limiting specific users on squid by time and username

[Louie Miranda]

Hello,

How will i be able to block certain users on using my proxy? They are all
connected via dhcp, and some i allowed full access and some i wish to deny
even on time ex.

user-full-access = NO LIMIT
user-1-access = 12pm to 1pm only access
user-2-access = 6pm to 7pm only access

is this possible?



-- -
Louie Miranda
http://www.axishift.com



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

Chroot install (was Re: making a floppy for essential modules)

[Karsten M. Self]

on Tue, Dec 09, 2003 at 11:31:16AM -0800, New Disorder Records ([EMAIL PROTECTED]) 
wrote:
>  On Tue, 9 Dec 2003, 
> Michael Martinell wrote:

> > On Tuesday, December 09, 2003, New Disorder Records
> > [mailto:[EMAIL PROTECTED] wrote:
> > > 
> > > I am trying to install debian on a dell poweredge, but I need
> > > drivers for the hard drives and ethernet card.  I have found the
> > > drivers, but I have no idea how to put them on a floppy disk so
> > > that debian will see them when I put the floppy in.  I don't have
> > > other debian machines in the office, so I have to make the
> > > floppies on a windows machine.  Any advice about how to put the
> > > drivers on a floppy so that debian will see them would be greatly
> > > appreciated.
> 
> > Here is how I set up my Dell 
> > Set up Debian 3.0r1 on Dell Poweredge 2400
> > 
> > 1.  Insert Debian disk and boot from it.
> > 2.  At boot: prompt type bf24 and press enter.
> > 3.  Complete rest of install normally.  Make sure to install the c

<...>

> thanks, but I can't install normally if I can't see the hard drive.
> The installer (any one of the flavors of installer, including bf24)
> doesn't see the harddrive on my 1750, so I need to install a driver at
> that point, I assume, in order to continue the installation.
>
> So, I can make it to step 2 of your steps, but not step 3.  I have a
> driver, I just need to know how to make a disk that the installer will
> recognize to load it.


A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

For clarity and to support conversational discussion style, please use
bottom-posting format:  your reply goes below the material cited.  Trim
your quotes appropriately and ensure your attributions are accurate.  

See: 

http://www.catb.org/~esr/jargon/html/email-style.html
http://www.faqs.org/rfcs/rfc1855.html
http://mailformat.dan.info/quoting/top-posting.html

Thank you.


My recommendation in cases such as this is to utilize a chroot install.
There are instructions in the standard Debian Installation Manual, or my
own (slightly divergent) instructions at:

http://twiki.iwethey.org/Main/DebianChrootInstall

The secret to a chroot is that you first boot a GNU/Linux system on your
hardware that _does_ support the core system:  console, disks, network.
Then you initiate a Debian installation within this environment as a
chroot (change root) process.  It's not terribly different from a stock
Debian installation, but it's far more flexible.

Choices for a booted system include Knoppix, Morphix, DemoLinux,
LNX-BBC, tomsrtbt, and other bootable GNU/Linux distributions (that is:
bootable on removeable media), _or_ a previously installed GNU/Linux
system.


Peace.

-- 
Karsten M. Self <[EMAIL PROTECTED]>http://kmself.home.netcom.com/
 What Part of "Gestalt" don't you understand?
Beware of bugs in the above code; I have only proved it correct, not
tried it.
- Donald Knuth


pgp0.pgp
Description: PGP signature

Re: Kernel 2.4.23

[Karsten M. Self]

on Tue, Dec 09, 2003 at 06:48:01AM -0800, Michael Montagne ([EMAIL PROTECTED]) wrote:

> I'm a little new at this.  I want to upgrade my kernel from my
> Libranet 2.4.21.  
>
> I have a RAID controller (IDE) that appears as a SCSI device for some
> reason.  

RAID controllers frequently do.  The disks themselves are IDE/ATA.
However the interface between the controller and the motherboard is
SCSI.

> So I need to use initrd to boot (correct?).  At least that's
> the way Libranet set it up for me.  

Either initrd or compiled-in support for your RAID card.

> That means that I can't use a pristine kernel, I need the
> cramfs/initrd patch.  I got a kernel panic when trying the pristine
> kernel.
>
> Since I can't find that anywhere perhaps I should use the Debian
> kernel.  I found that at backports.org.  I guess it would be ok even
> thought I'm not running a "stable" system.  There is a source and a
> .deb that looks already compiled.  Since I want to save my settings
> from before, I need to use the source (correct?).  Are my assumptions
> even close to reality?

Herbert Xu (Debian kernel packages maintainer) has posted prerelease,
unofficial, 2.4.23 debs to:

http://master.debian.org/~herbert/sid/


Peace.

-- 
Karsten M. Self <[EMAIL PROTECTED]>http://kmself.home.netcom.com/
 What Part of "Gestalt" don't you understand?
   NPR:  Radio for between the ears:  http://www.npr.org/


pgp0.pgp
Description: PGP signature

kernel recompilation - just copy over?

[Monique Y. Herman]

I've been running 2.4.21.  I had Debian's 2.4.21-5 and now see that
2.4.21-6 is available, with the fix that's been on everyone's mind.

Can I safely just copy a few files over from the previous build, or
might there be changes that would require a more careful procedure?  Can
I use `make oldconfig` when it's really still the same kernel version?

If it *is* as simple as copying over files, do I just need .config, or
is there more to the story?

-- 
monique


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

Re: xfree won't start w/ video card sis630

[emilp]

--On Tuesday, December 09, 2003 17:01:38 -0700 Terry 
<[EMAIL PROTECTED]> wrote:

XFree cannot start.  When I select the option to start the program to
make adjustments, the monitor flashes from black to a small window
with the monitor's settings and the message "out of range".
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a
subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]


Sounds like you got too high values for "HorizSync" and/or 
"VertRefresh" in your xconfig file (/etc/X11/XF86Config-4 or 
/etc/X11/XF86Config).  You can try to run (from console) "XFree86 
--configure" and see if it works, or you can adjust the settings your 
self.  Check the manual for your monitor, the ranges it's usually 
specified.  If it isn't (and configure above doesn't work) you can try 
with:

   HorizSync   30-50
   VertRefresh 50-70
which I _think_ should work for most monitors, although it may be far 
from optimal.

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

Getting kicker back

[Francisco Castellon]

Hello list:

I seem to be having a problem with my KDE desktop. Somehow my brother
managed to get rid of the kicker menu at the bottom of the screen. So
what I have now is the default blue screen of KDE and some icons in my
desktop, and I want to get kicker back. So this is what I am doing, I
right click anywhere on the desktop and I get the alternate menu, then I
select "Run command" and I type in "kicker" and press enter. Then the
kicker menu appears on the screen for less than a second and then it
disappears again, it does this twice then nothing happens, no kicker, no
nothing. Does anyone know how I can get it back?

As well, now my keyboard does not seem to be recognizing the "/"
character or any other character that is not a letter or number for that
matter, what should be the right configuration for my keyboard as an
input device? (I have wireless Logitech keyboard (Cordless Freedom Pro
Kit) my mouse is also wireless but it works fine, I don't know why my
keyboard is acting up now since it didn't before.

Any help would be appreciated, thanx!

Francisco


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

Re: Debian ZOPE default user & passward

[John Foster]

Derrick 'dman' Hudson wrote:

On Tue, Dec 09, 2003 at 02:17:45AM -0600, John Foster wrote:
| I have been going over all of the docs to no avail Yet!! If someone 
| knows what the first login user & the passward for an out of the box 
| debian version install of ZOPE (sid) is please advise.Seems strange that 
| this info would not be in the readme.deb.txt or some such.

Depending on which version of the package you installed, I would
expect one of the following to be true :
   1)  The installation told you to run 'zope-zpasswd' to create the
   account.
   2)  The installation contains a bug and no account was created,
   but you can still create it manually.  (this should have been
   fixed a while ago)
   3)  The installation prompted you, via debconf, for the username
   and password and set it up as you specified.
One solution is to run zope-zpasswd and (re)create the initial
"emergency" user account.
Alternatively, look in /var/lib/zope/instance/default.  There should
be a file named 'access'.  If it doesn't exist, create it with
onwership root:root, permissions 644.  With an editor, type
'user:pass' in the file.  (the password does not have to be in
cleartext; the zpasswd utility gives a couple different options for
encryption)
HTH,
-D
 

Well I used kpackage  to install it and it did not ask anything during 
the install so I guess there is a bug. The install did not create the 
'access' file so I followed you instructions and all is well & working . 
I have not tried to use zope before so I really appreciate the tip. This 
set of instructions should be included if they do not get the bug fixed 
soon.
Thanks!!

--
John Foster


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

Re: Kernel modules for AIC79xx scsi card?

[adam morley]

On Wed, Dec 10, 2003 at 02:14:33AM +, Colin Watson wrote:
> On Tue, Dec 09, 2003 at 05:13:45PM -0500, adam morley wrote:
> > However, I can't access the above page at people.debian.org because I
> > get a redirection loop in both Mozilla and w3m.  I reported this to
> > [EMAIL PROTECTED] on 3 Dec 2003, but have not received a
> > response.
> > 
> > I'm guessing people.debian.org is hosted on glucker (from the
> > redirection page I saw), which was affected by the recent compromise
> > --- is this why I'm seeing redirection loops?
> 
> Yeah, people == gluck. The apache configuration should be fixed fairly
> soon, I'd guess.
> 
> debian-www don't control the apache configuration on gluck, although
> they've been somewhat flooded by mails about it.

Then my next question would be: what's a stable, high performance kernel for SMP 
machines for a debian server?  I use WOLK on my laptop, but I've heard people have 
problems with it on SMP machines.  (random panics)

I'd prefer 2.4 --- some people have been suggesting the 2.6, but I just don't know 
about that.

-- 
adam

pgp key at:
http://fedora.cwru.edu/axm135.key


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

Re: Done yet?

[Nunya]

On Tue, Dec 09, 2003 at 07:59:09PM -0800, Paul Johnson wrote:
> 
> Yes.  You're acting like you're two.  FOAD.

Okay, you made your point.  Like I said, I'm completely comfortable with 
whatever you think of me.  Whatever you say, I refuse to express any 
futher opinions.


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

Re: Done yet?

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 07:19:37PM -0800, Tom wrote:
> On Tue, Dec 09, 2003 at 09:52:19PM -0500, ScruLoose wrote:
> > 
> > I will, however, try to remember not to bother with responding to what's
> > merely meaningless or wildly irrelevant.
> > 
> > Just so we're clear.
> 
> Gotcha!  (Got a problem with this?)

Yes.  You're acting like you're two.  FOAD.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1poNUzgNqloQMwcRArKQAJ4yy+qHLkt2AWWmEIV5YevvjRje+gCgk/6r
PwGjHqRPgFWa/whmr1uzFUo=
=jkLG
-END PGP SIGNATURE-


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

Re: Can spamassassin remove SIGs?

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 07:04:50PM -0800, Nunya wrote:
> I've decided SIGs are spam:

It appears you have a misunderstanding what spam is in the context of
email.  By proxy, this hinders the intuitiveness of SpamAssassin's name.

http://ursine.ca/jargon/html/S/spam.html

I think you meant to say "I've decided sigs are noise."

> For now, I've set my .muttrc to color signatures as "black/black", which 
> is doing a pretty good job.  But since most sigs follow quite regular 
> patterns structurally as well as content-wise, a bayesian filter should 
> be able to strip them effectively.

No filtering of any kind will do that.  Why would you think that?  You
want a non-interactive editor you can pipe in.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1pmCUzgNqloQMwcRAnBHAKDLJvF5mWFnvG90i+/fErzP3k/KoQCgk58J
nulBv71v0dc1CsaEO7ZYh5w=
=vlCa
-END PGP SIGNATURE-


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

Re: Dhcp and ssh

[Walter Dnes]

On Sat, Dec 06, 2003 at 11:55:29AM -0700, Monique Y. Herman wrote

> I'm pretty sure ssh still gets upset if a hostname doesn't match the
> expected IP  known_hosts records both the hostname and the IP,
> then gets upset if either changes in relation to the other.

  ??? I'm on an ADSL ISP, and I have a dynamic IP address to the outside
world.  For even ,ore fun, I'm behind two layers of NAT.  Yet I still
manage to ssh in daily to a remote machine to collect my email
(including this list).  Here's my setup...


   Dynamic publicly192.168.1.2/
   routable IP address 255.255.255.248
 v===<  Netgear Router   |
   ||^| Internal address |
 ^| 192.168.1.254/   |
 192.168.1.1/^| 255.255.255.248  |
 255.255.255.248 ^|_  ___  ___  ___  |
internal address>^  ||   ||   ||   ||
||   ||
||   ||
 ___||   ||
| |  ||
| Linux 192.168.1.249 |  ||
| mask 255.255.255.248|  ||
|_|  ||
 ||
  ___||
 | |
 | Linux 192.168.1.250 |
 | mask 255.255.255.248|
 |_|

  The ADSL modem has internal address 192.168.1.1 mask 255.255.255.248.
I could get away with netmask 255.255.255.252.  I set the router's
"external/WAN" address as 192.168.1.2 netmask 255.255.255.248 (netmask
255.255.255.252 would also work here.)  I also set the router's external
default gateway route to 192.168.1.1 (i.e. the modem's internal address).

  Apparently, the primary rule with double-NATting is not to overlap IP
address ranges.  So I decided to select the block defined by
192.168.1.248 netmask 255.255.255.248 (same as 192.168.1.248/29).  The
allocations I use are...

  192.168.1.248 base address (not supposed to have a machine here)
  192.168.1.249 \
  192.168.1.250  \ are for up to 4 machines to hook
  192.168.1.251  / up to the 4 ports on the router
  192.168.1.252 /
  192.168.1.253 I haven't quite figured out a use for this address
  192.168.1.254 The internal (LAN side) address used by the router.
This follows the default convention for router address.
  192.168.1.255 The broadcast address used by the router (syslog output).
This follows the default convention for broadcast address.

