhat and suse. I could use alien to install the
> rpm
> > packages but I'm not sure if this will work or solve my problem.
> > Should I try kernel 2.6 ?
> >
>
> I don't know what kind of kernel is 2.4.22-ow1, but i'll suggest you
> to try with one of the latest 2.4.x series (2.4.27 IIRC) instead of
> trying 2.6.x.
>
> There are strange error messages on syslog?
>
>
> Andrea
>
>


-ow is the openwall patch the kernel is 2.4.22
I would prefer a 2.4 kernel since there is no openwall patch for 2.6
No alas there is absolute no info in any log, the server just freezes.
I will give it a try with the latest 2.4.x kernel.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

multisession on CD-RW with cdrecord

[Gerard Robin]

hello,
with kernel-image 2.4.20, I burned a CD-RW in multisession mode.
I use mkisofs 2.0+a38-1 + cdrecord 2.0+a38-1.

My problem is that until the 19 th session all works fine, but at the 20 th
session when I eject, I mount and I do: "cd /cdrw/Debsept04; ls" I get:

Oct  4 12:44:09 mymachine kernel: scsi0: ERROR on channel 0, id 1, lun 0, CDB:
Request Sense 00 00 00 40 00 
Oct  4 12:44:09 mymachine kernel: Info fld=0x44d5f (nonstd), Current sd0b:01: 
sense key Medium Error
Oct  4 12:44:09 mymachine kernel: Additional sense indicates Unrecovered read 
error
Oct  4 12:44:09 mymachine kernel:  I/O error: dev 0b:01, sector 1127804
 
And on 62 files (packages Debian) only 26 files are copied on the CD-RW.

"isovfy /imgISO/image2.iso" displays:

Root at extent 44d5c, 2048 bytes
[0 0]
No error found

"cdrecord -dev=1,0 -msinfo"  displays: 281920, 290135

"du -s /cdrw" displays: 283008

Is there a problem in mode multisession with cdrecord and CD-RW ?

Thanks in advance for any help. 

-- 
Gerard


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Recomended dual head video card

[Johann Koenig]

On Tuesday October  5 at 02:44pm
Nick Hastings <[EMAIL PROTECTED]> wrote:

> My machines motherboard has and AGP slot so I thought an AGP card
> would be best. 

Another alternative is to run two video cards. I did that for a long
time, with an Nvidia GeForce 4 MX440 SE (AGP) and a cheap Matrox (PCI),
then replaced the Matrox with an Nvidia TNT2(PCI). Both setups worked
great, there was only a little trouble with the kernel switching to the
PCI card if I set the bios to use AGP, so I just set it to use PCI all
the time.

For the Nvidia cards, I used Nvidia's non-free drivers. They work
awesome. For the Matrox, I just used the in-tree kernel driver. Worked
just fine.

Also, don't try dual-booting this with Win2k. I think I got it working,
but it sure sucked.

But if money is not an obstacle, I would start with looking for used
Matrox cards on Ebay. You should be able to get something pretty decent
for less than retail.
-- 
-johann koenig
Today is Prickle-Prickle, the 55th day of Bureaucracy in the YOLD 3170
My public pgp key: http://mental-graffiti.com/pgp/


pgp9b53KSFY7m.pgp
Description: PGP signature

Re: Limiting CUPS "Completed Jobs" Listing

[Adam Funk]

On Thursday 30 September 2004 16:00, CW Harris wrote:

> # MaxJobs: maximum number of jobs to keep in memory (active and
> # completed.) Default is 500; the value 0 is used for no limit.
> #
> #MaxJobs 500
> MaxJobs 50

I'd also been trying to figure this out.
Thanks,
Adam


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[Rodney Richison]

Top posting leads to a big mess. Every non-technical person does it
and it ends up with them saying "let's use a forum or something"
because they can't follow the thread any longer. Surprise surprise, a
forum works with bottom posting and everybody seems to find that
normal.
 

I run a list for techs/consultants.  (ChannelVar.com)
They are a VERY qualified bunch of guys. (Admittedly not linux users 
though)  :)
The preferred method is top posting.
The preferred format is to allow html.

--

Rodney Richison
[EMAIL PROTECTED]
118 N. Broadway
PO Box 566
Cleveland, OK 74020
RCR Computing - RCRnet
http://www.rcrnet.net
tel: 918-358-
fax: 918-358-1112



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Woody or Sarge

[Brett Charbeneau]

Adi Linden <[EMAIL PROTECTED]> writes:

> My BIG issue is the short lived support in terms of security updates. Having
> to reinstall a server after 18 month is totally unacceptable.

I hear you.
I'm in the middle of moving all of my RedHat servers to Debian (yay!), 
but it's a time consuming process and some boxes have to wait their turn.
As a stop-gap solution, I managed to stumble across Progeny Transition 
Service:

http://transition.progeny.com/index.html

which rolls security updates in RPM format for RedHat 7.2, 7.3, 
8.0, and 9 and makes them available for a fee.
At $5 per month per server, it's definitely worth it to me, but perhaps 
there is a similiar group out there doing the work for free?

P.S. I have NO affiliation with Progeny except for being a customer grateful for 
their service.


-- 

Brett Charbeneau, Network Administrator Tel: 757-259-7750
Williamsburg Regional Library   FAX: 757-259-7798
7770 Croaker Road   [EMAIL PROTECTED]
Williamsburg, VA 23188-7064 http://www.wrl.org




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Adaptec 2100S: System freeze when trying to backup with tar

[Marc Mueller]

Hi,
I recently installed a Adaptec 2100S SCSI HW Raid controler and two HD which
are mirroerd.
Kernel version is 2.4.22-ow1 and I compiled the i2o module to the kernel and I
thought that everyithing works fine, well actualz it does work perfect till I
start a tar.
After a short while the system just hangs and I have to reboot it.
Try using the "dpt_i2o" driver (Adaptec I2O RAID support) instead of the 
i2o_* modules.

Marc
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Adaptec 2100S: System freeze when trying to backup with tar

[Cristian Livadaru]

Citat Marc Mueller <[EMAIL PROTECTED]>:


> Try using the "dpt_i2o" driver (Adaptec I2O RAID support) instead of the
> i2o_* modules.
>
> Marc
>


that's the one I'm using. sorry for the poor information, had no conection to my
server and didn't remember everything.

here is all information about my controler:

vwclub:/etc# cat /proc/scsi/dpt_i2o/0
Adaptec I2O RAID Driver Version: 2.4 Build 5

Vendor: Adaptec  Model: 2100SFW:370F
SCSI Host=scsi0  Control Node=/dev/dpti0  irq=11
post fifo size  = 255
reply fifo size = 255
sg table size   = 56

Devices:
ADAPTEC RAID-1   Rev: 370F
TID=517, (Channel=0, Target=4, Lun=0)  (online)


vwclub:/etc# cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 04 Lun: 00
  Vendor: ADAPTEC  Model: RAID-1   Rev: 370F
  Type:   Direct-AccessANSI SCSI revision: 02




RAIDUTIL  Version: 3.31  Date: 8/12/2002  LINUX CLI Configuration Utility
Adaptec ENGINE  Version: 3.31  Date: 8/12/2002  Adaptec LINUX SCSI Engine

#  b0 b1 b2  Controller Cache  FWNVRAM Serial Status
---
d0 --ADAP2100S  16MB   370F  CHNL 1.1  BF0B12620AHOptimal

Physical View
AddressType  Manufacturer/Model Capacity  Status
---
d0b0t4d0   Disk Drive (DASD) IBM  DDYS-T36950N  35003MB   Optimal
d0b0t8d0   Disk Drive (DASD) IBM  DDYS-T36950N  35003MB   Optimal

Logical View
Address   Type  Manufacturer/Model  Capacity  Status
---
d0b0t4d0  RAID 1 (Mirrored) ADAPTEC  RAID-1 35003MB   Optimal
 d0b0t4d0 Disk Drive (DASD) IBM  DDYS-T36950N   35003MB   Optimal
 d0b0t8d0 Disk Drive (DASD) IBM  DDYS-T36950N   35003MB   Optimal


AddressMax Speed  Actual Rate / Width
---
d0b0t4d0   Ultra3 160 MB/secwide
d0b0t8d0   Ultra3 160 MB/secwide

AddressManufacturer/ModelWrite Cache Mode
---
d0b0t4d0   ADAPTEC  RAID-1   Write Back
 d0b0t4d0  IBM  DDYS-T36950N Write Back
 d0b0t8d0  IBM  DDYS-T36950N Write Back

#  Controller Cache  FWNVRAM BIOS   SMOR  Serial
---
d0 ADAP2100S  16MB   370F  CHNL 1.1  1.62   1.12/79I  BF0B12620AH

#  Controller  Status Voltage  Current  Full Cap  Rem Cap  Rem Time
---
d0 ADAP2100S   No battery

AddressManufacturer/ModelFW  Serial123456789012
---
d0b0t4d0   IBM  DDYS-T36950N S93E 5FLB9465 ----XOX-
d0b0t8d0   IBM  DDYS-T36950N S93E VFLAP060 ----XOX-

Capabilities Map:  Column 1 = Soft Reset
   Column 2 = Cmd Queuing
   Column 3 = Linked Cmds
   Column 4 = Synchronous
   Column 5 = Wide 16
   Column 6 = Wide 32
   Column 7 = Relative Addr
   Column 8 = SCSI II
   Column 9 = S.M.A.R.T.
   Column 0 = SCAM
   Column 1 = SCSI-3
   Column 2 = SAF-TE
   X = Capability Exists, - = Capability does not exist, O = Not Supported



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Mozilla browsers and Flashplayer

[Freddy Freeloader]

Wim De Smet wrote:
Hi,
On Mon, 04 Oct 2004 13:03:51 -0700, Freddy Freeloader
<[EMAIL PROTECTED]> wrote:
 

I have a strange problem with Mozilla browsers, both Mozilla and
Firefox, and Flashplayer when I install sarge using the netinst cd.
If I install woody first, and then do a dist-upgrade from there all
sites that use flash work normally, but if I do a fresh install using
the new installer some of the sites, such as MLB's Gameday do not
display correctly.  At first I thought this was something to do with
just the browser and Flashplayer but then realized that this only
happens on fresh installs using the new installer, so it would seem to
me that it has to be related to something being left out or changed in
the system between woody and sarge.
What happens is that in the areas that normally display player stats for
the game display nothing expect for player names and then a total at the
bottom of where each column should be.  This is an accurate total of
hits, BB's, KO's RBI's, etc... to that point in the game there's just no
stats displayed above the totals, and the inning by inning scores are
all blank too along with all pitching stats.  In the box that shows each
pitch location the graphics are off there too.  It shows them, but
displays them differently than when the page works normally.
Has anyone else experienced this?
   

No, but I do seem to recall that flashplayer requires some fonts to be
installed. Maybe it is these that you are missing? It's documented in
its readme I think. (package gsfonts if I'm not mistaken)
greets,
Wim
 

Thanks for the reply
Those notes about the gsfonts are in the installer not the readme.  :)
I had the gsfonts package installed, just not the gsfonts-x11 package, 
so I'll see what happens when the baseball playoffs start and I use the 
MLB Gameday again. 

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installation recommendation

[Gerard Robin]

On Tue, Oct 05, 2004 at 04:32:12PM +0545, Ritesh Raj Sarraf wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1


> I am planning to use the following partition structure:
> 
> /boot -- 150mb
> /home -- 10GB
> /usr -- 15GB
> /usr/local -- 5GB

Have a look at  http://kmself.home.netcom.com/Linux/FAQs/partition.html
hth
 
-- 
Gerard


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installation recommendation

[s. keeling]

Incoming from Ritesh Raj Sarraf:
> 
> I've just purchased a laptop and will install debian onto it. I just
> wanted your generous suggestions regarding installation.
> 
> I am planning to use the following partition structure:
> 
> /boot -- 150mb
> /home -- 10GB
> /usr -- 15GB
> /usr/local -- 5GB

My /boot has 6.1 Mb used.  Personally, I think /boot as a separate ptn
is unnecessary, but it is nice to not have to mount it at all (defence
against LKM's?).