-- 
Walter Dnes <[EMAIL PROTECTED]>
I'm not repeating myself; I'm an X Window user, I'm an ex-Windows user


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

Re: make-kpkg

[Johann Koenig]

On Tuesday December  9 at 06:42pm
Paul Johnson <[EMAIL PROTECTED]> wrote:

> On Tue, Dec 09, 2003 at 10:09:48AM -0800, Michael Montagne wrote:
> > I use Grub as a bootloader.  After making a kernel .deb using
> > make-kpkg, I'm running dpkg -i  Near the end you are asked to if
> > you want to make a boot block.  What is this?  Is it just an entry
> > in Grub or LILO?  What I'm most concerned about is being able to
> > boot to my old kernel if I screwed this one up.
> 
> Since LILO is so much better documented, easy to configure, far more
> widely used, is generally the "assumed" bootloader, and just works,
> why not just use LILO?

I've had some annoying problems with lilo not liking my hard drives, and
weird booting problems that I can't really trace back to any definitive
source. So I switched to grub, and I'm pretty happy with it.
-- 
-johann koenig
Now Playing: The Mountain Goats - Going to Cleveland : Bitter Melon Farm
Today is Pungenday, the 51st day of The Aftermath in the YOLD 3169
My public pgp key: http://mental-graffiti.com/pgp/johannkoenig.pgp


pgp0.pgp
Description: PGP signature

Re: process limits (was: Spamassain question, whitelist?)

[Bill Goudie]

On Tue, Dec 09, 2003 at 06:06:52PM -0800, Karsten M. Self wrote:
> on Tue, Dec 09, 2003 at 10:51:26AM -0800, Vineet Kumar ([EMAIL PROTECTED]) wrote:
> > * Karsten M. Self ([EMAIL PROTECTED]) [031208 19:46]:
> > > on Mon, Dec 08, 2003 at 06:44:04PM -0800, Vineet Kumar ([EMAIL PROTECTED]) wrote:
> > > > * Karsten M. Self ([EMAIL PROTECTED]) [031208 16:52]:
> > > > > For performance reasons, I also have in /etc/security/limits:
> > > > > 
> > > > > mailhardnproc   30
> > > > > 
> > > > > ...to avoid runaway conditions when large mail loads hit.  Mail
> > > > > processing will be limited to a max of 30 processes (generally 10 exim
> > > > > processes, 10 spamassassin clients, and a bit of overhead), but the
> > > > > system as a whole won't be bogged.
> > > > 
> > > > So you have spamc running as mail, and not as the destination user
> > > > account?  
> > > 
> > > No.
> > 
> > As I understand the line you gave above, that limits the number of
> > processes being run as the mail user.  (I'm not using
> > /etc/security/limits.conf ; this is my understanding from reading the
> > comments in that file.)  
> 
> Correct.
> 
> > So how does this work?  Is it that spamd forks for each client, and
> > that's running as mail, and that's where the limit comes into play?
> 
> Yes.
> 
> > It looks like spamd's default behavior is to run as root.  
> 
> This is true, but its children run as 'mail'.  I think.

My understanding is that spamc will change its uid from root to the
calling user if "-u" is not given.  For example, I call spamc from
procmail without "-u".  Spamd logs:

   Dec  9 21:05:02 zeus spamd[1807]: connection from localhost [127.0.0.1] at port 
43665
   Dec  9 21:05:02 zeus spamd[20562]: info: setuid to bill succeeded
   Dec  9 21:05:02 zeus spamd[20562]: processing message <[EMAIL PROTECTED]> for 
bill:1000.


And while in action:

   UIDPID  PPID  C STIME TTY  TIME CMD
   root  1807 1  0 Dec03 ?00:00:10 /usr/sbin/spamd --create-prefs 
--max-children 10 --auto-whitelist -H -d --pidfile=/var/run/spamd.pid
   bill 20696 20682  0 21:16 ?00:00:00 formail -s procmail
   bill 20697 20696  0 21:16 ?00:00:00 procmail
   bill 20698 20697  1 21:16 ?00:00:00 procmail
   bill 20699 20698  0 21:16 ?00:00:00 spamc
   bill 20700  1807 47 21:16 ?00:00:00 /usr/sbin/spamd --create-prefs 
--max-children 10 --auto-whitelist -H -d --pidfile=/var/run/spamd.pid

So: formail -> procmail -> procmail -> spamc -> spamd(1807) -> spamd(20700)

Your earlier posts suggest to me you are doing the same.  Correct?

> What I know is that the above config *does* keep a box from spawning
> endless processes in response to spam swarms.

Well, you've place the resource control at the begginning of the
pipeline and indirectly control the number of spamd runs, I think.  The
number of spamd runs, however, still remains unlimited by the system but
restricted by the max-children arguement, right?

-- 
OS/2 must die!


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

Re: Info Problem

[Walter Dnes]

On Mon, Dec 08, 2003 at 04:58:26PM -0800, Karsten M. Self wrote

> Well, for values of "anywhere" equalling "take it up with Richard",
> yes.  But no specific movement.  Among the reasons for polishing
> my man vs.  info debate was preparation for a formal proposal.
> Which still hasn't happened.  But could.  Think it's worth a try?

  Sometimes, laughter is the best weapon.  I notice that "info" seems
to be like some some of the role-playing adventure games where you
wander up and down many levels in a maze to reach the final destination.
so here's my proposal...

  1) Create a trivial program that does nothing.  Call it "sucks".

  2) To invoke its info page, you'd have to type "info sucks".

  Have a *MASSIVE* info file for "sucks" with many levels and links
through its structure.  The opening page would be something like...




File: sucks.info,  Node: The beginning,  Next: ???  Prev: ???,  Up: Top

INFO sucks
**

   Admit it, we've all wasted more than our fair share of time trying to
find some minor little item in INFO.  We chose which way to jump, and
climbed up and down levels in our frantic search for the truth.  While
INFO may suck as an information-retreival engine, it has the beginnings
of a wonderful gaming engine.  You'll still end up wasting tons of time
with INFO, but at least you'll have fun doing so.  Without further ado,
here's the demo game to demonstrate the potential of INFO.

   You are in a maze of twisty passages, all alike...

* Menu:

* left::???
* right::   ???
* up::  ???
* down::???
* backwards::   ???
* forwards::???


-- 
Walter Dnes <[EMAIL PROTECTED]>
Email users are divided into two classes;
1) Those who have effective spam-blocking
2) Those who wish they did


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

Done yet?

[Tom]

On Tue, Dec 09, 2003 at 09:52:19PM -0500, ScruLoose wrote:
> 
> I will, however, try to remember not to bother with responding to what's
> merely meaningless or wildly irrelevant.
> 
>   Just so we're clear.

Gotcha!  (Got a problem with this?)


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

Re: make-kpkg

[Isaac To]

> "Paul" == Paul Johnson <[EMAIL PROTECTED]> writes:

>> I use Grub as a bootloader.  After making a kernel .deb using
>> make-kpkg, I'm running dpkg -i  Near the end you are asked to if
>> you want to make a boot block.  What is this?  Is it just an entry in
>> Grub or LILO?  What I'm most concerned about is being able to boot to
>> my old kernel if I screwed this one up.

If the old kernel is not yet deleted, you can edit the Grub entry at boot to
replace the kernel file by the old kernel, and boot.  The old kernel will be
booted allowing you to correct the problem.

Paul> Since LILO is so much better documented, easy to configure, far
Paul> more widely used, is generally the "assumed" bootloader, and just
Paul> works, why not just use LILO?

Because Grub provides a lot of functionalities that is not provided by LILO.
E.g., in Grub you can boot a test kernel once without running "grub-install"
(in LILO you must run "lilo" after rewriting lilo.conf).  Even filename
completion will work, so you can still find your kernel or initrd even if
you forget their exact names.  You can change any boot parameters before
that, so you don't need nearly as many "emergency" boot entries as in LILO.
Grub understand the filesystem, so you don't need to worry about any utility
moving the actual location of the kernel on the disk---you can safely cp
your kernel to somewhere else and mv it back (or run your defragmentor), and
there is no need to rerun "grub-install" as in LILO (rerun "lilo").  Grub
also allow some entries to be protected by a password, so you can have your
Windows boot entry without worrying about somebody boot to Windows and trash
your Linux partition (because only you, knowing the password, can boot to
Windows), and at the same time allow anybody to boot the Linux partition and
login there.

And I don't consider Grub to be much less documented than LILO.

Regards,
Isaac.


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

Re: Problems getting dialup to work

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 06:57:07PM -0500, Walter Dnes wrote:
>   I'll be moving soon, and temporarily slumming it on dialup, bleagh.  I
> set up an account on a low-cost dial-up ISP (they confirm connection is
> bog-standard, and no special software is required).  I also installed
> wvdial and set up the account.  Wvdial requires root to run !!!  Testing
> it from root, it sees the login/password prompts and responds.  The
> connection seems to be established, but nothing happens, and my external
> modem's lights stay solid.  The same machine+modem did work with Redhat
> 7.3.

dpkg --purge wvdial && apt-get install ppp pppconfig

Use pppconfig as root to twiddle your PPP settings.  Enable demand
mode and forget about it.  Might have to retry your connection after a
minute if the modem's on the hook when you try to use the net.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1o9eUzgNqloQMwcRAv+xAJ4orZeuK7pLSj+6+hcactWXYqb1qACdHsub
uOr/XHjFcw3JlbYwOu9nZAM=
=frMS
-END PGP SIGNATURE-


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

Re: Compiling a kernel without making a .deb package.

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 11:00:16PM +, Joseph Jones wrote:
> I can compile a kernel into a .deb package as described in the 
> newbiedoc, but I need to compile a kernel with drivers for my laptop's 
> NIC so I can make a rescue disc to do a network install from. Could 
> anyone advise me as to how I do this, if possible in relation to the 
> instructions in the newbiedoc?

http://ursine.ca/doc/HOWTO/en-html/Kernel-HOWTO/index.html

You will also find this and much more documentation in
/usr/share/doc/HOWTO if you have linux-doc-html installed.  Or you can
bookmark http://ursine.ca/doc/HOWTO/en-html/HOWTO-INDEX/howtos.html
for the HOWTO index.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1o6wUzgNqloQMwcRArjvAJ0YGz2v79NKi/RFSlJ5MBoaX1Z6kQCdHl96
vwrt2qfsniDjSNzVgu1M6CQ=
=VZv/
-END PGP SIGNATURE-


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

Can spamassassin remove SIGs?

[Nunya]

I've decided SIGs are spam: they are unrelated to what I'm reading this 
list for, and most of them bug the crap out of me.  Mostly because they 
break my technical concentration with some random quip of ego or dumb 
philosophy.

For now, I've set my .muttrc to color signatures as "black/black", which 
is doing a pretty good job.  But since most sigs follow quite regular 
patterns structurally as well as content-wise, a bayesian filter should 
be able to strip them effectively.

Does a tool exist to strip a message of random structral bits of spam 
within a message, and just pull out the "meat" ??

(I like most of the technical info here I just loathe reading the SIGs, 
regardless of its message or personal website links).  Tools to help me?


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

Re: Re: Newbie problem finding module

[Paul Schwartz]

* Paul Schwartz ([EMAIL PROTECTED]) [031208 10:28]:
I have just installed Woody on my system.  My first problem is that I 
can't start the X server.  Searching various archives I have determined 
that the problem is most likely associated with the i815 graphics chips 
on the MB.

lsmod does not show that agpgart is installed as a module [there is an 
entry /dev/agpgart]
modprobe -l agpgart
   doesn't return anything

the only files that show up in my installed file system are
agpgart.h
/dev/agpgart
Where do I find the files to install that module?
What kernel are you running?  If you don't know, 'uname -r' will tell
us.
 

Linux 2.4.17 i686 [ELF]

Paul Schwartz



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

Re: Tom Ballard, MSFT shill

[ScruLoose]

On Mon, Dec 08, 2003 at 02:07:17PM -0800, Tom wrote:
> On Mon, Dec 08, 2003 at 04:23:36PM -0500, ScruLoose wrote:
> > 
> > I will admit that I have pretty much given up on any expectation that
> > you have anything useful to add to this forum
> 
> I'm totally okay with that.  Leave me alone.

As seems to happen distressingly often, your conclusion does not follow
from the statements that lead up to it.

As long as you persist in posting internally inconsistent, logically
fallacious, misleading nonsense to this public forum, I and others may
respond with corrections, rebuttals, counter-arguments, etc.

I most certainly do _not_ agree to "leave you alone" to spread FUD
unchallenged.

I will, however, try to remember not to bother with responding to what's
merely meaningless or wildly irrelevant.

Just so we're clear.
-- 
---<>---
 I don't want to start any blasphemous rumours
 but I think that God's got a sick sense of humour
 - Depeche Mode
--<>--


pgp0.pgp
Description: PGP signature

Re: make-kpkg

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 10:09:48AM -0800, Michael Montagne wrote:
> I use Grub as a bootloader.  After making a kernel .deb using
> make-kpkg, I'm running dpkg -i  Near the end you are asked to if
> you want to make a boot block.  What is this?  Is it just an entry in
> Grub or LILO?  What I'm most concerned about is being able to boot to
> my old kernel if I screwed this one up.

Since LILO is so much better documented, easy to configure, far more
widely used, is generally the "assumed" bootloader, and just works,
why not just use LILO?

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1ogwUzgNqloQMwcRAvEfAKC5M7L7T6Lpgpz3KXLCDqJ6iqtyGwCfdtQt
Mc0nfXV5RlDb3XhO+/MILlI=
=dKbv
-END PGP SIGNATURE-


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

Re: New Debian installer?