> I don't think having /tmp, /var etc on a laptop would be much

When you do "apt-get upgrade", the packages are downloaded to
/var/cache/apt/archives.  If there isn't enough space there, you'll
have to do something creative (ln -s /var/cache/apt/archives
/somewhere/with/space).

Others insist a separate /tmp is a good security measure or system
management feature.  I don't know, but they're probably correct.

I would suggest the following:

/200 Mb
/usr as much as you can spare
/var 1 Gb (that's possibly excessive)
/homewhatever's left over (mine's 1 Gb, 300 Mb used)
/opt [if you use that sort of thing] 500 Mb?
/scratch useful for the "ln -s ..." trick above


-- 
Any technology distinguishable from magic is insufficiently advanced.
(*)   http://www.spots.ab.ca/~keeling 
- -


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[s. keeling]

Incoming from Rodney Richison:
> 
> >Top posting leads to a big mess. Every non-technical person does it
> >and it ends up with them saying "let's use a forum or something"
> >because they can't follow the thread any longer. Surprise surprise, a
> >forum works with bottom posting and everybody seems to find that
> >normal.
> 
> I run a list for techs/consultants.  (ChannelVar.com)
> They are a VERY qualified bunch of guys. (Admittedly not linux users 
> though)  :)
> The preferred method is top posting.
> The preferred format is to allow html.

Thanks for the warning.


-- 
Any technology distinguishable from magic is insufficiently advanced.
(*)   http://www.spots.ab.ca/~keeling 
- -


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[Daniel Goldsmith]

On Tue, 05 Oct 2004 08:24:10 -0500, Rodney Richison
<[EMAIL PROTECTED]> wrote:
> >
> I run a list for techs/consultants.  (ChannelVar.com)
> They are a VERY qualified bunch of guys. (Admittedly not linux users
> though)  :)
> The preferred method is top posting.
> The preferred format is to allow html.

Qualified they may be, but the sheer *waste* in what you are
describing seems appalling.

Why choke up bandwidth with  rubbish which all too often adds
very little to the actual discussion. Anything technical can usually
be best described with raw text, anything which needs bells and
whistles can be dealt with by linking to a static web-page. If it
isn't signal, then it is noise.

The growth in band-width waste with the development of broadband
services is probably the most regrettable trend of recent years.

pace the whole GMail issue - I use it for mailing lists, for pretty
much the same reasons to do with wastage. I simply cannot afford the
bandwidth to download much of the crud which populates mailing-lists -
Google Inc. can. They also provide a means for me to scan and search
the lists, looking for what I want or need.

Private email is handled elsewhere, without the issues alluded to
earlier in this thread.

Just my 2c.
-- 
Daniel Goldsmith.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Radius + wireless access

[Gabriel Granger]

Hi All,
I've got a couple of wireless AP around the office, and would like to 
mange the laptop with wireless cards using Radius for authentication.  
all of my AP have the ability to talk to a Radius server.  I have got 
my radius server up and working using some simple test accounts, I've 
done some looking around everything seems to point to EAP.  I might be 
wrong, but I thought EAP requires cert to be issued to the client 
machines.  I want a simple authentication system, so that office 
workers can gain access anywhere in the building but also so that 
trusted clients can gain access too with out having to had out 
certs(just give them a username password)

Am I thinking about this the wrong way? is it possible to just 
wirelessly authenticate users based on a username and password like you 
can with dialup users?

Any comments / suggestions / solutions greatly appreciated
Gabe
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why Grub? Must I Switch?

[Daniel L. Miller]

cr wrote:
On Mon, 04 Oct 2004 18:03, Daniel L. Miller wrote:
 

Paul Johnson wrote:
   

<#secure method=pgp mode=sign>
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
"Daniel L. Miller" <[EMAIL PROTECTED]> writes:
 

I don't understand that - unless your different kernel versions are
all using the same filename.
   

You have a basic shell.  You can type the name, and IIRC, there is tab
autocompletion.
 

So you can't have an unattended boot?
Daniel
   

You can certainly have an unattended boot - the GRUB menu comes up and, after 
a specified interval, it will go on to boot the first entry on the menu  
(unless, of course, you manually select another).

However, if you've just done a system upgrade or destroyed a kernel or 
whatever, or are just Grub-floppy-booting into a system, then you can bring 
up the  Grub command line which has, as Paul said, tab autocompletion - and 
very handy it is too for finding a kernel when you can't remember whether 
it's  vmlinuz2.4.18-bf2.4 or   vmlinuz2.4.18_bf2.4   

cr
 

That's not my idea of an unattended boot - which is where the comments 
on symlinks came from.  If you have a symlink to your current kernel 
version, with the grub config looking at the symlink - then you can 
upgrade the kernel, hit restart, and walk away.  When you come back, 
your machine is ready (assuming the kernel isn't messed up).

--
Daniel
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[William Ballard]

On Tue, Oct 05, 2004 at 05:05:22PM +0100, Daniel Goldsmith wrote:
> pace the whole GMail issue - I use it for mailing lists, for pretty
> much the same reasons to do with wastage. I simply cannot afford the
> bandwidth to download much of the crud which populates mailing-lists -
> Google Inc. can. They also provide a means for me to scan and search
> the lists, looking for what I want or need.

Congratulations, you've managed to reinvent Usenet.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Adding folded headers in Exim 4.

[Adam Funk]

I want to add a folded, multi-line header (RFC2822, Sec.2.2.3) like
this:

X-Comment: foo foo foo
   bar bar bar

using the headers_add option.  I've experimented with it a little and I
think this works:

   headers_add = X-Comment: foo foo foo\n   bar bar bar

and even this:

   headers_add = X-Comment: foo foo foo\n   bar \
bar bar bar\n  foobar \
foobar

for improved readability of very long comments.  But before I leave it
in my config file, I'd like to know if anyone else has tried these,
successfully or not.

Thanks,
Adam


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

set default paper size in firefox?

[Matt Price]

Hi,

the default paper size seems to be set to "A4" in my version of
firefox (0.9.3).  Anyone know how to change that setting?  I often
forget to reset the print option and end up with slightly-wrong page
lengths...  

I have tried messing with "prefs.js" but haven'th ad any luck.  
e.g. I have this in my prefs.js, which I wouldh ave thought would
work:  

user_pref("print.printer_PostScript/default.print_command", "lpr -o number-up=1 
${MOZ_PRINTER_NAME:+'-P'}${MOZ_PRINTER_NAME}"); 
user_pref("print.printer_PostScript/default.print_edge_bottom", 4); 
user_pref("print.printer_PostScript/default.print_edge_left", 4); 
user_pref("print.printer_PostScript/default.print_edge_right", 4); 
user_pref("print.printer_PostScript/default.print_edge_top", 4); 
user_pref("print.printer_PostScript/default.print_in_color", false); 
user_pref("print.printer_PostScript/default.print_paper_height", " 10.98"); 
user_pref("print.printer_PostScript/default.print_paper_name", "Letter"); 
user_pref("print.printer_PostScript/default.print_paper_size_type", 1); 
user_pref("print.printer_PostScript/default.print_paper_size_unit", 0); 
user_pref("print.printer_PostScript/default.print_paper_width", "  8.46"); 
user_pref("print.printer_gimp@:64.print_command", "lpr 
${MOZ_PRINTER_NAME:+'-P'}${MOZ_PRINTER_NAME}"); 
user_pref("print.printer_gimp@:64.print_edge_bottom", 4); 
user_pref("print.printer_gimp@:64.print_edge_left", 4); 
user_pref("print.printer_gimp@:64.print_edge_right", 4); 
user_pref("print.printer_gimp@:64.print_edge_top", 4); 
user_pref("print.printer_gimp@:64.print_in_color", true); 
user_pref("print.printer_gimp@:64.print_paper_height", "279.00"); 
user_pref("print.printer_gimp@:64.print_paper_name", "na-letter"); 
user_pref("print.printer_gimp@:64.print_paper_size_type", 1); 
user_pref("print.printer_gimp@:64.print_paper_size_unit", 1); 
user_pref("print.printer_gimp@:64.print_paper_width", "215.00"); 


anyway, appreciate the help.
m


--
 .''`.   Matt Price 
: :'  :  Debian User
`. `'`   & hemi-geek
  `- 
-- 
don't use these addresses:  [EMAIL PROTECTED], [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why Grub? Must I Switch?

[Andrea Vettorello]

On Tue, 05 Oct 2004 09:08:45 -0700, Daniel L. Miller <[EMAIL PROTECTED]> wrote:
> cr wrote:
> 
> >On Mon, 04 Oct 2004 18:03, Daniel L. Miller wrote:
> >
> >
> >>Paul Johnson wrote:
> >>
> >>
> >>><#secure method=pgp mode=sign>
> >>>-BEGIN PGP SIGNED MESSAGE-
> >>>Hash: SHA1
> >>>
> >>>"Daniel L. Miller" <[EMAIL PROTECTED]> writes:
> >>>
> >>>
> I don't understand that - unless your different kernel versions are
> all using the same filename.
> 
> 
> >>>You have a basic shell.  You can type the name, and IIRC, there is tab
> >>>autocompletion.
> >>>
> >>>
> >>So you can't have an unattended boot?
> >>
> >>Daniel
> >>
> >>
> >
> >You can certainly have an unattended boot - the GRUB menu comes up and, after
> >a specified interval, it will go on to boot the first entry on the menu
> >(unless, of course, you manually select another).
> >
> >However, if you've just done a system upgrade or destroyed a kernel or
> >whatever, or are just Grub-floppy-booting into a system, then you can bring
> >up the  Grub command line which has, as Paul said, tab autocompletion - and
> >very handy it is too for finding a kernel when you can't remember whether
> >it's  vmlinuz2.4.18-bf2.4 or   vmlinuz2.4.18_bf2.4
> >
> >cr
> >
> >
> That's not my idea of an unattended boot - which is where the comments
> on symlinks came from.  If you have a symlink to your current kernel
> version, with the grub config looking at the symlink - then you can
> upgrade the kernel, hit restart, and walk away.  When you come back,
> your machine is ready (assuming the kernel isn't messed up).
> 

You can do this with Grub and Lilo too, you can use some hook to
launch a command at installation time, IIRC modifying the
/etc/kernel-img.conf file. If you use grub add the line "postinst_hook
= /sbin/update-grub" and "postrm_hook = /sbin/update-grub". I'm not
sure, but this should be explained in the kernel-package
documentation...


Andrea


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Software RAID 0 and linux 2.6.8.1

[Paul Yeatman]

Well, definitely sounds like you know what you want :)

Seems like my confusion between you saying "software" and "hardware"
controller is, as Paolo was saying, the Promise FastTrak isn't a "true"
hardware controller apparently.  It still requires some software, ie.
kernel module, on the software side to operate.  This is, however,
coNSIDerably faster than using strictly software RAID where no card is
needed at all.  From the description of your system, however, this
wouldn't appear to be a problem.

I'll opt out of the conversation at this point because your
proposed setup is different enough from mine that I can't really
helpfully comment on it.  As I mentioned (and I believe even
tried), the "partial source code" from Promise for the FastTrak
doesn't compile (or at least isn't written to compile) for the 2.6
kernel.  If the drivers you know of are slow, maybe Paolo is right and
you should go with either strictly a hardware RAID controller or
strictly software RAID given that you have a slick system.

As for the boot loader issue, I'm not sure how a large BIOS is a
solution unless you can get your boot loader to use your BIOS which
I've never heard of.  Unless you are using a true hardware controller,
I'm not sure where/how your boot loader will be read with no other
disks other than the RAID 0'ed disks.

Best wishes,

Paul

->>In response to your message<<-
  --received from Niels--
>
> I shall explain my choises. I really want to buy a RAID 0 (striping) 
> controller, because my current harddisk is the bottleneck in my current 
> system. I have a dual XP-M 2800+ with 1Gbit NIC and 1GB memory. I have 
> read enough about RAID 0. I really want it! I know that it is not safe, 
> but I can backup my data.
> 
> I choose for a software RAID controller because it is cheap. An hardware 
> controller would cost me six times more, but would be at most 10% 
> faster. That's not worth it. Installing this RAID card is easy under Win 
> XP, because I can just put a disk in my pc with the drivers at setup 
> time and it will work.
> 
> I know this will be another story for debian. I've read stories of 
> people who got the Promise drivers working under 2.6.8, but those 
> drivers are not fast. I use 2.6.8 because I have a SMP setup. I want no 
> other drives than the two of my RAID setup in my pc because I hate noise.
> 
> I buy a software RAID controller because the price difference between a 
> software RAID SATA controller and a SATA controller is < €10,-
> 
> What are the troubles I will run into? I can load things in my BIOS, 
> because I have 2MB BIOS memory (MPX2).
> 
> The adress you gave me was wrong, 
> http://www.1U-Raid5.net/HowTO/SW-Raid-HOWTO.txt this one is correct.
> 
> Alvin Oga wrote:
> 
> >On Mon, 4 Oct 2004, Niels wrote:
> >
> > 
> >
> >>I want to buy a Promise Fasttrack TX2 plus RAID 0 controller with two 
> >>SATA hard disks. The controller is a software controller. I want no 
> >>other hard disks than these two in my PC. Can I boot of this setup with 
> >>the 2.6 kernel?
> >>   
> >>
> >
> >yes...
> >
> > 
> >
> >>Is it possible to put Windows XP and Debian Linux on the 
> >>same RAID 0 array?
> >>   
> >>
> >
> >yes
> >
> > 
> >
> >>Is it difficult to install?
> >>   
> >>
> >
> >no
> >
> >-- trick is to install windoze on the 1st partition so it thinks its C:
> >
> >--- bigger problem ...
> >
> >- why use 2 disks in raid0 ( aks stripping ) which might allow you to read
> > faster but also makes 2 disks look like 1 bigger disk
> > - if one disk dies... both disk are useless
> >
> >- i think you want to use (sw or hw ) raid1 ( mirror )
> > - if one disk dies ... all your data is still intact on the other
> > disk
> >
> >- gazillion raid howtos urls
> > http://1u_raid5.net/HowTo/SW-Raid-HOWTO.txt
> >
> >
> >c ya
> >alvin
> >
> >
> > 
> >
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact 
> [EMAIL PROTECTED]
> 

-- 
Paul Yeatman   (858) 534-9896[EMAIL PROTECTED]
 ==
 ==Proudly brought to you by Mutt==
 ==


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: set default paper size in firefox?

[Martin Dickopp]

Matt Price <[EMAIL PROTECTED]> writes:

> the default paper size seems to be set to "A4" in my version of
> firefox (0.9.3).  Anyone know how to change that setting?  I often
> forget to reset the print option and end up with slightly-wrong page
> lengths...  
>
> I have tried messing with "prefs.js" but haven'th ad any luck.  

In might depend on which printing method you use (postscript or xprint).
I use postscript printing, and for me, the following line in "user.js"
(or "prefs.js" or the "about:config" page...) works:

user_pref("print.postscript.paper_size", "A4");

(Of course, this sets the paper size *to* A4, so you should replace "A4"
with whatever paper size you want to use.)

Martin


-- 
   ,--.  ,= ,-_-. =.
  / ,- )Martin Dickopp, Dresden, Germany((_/)o o(\_))
  \ `-'http://www.zero-based.org/`-'(. .)`-'
   `-.   \_/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Postgresql upgrade!!!!

[D S]

Does anybody knows how to upgrade postgresql on Debian manually or by apt-get?
The debian security web site has only the old 7.2 version.
I tried to load it form other http sources but it can't find it.
I need upgrade it to version 7.4.
Does anybody have a init script that can run Postgresql 7.4 on Debian?
 
Thank you
Dmitry
		Do you Yahoo!?vote.yahoo.com - Register online to vote today!

Re: Postgresql upgrade!!!!

[Andrea Vettorello]

On Tue, 5 Oct 2004 10:44:48 -0700 (PDT), D S <[EMAIL PROTECTED]> wrote:
> 
> Does anybody knows how to upgrade postgresql on Debian manually or by
> apt-get? 
> The debian security web site has only the old 7.2 version. 
> I tried to load it form other http sources but it can't find it. 
> I need upgrade it to version 7.4. 
> Does anybody have a init script that can run Postgresql 7.4 on Debian? 
>   

If you are referring to the CAN-2004-0547 advisory (postgresql --
buffer overflow) the version in Woody (postgresql_7.2.1-2woody5) is
patched...


Andrea


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

How to change default partition scheme in chroot sid

[Henry Lenzi]

Hello --

 I am currently running Woody and I created a chrooted sid partition.
 Here's my current partition scheme for sid:

Filesystem   1k-blocks  Used Available Use% Mounted on
/dev/hda2  4806080   4711892 0 100% /
/dev/hda5  3844584119880   3529408   4% /var
/dev/hda4  2691228862848   1691672  34% /root
/dev/hda1   189339  1333178230   1% /boot
/dev/hda6  3844584   2486176   1163112  69% /usr
/dev/hda7  3844584   2864268785020  79% /home

 As you can see, I am already facing some space problems.
 I did not chose this partition scheme, it was the default scheme.
 I would like to know if there's a way to change this default when you're
installing sid. Let me rephrase that: how do I install chrooted sid with an
arbitrary partition scheme and what tools do I use for that?
 The documentation I folllowed in order to install a chrooted sid was:
 
 http://javierlinares.com/debian/docs/chroot/chroot.pdf

 Thank you.

 Cheers,

 Henry 






-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: question re: removing all traces of Windows ME OS

[Rich Rudnick]

On Mon, 2004-10-04 at 20:42 -0500, Jeff Golden wrote:
> > 
> > A basic rule of life is never burn bridges behind you. Retreat is often
> > a tactical necessity.
> > 
> 
> This is true but it is rather annoying to have to re-boot into windows
> to do simple things like banking or dial into your office.  My
> solution to the same problem is to utilize Win4Lin (commercial but it
> saves me time).   It allows me to do the minimal things I need to do
> for work and banking quite easily.  Also allows me to run Rhapsody
> music service.  Even supports my VPN client (nortel) and PC Anywhere. 
>  I would suggest having a look at that.
> 
> --Jeff
> 

I thought about Win4Lin back when I was converting our office to
staroffice 5.2 and linux (the advantage of being the CEO, CFO and CTO :)
Instead I networked an extra PC and rolled my chair back and forth for a
few months. Nowadays I generally run windows once a month, and for less
than 30 minutes, so Win4Lin would be overkill. 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[calvin]

On Tue, Oct 05, 2004 at 08:24:10AM -0500, Rodney Richison wrote:
>
>
>>Top posting leads to a big mess. Every non-technical person does it
>>and it ends up with them saying "let's use a forum or something"
>>because they can't follow the thread any longer. Surprise surprise, a
>>forum works with bottom posting and everybody seems to find that
>>normal.
>>
>> 
>>
>I run a list for techs/consultants.  (ChannelVar.com)
>They are a VERY qualified bunch of guys. (Admittedly not linux users 
>though)  :)

So qualified that the site is not even up...


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

GDM puts a semi-persistent "X" in the middle of the screen.

[Adam Funk]

When I log out of xfce4, the gdm login box appears normally.  Not long
afterwards, an "X" (like the multiplication symbol rather than the
letter) appears right in the middle of the screen.  When I log back in,
it stays on the screen until I run kmail or knode, either of which
makes it disappear.

I looked through the gdm bugreport page but didn't see anything.  I
suspect another package could cause the problem but I have no idea
where else to look.  Has anyone else noticed this?

Background: I currently have the following packages installed, although
this problem has persisted through upgrades for several months.
   gdm2.4.4.7-3
   kmail  3.3.0-3
   knode  3.2.2-2
   xfce4  4.0.5-1

Thanks,
Adam


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Nvidia + 2.6 kernel

[Noah Durell]

Hello,
I was hoping to try out a 2.6 kernel.  When I installed it (using
apt-get) I couldn't get the nvidia drivers to install (using
module-assisstant) nor by doing a manual install as per the
instructions of this webpage:
http://home.comcast.net/~andrex/Debian-nVidia/installation.html.
My computer is a dell dimension 8250 with a GeForce4 MX 420 video card.
Thanks,
Noah Durell


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Installing from net with floppy disks - system don´t recognize my hard disk

[Juan Stiller]

Hi, im trying to install Debian Unstable Version from
internet with the floppy disks, the installer can see
my ethernet card, and beggins downloadin some
packages, but when ends the installed tell me that i
have no removable media, i know the hdd works fine,
but i can´t seem to use it with this installer.

Any hints?





___
100mb gratis, Antivirus y Antispam
Correo Yahoo!, el mejor correo web del mundo
http://correo.yahoo.com.ar


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[Nate Duehr]

Rodney Richison wrote:

Top posting leads to a big mess. Every non-technical person does it
and it ends up with them saying "let's use a forum or something"
because they can't follow the thread any longer. Surprise surprise, a
forum works with bottom posting and everybody seems to find that
normal.
 

I run a list for techs/consultants.  (ChannelVar.com)
They are a VERY qualified bunch of guys. (Admittedly not linux users 
though)  :)
The preferred method is top posting.
The preferred format is to allow html.


Consultants always do things the hard way.  It results in more billable 
hours.

Let's see them read a newspaper that's been printed "top-posted".  It's 
retarded.

HTML in e-mail well, that's a completely different flamefest, but 
suffice it to say it's rare that HTML formatting can add any information 
to the message being conveyed.  There are exceptions, but for 99% of 
e-mail, it's not necessary. 

Just like most consultants.
Nate
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Nvidia + 2.6 kernel

[Alexander Schmehl]

* Noah Durell <[EMAIL PROTECTED]> [041005 20:49]:

> I was hoping to try out a 2.6 kernel.  When I installed it (using
> apt-get) I couldn't get the nvidia drivers to install (using
> module-assisstant) nor by doing a manual install as per the
> instructions of this webpage:
> http://home.comcast.net/~andrex/Debian-nVidia/installation.html.
> My computer is a dell dimension 8250 with a GeForce4 MX 420 video card.

Please read http://www.catb.org/~esr/faqs/smart-questions.html


Yours sincerely,
  Alexander


PS: Would be very usefull, if you could us, what went wrong and why you
couldn't install it.


signature.asc
Description: Digital signature

"linmodem" works but hangs up when idle

[Eric Dickner]

I got a BCMSM 4212 to work on Debian 2.2.20-idepci
with the driver .rpm from Dell but it seems to hang up
after a minute or so of idle time.  I used "pppconfig"
to set no hang-up at all for idle time, and it seems
to be sort of a random period, ranging from 50 seconds
to two minutes.

If it's doing something with the connection, like a
download, it works like a champ, but reading webpages
means a redial everytime I pause to take the content
in.

Anyone know what this might be caused by?



___
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: set default paper size in firefox?

[Matt Price]

Martin Dickopp wrote:
Matt Price <[EMAIL PROTECTED]> writes:

the default paper size seems to be set to "A4" in my version of
firefox (0.9.3).  Anyone know how to change that setting?  I often
forget to reset the print option and end up with slightly-wrong page
lengths...  

I have tried messing with "prefs.js" but haven'th ad any luck.  

In might depend on which printing method you use (postscript or xprint).
I use postscript printing, and for me, the following line in "user.js"
(or "prefs.js" or the "about:config" page...) works:
user_pref("print.postscript.paper_size", "A4");
hmm... so this had already been set to "Letter" -- I tried "na-letter" 
as well and neither seems to have any effect.  I don't know whether I 
use xprint or postscript -- I don't remember messing with the print 
system, so I expect firefox goes through CUPS to getthe printer names. 
Is this the default behaviour?  Any other suggestions on where to set 
this value:
thx,
matt


(Of course, this sets the paper size *to* A4, so you should replace "A4"
with whatever paper size you want to use.)
Martin


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[Clive Menzies]

On (05/10/04 12:52), Nate Duehr wrote:
> Rodney Richison wrote:
> 
> >
> >
> >>Top posting leads to a big mess. Every non-technical person does it
> >>and it ends up with them saying "let's use a forum or something"
> >>because they can't follow the thread any longer. Surprise surprise, a
> >>forum works with bottom posting and everybody seems to find that
> >>normal.
> >> 
> >>
> >I run a list for techs/consultants.  (ChannelVar.com)
> >They are a VERY qualified bunch of guys. (Admittedly not linux users 
> >though)  :)
> >The preferred method is top posting.
> >The preferred format is to allow html.
> >
> >
> Consultants always do things the hard way.  It results in more billable 
> hours.
> 
> Let's see them read a newspaper that's been printed "top-posted".  It's 
> retarded.
> 
> HTML in e-mail well, that's a completely different flamefest, but 
> suffice it to say it's rare that HTML formatting can add any information 
> to the message being conveyed.  There are exceptions, but for 99% of 
> e-mail, it's not necessary. 
> 
> Just like most consultants.
Ouch !