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 06:30:06PM +0100, Alexander Fitterling wrote:
> I would like to know where I could get a new Debian installer - that
> one I heard wispering about. Interesting to me seems the question if
> it comes already with Sarge/testing, if downloading using the jigdo
> utility. Or must I take a look some place else at?

Nope, that's it.  That's how you get it.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1ofQUzgNqloQMwcRAq9hAKDBZUwioDrq+ZOwNSbmycFCYJq3bwCffNj6
W334mIno2HBbEhSOlo9oodc=
=SgZk
-END PGP SIGNATURE-


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

Re: Kernel modules for AIC79xx scsi card?

[Colin Watson]

On Tue, Dec 09, 2003 at 05:13:45PM -0500, adam morley wrote:
> However, I can't access the above page at people.debian.org because I
> get a redirection loop in both Mozilla and w3m.  I reported this to
> [EMAIL PROTECTED] on 3 Dec 2003, but have not received a
> response.
> 
> I'm guessing people.debian.org is hosted on glucker (from the
> redirection page I saw), which was affected by the recent compromise
> --- is this why I'm seeing redirection loops?

Yeah, people == gluck. The apache configuration should be fixed fairly
soon, I'd guess.

debian-www don't control the apache configuration on gluck, although
they've been somewhat flooded by mails about it.

Cheers,

-- 
Colin Watson  [EMAIL PROTECTED]


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

Re: Mixing woody and sarge

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Wed, Dec 10, 2003 at 12:40:21AM +0100, Benedict Verheyen wrote:
> I wonder if you could get away with this by pinning libc6 and say
> perl and some other important stuff. Or would that just result in
> apt not wanting to install the package from testing since one of
> it's dependencies aren't met?

That's it.  Then nothing from testing would work.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1oJNUzgNqloQMwcRAoMPAJ9ZxASa5pej6djaFE8FJGbbqCPI9ACgsOvQ
ZfgLXTrC6rIo41KPgSM1O/E=
=Ow1g
-END PGP SIGNATURE-


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

Re: Unable to start kde session on display :1

[Karsten M. Self]

on Tue, Dec 09, 2003 at 04:43:55PM +0100, Steinar Bang ([EMAIL PROTECTED]) wrote:
> > Mauro Darida <[EMAIL PROTECTED]>:
> 
> > Try (look at spacing):
> > startx -- :2 vt8
> 
> Does the same as the same command without the "vt8", ie. starts an X
> display with no WM, three xterms and an xclock.

That's going to depend on the existence and/or contents of ~/.xsession
or ~/.xinitrc.


Peace.

-- 
Karsten M. Self <[EMAIL PROTECTED]>http://kmself.home.netcom.com/
 What Part of "Gestalt" don't you understand?
I'm no one to be trifled with. That is all you ever need know.
- Princess Bride


pgp0.pgp
Description: PGP signature

Re: Mixing woody and sarge

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 09:59:44PM +0530, Rajkumar S wrote:
> What will happen when I add testing lines also in sources.list of a 
> stable (woody) box and apt-get  a package available in testing?

This is *way* too complex to answer in email.  Generally speaking,
this will cause you to pull a new libc6, which can make things break
if you're not updating everything.  Stable does *NOT* mix with newer
distros, avoid the temptation of using apt pinning unless you
absolutely know what the ramifications are.

Just use apt-get.org or backports.org instead.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1oInUzgNqloQMwcRAr3yAKDPAg9wcF1NinKIZOhPkEsmTJrPpACfR3xy
ZxnAchSu5iSFTuc6uEBbRQQ=
=ba13
-END PGP SIGNATURE-


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

Re: Mutt 1.3.28i and attachments

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 10:18:07AM -0700, Monique Y. Herman wrote:
> If I'm not mistaken, it only says such things if it's had to use an
> external program to display the email (such as lynx for html) ..?

No, only when reading a MIME email with tty-friendly mimetypes.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1oDUUzgNqloQMwcRAoh+AKCSYeYhrQucsSlXbKzRsNQnjAZvEACdG0Sz
lFJXd3g5UDlLSeeIEtZ4RvY=
=RJUU
-END PGP SIGNATURE-


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

Re: Mutt 1.3.28i and attachments

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 10:45:37AM +0100, Mauro Darida wrote:
> What kind of settings should I put in .muttrc to avoit Mutt displaying
> such annoying things:
> [-- Attachment #1 --]
> [-- Type: multipart/signed, Encoding:7bit, Size:1.2K --]
> 
> [-- Attachment #1 --]
> [-- Type: text/plain, Encoding:quoted-printable, Size:0.7K --]

That's just mutt's way of showing you each part of a MIME message.
I'm not sure if you can disable that, but doing so could be confusing.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1oCfUzgNqloQMwcRAsy7AKCPMxAVk4LU5yA+8B+Z/Wlv+LOb2ACeMMmc
oaqiE6JPn+AI/D0jrnXh+ZQ=
=hfMn
-END PGP SIGNATURE-


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

Re: Installing Debian question.

[Paul Johnson]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Tue, Dec 09, 2003 at 03:30:48PM +0800, Stephen Liu wrote:
> I have not downloaded other ISO CD images.  Can I install a stardard
> Debian-Linux workstation only from these 2 CDs

RTFM.  You only need the first one.
http://www.debian.org/releases/stable/installmanual

> Has any folk tried install Debian from "Anaconda for Debian"?  Does it
> install Debian from Internet?  If YES would it take lengthy to complete
> compiling from source code similar to Gentoo?

Just use the official installer.  I'm not sure the installer folks
particularly care about Anaconda, I don't follow that group and they
were already well on the way to making a better installer before RH
opened Anaconda.

- -- 
 .''`. Paul Johnson <[EMAIL PROTECTED]>
: :'  :
`. `'` proud Debian admin and user
  `-  Debian - when you have better things to do than fix a system
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/1oAMUzgNqloQMwcRAqtOAKDP4oB7HbJbWcQ5XsS7nfGZw/Bt1ACeOgwT
vahbXv9i+OVH8F9n2hGpnqc=
=DqzD
-END PGP SIGNATURE-


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

Re: process limits (was: Spamassain question, whitelist?)

[Karsten M. Self]

on Tue, Dec 09, 2003 at 10:51:26AM -0800, Vineet Kumar ([EMAIL PROTECTED]) wrote:
> * Karsten M. Self ([EMAIL PROTECTED]) [031208 19:46]:
> > on Mon, Dec 08, 2003 at 06:44:04PM -0800, Vineet Kumar ([EMAIL PROTECTED]) wrote:
> > > * Karsten M. Self ([EMAIL PROTECTED]) [031208 16:52]:
> > > > For performance reasons, I also have in /etc/security/limits:
> > > > 
> > > > mailhardnproc   30
> > > > 
> > > > ...to avoid runaway conditions when large mail loads hit.  Mail
> > > > processing will be limited to a max of 30 processes (generally 10 exim
> > > > processes, 10 spamassassin clients, and a bit of overhead), but the
> > > > system as a whole won't be bogged.
> > > 
> > > So you have spamc running as mail, and not as the destination user
> > > account?  
> > 
> > No.
> 
> As I understand the line you gave above, that limits the number of
> processes being run as the mail user.  (I'm not using
> /etc/security/limits.conf ; this is my understanding from reading the
> comments in that file.)  

Correct.

> So how does this work?  Is it that spamd forks for each client, and
> that's running as mail, and that's where the limit comes into play?

Yes.

> It looks like spamd's default behavior is to run as root.  

This is true, but its children run as 'mail'.  I think.

What I know is that the above config *does* keep a box from spawning
endless processes in response to spam swarms.


> ISTR it needs this to be able to maintain users' ~/.spamassassin files
> (auto-whitelists, Bayes DBs, etc.).

Possibly handed to the children by the root process?  I'm not sure of
the guts here.

> I'm trying to understand this better since I'm interested in setting
> this up on one of my systems, which has, in the past, fallen victim to
> what was essentially a spamassassin fork-bomb (a big sa-learn job in
> the middle of the day, without nice).

The above should help.

> good times,
> Vineet
> -- 
> http://www.doorstop.net/
> -- 
> One nation, indivisible, with equality, liberty, and justice for all.

Amen ;-)


Peace.

-- 
Karsten M. Self <[EMAIL PROTECTED]>http://kmself.home.netcom.com/
 What Part of "Gestalt" don't you understand?
   Unless you are very rich and very eccentric, you will not enjoy the
   luxury of having a computer in your own home.
 -- Ed Yourdon, _Techniques of Program Structure and Design_, 1975


pgp0.pgp
Description: PGP signature

RE: Initrd problems

[Lewis Shobbrook]

 
Thanks Herbert that did the trick,  should have guessed that from the
example...
> > # Examples:
> > #
> > #  ext2
> > #  wd io=0x300
> > FastTrak.o
> 
> Try just "FastTrak".

Cheers,

Lewis


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

Re: Help with installation

[Luigidigi]

Colin Watson wrote:
On Tue, Dec 02, 2003 at 05:52:30PM -0500, H. S. wrote:

I am new to Debian, and during installation it asks if you want to use 
two package installers, I used the first one and not the second one. The 
second one, I think it is dpkg, is way too tedious for a first timer.


That's dselect; dpkg is something you can't avoid using, even if it's
wrapped up in other tools. Some of us like dselect, actually. :) I think
the sarge installer offers aptitude, which is reputed to be rather
better. Haven't got round to trying it myself.

I wish you all the luck in the world during installation :) Once you 
have it installed, it is not that bad, actually it is pretty nice. BTW, 
if only Debian had a better installer, a HUGE bunch of Redhat users 
would now be using Debian.


There are always "if only"s, unfortunately ...

Cheers,
Try this:

http://osnews.com/story.php?news_id=2016

salud,
mario


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

Re: Initrd problems

[Herbert Xu]

On Wed, Dec 10, 2003 at 10:23:14AM +1100, Lewis Shobbrook wrote:
>
> # Examples:
> #
> #  ext2
> #  wd io=0x300
> FastTrak.o

Try just "FastTrak".
--
Debian GNU/Linux 3.0 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <[EMAIL PROTECTED]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt


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

Re: PCMCIA Woes

[Marc Wilson]

On Tue, Dec 09, 2003 at 05:17:07PM -0600, Peter Quackenbush wrote:
> I have a compaq armada m300 trying to configure a 3Com 10/100 LAN Cardbus.
> Model 3CCFE575BT.  I have tried to get pcmcia working on sarge, but to no
> avail.

It's supported by the 3c59x driver.  You *do* have that one, don't you?

> I would like it to work with dhcp. I have the same problem working
> with every kernel I've tried (even ones I've built myself.) Running
> /etc/init.d/pcmcia start activates the card and cardctl status and cardctl
> ident both show the correct card. lspci -v also shows it.

It's not controlled by PCMCIA.  Not without a lot of hackery.  It's
controlled by Kernel Card Services (most or all CardBus cards seem to be).
If that's not enabled in your kernel, it needs to be.  You need
CONFIG_HOTPLUG, CONFIG_PCMCIA, but not CONFIG_HOTPLUG_PCI.

> But when I run ifconfig, it only shows the loopback interface. I cannot
> get eth0 to show up.

It's there, it's just not configured yet.  You can see it if you explicitly
ask for eth0.

Install hotplug, read README.Debian, configure /etc/network/interfaces
appropriately, you're done.

I'd like to hear about it if you get yours to survive a suspend.  My 575 on
a Thinkpad 600E does not.  I have apmd eject it whenever the laptop goes to
sleep.  In fact, no CardBus cards do.

-- 
 Marc Wilson | "Ubi non accusator, ibi non judex."  (Where there is
 [EMAIL PROTECTED] | no police, there is no speed limit.)  -- Roman Law,
 | trans. Petr Beckmann (1971)


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

Re: Problems getting dialup to work

[Paul Morgan]

On Tue, 09 Dec 2003 18:57:07 -0500, Walter Dnes wrote:

>   I'll be moving soon, and temporarily slumming it on dialup, bleagh.  I
> set up an account on a low-cost dial-up ISP (they confirm connection is
> bog-standard, and no special software is required).  I also installed
> wvdial and set up the account.  Wvdial requires root to run !!!  Testing
> it from root, it sees the login/password prompts and responds.  The
> connection seems to be established, but nothing happens, and my external
> modem's lights stay solid.  The same machine+modem did work with Redhat
> 7.3.
> 
>   Here is the output of route -n with adsl running...
> 
> [18:34:34][/root] route -n
> Kernel IP routing table
> Destination Gateway Genmask Flags Metric RefUse Iface
> 192.168.1.248   0.0.0.0 255.255.255.248 U 0  00 eth0
> 0.0.0.0 192.168.1.254   0.0.0.0 UG0  00 eth0
> 
>   Here it is with the dialup connection established...
> 
> [18:34:31][/root] route -n
> Kernel IP routing table
> Destination Gateway Genmask Flags Metric RefUse Iface
> 64.7.158.21 0.0.0.0 255.255.255.255 UH0  00 ppp0
> 192.168.1.248   0.0.0.0 255.255.255.248 U 0  00 eth0
> 0.0.0.0 192.168.1.254   0.0.0.0 UG0  00 eth0
> 
>   Here is the wvdial dialog with userid and phone# X'd out.  Any ideas
> anyone ?
> 
> [18:33:08][/root] wvdial
> --> WvDial: Internet dialer version 1.53
> --> Initializing modem.
> --> Sending: ATZ
> ATZ
> OK
> --> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0
> ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0
> OK
> --> Modem initialized.
> --> Sending: ATDTxx
> --> Waiting for carrier.
> ATDTxx
> CONNECT 50666/ARQ/V90/LAPM/V42BIS
> --> Carrier detected.  Waiting for prompt.
> login:
> login:
> --> Looks like a login prompt.
> --> Sending: 
> 
> Password:
> --> Looks like a password prompt.
> --> Sending: (password)
> PPP session from (64.7.158.21) to 38.112.103.21 [EMAIL PROTECTED] }4}"}&} } } } 
> }%}&u}!Vs}'}"}(}"}3K~
> --> PPP negotiation detected.
> --> Starting pppd at Tue Dec  9 18:33:41 2003
> --> pid of pppd: 673
> 
>   It stalls and just sits here until I hit CTRL-C
> 
> --> Disconnecting at Tue Dec  9 18:35:04 2003
> [18:35:04][/root]

I am *not* an expert on this, but two observations:

1.  I think that maybe your routing table should have two ppp0
entries:

Destination Gateway Genmask Flags Metric RefUse Iface
64.7.158.21 0.0.0.0 255.255.255.255 UH0  00 ppp0
0.0.0.0 64.7.158.21 0.0.0.0 UG0  00 ppp0

2.  I've not used wvdial, but I have never had problems setting up dialup
accounts through pppconfig, and using pon and poff - it's very simple.