;-)

Regards

Clive


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: GDM puts a semi-persistent "X" in the middle of the screen.

[Andrea Vettorello]

On Tue, 05 Oct 2004 18:28:29 GMT, Adam Funk <[EMAIL PROTECTED]> wrote:
> When I log out of xfce4, the gdm login box appears normally.  Not long
> afterwards, an "X" (like the multiplication symbol rather than the
> letter) appears right in the middle of the screen.  When I log back in,
> it stays on the screen until I run kmail or knode, either of which
> makes it disappear.
> 
> I looked through the gdm bugreport page but didn't see anything.  I
> suspect another package could cause the problem but I have no idea
> where else to look.  Has anyone else noticed this?
> 
> Background: I currently have the following packages installed, although
> this problem has persisted through upgrades for several months.
>gdm2.4.4.7-3
>kmail  3.3.0-3
>knode  3.2.2-2
>xfce4  4.0.5-1
> 

The X you see is probably the default xfree86 mouse pointer. When the
X appear, can you still see, and move, the usual arrow mouse pointer?


Andrea


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing from net with floppy disks - syst em don´t recognize my hard disk

[Andrea Vettorello]

On Tue, 5 Oct 2004 15:51:10 -0300 (ART), Juan Stiller
<[EMAIL PROTECTED]> wrote:
> Hi, im trying to install Debian Unstable Version from
> internet with the floppy disks, the installer can see
> my ethernet card, and beggins downloadin some
> packages, but when ends the installed tell me that i
> have no removable media, i know the hdd works fine,
> but i can´t seem to use it with this installer.
> 

I'm going to do a little nitpick here, but doesn't exist a Debian
installer for the unstable (Sid) branch, probably you're using one of
the testing release candidate installer. =)

Your HD use some exotic RAID or SCSI controller?


Andrea

Re: Radius + wireless access

[Alvin Oga]

hi ya

On Tue, 5 Oct 2004, Gabriel Granger wrote:

> I've got a couple of wireless AP around the office, and would like to 
> mange the laptop with wireless cards using Radius for authentication.  
> all of my AP have the ability to talk to a Radius server.  I have got 
> my radius server up and working using some simple test accounts, I've 
> done some looking around everything seems to point to EAP.  I might be 
> wrong, but I thought EAP requires cert to be issued to the client 
> machines.  I want a simple authentication system, so that office 
> workers can gain access anywhere in the building but also so that 
> trusted clients can gain access too with out having to had out 
> certs(just give them a username password)
> 
> Am I thinking about this the wrong way? is it possible to just 
> wirelessly authenticate users based on a username and password like you 
> can with dialup users?

from what i seen ... depending on the wireless setup

- once a wl card finds an ap and associates with it ...
- it can go live on the wire unless there's soemthing else
blocking its connections
( live on the wire is surf the web w/o any login info )

- assumes that you have dhcp runnng as most people do
( a extreme bad idea for the above reason )

- if the association to the AP is step 1, and you need authentication
  to go live, means that you turn on ssh and a firewall to allow
  that new host to go live
- radius adds more complications .. too much work 

- ssh login is good enough
- i use static ip#  ... i like to know who it is

- once ssh presents a login: screen ..
than it can updates the firewall rules or even a web-based
login 

- remember... they can surf the web or see your wl gateway 
webserver without logging in unless:
- you prevent that by not using dhcp 
- you prevent it by using a dynamic firewall
- you prevent it by a login requirement first

- they are able to see all packets ... 

- use a linux based AP ... so you have the flexibility
and control vs the generic AP off the shelf which allows
anybody to do anything via the web gui that most people
never change ( 1/2 the people, aka average consumers )


- if you want eap ... it implies you are using WPA for the
  wireless encrypted packets instead of WEP encrypted packets

- too much work ... and too many incompatibilities

- use a netgear AP w/ wpa with a netgear pci cards
and do not mix manufacturers unless you like being
their testing and qa lab for them
same for linksys, dlink, cisco, ...

- wep or WPA makes no difference ...
- always use ssh ... so if they do break your wireless
encryption, all they sse is oyur ssh encrypted traffic

- always use secure pop3, secure imap, scp(not ftpp)
... on and on ..

for more wireless fun ... 
www.linux-wireless.org

c ya
alvin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Nvidia + 2.6 kernel

[Noah Durell]

Sorry for the lack of details.  I couldn't find the log file from the
previous install.  Anyway I tried installing the nvidia drivers again
by entering the command:

module-assistant auto-install nvidia

I get a warning about the rivafb module and it says there is no rule
to make target kernel-headers-2.6.8-1-686.  Let me know if you need
more information.
-Noah
On Tue, 5 Oct 2004 20:59:57 +0200, Alexander Schmehl
<[EMAIL PROTECTED]> wrote:
> * Noah Durell <[EMAIL PROTECTED]> [041005 20:49]:
> 
> 
> 
> > I was hoping to try out a 2.6 kernel.  When I installed it (using
> > apt-get) I couldn't get the nvidia drivers to install (using
> > module-assisstant) nor by doing a manual install as per the
> > instructions of this webpage:
> > http://home.comcast.net/~andrex/Debian-nVidia/installation.html.
> > My computer is a dell dimension 8250 with a GeForce4 MX 420 video card.
> 
> Please read http://www.catb.org/~esr/faqs/smart-questions.html
> 
> Yours sincerely,
>   Alexander
> 
> PS: Would be very usefull, if you could us, what went wrong and why you
> couldn't install it.
> 
> 
>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing from net with floppy disks - syst em don´t recognize my hard disk

[Juan Stiller]

 --- Andrea Vettorello <[EMAIL PROTECTED]>
escribió: 
> I'm going to do a little nitpick here, but doesn't
> exist a Debian
> installer for the unstable (Sid) branch, probably
> you're using one of
> the testing release candidate installer. =)
> 
> Your HD use some exotic RAID or SCSI controller?
> 
> 
> Andrea
> 
>  
Thanks for responding

im quite new to debian so i search on debian.org for
the floppy images and make the disks.

No, i got a pc with only one ide hd setted as primary
master, no raid or scsi drives.





___ 
¡Llevate a Yahoo! en tu Unifón! 
Ahora podés usar Yahoo! Messenger en tu Unifón, en cualquier momento y lugar. 
Encontrá más información en: http://ar.mobile.yahoo.com/sms.html 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Headless Installation

[Richard Lyons]

On Tuesday 05 October 2004 02:31, Adi Linden wrote:
> > You -could- (I stress could here) use 'fai', but I suspect that you'd
> > consider that overkill on you part. Indeed, I'd agree. What you'd 
really
> > want to do is use 'deboostrap' on the headless server. There's plenty 
of
> > documentation on how to do this.
> 
> To use 'deboostrap' I already need a working Unix/Linux system (reading 
> ). I 
am 
> looking for something more along the line of a regular installation CD 
> that is non-interactive but installs a bare Debian system automatically.

Wouldn't that be possible using a live cd (mepis/knoppix/etc)?

I've not done this myself, so it's only a thought...

-- 
richard


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Nvidia + 2.6 kernel

[Noah Durell]

Here is log file from the module-assistant.

Build log starting, file:
/var/cache/modass/nvidia-kernel-source.buildlog.2.6.8-1-686.1097003781
Date: Tue, 05 Oct 2004 15:16:22 -0400

echo "ROOT_CMD = "
ROOT_CMD = 
/usr/bin/make  -f debian/rules binary_modules
make[1]: Entering directory `/usr/src/modules/nvidia-kernel'
# select which makefile to use.
rm -f /usr/src/modules/nvidia-kernel/nv/Makefile || true
if [ 6 = 6  ]; then \
 cd /usr/src/modules/nvidia-kernel/nv ; \
 ln -s Makefile.kbuild Makefile ; \
 cd .. ; \
 if [ 0 = 1 ]; then \
dpatch apply 01_sysfs ; \
dpatch status 01_sysfs >patch-stamp ; \
dpatch apply 02_pcialias ; \
dpatch status 02_pcialias >>patch-stamp ; \
 fi \
fi
if [  6 = 4  ]; then \
 cd /usr/src/modules/nvidia-kernel/nv ; \
 ln -s Makefile.nvidia Makefile ; \
 cd .. ; \
fi
#nothing here anymore
touch configure-stamp
if [ -f /usr/src/modules/nvidia-kernel/debian/control.template ]; then \
cp  /usr/src/modules/nvidia-kernel/debian/control.template
/usr/src/modules/nvidia-kernel/debian/control; \
fi
dh_testdir
dh_testroot
PATCHLEVEL = 6 
Kernel compiler version : 3.3.4
Detected compiler version : 3.3.4
Using compiler gcc-3.3 version 3.3.4
touch /usr/src/modules/nvidia-kernel/nv/gcc-check
touch /usr/src/modules/nvidia-kernel/nv/cc-sanity-check
## Main Make ##
IGNORE_CC_MISMATCH=1 CC="gcc-3.3"  /usr/bin/make -C
/usr/src/modules/nvidia-kernel/nv -f Makefile SYSSRC=/usr/src/linux  
KBUILD_PARAMS="-C /usr/src/linux
SUBDIRS=/usr/src/modules/nvidia-kernel/nv" module;
make[2]: Entering directory `/usr/src/modules/nvidia-kernel/nv'
Your kernel was configured to include rivafb support as
a loadable kernel module.

The rivafb driver conflicts with the NVIDIA driver; the
NVIDIA kernel module will still be built and installed,
but be aware that the NVIDIA driver will not be able to
function properly if the rivafb module is loaded!

*** Failed rivafb module sanity check, but continuing! ***


NVIDIA: calling KBUILD...
make[3]: Entering directory `/usr/src/kernel-headers-2.6.8-1-686'
Makefile:434: /usr/src/kernel-headers-2.6.8-1-686/arch/linux/Makefile:
No such file or directory
make[3]: *** No rule to make target
`/usr/src/kernel-headers-2.6.8-1-686/arch/linux/Makefile'.  Stop.
make[3]: Leaving directory `/usr/src/kernel-headers-2.6.8-1-686'
NVIDIA: left KBUILD.
nvidia.ko failed to build!
make[2]: *** [module] Error 1
make[2]: Leaving directory `/usr/src/modules/nvidia-kernel/nv'
make[1]: *** [build-stamp] Error 2
make[1]: Leaving directory `/usr/src/modules/nvidia-kernel'
make: *** [kdist_image] Error 2

Thanks,
Noah

On Tue, 5 Oct 2004 20:59:57 +0200, Alexander Schmehl
<[EMAIL PROTECTED]> wrote:
> * Noah Durell <[EMAIL PROTECTED]> [041005 20:49]:
> 
> 
> 
> > I was hoping to try out a 2.6 kernel.  When I installed it (using
> > apt-get) I couldn't get the nvidia drivers to install (using
> > module-assisstant) nor by doing a manual install as per the
> > instructions of this webpage:
> > http://home.comcast.net/~andrex/Debian-nVidia/installation.html.
> > My computer is a dell dimension 8250 with a GeForce4 MX 420 video card.
> 
> Please read http://www.catb.org/~esr/faqs/smart-questions.html
> 
> Yours sincerely,
>   Alexander
> 
> PS: Would be very usefull, if you could us, what went wrong and why you
> couldn't install it.
> 
> 
>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Nvidia + 2.6 kernel

[Andrew Schulman]

> Sorry for the lack of details.  I couldn't find the log file from the
> previous install.  Anyway I tried installing the nvidia drivers again
> by entering the command:
> 
> module-assistant auto-install nvidia
> 
> I get a warning about the rivafb module and it says there is no rule
> to make target kernel-headers-2.6.8-1-686.  Let me know if you need
> more information.
> > 
> > > I was hoping to try out a 2.6 kernel.  When I installed it (using
> > > apt-get) I couldn't get the nvidia drivers to install (using
> > > module-assisstant) nor by doing a manual install as per the
> > > instructions of this webpage:
> > > http://home.comcast.net/~andrex/Debian-nVidia/installation.html.
> > > My computer is a dell dimension 8250 with a GeForce4 MX 420 video card.