Hope this helps.

-- 
paul

I'll tell you what kind of guy I was. If you ordered a boxcar full of
sons-of-bitches and opened the door and only found me inside, you could
consider the order filled.
  -- Robert Mitchum



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

xfree won't start w/ video card sis630

[Terry]

XFree cannot start.  When I select the option to start the program to 
make adjustments, the monitor flashes from black to a small window with 
the monitor's settings and the message "out of range".

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

Re: Bug#223372: apt-zip: dynamic computation of what files we need

[Dan Jacobson]

OK, the idea is we depart our faraway Antarctic (or mere hilltop, or
Space Station) debian machine with as little information possible that
describes its state, in terms of which of sid/sarge/woody, and which
versions of packages we have installed:
COLUMNS=222 dpkg -l|awk 'NR==1,/===/{next};{print $1,$2,$3}
Also we would have a list of packages we want to add, but didn't have
the bandwidth do download.

When we arrive at the well connected site, oh weeks later, we furnish
the above information and create a, oh, CDROM or two, of the same
files we would get if we had done a 'apt-get dist-upgrade' back in
Antarctica, however with current Packages files instead.

This gives us weeks fresher packages than if we had computed what we
needed before we left the faraway site.

We return to the faraway site and install the packages.

When we were at the well connected site, we told apt: given these
packages installed, and these available, etc. etc., please compute
what I need for a dist-upgrade etc.

Apt programs like to have a "status" file.  This we can construct from
the minimal
COLUMNS=222 dpkg -l|awk 'NR==1,/===/{next};{print $1,$2,$3}
that we left home with.

But what happens when we arrive at the well connected site only to
find that they all use some other leading brand operating system... no
apt anywhere to compute what we need.

No problem, we just slide our pitiful floppy with the dpkg -l,
sid/sarge/woody choice, and additional wanted packages list, into
their computer and tell the apt-computations website to compute what
we need for download, (and ask about problem resolution too.)

Of course one might ask, why not take the whole computer or the hard
disk to the well connected site... well, as we take a kayak thru a bad
neighborhood to get there we could likely loose all along the way,
plus mom needs to use it while we are gone.

> "G" == Giacomo A Catenazzi <[EMAIL PROTECTED]> writes:

G> Dan Jacobson wrote:
>> Package: apt-zip
>> Version: 0.13.2
>> Severity: wishlist
>> Let's say I leave home with fetch-script-wget-jidanni.org for a two
>> week trip to a place with fast connections, I'll return home with
>> files no newer than two weeks.
>> If however, I left home with an extract of the dpkg-status file, and
>> some other vital snippets, I could be able to get the freshest
>> versions available for my system... but how... probably thru various
>> apt config stuff...

G> Hello.

G> I used a script that download the Packages file and from this it
G> extracts the newer version of package to download.  So I could
G> update unstable with few problems. It is no perfect, but it can
G> help.  Maybe I can implement an other script to put the complete
G> package list on zip and than remotely to see what new packages are
G> needed. This methos would have nearly no more dependence
G> problem. (but in case of packages split, packages that depends on
G> new created packages. To solve this it would to complicates, but if
G> we use the next methods)

G> apt-get documentation give some hint about use a remote debian machine to
G> calculate the new list of packages of download.

G> These methods are more complex as the actual methods, and the
G> remote machine (fast connected machine) maybe doesn't have the
G> needed tools (shell scripts are not so portable in the various unix
G> because of "creative" implementation of some features).  I'm
G> working on doing some scripts so that user can use powerfull script
G> on GNU/Linux machines and simple scripts on some old unix machines
G> (and maybe also a .bat for dos/windoze).

G> ciao
G>  giacomo

G> PS: sorry about the big delay of fixing apt-zip bugs :-( . Really I
G> should/want to correct shortly the apt-zip, and in a second step
G> upload the newfeatured apt-zip


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

PCMCIA Woes

[Peter Quackenbush]

Hi.

I have a compaq armada m300 trying to configure a 3Com 10/100 LAN Cardbus.
Model 3CCFE575BT.  I have tried to get pcmcia working on sarge, but to no
avail. I would like it to work with dhcp. I have the same problem working
with every kernel I've tried (even ones I've built myself.) Running
/etc/init.d/pcmcia start activates the card and cardctl status and cardctl
ident both show the correct card. lspci -v also shows it.

But when I run ifconfig, it only shows the loopback interface. I cannot
get eth0 to show up. Its supposed to be on dhcp.  ifconfig eth0 outputs

My /etc/network/interfaces reads,

auto lo
iface lo inet loopback

#This entry was created during the Debian Installation
auto eth0
iface eth0 inet dhcp

I did notice that when I run "/etc/pcmcia/network start eth0" I only
get (and I'm not sure if this is useful.  I can't decipher what the
script is doing, but maybe someone out there can);

usage: /etc/pcmcia/network [action] [device name]
actions: start check stop suspend resume

In other words, its acting as if I never put in any arguments in the
first place.

Running "sh -x /etc/pcmcia/network start eth0" as per the
PCMCIA-HOWTO troubleshooting part recommends gives the following output.

ACTION=start
DEVICE=eth0
NEW_SCHEME=
get_info eth0
'[' -d /var/lib/pcmcia ']'
STAB=/var/run/stab
'[' -f /var/lib/pcmcia/scheme ']'
'[' -f /var/run/pcmcia-scheme ']'
cat /var/lib/misc/pcmcia-scheme
SCHEME=
'[' -z '' '['
SCHEME=default
grep_stab eth0
local CLASS DEV
read SOCKET CLASS DRIVER INSTANCE DEV MAJOR MINOR
'[' Socket '!=' Socket -a eth0 = device '['
read SOCKET CLASS DRIVER INSTANCE DEV MAJOR MINOR
return 1
usage
echo 'usage: /etc/pcmcia/network [action] [device name]'
usage: /etc/pcmcia/network [action] [device name]
echo ' actions: start check stop suspend resume'
actions: start check stop suspend resume
exit 1

I'm guessing that the line that says, '[' Socket '!=' Socket -a
eth0 = device '[' is where the problem is, but I have no idea how to
rectify this.

Anyone have any ideas how I can get my PCMCIA card working?

Thanks in advance.

_
Peter Quackenbush


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

RE: Initrd problems

[Lewis Shobbrook]

Hi Herbert,

thanks for the reply, here's the relevant files as requested...

# /etc/mkinitrd/mkinitrd.conf:
#  Configuration file for mkinitrd(8).  See mkinitrd.conf(5).
#
# This file is meant to be parsed as a shell script.

# What modules to install.
MODULES=most

# The length (in seconds) of the startup delay during which linuxrc may
be
# interrupted.
DELAY=0

# If this is set to probe mkinitrd will try to figure out what's needed
to
# mount the root file system.  This is equivalent to the old PROBE=on
setting.
ROOT=probe

# This controls the permission of the resulting initrd image.
UMASK=022

# Command to generate the initrd image.
MKIMAGE='mkcramfs %s %s > /dev/null'
#MKIMAGE='mkext2fs %s %s > /dev/null'
# Set this to yes if you want to use busybox(1).
BUSYBOX=no

# Set this to no if you want to disable /usr/share/initrd-tools/scripts.
PKGSCRIPTS=yes 


# /etc/mkinitrd/modules: Kernel modules to load for initrd.
#
# This file should contain the names of kernel modules and their
arguments
# (if any) that are needed to mount the root file system, one per line.
# Comments begin with a `#', and everything on the line after them are
ignored.
#
# You must run mkinitrd(8) to effect this change.
#
# Examples:
#
#  ext2
#  wd io=0x300
FastTrak.o

> -Original Message-
> From: Herbert Xu [mailto:[EMAIL PROTECTED] 
> Sent: Tuesday, 9 December 2003 06:08
> To: Lewis Shobbrook
> Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
> Subject: Re: Initrd problems
> 
> On Tue, Dec 09, 2003 at 11:04:22AM +1100, Lewis Shobbrook wrote:
> > 
> > Basically the driver module is not loading and we can't 
> mount the root 
> > filesystem. The module FastTrak.o can't be found, but I've 
> mounted the 
> > rd image and can verify the inclusion. I've adjusted the 
> modules file 
> > within the /etc/mkinitrd folder, and added the module as 
> > /lib/modules/2.4.22/kernel/drivers/scsi/FastTrak.o and simply 
> > FastTrak.o, the result is the same.  It looks close, but 
> I'm unable to 
> > get around the failure to locate the module.
> > I've scrounged through what little info I can find and have 
> struck a 
> > bit of an impasse.
> > 
> > Can anyone offer any advice on how to do this correctly?
> 
> Please show me the contents of the files that you've modified 
> under /etc/mkinitrd.
> --

Cheers,

Lewis 


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

Problems getting dialup to work

[Walter Dnes]

  I'll be moving soon, and temporarily slumming it on dialup, bleagh.  I
set up an account on a low-cost dial-up ISP (they confirm connection is
bog-standard, and no special software is required).  I also installed
wvdial and set up the account.  Wvdial requires root to run !!!  Testing
it from root, it sees the login/password prompts and responds.  The
connection seems to be established, but nothing happens, and my external
modem's lights stay solid.  The same machine+modem did work with Redhat
7.3.

  Here is the output of route -n with adsl running...

[18:34:34][/root] route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric RefUse Iface
192.168.1.248   0.0.0.0 255.255.255.248 U 0  00 eth0
0.0.0.0 192.168.1.254   0.0.0.0 UG0  00 eth0

  Here it is with the dialup connection established...

[18:34:31][/root] route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric RefUse Iface
64.7.158.21 0.0.0.0 255.255.255.255 UH0  00 ppp0
192.168.1.248   0.0.0.0 255.255.255.248 U 0  00 eth0
0.0.0.0 192.168.1.254   0.0.0.0 UG0  00 eth0

  Here is the wvdial dialog with userid and phone# X'd out.  Any ideas
anyone ?

[18:33:08][/root] wvdial
--> WvDial: Internet dialer version 1.53
--> Initializing modem.
--> Sending: ATZ
ATZ
OK
--> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0
ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0
OK
--> Modem initialized.
--> Sending: ATDTxx
--> Waiting for carrier.
ATDTxx
CONNECT 50666/ARQ/V90/LAPM/V42BIS
--> Carrier detected.  Waiting for prompt.
login:
login:
--> Looks like a login prompt.
--> Sending: 

Password:
--> Looks like a password prompt.
--> Sending: (password)
PPP session from (64.7.158.21) to 38.112.103.21 [EMAIL PROTECTED] }4}"}&} } } } 
}%}&u}!Vs}'}"}(}"}3K~
--> PPP negotiation detected.
--> Starting pppd at Tue Dec  9 18:33:41 2003
--> pid of pppd: 673

  It stalls and just sits here until I hit CTRL-C

--> Disconnecting at Tue Dec  9 18:35:04 2003
[18:35:04][/root]


-- 
Walter Dnes <[EMAIL PROTECTED]>
Email users are divided into two classes;
1) Those who have effective spam-blocking
2) Those who wish they did


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

Re: debian install over ssh?

[Benedict Verheyen]

- Original Message -
>From: "Vineet Kumar" <[EMAIL PROTECTED]>
>To: <[EMAIL PROTECTED]>
>Sent: Tuesday, December 09, 2003 11:40 PM
>Subject: Re: debian install over ssh?
>
>* [EMAIL PROTECTED] ([EMAIL PROTECTED]) [031209 14:19]:
>> Hi, here's a challenge that I'm trying to do.
>>
>> I have an old remote P2 box that has no monitor, mouse, or keyboard
>> attached. It currently has sid installed, and I can access it through
>> ssh. What I'd like to do is install debian (say, Woody or Sarge) onto
>> another partition/harddrive on the same machine, and have it running
ssh
>> so that I have access when I reboot.
>
>You'll want to do a "chroot install".  It's really quite easy.  Try
>googling for "chroot install" -- I think Karsten has written up some
>good directions.
>
>good times,
>Vineet

I think Vineet is referring to this document:
http://twiki.iwethey.org/twiki/bin/view/Main/DebianChrootInstall

I think i will also try this once for another woody on the system
i'm currently using. Then i'll have a woody to play with (Is it me
or does this sound funny ;p )

Regards,
Benedict



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

Re: Mixing woody and sarge

[Benedict Verheyen]

- Original Message -
From: "Jamin W. Collins" <[EMAIL PROTECTED]>
To: "Debian User" <[EMAIL PROTECTED]>
Sent: Tuesday, December 09, 2003 5:53 PM
Subject: Re: Mixing woody and sarge


> If you don't set a Default Release or any other pinning configuration,
> the next time you do an upgrade many of your installed packages will
be
> upgraded to the version available in testing, provided they don't
> require the installation of any additional packages or the removal of
> any.  However, if a dist-upgrade is used instead your system will be
> wholely upgraded to testing.