The rivafb module is incompatible with the nvidia driver; see 
/usr/share/doc/nvidia-kernel-source/README.Debian.  This shouldn't be a 
problem if you never load that module, but if it does ever get loaded 
for some reason, your monitor could explode (figuratively speaking).  I 
suggest deleting /lib/modules/2.6.8-1-
686/kernel/drivers/video/riva/rivafb.ko.  Don't worry about deleting it; 
if you ever need it back, you can always reinstall your kernel image 
package.  OTOH, if you ever upgrade to a new version of that kernel, 
that file will be replaced, and you'll want to remove it again to be 
safe.

I haven't seen the error about no target to make kernel-headers-* 
before.  Normally module-assistant will find and install these for you, 
but in this case it seems to have failed.  Try running 'apt-get install 
kernel-headers-2.6.8-1-686', then rerunning module-assistant.  If you 
succeed in building the nvidia module, then you've found a bug in 
module-assistant and you should report it.

Good luck,
Andrew.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Software RAID 0 and linux 2.6.8.1

[Niels]

Promise Netherlands told me today that the kernel 2.6.x.x drivers 
probably come next year. On a dutch forum they told me (2 secs ago) that 
the name of a non-true hardware controller is a firmware controller. I 
created a picture 
http://www.xs4all.nl/~mghvl404/bin/fotos/got/RAID0uitleg.gif to explain 
what I want with the promise fasttrack TX2 plus firmware raid 
controller. How can I do this?

Paul Yeatman wrote:
Well, definitely sounds like you know what you want :)
Seems like my confusion between you saying "software" and "hardware"
controller is, as Paolo was saying, the Promise FastTrak isn't a "true"
hardware controller apparently.  It still requires some software, ie.
kernel module, on the software side to operate.  This is, however,
coNSIDerably faster than using strictly software RAID where no card is
needed at all.  From the description of your system, however, this
wouldn't appear to be a problem.
I'll opt out of the conversation at this point because your
proposed setup is different enough from mine that I can't really
helpfully comment on it.  As I mentioned (and I believe even
tried), the "partial source code" from Promise for the FastTrak
doesn't compile (or at least isn't written to compile) for the 2.6
kernel.  If the drivers you know of are slow, maybe Paolo is right and
you should go with either strictly a hardware RAID controller or
strictly software RAID given that you have a slick system.
As for the boot loader issue, I'm not sure how a large BIOS is a
solution unless you can get your boot loader to use your BIOS which
I've never heard of.  Unless you are using a true hardware controller,
I'm not sure where/how your boot loader will be read with no other
disks other than the RAID 0'ed disks.
Best wishes,
Paul
->>In response to your message<<-
 --received from Niels--
 

I shall explain my choises. I really want to buy a RAID 0 (striping) 
controller, because my current harddisk is the bottleneck in my current 
system. I have a dual XP-M 2800+ with 1Gbit NIC and 1GB memory. I have 
read enough about RAID 0. I really want it! I know that it is not safe, 
but I can backup my data.

I choose for a software RAID controller because it is cheap. An hardware 
controller would cost me six times more, but would be at most 10% 
faster. That's not worth it. Installing this RAID card is easy under Win 
XP, because I can just put a disk in my pc with the drivers at setup 
time and it will work.

I know this will be another story for debian. I've read stories of 
people who got the Promise drivers working under 2.6.8, but those 
drivers are not fast. I use 2.6.8 because I have a SMP setup. I want no 
other drives than the two of my RAID setup in my pc because I hate noise.

I buy a software RAID controller because the price difference between a 
software RAID SATA controller and a SATA controller is < ï10,-

What are the troubles I will run into? I can load things in my BIOS, 
because I have 2MB BIOS memory (MPX2).

The adress you gave me was wrong, 
http://www.1U-Raid5.net/HowTO/SW-Raid-HOWTO.txt this one is correct.

Alvin Oga wrote:
   

On Mon, 4 Oct 2004, Niels wrote:

 

I want to buy a Promise Fasttrack TX2 plus RAID 0 controller with two 
SATA hard disks. The controller is a software controller. I want no 
other hard disks than these two in my PC. Can I boot of this setup with 
the 2.6 kernel?
 

   

yes...

 

Is it possible to put Windows XP and Debian Linux on the 
same RAID 0 array?
 

   

yes

 

Is it difficult to install?
 

   

no
-- trick is to install windoze on the 1st partition so it thinks its C:
--- bigger problem ...
- why use 2 disks in raid0 ( aks stripping ) which might allow you to read
faster but also makes 2 disks look like 1 bigger disk
- if one disk dies... both disk are useless
- i think you want to use (sw or hw ) raid1 ( mirror )
- if one disk dies ... all your data is still intact on the other
disk
- gazillion raid howtos urls
http://1u_raid5.net/HowTo/SW-Raid-HOWTO.txt
c ya
alvin

 

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact 
[EMAIL PROTECTED]

   

 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Mozilla browsers and Flashplayer

[Juha Siltala]

On 2004-10-05, Freddy Freeloader <[EMAIL PROTECTED]> wrote:

> so I'll see what happens when the baseball playoffs start and I use the 

Grown-up men will loiter around, spit on the grass, and adjust their
bollocks, while giving each other really mean looks. Then occasionally
someone will try and hit a ball, misses and curses. People will applause.
Then they'll loiter, spit, and adjust their bollocks some more because
their balls are apparently very difficult to get just right. Millions of
otherwise reasonably sane people will watch that live on their TV sets for
hours and hours. You will sit at the bar and people will hiss at you when
you try to start a conversation.

Yes, I've been to the US during playoffs :)

-- 
Juha Siltala
http://www.edu.helsinki.fi/activity/people/jsiltala/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: question re: removing all traces of Windows ME OS

[Alan Chandler]

On Monday 04 October 2004 22:26, Ulisses Reina Montenegro de Albuquerque 
wrote:
> Kent West wrote:
...
> > About the only disadvantages of multiple platforms:
> >
> > * If you size them wrong to begin with (too small, and they'll fill
> > up; too large, and it's wasted disk space), it can be problematic to
> > resize them.
> >
> > * It's a little more complex to set up (plan the sizes, names, make
> > sure /etc/fstab is correct, etc).
>
> Since the new Debian installer (pre-rc2) supports boot setups using LVM,
> these are no longer a big problem. Many people do not use LVM on desktop
> systems, because they associate it with servers (and their huge storage
> requirements, with multi-disk setups and such), but LVM can offer some
> real benefits for desktop users, too.

Just to take this discussion a step further.

I use /boot with grub and the kernel images as a small 30MB partition at the 
front of the disk.  This is formatted using ext2

The root partition and all other partitions are reiserfs.  Reiserfs can be 
grown without dismounting it (assuming space),  It can be contracted by 
dismounting

All other partitions other than root I use as Logical Partitions within an LVM 
volume group.  This way you can pretty well grow or shrink these at will (and 
create new ones if there is a new subdirectory with a lot of data to add, or 
remove them when done with them)


-- 
Alan Chandler
[EMAIL PROTECTED]
First they ignore you, then they laugh at you,
 then they fight you, then you win. --Gandhi


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Sound recording in ALSA

[Kirk Strauser]

I'm using Sid with Debian's 2.6.8-1-686 kernel package on a Dell PC with an 
Intel 8x0 sound device and ALSA.  My ultimate goal is to get liveice to 
read from the line in on my sound card, but I'm trying to take baby steps 
to get there.

Right now I'm attempting to record by:

   1) Going into KMix and selecting "Line" as the capture device (verified 
with amixer).
   2) Using something like "arecord -t wav -d 10 /tmp/foo.wav" to attempt to 
record the sound that's being played through my speakers to a file.

However, that file seems to be silent on playback, regardless of what 
permutations of the steps above I've messed with.  Any pointers on how I 
can get started with this?
-- 
Kirk Strauser



pgpe5wgwI6czz.pgp
Description: PGP signature

Re: Software RAID 0 and linux 2.6.8.1

[Alvin Oga]

hi ya

On Tue, 5 Oct 2004, Niels wrote:

> Promise Netherlands told me today that the kernel 2.6.x.x drivers 
> probably come next year. On a dutch forum they told me (2 secs ago) that 
> the name of a non-true hardware controller is a firmware controller.

dependin on your ideas/view...
hw raid from promise is okay... its cheap .. but.. ymmv

but, none of the onboard ( motherboard-based bios) raid stuff
works
( "works" is you unply any disk at anytime and you do not need
( to do anything to recover the system and no data loss

hardware raid controllers

http://1U-Raid5.net/HW

3ware is okay ..but is $300 - $500+ depending on which one

( for that price you can have redundant disks, redundant
( motherboard, redundant cpu, redundant power supply, 
( redundant nic, redundant fan ...

> >Seems like my confusion between you saying "software" and "hardware"
> >controller is, as Paolo was saying, the Promise FastTrak isn't a "true"
> >hardware controller apparently.

all hardware raid controllers has its own microcontroller and firmware

>  It still requires some software, ie.
> >kernel module, on the software side to operate.

all hardware controllers ( raid or not ) need a linux driver 

>  This is, however,
> >coNSIDerably faster than using strictly software RAID where no card is
> >needed at all.  From the description of your system, however, this
> >wouldn't appear to be a problem.

whether hw or sw raid is faster would depend on your test setup
and your software you use

"test sw" is not the same as real world applications doing its normal
workload  w/ random "do this now for me cpu interrupts"

> >>I shall explain my choises. I really want to buy a RAID 0 (striping) 
> >>controller, because my current harddisk is the bottleneck in my current 
> >>system. I have a dual XP-M 2800+ with 1Gbit NIC and 1GB memory. I have 
> >>read enough about RAID 0. I really want it! I know that it is not safe, 
> >>but I can backup my data.

if the disk is a bottle neck...

- use one ide disk per cable 
=

- NEVER ever put cdrom ( ata-33 ) on the same ide cable as a 
disk drive ( ata-100, ata-133 )

- use disk with 8MB buffers  instead of the old fashion 2mb

- some botherboards can sustain 40MByte/sec disk transfers...
while the same disk with different motherboard can sustain
a measly 20MB/sec 

( tar cf sourceDisk | ( /mnt/targetDisk; tar xvfp - )

where both disk on the same motherboard

( remember the disks are marketed at 100Mb/sec or 133Mbyte/sec
( or 320MB/sec for scsi... 

> >>I buy a software RAID controller because the price difference between a 
> >>software RAID SATA controller and a SATA controller is < �10,-

you do NOT need to buy anything for sw raid ...

- its all in the /etc/raidtab file and sw raid is in the linux
kernel

you might need to buy a $10 or $20 generic IDE controller so you can have
one disk per ide cable ...