I wonder if you could get away with this by pinning libc6 and say perl
and some other important stuff. Or would that just result in apt not
wanting to install the package from testing since one of it's
dependencies
aren't met?

Regards,
Benedict



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

Re: Mixing woody and sarge

[Benedict Verheyen]

- Original Message -
From: "Rajkumar S" <[EMAIL PROTECTED]>
To: "Derrick 'dman' Hudson" <[EMAIL PROTECTED]>
Sent: Tuesday, December 09, 2003 6:41 PM
Subject: Re: Mixing woody and sarge

> One of the most urgent need is samhain (a file integrity checker which
> can detect kernel root kits) With debian servers getting rooted this
>   has become a necessity.

I also compiled it from source. I do use aide too. Put both the
databases
from aide and samhain on a floppy and mount that readonly.
(If you already knew that, sorry for creating the noise)

Regards,
Benedict




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

Re: compiling kernel for another system

[Magnus von Koeller]

On Tuesday 09 December 2003 09:46, John L. Fjellstad wrote:
> Yeah, I figured that out after looking at the dependency list of
> the kernel-packages.  I'm wondering though if using gcc3.3 would
> have any impact (since everything else on the stable system would
> have the programs compiled with gcc2.95)

Well, that's what I've been thinking. But isn't the kernel basically 
one self-contained unit? I mean sure: The modules depend on the 
kernel but they're both compiled with the same compiler anyway, 
right?

Apart from that: The kernel is pure C, isn't it? The changees from 
gcc2.95 to gcc3.2 were in the C++ ABI, they didn't affect C 
compatibility IIRC. So it really doesn't make much sense to have any 
unresolved symbols here - and, as I said, at least for me it worked.

Anyways, I'm really no expert on this. But it's been really 
interesting to discuss it - always something to learn on this list!

Best regards,
Magnus

-- 
---  Magnus von Koeller ---
email:[EMAIL PROTECTED]
address:  International University
  Campus 9, App. 13
  D-76646 Bruchsal / Germany
phone:+49-7251-700-659
mobile:   +49-179-4562940
web:  http://www.vonkoeller.de   


pgp0.pgp
Description: signature

Re: Debian IPv6 Ready?

[Robert L. Harris]

If you run the stock kernels that may be, sorry I run custom kernels and
didn't compile in the IPv6 until I was ready to mess with it.


Thus spake Vineet Kumar ([EMAIL PROTECTED]):

> * Robert L. Harris ([EMAIL PROTECTED]) [031209 14:52]:
> > 
> > You'll need to compile and IPv6 kernel, you don't need any patches, it
> > works fine out of the box.  While waiting on this to compile poke around
> > tdoi.org and ipv6.org.
> 
> Is this so?  On my laptop I just modprobed ipv6 and I was off and
> running -- no recompilation needed.
> 
> good times,
> Vineet
> -- 
> http://www.doorstop.net/
> -- 
> I am mortified to be told that, in the United States of America, the sale of a
> book can become a subject of inquiry, and of criminal inquiry too.
>   -- President Thomas Jefferson



:wq!
---
Robert L. Harris | GPG Key ID: E344DA3B
 @ x-hkp://pgp.mit.edu
DISCLAIMER:
  These are MY OPINIONS ALONE.  I speak for no-one else.

Life is not a destination, it's a journey.
  Microsoft produces 15 car pileups on the highway.
Don't stop traffic to stand and gawk at the tragedy.


signature.asc
Description: Digital signature

Re: Debian IPv6 Ready?

[Vineet Kumar]

* Robert L. Harris ([EMAIL PROTECTED]) [031209 14:52]:
> 
> You'll need to compile and IPv6 kernel, you don't need any patches, it
> works fine out of the box.  While waiting on this to compile poke around
> tdoi.org and ipv6.org.

Is this so?  On my laptop I just modprobed ipv6 and I was off and
running -- no recompilation needed.

good times,
Vineet
-- 
http://www.doorstop.net/
-- 
I am mortified to be told that, in the United States of America, the sale of a
book can become a subject of inquiry, and of criminal inquiry too.
-- President Thomas Jefferson


signature.asc
Description: Digital signature

Re: Could not load OpenGL library

[Arnt Karlsen]

On Tue, 09 Dec 2003 20:50:25 +0100, 
Demonen of z0rZ <[EMAIL PROTECTED]> wrote in message 
<[EMAIL PROTECTED]>:

> I'm not a subscriber to this list, but I've found a fix for this
> problem and I want to contribute.
> I see ALOT of people have this problem!
> 
> As you can see HERE:
> http://www.nvnews.net/vbulletin/showthread.php?s=032e6e5b1ad658b6060d2719b06665cb&threadid=20524&highlight=Could+not+load+OpenGL+library
> 
> the following command (as root) will fix it:
> 
> rm -rf /usr/lib/tls
> 
> That's it!
> There are some conflicting libs in there that cause it not to
> function. Or something.
> 
> Spread the word!
> 
> -
> Fredrik Vold
> aka Demonen
> 

..a wee word of caution, this forum is the "nV News Forums > 
Linux Forums > Linux And NVIDIA Graphics > "Could not
load OpenGL library" with 4496 ", you might wanna consider 
trying 'mv -vf /usr/lib/tls /usr/lib/TEMPORARILY-DISABLED-tls ' 
or somesuch, before the more brutal rm -rf.


-- 
..med vennlig hilsen = with Kind Regards from Arnt... ;-)
...with a number of polar bear hunters in his ancestry...
  Scenarios always come in sets of three: 
  best case, worst case, and just in case.



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

Re: kernel-images

[Geoff Thurman]

On Tuesday 09 December 2003 9:08 pm, Andreas Janssen wrote:
> Hello
>
> Geoff Thurman (<[EMAIL PROTECTED]>) wrote:
> > Is it possible for the unwitting to install a kernel-image
> > downloaded from official debian sources that hasn't been patched
> > for the recent exploit, or can all the currently downloadable
> > images (and kernel source packages too, for that matter) be taken
> > to be safe from it? I've
> > switched to woody, and have today installed image-2-4-18-k6  #1,
> > dated Apr 14 2002. Clearly the date suggests no patch has been
> > applied, so is this kernel vulnerable to the exploit, please, or
> > does it not arise in this branch?
>
> Your Kernel is vulnerable. When the ptrace bug was fixed, the
> packages became incompatible to modules compiled for older versions,
> and they were renamed. Install kernel-image-2.4.18-1-k6 from
> security.debian.org. The current version from ftp.debian.org (Woody
> r2) does /not/ fix all vulnerabilities (I even think it is still the
> same one as in Woody r1 because newer packages were rejected from r2
> for some reasons).
>
> best regards
> Andreas Janssen

Sheesh. Thank you. The new one is downloading now.

Cheers,

Geoff


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

Kernel modules for AIC79xx scsi card?

[adam morley]

Hi,

I'm planning on installing Debian on a Sun Fire V65x to see how I like Debian.  I 
noticed that the default kernel for Debian stable doesn't include support for the 
ethernet card (e1000) or the SCSI card (aic79xx).  This is understandable, given the 
age of the kernel (aic79xx wasn't in 2.4.18).

I (luckily) found this page:

http://people.debian.org/~blade/install/preload/

Which (supposedly) contains handy modules for installation, as mentioned here:

http://lists.debian.org/debian-user/2003/debian-user-200307/msg00413.html

However, I can't access the above page at people.debian.org because I get a 
redirection loop in both Mozilla and w3m.  I reported this to [EMAIL PROTECTED] on 3 
Dec 2003, but have not received a response.

I'm guessing people.debian.org is hosted on glucker (from the redirection page I saw), 
which was affected by the recent compromise --- is this why I'm seeing redirection 
loops?

Is there a mirror of the modules page somewhere?  Or will it be back online sometime 
this month?  Or should I just go build my own modules?  If I should build my own 
modules, where can I get the source for the Debian bf24 kernel on the stable media --- 
I'm on a gentoo box right now, so I don't have the debian tools but I can probably 
compile them if needed.

Thanks!

-- 
adam

pgp key at:
http://fedora.cwru.edu/axm135.key


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

Re: Info Problem

[Arnt Karlsen]

On Tue, 9 Dec 2003 21:02:46 +, 
Colin Watson <[EMAIL PROTECTED]> wrote in message 
<[EMAIL PROTECTED]>:

> On Tue, Dec 09, 2003 at 08:23:19PM +0100, Arnt Karlsen wrote:
> > On Tue, 9 Dec 2003 17:59:09 +, 
> > Colin Watson <[EMAIL PROTECTED]> wrote in message 
> > <[EMAIL PROTECTED]>:
> > > Eh? The code is GPL. The person in question was perfectly aware of
> > > this and licensed his code the same way. *No* liberties are being
> > > taken! 
> > 
> > .._we_ know that.  His _heirs_, does not neccessarily know,
> > understand, or appreciate, this fact.  Which again _may_ cause such
> > bad feelings. 
> > 
> > ..myself, I prefer to err on the side of caution, such bad feelings
> > _can_ be prevented, both by removal of dead mens code, and 
> > by asking the heirs.  Removal and rewriting, is of course the 
> > safest approach, but forces "re-inventing the wheel", I like to see
> > re-innovation happen on technological merits alone, not just 
> > because someone dies.
> 
> Now I wish I'd never mentioned this. The case is no different from
> code contributed in any other way. Go and dig up copies from
> /usr/share/doc/man-db/copyright from back when Fabrizio was
> maintaining it if you care; I don't have an interest in discussing

..I do care, I prefer us chewing each other out than risk anyone doubt
anyones copyright unwarrantedly, it also has the benefit of making our
position on it clear as glass. ;-)

> this further when you haven't read the text he himself wrote which
> makes the legal situation perfectly clear.

..I agree, he wrote: "The primary copyright holder is Graeme Wilford.".

..http://www.google.com/search?num=100&q=copyright+%22Fabrizio+Polacco+
%3Cfpolacco%40debian.org%3E%22+%22man-db%22++-%22the+late+%22
produced  /usr/share/doc/man-db/copyright files like this below:
"This is the Debian GNU/Linux prepackaged version of the "man_db" Manual
Page Browser Utility Package Version 2.3.12 ; it was written by Graeme
Wilford<[EMAIL PROTECTED]>, with documentation reorganized by
Nils Magnus <[EMAIL PROTECTED]> .

The author has stoped work on it and it is actually maintained by
Fabrizio Polacco <[EMAIL PROTECTED]>.

This package is a native Debian package.

This program is under the GNU General Public License, Version 2.
The primary copyright holder is Graeme Wilford. Portions are copyrighted
by Free Software Foundation, Inc. Portions are copyrighted by John W.
Eaton.

   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; version 2 dated June, 1991.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   A copy of the GNU General Public License is available in
   /usr/share/common-license/GPL (as installed by package base-files);
   if not, write to the Free Software Foundation, Inc.,
   59 Temple Place - Suite 330, Boston, MA 02111-1307, USA."

..above file quote is a wee bit mangled by my 72 character line wraps.

> EOThread for me.

..agreed, my note of disagrement on the appropriateness of approaching
copyright heirs is not specific to man-db, we should always make sure
such heirs knows our position on their copyrights future use.  

..and, as John Hasler <[EMAIL PROTECTED]> wrote on Tue, 09 Dec 2003
10:59:40 -0600, in message <[EMAIL PROTECTED]>:

> I wrote:
> > I don't think [the right of reversion] has any effect at all [on
> > Free Software].  I believe it only affects transfers, not licenses.
> 
> I just did some research and it appears that I was wrong. See
> 
> 
> Looks like the best thing to do about this damn thing is to leave your
> copyrights to the FSF in your will.

-- 
..med vennlig hilsen = with Kind Regards from Arnt... ;-)
...with a number of polar bear hunters in his ancestry...
  Scenarios always come in sets of three: 
  best case, worst case, and just in case.


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

Compiling a kernel without making a .deb package.

[Joseph Jones]

I can compile a kernel into a .deb package as described in the 
newbiedoc, but I need to compile a kernel with drivers for my laptop's 
NIC so I can make a rescue disc to do a network install from. Could 
anyone advise me as to how I do this, if possible in relation to the 
instructions in the newbiedoc?

Thanks :)

Joe

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

Re: Debian IPv6 Ready?

[Robert L. Harris]

You'll need to compile and IPv6 kernel, you don't need any patches, it
works fine out of the box.  While waiting on this to compile poke around
tdoi.org and ipv6.org.

Then you need to add this to your apt-sources:

# http://debian.fabbione.net/ <- read as well
deb http://debian.concepts.nl/debian-ipv6 woody ipv6
deb http://debian.concepts.nl/debian-ipv6 sarge ipv6
deb http://debian.concepts.nl/debian-ipv6 sid ipv6





Thus spake Frank A. Uepping ([EMAIL PROTECTED]):

> Hello,
> is Debian IPv6 ready?
> What additional packages need I to install?
> 
> /FAU
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 

:wq!
---
Robert L. Harris | GPG Key ID: E344DA3B
 @ x-hkp://pgp.mit.edu
DISCLAIMER:
  These are MY OPINIONS ALONE.  I speak for no-one else.

Life is not a destination, it's a journey.
  Microsoft produces 15 car pileups on the highway.
Don't stop traffic to stand and gawk at the tragedy.


signature.asc
Description: Digital signature

Re: Virtual PC 5.2

[Kent West]

Derrick 'dman' Hudson wrote:

On Tue, Dec 09, 2003 at 01:05:01PM -0600, Michael Martinell wrote:
| I have tried the vga driver with no success.
| 
| How do you get the busID?  I have tried searching google but have came up
| empty.

lspci.