- one disk per ide cable is critical 

> >>The adress you gave me was wrong, 
> >>http://www.1U-Raid5.net/HowTO/SW-Raid-HOWTO.txt this one is correct.

oopps...  thanx for the _ vs -

c ya
alvin

Samba

[Peter Quackenbush]

I'm having some problems working with Samba.  Here is by /etc/samba/smb.conf

# Samba config file created using SWAT
# from 192.168.0.100 (192.168.0.100)
# Date: 2004/10/04 01:03:28

# Global parameters
[global]
workgroup = MSHOME
server string = %h server (Samba %v)
passdb backend = tdbsam, guest
passwd program = /usr/bin/passwd %u
passwd chat = *Enter\snew\sUNIX\spassword:* %n\n
*Retype\snew\sUNIX\spassword:* %n\n .
syslog = 0
log file = /var/log/samba/log.%m
max log size = 1000
dns proxy = No
panic action = /usr/share/samba/panic-action %d
invalid users = root
unix password sync = yes
dns proxy = no
ldap ssl = no
[homes]
comment = Home Directories
read only = no
create mask = 0700
directory mask = 0700
browseable = No

[printers]
comment = All Printers
path = /tmp
create mask = 0700
printable = Yes
browseable = No

[print$]
comment = Printer Drivers
path = /var/lib/samba/printers

[lp]
comment = Samsung ML-1430 Series (SPL)
path = /tmp
read only = No
create mask = 0700
guest ok = Yes
printable = Yes
printer name = lp
use client driver = Yes
oplocks = No
share modes = No


Here is tail /var/log/samba/log.nmbd

  Samba name server QUACK-ATTACK has stopped being a local master
browser for workgroup MSHOME on subnet 192.168.0.2

  *
[2004/10/05 16:49:25, 0] nmbd/nmbd_become_lmb.c:become_local_master_stage2(396)
  *

  Samba name server QUACK-ATTACK is now a local master browser for
workgroup MSHOME on subnet 192.168.0.2

  *

And log.smbd
[2004/10/04 00:59:57, 0] smbd/server.c:main(760)
  smbd version 3.0.7-Debian started.
  Copyright Andrew Tridgell and the Samba Team 1992-2004
[2004/10/04 01:04:37, 0] smbd/server.c:main(760)
  smbd version 3.0.7-Debian started.
  Copyright Andrew Tridgell and the Samba Team 1992-2004
[2004/10/05 07:26:24, 0] smbd/server.c:main(760)
  smbd version 3.0.7-Debian started.
  Copyright Andrew Tridgell and the Samba Team 1992-2004


And lastly log.winbindd
[2004/10/04 01:04:45, 1] nsswitch/winbindd_util.c:init_domain_list(327)
  Could not fetch sid for our domain MSHOME
[2004/10/05 07:26:29, 1] nsswitch/winbindd.c:main(854)
  winbindd version 3.0.7-Debian started.
  Copyright The Samba Team 2000-2004
[2004/10/05 07:26:29, 0] nsswitch/winbindd_util.c:winbindd_param_init(560)
  winbindd: idmap uid range missing or invalid
[2004/10/05 07:26:29, 0] nsswitch/winbindd_util.c:winbindd_param_init(561)
  winbindd: cannot continue, exiting.
[2004/10/05 07:26:29, 1] nsswitch/winbindd.c:main(887)
  Could not init idmap -- netlogon proxy only
[2004/10/05 07:26:30, 1] nsswitch/winbindd_util.c:init_domain_list(327)
  Could not fetch sid for our domain MSHOME

Here's the setup.
I have 3 people with 3 windows xp sp2 computers trying to connect to
one server, called quack-attack.  Server is running debian sarge. 
That's samba 3.0.7-1, or whatever the most

Here's the problems. 
1.  Incosistent connections.  I often get the error "\\quack-attack is
not accessible.  You might not have permission to use this network
resource. Contact the administrator of this server to find out if you
have permissions.  Access is denied."  when trying to connect to
quack-attack.  Also note that sometimes, when quack-attack is the smb
network master, I'll get the error as well.  Just with MSHOME, and not
quack-attack.


2.  Printing is slow.  If I try to print a document when problem #1 is
not occuring (about 25% of the time), it a long time to send.  I can
see my network card send away data and I can see the server's card
accept it.  Doing a ifconfig before and after suggests that about 10
megabytes of data are sent, sometimes just to print a plain ascii
document.  If I click on the printer status in windows, it opens the
printer very slowly.  For example, I'll open the printer, about 10
seconds later the printer queue box will open with title saying
"Printer -- Initializing" then "Printer -- Refereshing" and about 10
seconds later, "Printer -- Ready"

Anyone have ideas how to fix this?

Thanks!

-- 
Peter Quackenbush
Welcome thy neighbor int-o thy fallout shelter.  He'll come in handy
if you run out of food.
 --Dean McLaughlin


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

kernel update, initrd image help

[Ian L]

I'm trying to install kernel 2.6.8 and during the apt process it says i 
need to configure lilo to use initrd.

Can someone either explain the process for doing this or point me to a link 
that does? I found a default mkinitrd.conf file, ran mkinitrd and it 
created an image file (it complaining about there being no module 
directory). But i'm not sure what settings in the config file i should set.

The main reason i want to upgrade the kernel is to try and get my network 
cards working (nforce2 and intel pro 100 s).

Thanks,
Ian

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel update, initrd image help

[s. keeling]

Incoming from Ian L:
> I'm trying to install kernel 2.6.8 and during the apt process it says i 
> need to configure lilo to use initrd.

  http://lists.debian.org/debian-user/2004/07/msg00582.html

See down at the bottom.


-- 
Any technology distinguishable from magic is insufficiently advanced.
(*)   http://www.spots.ab.ca/~keeling 
- -


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

RE: linmodem works but hangs up when idle

[[EMAIL PROTECTED]]

Eric

The Broadcom code has not been updated for some time.
   A fix by Giacomo Comes must be used for current 2.4 kernels
   http://linmodems.technion.ac.il/archive-third/msg01652.html

Giacomo is however no longer using this modem type and has volunteered that
an upgrade to  2.6.n kernels is probably impossible without Broadcom
participation.

Please forward to the Debian-user List.  Not being a current member, my
email would be rejected

MarvS


Original Message:
-
From: Eric Dickner [EMAIL PROTECTED]
Date: Tue, 5 Oct 2004 12:01:31 -0700 (PDT)
To: [EMAIL PROTECTED], [EMAIL PROTECTED]
Subject: "linmodem" works but hangs up when idle


I got a BCMSM 4212 to work on Debian 2.2.20-idepci
with the driver .rpm from Dell but it seems to hang up
after a minute or so of idle time.  I used "pppconfig"
to set no hang-up at all for idle time, and it seems
to be sort of a random period, ranging from 50 seconds
to two minutes.

If it's doing something with the connection, like a
download, it works like a champ, but reading webpages
means a redial everytime I pause to take the content
in.

Anyone know what this might be caused by?



___
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com


mail2web - Check your email from the web at
http://mail2web.com/ .

Squid and increaing from "4096 file descriptors"

[Jeff Green]

In running a squid on a large and capable (Dual Xeon 4GB RAM) server to 
supply furniture graphics for a large website we are unable to find any way 
to increase the number of available file descriptors beyond 4096. The server 
has the following ulimits:-
ulimit -a
core file size(blocks, -c) 0
data seg size (kbytes, -d) unlimited
file size (blocks, -f) unlimited
max locked memory (kbytes, -l) unlimited
max memory size   (kbytes, -m) unlimited
open files(-n) 32768
pipe size  (512 bytes, -p) 8
stack size(kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes(-u) unlimited
virtual memory(kbytes, -v) unlimited

cat /proc/sys/fs/file-max gives
206283
and /etc/default/squidcontains
SQUID_MAXFD=32768
I assume we need to rebuild the package after changing the appropriate files 
but does anyone have any clue what changes we need to make to which files?

Jeff
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel update, initrd image help

[Clive Menzies]

On (05/10/04 14:27), Ian L wrote:
> To: [EMAIL PROTECTED]
> From: Ian L <[EMAIL PROTECTED]>
> Date: Tue, 05 Oct 2004 14:27:35 -0700
> Subject: kernel update, initrd image help
> 
> I'm trying to install kernel 2.6.8 and during the apt process it says i 
> need to configure lilo to use initrd.
> 
> Can someone either explain the process for doing this or point me to a link 
> that does? I found a default mkinitrd.conf file, ran mkinitrd and it 
> created an image file (it complaining about there being no module 
> directory). But i'm not sure what settings in the config file i should set.
> 
> The main reason i want to upgrade the kernel is to try and get my network 
> cards working (nforce2 and intel pro 100 s).

Hi Ian

You need to edit /etc/lilo.conf (as root) ...be careful ;)

It is worth looking at man lilo.conf

FWIW the kernel options in my lilo.conf looks like this:

# Kernel command line options that apply to all installed images go
# here.  See: The `boot-prompt-HOWO' and `kernel-parameters.txt' in
# the Linux kernel `Documentation' directory.
#
# append=""

# Boot up Linux by default.
#
default=Linux
initrd=/initrd.img
image=/vmlinuz
label=Linux
read-only
#   restricted
#   alias=1

image=/vmlinuz.old
label=LinuxOLD
read-only
optional
#   restricted
#   alias=2

# If you have another OS on this
# machine to boot, you can
# uncomment the
# following lines, changing the
# device name on the `other'
# line to
# where your other OS' partition
# is.
#
# other=/dev/hda4
#   label=HURD
#   restricted
#   alias=3
~

YMMV

Clive




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel update, initrd image help

[Ian L]

At 02:43 PM 10/5/2004, you wrote:
Incoming from Ian L:
> I'm trying to install kernel 2.6.8 and during the apt process it says i
> need to configure lilo to use initrd.
  http://lists.debian.org/debian-user/2004/07/msg00582.html
See down at the bottom.
Thanks ... but i screwed something up. I'm wondering if there's something 
about my hardware that doesnt work with 2.6.8 because the error i am 
getting now is i think the same error i was getting when i tried to do the 
latest sarge install.

i ran mkinitrd, then ran apt, it created its own initrd img file. I edited 
lilo to point to the new file, ran lilo -v, it ran without any errors.

I rebooted assuming lilo would give me a choice as to which kernel i wanted 
to boot into. It didnt. Linux starts to load and then fails with the 
following lines:

nforce2 ide controller at pci slot :00:09.0
nforce2 chipset revision 162
nforce2 not 100% native mode, will probe irqs later
nforce2 bios didnt set cable bits correctly, enabling workaround

/dev/ide/host0/bus0/target0/lun0: unknown partition table
NTFS driver 2.1.15 flags R/O MODULE
pivot_root: no such file or directory
/sbin/init: 426: cannot open dev/console: no such file
kernel panic: attempted to kill init!
I'll do some more searching on nforce2 and chipset 162. Any other suggestions?
thanks for the help,
Ian

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel update, initrd image help

[Ian L]

# Boot up Linux by default.
#
default=Linux
initrd=/initrd.img
image=/vmlinuz
label=Linux
read-only
#   restricted
#   alias=1
This is similiar to what my lilo.conf looked like after i edited it, except:
image=/vmlinuz
label=linux
read-only
initrd=/initrd/img
The main difference being that initrd was under the image= section. I'm 
typing this from memory since right now my system is hosed and doesnt boot. 
I dont care too much since its a new system. Just a pita to reinstall but 
no lost data or anything.

I'm right now thinking there's some sort of incomparability between my 
chipset and the kernel. Its an asus a7n8x with nforce2 (chipset revision 
162 according to the boot screen).

thanks,
ian 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel update, initrd image help

[Ian L]

one quick question:
apt-get install kernel-image-x
Thats going to install that specific kernel image alongside the existing 
kernel image, correct? i.e. its not going to replace it.

if so, how do i get lilo to offer me the choice to boot into both? I 
thought the linuxOLD section of my lilo.conf woudl be for the older kernel 
image. So i wasnt too concerned about the new kernel loading since i 
thought i could just boot back into the old one. Right now i'm going to 
have to reinstall the whole thing again since i dont have a choice as to 
which kernel loads. (not that big a deal since its a new system, just 
getting tired of all the re-installs)

thanks for the help,
Ian
At 02:43 PM 10/5/2004, you wrote:
Incoming from Ian L:
> I'm trying to install kernel 2.6.8 and during the apt process it says i
> need to configure lilo to use initrd.
  http://lists.debian.org/debian-user/2004/07/msg00582.html
See down at the bottom.
--
Any technology distinguishable from magic is insufficiently advanced.
(*)   http://www.spots.ab.ca/~keeling
- -
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: raid barfs at startup with 2.6 kernel

[Paul Yeatman]

->>In response to your message<<-
  --received from Justin Guerin--
>
> On Tuesday 14 September 2004 07:21, Blake Swadling wrote:
> > OK, firstly i will say i know bugger all about the way raid works.
> > however i have managed to set up, both running software raid (raid0).
> > one uses 2 SATA drives and the other has 2 ATA drives
> >
> > When I upgrade to kernel 2.6.* the raid barfs on both machines with the
> > following errors
> >
> > [snip error messages]
> >
> > If i manually start run mdadm and raid2 and then do mount -a on both
> > machines the scripts after logging in then all is fine.  mdadm-raid
> > always fails hence the use of raidtools. All packages are up to date.
> >
> > It appears to me that the raid as trying to start before the modules are
> > present. Has anyone seen this?   how do i get it to load correctly now.
> >
> > Cheers
> >
> > Blake
> 
> You don't indicate how you are upgrading the kernel, but if you're using the 
> precompiled Debian kernels, you will need to make sure the raid drivers are 
> included in the initial ram disk.  "man mkinitrd" will help you here.

I agree.  Chances are you are using the Debian kernel-image
packages.  A simple

mkinitrd -o /boot/initrd.img-

with telling your boot loader to use this init image during boot should
get you back up and running if indeed it is a problem with modules not
getting loaded soon enough.

> 
> If you have compiled your kernel yourself, did you either A) compile in all 
> raid / filesystem support you need or B) add the modules you need to the 
> initial ram disk?
> 
> If you've done the above and are still seeing problems, check out the list 
> archives for threads on moving the raid start back in the boot sequence.  
> I'm sorry I don't have the link available.
> 
> If none of this helps, please be more specific with your configuration, 
> differences between what works and what doesn't, and what you've tried so 
> far.
> 
> Justin Guerin
> 
> 

Paul

-- 
Paul Yeatman   (858) 534-9896[EMAIL PROTECTED]
 ==
 ==Proudly brought to you by Mutt==
 ==


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[Bob Underwood]

On Tuesday 05 October 2004 02:32 pm, calvin wrote:
> On Tue, Oct 05, 2004 at 08:24:10AM -0500, Rodney Richison wrote:
> >>Top posting leads to a big mess. Every non-technical person does it
> >>and it ends up with them saying "let's use a forum or something"
> >>because they can't follow the thread any longer. Surprise surprise, a
> >>forum works with bottom posting and everybody seems to find that
> >>normal.
> >
> >I run a list for techs/consultants.  (ChannelVar.com)
> >They are a VERY qualified bunch of guys. (Admittedly not linux users
> >though)  :)
>
> So qualified that the site is not even up...

www.channelvar.com is NOT up.  channelvar.com (no www!) is!  just had to check 
it for myself.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel update, initrd image help

[Ian L]

nevermind, found out about hitting shift key during boot. still need help 
getting 2.6.8 working though :)

Ian
At 03:08 PM 10/5/2004, you wrote:
one quick question:
apt-get install kernel-image-x
Thats going to install that specific kernel image alongside the existing 
kernel image, correct? i.e. its not going to replace it.

if so, how do i get lilo to offer me the choice to boot into both? I 
thought the linuxOLD section of my lilo.conf woudl be for the older kernel 
image. So i wasnt too concerned about the new kernel loading since i 
thought i could just boot back into the old one. Right now i'm going to 
have to reinstall the whole thing again since i dont have a choice as to 
which kernel loads. (not that big a deal since its a new system, just 
getting tired of all the re-installs)

thanks for the help,
Ian
At 02:43 PM 10/5/2004, you wrote:
Incoming from Ian L:
> I'm trying to install kernel 2.6.8 and during the apt process it says i
> need to configure lilo to use initrd.
  http://lists.debian.org/debian-user/2004/07/msg00582.html
See down at the bottom.
--
Any technology distinguishable from magic is insufficiently advanced.
(*)   http://www.spots.ab.ca/~keeling
- -
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a 
subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: kernel update, initrd image help

[Clive Menzies]

On (05/10/04 15:08), Ian L wrote:
> one quick question:
> 
> apt-get install kernel-image-x
> 
> Thats going to install that specific kernel image alongside the existing 
> kernel image, correct? i.e. its not going to replace it.
> 
> if so, how do i get lilo to offer me the choice to boot into both? I 
> thought the linuxOLD section of my lilo.conf woudl be for the older kernel 
> image. So i wasnt too concerned about the new kernel loading since i 
> thought i could just boot back into the old one. Right now i'm going to 
> have to reinstall the whole thing again since i dont have a choice as to 
> which kernel loads. (not that big a deal since its a new system, just 
> getting tired of all the re-installs)
>From man lilo.conf :

When  booting,  the boot loader will issue its boot: prompt and wait for
you to enter the label of the kernel (and any options) which you wish to
boot.  At  any time you may hit [Tab] to see a list of kernel/other
labels.  Alter­ nately, if the boot-menu.b boot loader is installed, a
menu of boot  options will  be presented for your selection.  The title
of this menu is overridden with the menu title specification in this
configuration file.  If you enter nothing,  then  the  default kernel
image, the first mentioned, (/boot/zIm­ age-1.5.99) will be booted after
a timeout of 15 seconds (150 deciseconds).  There can be up to 16 images
mentioned in lilo.conf.

Regards

Clive


> 
> thanks for the help,
> 
> Ian
> 
> At 02:43 PM 10/5/2004, you wrote:
> >Incoming from Ian L:
> >> I'm trying to install kernel 2.6.8 and during the apt process it says i
> >> need to configure lilo to use initrd.
> >
> >  http://lists.debian.org/debian-user/2004/07/msg00582.html
> >
> >See down at the bottom.
> >
> >
> >--
> >Any technology distinguishable from magic is insufficiently advanced.
> >(*)   http://www.spots.ab.ca/~keeling
> >- -
> >
> >
> >--
> >To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> >with a subject of "unsubscribe". Trouble? Contact 
> >[EMAIL PROTECTED]
> 
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact 
> [EMAIL PROTECTED]
> 

-- 
Byte-Night 2004 ... helping the young homeless .. please visit:
https://www.bmycharity.com/clive


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Postgresql upgrade!!!!

[Katipo]

Andrea Vettorello wrote:
On Tue, 5 Oct 2004 10:44:48 -0700 (PDT), D S <[EMAIL PROTECTED]> wrote:
 

Does anybody knows how to upgrade postgresql on Debian manually or by
apt-get? 
The debian security web site has only the old 7.2 version. 
I tried to load it form other http sources but it can't find it. 
I need upgrade it to version 7.4. 
Does anybody have a init script that can run Postgresql 7.4 on Debian? 
 
   

If you are referring to the CAN-2004-0547 advisory (postgresql --
buffer overflow) the version in Woody (postgresql_7.2.1-2woody5) is
patched...
 

If you're running Sarge, you have postgresql 7.4.3-3.
Regards,
david.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[Grant]

Bob Underwood wrote:
On Tuesday 05 October 2004 02:32 pm, calvin wrote:
 

On Tue, Oct 05, 2004 at 08:24:10AM -0500, Rodney Richison wrote:
   

Top posting leads to a big mess. Every non-technical person does it
and it ends up with them saying "let's use a forum or something"
because they can't follow the thread any longer. Surprise surprise, a
forum works with bottom posting and everybody seems to find that
normal.
   

I run a list for techs/consultants.  (ChannelVar.com)
They are a VERY qualified bunch of guys. (Admittedly not linux users
though)  :)
 

So qualified that the site is not even up...
   

www.channelvar.com is NOT up.  channelvar.com (no www!) is!  just had to check 
it for myself.

 

lol seems i am special! both work for me :)

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: How to change default partition scheme in chroot sid

[Cameron Hutchison]

Once upon a time Henry Lenzi said...
> 
>  I am currently running Woody and I created a chrooted sid partition.
>  Here's my current partition scheme for sid:

If sid is in a chroot, it does not have a partitioning scheme. It lives
on the filesystem on which woody was installed.

> Filesystem   1k-blocks  Used Available Use% Mounted on
> /dev/hda2  4806080   4711892 0 100% /
> /dev/hda5  3844584119880   3529408   4% /var
> /dev/hda4  2691228862848   1691672  34% /root
> /dev/hda1   189339  1333178230   1% /boot
> /dev/hda6  3844584   2486176   1163112  69% /usr
> /dev/hda7  3844584   2864268785020  79% /home
> 
>  I would like to know if there's a way to change this default when you're
> installing sid. Let me rephrase that: how do I install chrooted sid with an
> arbitrary partition scheme and what tools do I use for that?

Basically, you can't change the partitioning scheme. You have only one
set of partitions regardless of how many chroot installations you have.
The chroot installations live on the existing partitions that were
created when you set up woody.

If I were in your situation, I would probably backup /var to somewhere
on /home and very very carefully repartition the drive, taking note of
*exactly* which cylinder each partition starts on. Since hda2 (root)
and hda5 (/var) *in your setup* are contiguous, it should be possible
to jiggle the partitions so that you end up with more space on your root
filesystem and less on /var.

However, I dont recommend this unless you can afford to reload everything
in case it mucks up, and only if you are sure of what you are doing.

Others may come up with better solutions - I'd wait to see what else
comes up on this thread. Maybe there's a partitioning tool that will do
it with little to no risk.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: GDM puts a semi-persistent "X" in the middle of the screen.

[Cameron Hutchison]

Once upon a time Adam Funk said...
> When I log out of xfce4, the gdm login box appears normally.  Not long
> afterwards, an "X" (like the multiplication symbol rather than the
> letter) appears right in the middle of the screen.  When I log back in,
> it stays on the screen until I run kmail or knode, either of which
> makes it disappear.

That's the default X11 cursor when it starts up.

I had a similar problem on my Toshiba laptop, using the "savage" X11
driver. I had to turn off the hardware cursor to get it fixed. I looked
in the man page for my X11 video driver to find the option. (man savage
in my case).


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Important

[dave]

From: [EMAIL PROTECTED]
Subject: From David Knell: E-mail address changed

Hi -
Owing to the unbelievable amount of spam coming in to this account (6000+ pieces over 
the space of one weekend) I've changed my e-mail address to [EMAIL PROTECTED]
Please forward the e-mail you've just sent me to my new address, and update your 
address book.  Apologies for the inconvenience.
--Dave
 Original Message 

> Important textfile!
> 
> 



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [Way OT] Re: GMAIL Invites..!

[Cybe R. Wizard]

On Tue, 05 Oct 2004 08:24:10 -0500
Rodney Richison <[EMAIL PROTECTED]> wrote:

...

> I run a list for techs/consultants.  (ChannelVar.com)
> They are a VERY qualified bunch of guys. (Admittedly not linux users 
> though)  :)
> The preferred method is top posting.
> The preferred format is to allow html.
> 
Who are you trying to fool?  Your site is a pathetic bare bones sample,
probably not even paid for yet as evidenced by the little, "default
theme sample" tag in the lower left corner of /every page/.  There is
/no/ content yet you expect people on this list to listen to you about
your hypothetical top posting tech gurus?

Cybe R. Wizard -wants summa what you been smokin'
-- 
Unofficial "Wizard of Odds," A.H.P.
Original PORG "Water Wizard," R.P.
"Wize(ned) Wizard," A.P.F-P-Y.
Barely Tolerated Wizard, A.J.L & A.A.L


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Compiler error for kernel 2.4.19- missing terminating " character

[Dayong Gu]

Hi, 
 I've compiled 
a 2.4.19 kernel on my debian system (2.2.20-idepci), but I got such kind of 
error: 
 
 gcc -D__KERNEL__ -I/usr/src/linux-2.4.19/include -Wall 
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common 
-fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=athlon -nostdinc 
-I /usr/lib/gcc-lib/i586-mandrake-linux-gnu/3.3.1/include 
-DKBUILD_BASENAME=semaphore -c -o semaphore.o semaphore.csemaphore.c:265:1: 
missing terminating " charactersemaphore.c:290:1: missing terminating " 
charactermake[1]: *** [semaphore.o] Error 1make[1]: Leaving directory 
`/usr/src/linux-2.4.19/arch/i386/kernel'make: *** [_dir_arch/i386/kernel] 
Error 2 
 
who can enlight me, what's wrong here? 
thanks!
 