For example, a machine here at work needed me to specify the busID.
lspci on that machine reports :
01:00.0 Display controller: Texas Instruments TVP4020 [Permedia 2] (rev 01)
^^^
On that machine, in XF86Config, I have
   Section "Device"
   Identifier  "Standard VGA"
   VendorName  "Unknown"
   BoardName   "Unknown"
   Driver "glint"
   #Chipset"ti_pm2"
   #BusID  "PCI:01:00:0"
   #VideoRam   8192
   #Clocks 25.2 28.3
   EndSection
 

And be aware that one of the numbers is in hex while the other is in 
decimal; I forget which is which. (In this case it wouldn't matter, but 
in some it would.)

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

Re: debian install over ssh?

[Vineet Kumar]

* [EMAIL PROTECTED] ([EMAIL PROTECTED]) [031209 14:19]:
> Hi, here's a challenge that I'm trying to do.
> 
> I have an old remote P2 box that has no monitor, mouse, or keyboard 
> attached. It currently has sid installed, and I can access it through 
> ssh. What I'd like to do is install debian (say, Woody or Sarge) onto 
> another partition/harddrive on the same machine, and have it running ssh 
> so that I have access when I reboot.

You'll want to do a "chroot install".  It's really quite easy.  Try
googling for "chroot install" -- I think Karsten has written up some
good directions.

good times,
Vineet
-- 
http://www.doorstop.net/
-- 
I am mortified to be told that, in the United States of America, the sale of a
book can become a subject of inquiry, and of criminal inquiry too.
-- President Thomas Jefferson


signature.asc
Description: Digital signature

debian install over ssh?

[[EMAIL PROTECTED]]

Hi, here's a challenge that I'm trying to do.

I have an old remote P2 box that has no monitor, mouse, or keyboard 
attached. It currently has sid installed, and I can access it through 
ssh. What I'd like to do is install debian (say, Woody or Sarge) onto 
another partition/harddrive on the same machine, and have it running ssh 
so that I have access when I reboot.

I have user AND root accounts on both machines (P2 and box I'm using) 
and can install/configure anything anyway I need. (i.e. putting 
installer programs on remote machine) I don't have physical access to 
the P2 though.

Any ideas?

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

Debian IPv6 Ready?

[Frank A. Uepping]

Hello,
is Debian IPv6 ready?
What additional packages need I to install?

/FAU


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

Re: kernel-images

[Andreas Janssen]

Hello

Geoff Thurman (<[EMAIL PROTECTED]>) wrote:

> Is it possible for the unwitting to install a kernel-image downloaded
> from official debian sources that hasn't been patched for the recent
> exploit, or can all the currently downloadable images (and kernel
> source packages too, for that matter) be taken to be safe from it?
> I've
> switched to woody, and have today installed image-2-4-18-k6  #1, dated
> Apr 14 2002. Clearly the date suggests no patch has been applied, so
> is this kernel vulnerable to the exploit, please, or does it not arise
> in this branch?

Your Kernel is vulnerable. When the ptrace bug was fixed, the packages
became incompatible to modules compiled for older versions, and they
were renamed. Install kernel-image-2.4.18-1-k6 from
security.debian.org. The current version from ftp.debian.org (Woody r2)
does /not/ fix all vulnerabilities (I even think it is still the same
one as in Woody r1 because newer packages were rejected from r2 for
some reasons).

best regards
Andreas Janssen

-- 
Andreas Janssen
[EMAIL PROTECTED]
PGP-Key-ID: 0xDC801674
Registered Linux User #267976


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

weird problem: X restarts if I click in an ssd'ed xterm

[H. S.]

Hi,

Last night I tried to work from home on my university's network. First 
the command "ssh -f [EMAIL PROTECTED] xterm -sl 2048 -fg beige" didn't work. 
I got "can't open display" error message. So I did some google search 
and tried "ssh -X [EMAIL PROTECTED] xterm -sl 2048 -fg beige". But when I 
right-click on the xterm that is opened, surprisingly and unexpectedly X 
restarts -- well actually I get logged out and the nvidia screen 
splashed and I am given the Debian login screen again, just as if I had 
logged out.

I am running 2.4.22 kernel on Sarge, compiled for the nvidia 4496 
driver. I am not sure where to start debugging this thing, so if I am 
missing any diagnostics I should have posted, just let me know. I will 
post them next time I log into Debian, at present I am in Fedora -- I 
*got* to get smoe work done, right? Can't keep on struggling Debain all 
the time hee hee :))

regards,
->HS
--
(Remove all underscores from my email address to get the correct one. 
Apologies for the inconvenience, but this is to reduce spam.)



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

Re: making a floppy for essential modules

[James Williamson]

On Tuesday 09 Dec 2003 7:31 pm, New Disorder Records wrote:
> thanks, but I can't install normally if I can't see the hard drive.  The
> installer (any one of the flavors of installer, including bf24) doesn't
> see the harddrive on my 1750, so I need to install a driver at that point,
> I assume, in order to continue the installation.
>   So, I can make it to step 2 of your steps, but not step 3.  I have a
> driver, I just need to know how to make a disk that the installer will
> recognize to load it.

I've been struggling with the same thing, 1750's come with Fusion
MPT scsi host adapters which none of the woody install kernels understand. I 
tried the 'Preload...modules' install bit with a MPT module and I got a load 
of undefined modules messages so I ended up replacing the 
kernel on the rescue floppy with one with Fusion MPT support compiled in. My 
scsi drives are now detected but getting the Broadcom network cards working 
is proving a serious headache. I've tried using a kernel with support for the 
Broadcom driver built in (using the Dell provided patch) and as a module, all
I'm getting is messages saying 'No such device' when I insmod it. I'd really 
appreciate it if anyone out there has got a 1750 with Debian installed would
let me know how they managed it.

James 

>
>  On Tue, 9 Dec 2003,
>
> Michael Martinell wrote:
> > Here is how I set up my Dell
> > Set up Debian 3.0r1 on Dell Poweredge 2400
> >
> > 1.  Insert Debian disk and boot from it.
> > 2.  At boot: prompt type bf24 and press enter.
> > 3.  Complete rest of install normally.  Make sure to install the c
> > compiler.
> > 4.  Download most current kernel.  For this example it was 2.4.20
> > 5.  Copy kernel to /usr/src
> > 6.  tar xzvf linux-2.4.20.tar.gz
> > 7.  rm linux
> > 8.  ln -s linux-2.4.20 linux
> > 9.  ls -al  (verify symbolic link you just created)
> > 10. cd /usr/src/linux
> > 11. make menuconfig
> > 12. Select your options in the menu
> > 13. Store configuration to file.  This is your backup.
> > 14. make dep
> > 15. make clean
> > 16. make bzImage  (ignore warning message)
> > 17. make modules
> > 18. make modules_install
> > 19. make install
> > 20. vi /etc/lilo.conf
> > 21. Verify settings are correct.
> > lilo.conf
> > Samba:/etc# more lilo.conf
> > # /etc/lilo.conf - See: `lilo(8)' and `lilo.conf(5)',
> > # ---   `install-mbr(8)', `/usr/share/doc/lilo/',
> > #   and `/usr/share/doc/mbr/'.
> > # +---+
> > # |!! Reminder !! |
> > # |   |
> > # | Don't forget to run `lilo' after you make changes to this |
> > # | conffile, `/boot/bootmess.txt', or install a new kernel.  The |
> > # | computer will most likely fail to boot if a kernel-image  |
> > # | post-install script or you don't remember to run `lilo'.  |
> > # |   |
> > # +---+
> > # Support LBA for large hard disks.
> > #
> > lba32
> > # Overrides the default mapping between harddisk names and the BIOS'
> > # harddisk order. Use with caution.
> > #disk=/dev/hde
> > #bios=0x81
> > #disk=/dev/sda
> > #bios=0x80
> > # Specifies the boot device.  This is where Lilo installs its boot
> > # block.  It can be either a partition, or the raw device, in which
> > # case it installs in the MBR, and will overwrite the current MBR.
> > #
> > boot=/dev/sda
> > # Specifies the device that should be mounted as root. (`/')
> > #
> > root=/dev/sda1
> > # Enable map compaction:
> > # Tries to merge read requests for adjacent sectors into a single
> > # read request. This drastically reduces load time and keeps the
> > # map smaller.  Using `compact' is especially recommended when
> > # booting from a floppy disk.  It is disabled here by default
> > # because it doesn't always work.
> > #
> > # compact
> > # Installs the specified file as the new boot sector
> > # You have the choice between: bmp, compat, menu and text
> > # Look in /boot/ and in lilo.conf(5) manpage for details
> > #
> > install=/boot/boot.b
> > # Specifies the location of the map file
> > #
> > map=/boot/map
> > # You can set a password here, and uncomment the `restricted' lines
> > # in the image definitions below to make it so that a password must
> > # be typed to boot anything but a default configuration.  If a
> > # command line is given, other than one specified by an `append'
> > # statement in `lilo.conf', the password will be required, but a
> > # standard default boot will not require one.
> > #
> > # This will, for instance, prevent anyone with access to the
> > # console from booting with something like `Linux init=/bin/sh',
> > # and thus becoming `root' without proper authorization.
> > #
> > # Note that if you really need this type of security, you will
> > # Note

Re: Virtual PC 5.2

[Derrick 'dman' Hudson]

On Tue, Dec 09, 2003 at 01:05:01PM -0600, Michael Martinell wrote:
| I have tried the vga driver with no success.
| 
| How do you get the busID?  I have tried searching google but have came up
| empty.

lspci.

For example, a machine here at work needed me to specify the busID.
lspci on that machine reports :

01:00.0 Display controller: Texas Instruments TVP4020 [Permedia 2] (rev 01)
^^^

On that machine, in XF86Config, I have
Section "Device"
Identifier  "Standard VGA"
VendorName  "Unknown"
BoardName   "Unknown"
Driver "glint"
#Chipset"ti_pm2"
#BusID  "PCI:01:00:0"
#VideoRam   8192
#Clocks 25.2 28.3
EndSection

Hmm, I guess it doesn't need the BusID right now.  ISTR it compalined
about not having it when I set it up the first time.


As for downgrading, I don't know.  I would expect the 'vga' driver to
work, in VGA mode, on any VGA-compatible hardware.  I'm surprised that
it doesn't work at all on your VirtualPC.  I've never put linux in a
VirtualPC, so I can't help with experience.

-D

-- 
Reckless words pierce like a sword,
but the tongue of the wise brings healing.
Proverbs 12:18
 
www: http://dman13.dyndns.org/~dman/jabber: [EMAIL PROTECTED]


pgp0.pgp
Description: PGP signature

Re: X configuration error

[Andreas Janssen]

Hello

Vivek Kumar (<[EMAIL PROTECTED]>) wrote:

> I was trying to configure X and I got following error message in my
> log:
> 
> (EE) Screen(s) found, but none have a usable configuration.
> 
> Fatal server error:
> no screens found

You must give us more information, including:

the version of XFree you are using
your graphic card
the "Device" and "Screen" sections of your configuration file
more of your output from startx, look for lines beginning with (EE)

If you want to configure XFree, I recommend using debconf:

dpkg-reconfigure xserver-xfree86

Note: I did get that error message because I selected "Use Kernel
Framebuffer".

best regards
Andreas Janssen

-- 
Andreas Janssen
[EMAIL PROTECTED]
PGP-Key-ID: 0xDC801674
Registered Linux User #267976


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

Re: how long it takes an unstable package to come into testing?

[Greg Folkert]

On Tue, 2003-12-09 at 18:48, J.S.Sahambi wrote:
> I was going to install Debian on a new machine. Currently I am using 
> unstable. I would  like to know how long (I mean days) it takes a 
> package to be moved from Unstable to Testing?
> 
> Just wanted to know this information to have an idea how long I will 
> have to wait for new packages if I am using Testing.
Varies widely from a few days to never.

It is not a matter of how long... but if everything is met on deps and
then it builds correctly. Some packages are at the end of the chain in a
long series of deps... there are a few packages... that are still
waiting since early last month (that *I* know of).

Recently Evolution wasn't installable in testing, dunno if it is still
that way or not... but in Unstable it's been 1.4.5-3 for quite a while.

-- 
[EMAIL PROTECTED]
REMEMBER ED CURRY! http://www.iwethey.org/ed_curry


signature.asc
Description: This is a digitally signed message part

Re: Kde3.2beta for Sarge/testing: any apt repositories?

[Colin Watson]

On Tue, Dec 09, 2003 at 06:35:10PM +0100, Alexander Fitterling wrote:
> Where could I get binaries from Kde3.2beta for Sarge/testing? I'd wish to take 
> mirrors from as I use a recursive wget and prepare iso images out of it. This 
> actually worked great so far. But I can't find a valid apt-repository for 
> Sarge and Kde3.2. By the way will Kde3.2 be part of Sarge if becoming stable?

Not unless the KDE maintainers get a move on in a big way ... they
haven't even got round to getting 3.1.4 working properly yet. I suspect
3.2 will have to wait for sarge+1.

-- 
Colin Watson  [EMAIL PROTECTED]


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

Re: Info Problem

[Colin Watson]

On Tue, Dec 09, 2003 at 08:23:19PM +0100, Arnt Karlsen wrote:
> On Tue, 9 Dec 2003 17:59:09 +, 
> Colin Watson <[EMAIL PROTECTED]> wrote in message 
> <[EMAIL PROTECTED]>:
> > Eh? The code is GPL. The person in question was perfectly aware of
> > this and licensed his code the same way. *No* liberties are being
> > taken! 
> 
> .._we_ know that.  His _heirs_, does not neccessarily know, understand,
> or appreciate, this fact.  Which again _may_ cause such bad feelings. 
> 
> ..myself, I prefer to err on the side of caution, such bad feelings
> _can_ be prevented, both by removal of dead mens code, and 
> by asking the heirs.  Removal and rewriting, is of course the 
> safest approach, but forces "re-inventing the wheel", I like to see
> re-innovation happen on technological merits alone, not just 
> because someone dies.