Cheers,
 
Dayong

downloading selected source packages automatically

[ROBERTOJIMENOCA]

I have to compile many packages from source because some of my patches
are not applied upstream. So I download the source package (dsc, diff.gz
& tar.gz), dpkg-source it, apply my patches, dpkg-buildpackage
-rfakeroot, install it and mark it for hold so when upgrading a newer
version isn't installed.

I'd like to have this process automated as much as possible so I'd like
to have the source packages I need downloaded when there are new
versions and the old sources removed (as I won't use them anymore)

So I have, in a folder,  something like:
glibc_2.3.2.ds1-16.diff.gz
glibc_2.3.2.ds1-16.dsc
glibc_2.3.2.ds1.orig.tar.gz

and just running a command have that files upgraded to the latest
version.

Is there a command to do that?
What about automating the whole process as much as possible?



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

HP 2210 All in Wonder USB storage devices

[Tom Allison]

um
How do I mount the usb storage devices?
I have usbfs on /proc/bus/usb type usbfs (rw)
but when I insert the CF card, I don't get anything on the syslog 
telling me that a device was found (I expected this for USB devices).

I know I had this working under Linux at some point, but that was about 
6 months ago under SuSE.

It prints just fine.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Sound recording in ALSA

[robin]

Kirk Strauser wrote:
I'm using Sid with Debian's 2.6.8-1-686 kernel package on a Dell PC with an 
Intel 8x0 sound device and ALSA.  My ultimate goal is to get liveice to 
read from the line in on my sound card, but I'm trying to take baby steps 
to get there.

Right now I'm attempting to record by:
  1) Going into KMix and selecting "Line" as the capture device (verified 
with amixer).
  2) Using something like "arecord -t wav -d 10 /tmp/foo.wav" to attempt to 
record the sound that's being played through my speakers to a file.

However, that file seems to be silent on playback, regardless of what 
permutations of the steps above I've messed with.  Any pointers on how I 
can get started with this?
 

Check out www.agnula.org documentation section. Agnula Demudi is a 
deb-based linux audio distribution. You may find what you need. Another 
site is http://ccrma.stanford.edu/planetccrma/software/

Robin
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

DVD standard "+" or "-"?

[Curt Howland]

I am interested in getting an opinion from folks. In the DVD standards 
wars, who do you think is winning, "DVD+" or "DVD-"?

Here's why I ask: The spankie CD/DVD drive in this Sony is both "+" 
and "-", and it and Linux write to whatever blank media I choose to 
put in it. But I do not know if in the future any new machine I can 
get will have as spankie a drive as this one. Therefore, if there is 
a trend toward "+" or "-" becoming predominant, I would prefer to buy 
that media so it will be easier to read in the future.

So what do y'all think?

Curt-


-- 
September 11th, 2001
The proudest day for gun control and central 
planning advocates in American history


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Nvidia + 2.6 kernel

[Tim Connors]

Andrew Schulman <[EMAIL PROTECTED]> said on Tue, 5 Oct 2004 15:56:37 -0400:
> The rivafb module is incompatible with the nvidia driver; see 
> /usr/share/doc/nvidia-kernel-source/README.Debian.  This shouldn't be a 
> problem if you never load that module, but if it does ever get loaded 
> for some reason, your monitor could explode (figuratively speaking).  I 

I was reading this just the other day, trying to decide whther I want
to try getting my old nvidia card working again (I don't even have
non-accelarated GLX in its current state using the nv driver -- when I
was using nvidia, I had GLX, and managed about 2 frames a second in
glxgears, before the machine would crash -- yay!).

I have successfully gotten rivafb to talk with my video card
simulataneously with the nv X driver. I need this, because I have a
fixed freq monitor, and so need modelines that vesafb can't provide
when at the console. Do you know of any other way? Presumably, I could
use framebuffer X (Option "UseFBDev" "true"), but then I would have to
put up with even slower X with no accelaration.


Damn, I should have asked DS this lastnight over the pizza (although
he may be sick of talking about work) :)

-- 
TimC -- http://astronomy.swin.edu.au/staff/tconnors/
Sufficiently advanced incompetence is indistinguishable from malice.
-- Hamilcar Barca @ comp.os.linux.advocacy


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Installing from net with floppy disks - syst em don´t recognize my hard disk

[Joey Hess]

Juan Stiller wrote:
> im quite new to debian so i search on debian.org for
> the floppy images and make the disks.

Please use the floppy images linked to from
http://www.debian.org/devel/debian-installer/

-- 
see shy jo


signature.asc
Description: Digital signature

Re: HP 2210 All in Wonder USB storage devices

[robin]

Tom Allison wrote:
um
How do I mount the usb storage devices?
I have usbfs on /proc/bus/usb type usbfs (rw)
but when I insert the CF card, I don't get anything on the syslog 
telling me that a device was found (I expected this for USB devices).

I know I had this working under Linux at some point, but that was 
about 6 months ago under SuSE.

It prints just fine.

Load the card and run dmesg to see if  any output generated.
Normally it's a case of:  mount /dev/sda1 -t msdos  /your/mount/point 
(create if necessary). Add to /etc/fstab if that works, then next time : 
mount /your/mount/point should work

Robin
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Squid and increaing from "4096 file descriptors"

[robin]

Jeff Green wrote:
In running a squid on a large and capable (Dual Xeon 4GB RAM) server 
to supply furniture graphics for a large website we are unable to find 
any way to increase the number of available file descriptors beyond 
4096. The server has the following ulimits:-
ulimit -a
core file size (blocks, -c) 0
data seg size (kbytes, -d) unlimited
file size (blocks, -f) unlimited
max locked memory (kbytes, -l) unlimited
max memory size (kbytes, -m) unlimited
open files (-n) 32768
pipe size (512 bytes, -p) 8
stack size (kbytes, -s) 8192
cpu time (seconds, -t) unlimited
max user processes (-u) unlimited
virtual memory (kbytes, -v) unlimited

cat /proc/sys/fs/file-max gives
206283
and /etc/default/squid contains
SQUID_MAXFD=32768
I assume we need to rebuild the package after changing the appropriate 
files but does anyone have any clue what changes we need to make to 
which files?

Jeff
Silly question possibly but have you checked the squid site for 
configuration information. 
http://squid.visolve.com/squid/squid24s1/contents.htm.
I could be barking up the wrong tree but there seemed to possibly valid 
info there.
A "just a thought" response
from
Robin

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Recomended dual head video card

[Dana J. Laude]

On Monday 04 October 2004 23:44, Nick Hastings wrote:
> Hi all,
>
>for reasons I won't bother going into I'm about turn a server
> machine into a workstation (and perhaps occasional game machine).
>
> There are two nice 1280x1024 17 inch digital LCD displays available
> (Mitsubishi RTD179S) [1], so I thought I'd go out and get myself a
> nice dual head video card. But which to choose? My budget is fairly
> generous.
>
> My machines motherboard has and AGP slot so I thought an AGP card
> would be best.
>
> I've had a bit of a look at the ATI web site and it seems that the
> "FireGL Z1-128" [2] might suit me well.

I have a matrox G450 that I picked off eBay for $35 and it works quite
well and the price was right. It's a older card though, so for gaming 
I wouldn't really recommend it... although the card works "ok", 3D
stuff is a tad on the slow side.  For normal stuff it works great 
though, and 2 monitors are very cool.. which I was running under 
debian unstable with windowmaker.

I'm sure there is something better out there, but I've never had
a problem with Matrox cards under linux.  Hopefully someone
can ring in here about some of the newer cards available.

Dana


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: GDM puts a semi-persistent "X" in the middle of the screen.

[yo mero]

the problem is the hardware cursor enabled

this is the setting on my  XF86Config-4


Section "Device"
Identifier  "Generic Video Card"
Driver  "savage"
Option  "SWCursor"  "on"
Option  "HWCursor"  "off"
EndSection


all solved

Leonel



On Tue, 05 Oct 2004 18:28:29 GMT, Adam Funk <[EMAIL PROTECTED]> wrote:
> When I log out of xfce4, the gdm login box appears normally.  Not long
> afterwards, an "X" (like the multiplication symbol rather than the
> letter) appears right in the middle of the screen.  When I log back in,
> it stays on the screen until I run kmail or knode, either of which
> makes it disappear.
> 
> I looked through the gdm bugreport page but didn't see anything.  I
> suspect another package could cause the problem but I have no idea
> where else to look.  Has anyone else noticed this?
> 
> Background: I currently have the following packages installed, although
> this problem has persisted through upgrades for several months.
>gdm2.4.4.7-3
>kmail  3.3.0-3
>knode  3.2.2-2
>xfce4  4.0.5-1
> 
> Thanks,
> Adam
> 
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 
>


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [OT] Video card recommendations.

[Scotty Fitzgerald]

I'd just like to know what video card works best "out of the box" with
debian.


This is because I have just got woody, and aparently I, as a newbie,
have to figure out how to install & compile a special driver so woody
can talk to the on-board four mb of memory on my Intel 82815 video
card.  I sure would like to just be able to have base-config just get
it up right as an xfree86 terminal.

I'm not freaking, tho, as I have knoppix running smooth as silk on
another partition,

BTW, what is VESA?!
---
Scotty

On Sat, 02 Oct 2004 04:50:08 +0200, Paul Johnson
<[EMAIL PROTECTED]> wrote:

><#secure method=pgp mode=sign>
>-BEGIN PGP SIGNED MESSAGE-
>Hash: SHA1
>
>Roberto Sanchez <[EMAIL PROTECTED]> writes:
>
>> What I would like to do is get an nVidia card.
>
>Mistake du jour!  You'll only spend more on the nVidia for even
>crappier drivers.  Just get a newer Radeon if you want graphics
>performance.  I believe there's actually working open drivers for the
>ATI adapters.
>-BEGIN PGP SIGNATURE-
>Version: GnuPG v1.2.5 (GNU/Linux)
>
>iD8DBQFBXhTLUzgNqloQMwcRAmG4AKCzn9tbWYoStwS0rirgIdZ2TE7a1QCeLRC3
>/cWojlUVlmy88q8qpdHJLfY=
>=WxKm
>-END PGP SIGNATURE-
>
>
>-- 
>To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
>with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [OT] Video card recommendations.

[William Ballard]

On Wed, Oct 06, 2004 at 01:28:54AM +, Scotty Fitzgerald wrote:
> I'd just like to know what video card works best "out of the box" with
> debian.

Do dpkg-reconfigure xserver-xfree86.  Look at the choices for video 
drivers which are there.  Google for video cards which work with those 
drivers.  These will video cards will "just work."
> 
> BTW, what is VESA?!

Unaccelerated "it just works" mode that all video cards support.  
However, nothing will be accelerated.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [OT] Video card recommendations.

[Carl Fink]

On Wed, Oct 06, 2004 at 01:28:54AM +, Scotty Fitzgerald wrote:

> This is because I have just got woody, and aparently I, as a newbie,
> have to figure out how to install & compile a special driver so woody
> can talk to the on-board four mb of memory on my Intel 82815 video
> card.  I sure would like to just be able to have base-config just get
> it up right as an xfree86 terminal.

Don't use Woody.  Woody is years out of date.  Sarge detects most video
cards flawlessly.
--  
Carl Fink [EMAIL PROTECTED]
Jabootu's Minister of Proofreading
http://www.jabootu.com


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

java plugin works, programs do not run (exception in thread main)

[Rohan Deshpande]

Hello,

I'm getting an error when tryign to run any java apps using ANY j2re
from blackdown's site (sun and blackdown versions.) Plugin works fine,
but I try to install LimeWire for java and get the following
exception:

Exception in thread "main" java.lang.NoClassDefFoundError: com/zerog/lax/LAX 

I've tried googling to no avail.  I set the JAVA_HOME variable and
still nothing.  Am I missing some libs or something? i assume the
blackdown srces would pick it up.

Thanks,
Rohan


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]