Now I wish I'd never mentioned this. The case is no different from code
contributed in any other way. Go and dig up copies from
/usr/share/doc/man-db/copyright from back when Fabrizio was maintaining
it if you care; I don't have an interest in discussing this further when
you haven't read the text he himself wrote which makes the legal
situation perfectly clear.

EOThread for me.

Cheers,

-- 
Colin Watson  [EMAIL PROTECTED]


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

Re: sudo su gives root without prompting for a password

[Stephen Touset]

On Tue, 2003-12-09 at 09:16, Benedict Verheyen wrote:
> > On Sun, 2003-12-07 at 19:33, Stephen Touset wrote:
> 
> > My suggestion? If being able to use "su" without a password gives you
> > the heebie-jeebies (as well it should), then be far more restrictive in
> > what you allow in /etc/sudoers. After all, if you're just going to allow
> > complete access with "sudo", you might as well just use "su".
> 
> Well, that's my problem: i don't know how i could tighten this some more
> My /etc/sudoers file looks like this:
>root ALL=(ALL)ALL
>benedict ALL= NOPASSWD: /usr/bin/find, /bin/cpio,
>   /home/benedict/scripts/backup, /bin/echo, PASSWD: ALL

Yes, but *benedict's* password--not root's. Also, the way sudo is
designed, it will only ask you your own password once (until you close
the shell). If you want to tighten it, remove the "PASSWD: ALL" portion.
What that clause says is essentially, "Allow the user to run any command
as root, as long as his password was specified to sudo at least once in
the current shell."

If you do want the functionality of being able to sudo any command
(you're using sudo just so benedict can execute find, cpio, echo, and
backup passwordless), then your current setup should be alright. If
someone were to theoretically get into benedict's account, and attempt
to use sudo to execute commands as root, he would still be required to
enter your password at least once.

> For my regular user only the commands find, cpio, backip and echo are
> allowed without a password. To execute the other commands a password is
> needed.
> The only way i can see right now to tighten this is to remove the (ALL)ALL
> from root and specify what commands can be run there.
> Or am i seeing this wrong? I have to admit i'm confused about this.
> Sudo seems to give easier access to root as opposed to when i didn't have
> it installed and used su to become root to install/manage things.
> 
> Any suggestions or references to docs showing how to tighten things up
> with sudo are welcome.

I don't know any off the top of my head. However, what I said above
should suffice. The offending part is the "PASSWD: ALL" clause.

> Thanks,
> Benedict
-- 
Stephen Touset <[EMAIL PROTECTED]>


signature.asc
Description: This is a digitally signed message part

Re: backports.org routing fixed

[Nate Duehr]

On Tue, Dec 09, 2003 at 01:34:17PM -0500, Mike Dresser wrote:
> > Hopefully this will clearn up the routing issues for anyone else still
> > having problems.
> 
> It's working here now from this host.
> 
> However, I'm still seeing it drop in the i-p-x.de addresses from a
> cgocable.net address.  It's working when it goes through the eurorings.net
> network fine though.
> 
> ge-7-0.c12008gsr-1.nbg-dhk1.core.i-p-x.de (212.123.97.70) is the last hop
> I get to.

Darn.  Since it's working from both my network connections now I have no
reason to bother the routing guys at either ISP anymore, but it does
appear from your post that there are still some route problems over
there in that network.  Perhaps someone who's a customer can report it.

-- 
Nate Duehr <[EMAIL PROTECTED]>


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

Re: Could not load OpenGL library

[Demonen of z0rZ]

I'm not a subscriber to this list, but I've found a fix for this problem
and I want to contribute.
I see ALOT of people have this problem!

As you can see HERE:
http://www.nvnews.net/vbulletin/showthread.php?s=032e6e5b1ad658b6060d2719b06665cb&threadid=20524&highlight=Could+not+load+OpenGL+library

the following command (as root) will fix it:

rm -rf /usr/lib/tls

That's it!
There are some conflicting libs in there that cause it not to function. 
Or something.

Spread the word!

-
Fredrik Vold
aka Demonen


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

Re: In my defense! (was: Getting rid of MS XP)

[Nate Duehr]

On Tue, Dec 09, 2003 at 09:32:15AM -0600, Hugo Vanwoerkom wrote:
> What is a *good* sysadmin database that records all actions and reasons 
> of the admin, with roadmap? I use notebooks, files, html, (example attached)

I think CVS is an excellent way to track server changes, surprisingly.

Check in every change you make with appropriate comments -- maybe in a
multi-server environment using a centralized pserver CVS box and each
mahine having its own project name.

Then when you wonder when and what you did you can peruse the CVS
history and comments and find out exactly what you were thinking...

-- 
Nate Duehr <[EMAIL PROTECTED]>


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

kernel-images

[Geoff Thurman]

Hello,

Is it possible for the unwitting to install a kernel-image downloaded 
from official debian sources that hasn't been patched for the recent 
exploit, or can all the currently downloadable images (and kernel 
source packages too, for that matter) be taken to be safe from it? I've 
switched to woody, and have today installed image-2-4-18-k6  #1, dated 
Apr 14 2002. Clearly the date suggests no patch has been applied, so is 
this kernel vulnerable to the exploit, please, or does it not arise in 
this branch?

Cheers,

Geoff


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

No email for some users

[Jeremy Brooks]

hello,
I have a woody box that acts as a web/email server.  I host some sites
for other people, using the ~username functionality of apache.  By
default, those users can send and receive email.  I would prefer that
they do not have access to email...is there a way to specify that
certain users do not have email access?  I'm using exim 3 and qpopper to
allow mail retrieval.

Thanks.
jeremy




signature.asc
Description: This is a digitally signed message part

Re: start program at login

[s. keeling]

Incoming from John L. Fjellstad:
> Wayne Topa wrote:
> 
> > Have you tried adding it to your .xinitrc?
> 
> I tried. I forgot to mention that I'm logging in through kdm.  As far as I
> know, neither local .xsession nor .xinitrc is called during login.

Speaking of which, how does this work?  I've been trying to:

   xconsole -file /dev/xconsole -daemon -notify -exitOnFail &

I put it into my ~/.xsession, to no effect.  I've tried to tell Gnome
to manage it instead, to no effect.  The only way I can make it work
is "su -c ..."

Why do all these windowing environments insist on doing things
differently?  Shouldn't Debian's X startup stuff all just consult the
same damn startup files, instead of picking their own?  What do I work
with in Gnome?  .xsession, .xinitrc, .xserverrc, or something else?
What happens if I (well, it _could_ happen) login with kde instead?
Will kde know anything about the startup preference changes I've made?

What's the rationale behind this failure to communicate?  If this is
annoying to me (a long time user), what's it look like to a new user?


-- 
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: start program at login

[John L. Fjellstad]

Chris Everts wrote:

> I sometimes use the session for that. Open the app you want to start at
> login, save session. Next time WindowMaker you login to wm that app will
> start too. You can use "attributes | icon&initial workspace" to sent it
> to any workspace you want to.

Wouldn't this be windowmanager specific? I know GNOME has a save the
session, although I don't think it saves console apps.  Problem is that I
haven't found a windowsmanager I really like yet, so I tend to jump between
GNOME, KDE, Enlightenment and WindowMaker (depending on which one irritates
me the least this week).

-- 
John L. Fjellstad
web: http://www.fjellstad.org/  Quis custodiet ipsos custodes


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

Re: clustering software

[Alvin Oga]

On Tue, 9 Dec 2003, Rus Foster wrote:

> Hi,
> Does anyone know if there is any clustering software for 2.6? I know mosix
> but there doesn't seem to be 2.6 support

2.4 or 2.6 really shouldnt make much difference...

cluster[ing] sw
http://www.Linux-Consulting.com/Cluster

c ya
alvin


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

Re: start program at login

[John L. Fjellstad]

Oliver Elphick wrote:

> It should go in ~/.xsession or ~/.xinitrc (depending on how X is
> started).  Generally those files are linked.

Neither is called when using a (k)dm (forgot to mention that I used that,
sorry).  
I did find a solution though.  Looking at the startup procedure, all scripts
in /etc/X11/xsession.d is called, so I could just put in a script called
xstartup which call a custom script in the userdirectory which does what I
want.  Seems to be a good solution.

-- 
John L. Fjellstad
web: http://www.fjellstad.org/  Quis custodiet ipsos custodes


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

Re: Mixing woody and sarge

[Hugo Vanwoerkom]

Rajkumar S wrote:
Hi,

What will happen when I add testing lines also in sources.list of a 
stable (woody) box and apt-get  a package available in testing? For 
example ulogd.

After that will the box be stable (woody), with just that package (and 
dependencies) from testing? What happens when a security updates comes 
in security.debian.org for that package? Will it gets installed or will 
it take the updates from testing?

What happens when I do apt-get upgrade, will all packages get updated to 
testing? what about apt-get dist-upgrade?

As you can see I am a bit confused about the whole thing, Can some one 
throw some light on this?

raj


Stay with one flavor as was pointed out already and if needed use a 
backport of one particular package. Was that pointed out?  Did I miss 
it? :-) But you can google for scores of examples of this.

Hugo.



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

Re: make-kpkg

[David Z Maze]

Michael Montagne <[EMAIL PROTECTED]> writes:

> I use Grub as a bootloader.  After making a kernel .deb using
> make-kpkg, I'm running dpkg -i  Near the end you are asked to if
> you want to make a boot block.  What is this?  Is it just an entry in
> Grub or LILO?  What I'm most concerned about is being able to boot to
> my old kernel if I screwed this one up.

I think that's specifically a hook to LILO; I always say "no", and if
you're using GRUB, you don't need to reinstall the boot block when you
update kernels.  You might leave an entry in /boot/grub/grub.conf that
boots 'kernel /vmlinuz' and 'initrd /initrd.img', and a second one
that does /vmlinuz.old and /initrd.img.old; that way you should, in
principle, always be able to boot your current and previous kernels
without ever touching your GRUB configuration.

-- 
David Maze [EMAIL PROTECTED]  http://people.debian.org/~dmaze/
"Theoretical politics is interesting.  Politicking should be illegal."
-- Abra Mitchell


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

RE: making a floppy for essential modules

[Keith O'Brien]

I'm in the process of doing the same thing for an HP-DL360. 

Here is the websites I'm working off of:
http://lists.debian.org/debian-user/2003/debian-user-200310/msg05875.html
http://www.debian.org/releases/stable/i386/ch-boot-floppy-techinfo.en.html#s
-rescue-replace-kernel

I can't seem to get the kernel small enough to fit on a floppy. 
Setup is 4651 bytes.
System is 1052 kB
warning: kernel is too big for standalone boot from floppy. 
I'm still working on it so I can't provide much more help. GL

If anyone has install floppies for dl-360/380 with cciss driver could you
post a link.

Thanks, 
Keith. 



> -Original Message-
> From: New Disorder Records [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, December 09, 2003 2:38 PM
> To: Michael Martinell
> Cc: [EMAIL PROTECTED]
> Subject: RE: making a floppy for essential modules
> 
> 
> thanks, but I can't install normally if I can't see the hard 
> drive.  The 
> installer (any one of the flavors of installer, including 
> bf24) doesn't 
> see the harddrive on my 1750, so I need to install a driver 
> at that point, 
> I assume, in order to continue the installation.
>   So, I can make it to step 2 of your steps, but not step 3.  
> I have a 
> driver, I just need to know how to make a disk that the 
> installer will 
> recognize to load it.
> 
>  On Tue, 9 Dec 2003, 
> Michael Martinell wrote:
> 
> > Here is how I set up my Dell 
> > Set up Debian 3.0r1 on Dell Poweredge 2400
> > 
> > 1.  Insert Debian disk and boot from it.
> > 2.  At boot: prompt type bf24 and press enter.
> > 3.  Complete rest of install normally.  Make sure to install the c
> > compiler.
> > 4.  Download most current kernel.  For this example it was 2.4.20
> > 5.  Copy kernel to /usr/src
> > 6.  tar xzvf linux-2.4.20.tar.gz
> > 7.  rm linux
> > 8.  ln -s linux-2.4.20 linux
> > 9.  ls -al  (verify symbolic link you just created)
> > 10. cd /usr/src/linux
> > 11. make menuconfig
> > 12. Select your options in the menu
> > 13. Store configuration to file.  This is your backup.
> > 14. make dep
> > 15. make clean
> > 16. make bzImage  (ignore warning message)
> > 17. make modules
> > 18. make modules_install
> > 19. make install
> > 20. vi /etc/lilo.conf
> > 21. Verify settings are correct.
> > lilo.conf
> > Samba:/etc# more lilo.conf
> > # /etc/lilo.conf - See: `lilo(8)' and `lilo.conf(5)',
> > # ---   `install-mbr(8)', `/usr/share/doc/lilo/',
> > #   and `/usr/share/doc/mbr/'.
> > # +---+
> > # |!! Reminder !! |
> > # |   |
> > # | Don't forget to run `lilo' after you make changes to this |
> > # | conffile, `/boot/bootmess.txt', or install a new kernel.  The |
> > # | computer will most likely fail to boot if a kernel-image  |
> > # | post-install script or you don't remember to run `lilo'.  |
> > # |   |
> > # +---+
> > # Support LBA for large hard disks.
> > #
> > lba32
> > # Overrides the default mapping between harddisk names and the BIOS'
> > # harddisk order. Use with caution.
> > #disk=/dev/hde
> > #bios=0x81
> > #disk=/dev/sda
> > #bios=0x80
> > # Specifies the boot device.  This is where Lilo installs its boot
> > # block.  It can be either a partition, or the raw device, in which
> > # case it installs in the MBR, and will overwrite the current MBR.
> > #
> > boot=/dev/sda
> > # Specifies the device that should be mounted as root. (`/')
> > #
> > root=/dev/sda1
> > # Enable map compaction:
> > # Tries to merge read requests for adjacent sectors into a single
> > # read request. This drastically reduces load time and keeps the
> > # map smaller.  Using `compact' is especially recommended when
> > # booting from a floppy disk.  It is disabled here by default
> > # because it doesn't always work.
> > #
> > # compact
> > # Installs the specified file as the new boot sector
> > # You have the choice between: bmp, compat, menu and text
> > # Look in /boot/ and in lilo.conf(5) manpage for details
> > #
> > install=/boot/boot.b
> > # Specifies the location of the map file
> > #
> > map=/boot/map
> > # You can set a password here, and uncomment the `restricted' lines
> > # in the image definitions below to make it so that a password must
> > # be typed to boot anything but a default configuration.  If a
> > # command line is given, other than one specified by an `append'
> > # statement in `lilo.conf', the password will be required, but a
> > # standard default boot will not require one.
> > #
> > # This will, for instance, prevent anyone with access to the
> > # console from booting with something like `Linux init=/bin/sh',
> > # and thus becoming `root' without proper authorization.
> > #
> > # Note that if you really need this type of securit

Re: clustering software

[Arnt Karlsen]

On Tue, 9 Dec 2003 18:26:47 + (GMT), 
Rus Foster <[EMAIL PROTECTED]> wrote in message 
<[EMAIL PROTECTED]>:

> Hi,
> Does anyone know if there is any clustering software for 2.6? I know
> mosix but there doesn't seem to be 2.6 support

..mosix is basically history, most mosix people use openmosix, 
unless you haven't, check out http://openmosix.sourceforge.net/ 
and http://openmosix.sourceforge.net/mailing.html , and you're 
right AFAICT, the guys work on 2.4.22 and 2.4.23.

-- 
..med vennlig hilsen = with Kind Regards from Arnt... ;-)
...with a number of polar bear hunters in his ancestry...
  Scenarios always come in sets of three: 
  best case, worst case, and just in case.


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

Re: start program at login

[John L. Fjellstad]

Wayne Topa wrote:

> Have you tried adding it to your .xinitrc?

I tried. I forgot to mention that I'm logging in through kdm.  As far as I
know, neither local .xsession nor .xinitrc is called during login.

-- 
John L. Fjellstad
web: http://www.fjellstad.org/  Quis custodiet ipsos custodes


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

Re: make-kpkg

[Hugo Vanwoerkom]

Michael Montagne wrote:
I use Grub as a bootloader.  After making a kernel .deb using
make-kpkg, I'm running dpkg -i  Near the end you are asked to if
you want to make a boot block.  What is this?  Is it just an entry in
Grub or LILO?  What I'm most concerned about is being able to boot to
my old kernel if I screwed this one up.

It actually runs lilo by default. Lilo will install a boot block. Or if 
lilo fails the package that you tried to install ( the .deb) will remain half-installed and apt-get or dpkg will complain 
until you fix it, by fixing the /etc/lilo.conf file or its dependencies, 
like other partitions to be mounted.

From man make-kpkg:

"On installation, it also offers to run the Linux loader, LILO (or 
alternates like loadlin, SILO, QUIK, VMELILO, ZIPL, yaboot, PALO or 
GRUB), creating a configuration file for supported boot loaders
if needed."

From man kernel-img.conf:

"postinst_hook
Set this variable to a script to be executed during installation after 
all the symbolic links are created, but before running the bootloader or 
offering to create a floppy. This script shall be called with two 
arguments, the first being the version of the kernel image, and the 
second argument being the location of the kernel image itself. Errors in 
the script shall produce a warning message, but shall be otherwise 
ignored. An example script for grub users is present in 
/usr/share/doc/kernel-package/ directory."

So in your case you should follow the advice in the last sentence.

I believe that applicable, though somewhat brusque, phrase is RTFM. That 
is why Debian is a *super* system. ;-)

Hugo.

















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

X configuration error

[Vivek Kumar]

Hi there,

I was trying to configure X and I got following error message in my log:

(EE) Screen(s) found, but none have a usable configuration.

Fatal server error:
no screens found

When reporting a problem related to a server crash, please send  
the full server output, not just the last messages.
This can be found in the log file "/var/log/XFree86.0.log".
Please report problems to [EMAIL PROTECTED]

Kindly HELP !!


-- 
Vivek Kumar



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

Re: problems with networking on boot

[Lou Losee]

* Andrea Tasso <[EMAIL PROTECTED]> [2003-12-09 14:32]:
> On Tue, Dec 09, 2003 at 02:02:49PM -0500 or thereabouts, Lou Losee wrote:
> > I am having problems getting networking to start up on boot.  I can get
> > it running manually, but it seems like there is an interface problem
> > between ifup/down and ifconfig.
> 
> what do you have in /etc/network/interfaces ?
> 
Here it is:

# /etc/network/interfaces -- configuration file for ifup(8), ifdown(8)

# The loopback interface
# automatically added when upgrading
auto lo
iface lo inet loopback
auto eth0
iface eth0 inet dhcp

Lou


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

Re: compiling kernel for another system

[John L. Fjellstad]

Colin Watson wrote:

> The kernel doesn't use the libc, so I expect that this is a red herring.
> (I don't know the real answer, though.)

Yeah, I figured that out after looking at the dependency list of the
kernel-packages.  I'm wondering though if using gcc3.3 would have any
impact (since everything else on the stable system would have the programs
compiled with gcc2.95)

This thread has been very educational, and I suspect Magnus is right.  I'm
almost tempted to recompile my kernel again, just to test it out (it's one
of those, good to know, things).

-- 
John L. Fjellstad
web: http://www.fjellstad.org/  Quis custodiet ipsos custodes


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

Re: Limit network speed

[Johann Koenig]

On Tuesday December  9 at 04:31pm
Ron Rademaker <[EMAIL PROTECTED]> wrote:

> The reason I want this is because when I use BitTorrent it
> uses a lot of data traffic, something I don't want (because of the
> upload limit)

Check out the option for bittorrent:

--max_upload_rate kbytes
maximum rate to upload at in kilobytes, 0 means no limit
(default 0)

I use 'btlaunchmanycurses . --max_upload_rate 10' in my Bittorrent
directory. Keep in mind that it's the max for each torrent, so 2
torrents can use 20kbps
-- 
-johann koenig
Now Playing: The OC Supertones - Another Show : Safety First
Today is Pungenday, the 51st day of The Aftermath in the YOLD 3169
My public pgp key: http://mental-graffiti.com/pgp/johannkoenig.pgp


pgp0.pgp
Description: PGP signature

RE: making a floppy for essential modules

[New Disorder Records]

thanks, but I can't install normally if I can't see the hard drive.  The 
installer (any one of the flavors of installer, including bf24) doesn't 
see the harddrive on my 1750, so I need to install a driver at that point, 
I assume, in order to continue the installation.
  So, I can make it to step 2 of your steps, but not step 3.  I have a 
driver, I just need to know how to make a disk that the installer will 
recognize to load it.

 On Tue, 9 Dec 2003, 
Michael Martinell wrote:

> Here is how I set up my Dell 
> Set up Debian 3.0r1 on Dell Poweredge 2400
> 
> 1.Insert Debian disk and boot from it.
> 2.At boot: prompt type bf24 and press enter.
> 3.Complete rest of install normally.  Make sure to install the c
> compiler.
> 4.Download most current kernel.  For this example it was 2.4.20
> 5.Copy kernel to /usr/src
> 6.tar xzvf linux-2.4.20.tar.gz
> 7.rm linux
> 8.ln -s linux-2.4.20 linux
> 9.ls -al  (verify symbolic link you just created)
> 10.   cd /usr/src/linux
> 11.   make menuconfig
> 12.   Select your options in the menu
> 13.   Store configuration to file.  This is your backup.
> 14.   make dep
> 15.   make clean
> 16.   make bzImage  (ignore warning message)
> 17.   make modules
> 18.   make modules_install
> 19.   make install
> 20.   vi /etc/lilo.conf
> 21.   Verify settings are correct.
> lilo.conf
> Samba:/etc# more lilo.conf
> # /etc/lilo.conf - See: `lilo(8)' and `lilo.conf(5)',
> # ---   `install-mbr(8)', `/usr/share/doc/lilo/',
> #   and `/usr/share/doc/mbr/'.
> # +---+
> # |!! Reminder !! |
> # |   |
> # | Don't forget to run `lilo' after you make changes to this |
> # | conffile, `/boot/bootmess.txt', or install a new kernel.  The |
> # | computer will most likely fail to boot if a kernel-image  |
> # | post-install script or you don't remember to run `lilo'.  |
> # |   |
> # +---+
> # Support LBA for large hard disks.
> #
> lba32
> # Overrides the default mapping between harddisk names and the BIOS'
> # harddisk order. Use with caution.
> #disk=/dev/hde
> #bios=0x81
> #disk=/dev/sda
> #bios=0x80
> # Specifies the boot device.  This is where Lilo installs its boot
> # block.  It can be either a partition, or the raw device, in which
> # case it installs in the MBR, and will overwrite the current MBR.
> #
> boot=/dev/sda
> # Specifies the device that should be mounted as root. (`/')
> #
> root=/dev/sda1
> # Enable map compaction:
> # Tries to merge read requests for adjacent sectors into a single
> # read request. This drastically reduces load time and keeps the
> # map smaller.  Using `compact' is especially recommended when
> # booting from a floppy disk.  It is disabled here by default
> # because it doesn't always work.
> #
> # compact
> # Installs the specified file as the new boot sector
> # You have the choice between: bmp, compat, menu and text
> # Look in /boot/ and in lilo.conf(5) manpage for details
> #
> install=/boot/boot.b
> # Specifies the location of the map file
> #
> map=/boot/map
> # You can set a password here, and uncomment the `restricted' lines
> # in the image definitions below to make it so that a password must
> # be typed to boot anything but a default configuration.  If a
> # command line is given, other than one specified by an `append'
> # statement in `lilo.conf', the password will be required, but a
> # standard default boot will not require one.
> #
> # This will, for instance, prevent anyone with access to the
> # console from booting with something like `Linux init=/bin/sh',
> # and thus becoming `root' without proper authorization.
> #
> # Note that if you really need this type of security, you will
> # Note that if you really need this type of security, you will
> # likely also want to use `install-mbr' to reconfigure the MBR
> # program, as well as set up your BIOS to disallow booting from
> # removable disk or CD-ROM, then put a password on getting into the
> # BIOS configuration as well.  Please RTFM `install-mbr(8)'.
> #
> # password=tatercounter2000
> # Specifies the number of deciseconds (0.1 seconds) LILO should
> # wait before booting the first image.
> #
> delay=100
> # You can put a customized boot message up if you like.  If you use
> # `prompt', and this computer may need to reboot unattended, you
> # must specify a `timeout', or it will sit there forever waiting
> # for a keypress.  `single-key' goes with the `alias' lines in the
> # `image' configurations below.  eg: You can press `1' to boot
> # `Linux', `2' to boot `LinuxOLD', if you uncomment the `alias'.
> #
> # message=/boot/bootmess.txt
> prompt
> #   single-key
> #   single-key
> #   delay=100
> 

RE: making a floppy for essential modules

[Michael Martinell]

Here is how I set up my Dell 
Set up Debian 3.0r1 on Dell Poweredge 2400

1.  Insert Debian disk and boot from it.
2.  At boot: prompt type bf24 and press enter.
3.  Complete rest of install normally.  Make sure to install the c
compiler.
4.  Download most current kernel.  For this example it was 2.4.20
5.  Copy kernel to /usr/src
6.  tar xzvf linux-2.4.20.tar.gz
7.  rm linux
8.  ln -s linux-2.4.20 linux
9.  ls -al  (verify symbolic link you just created)
10. cd /usr/src/linux
11. make menuconfig
12. Select your options in the menu
13. Store configuration to file.  This is your backup.
14. make dep
15. make clean
16. make bzImage  (ignore warning message)
17. make modules
18. make modules_install
19. make install
20. vi /etc/lilo.conf
21. Verify settings are correct.
lilo.conf
Samba:/etc# more lilo.conf
# /etc/lilo.conf - See: `lilo(8)' and `lilo.conf(5)',
# ---   `install-mbr(8)', `/usr/share/doc/lilo/',
#   and `/usr/share/doc/mbr/'.
# +---+
# |!! Reminder !! |
# |   |
# | Don't forget to run `lilo' after you make changes to this |
# | conffile, `/boot/bootmess.txt', or install a new kernel.  The |
# | computer will most likely fail to boot if a kernel-image  |
# | post-install script or you don't remember to run `lilo'.  |
# |   |
# +---+
# Support LBA for large hard disks.
#
lba32
# Overrides the default mapping between harddisk names and the BIOS'
# harddisk order. Use with caution.
#disk=/dev/hde
#bios=0x81
#disk=/dev/sda
#bios=0x80
# Specifies the boot device.  This is where Lilo installs its boot
# block.  It can be either a partition, or the raw device, in which
# case it installs in the MBR, and will overwrite the current MBR.
#
boot=/dev/sda
# Specifies the device that should be mounted as root. (`/')
#
root=/dev/sda1
# Enable map compaction:
# Tries to merge read requests for adjacent sectors into a single
# read request. This drastically reduces load time and keeps the
# map smaller.  Using `compact' is especially recommended when
# booting from a floppy disk.  It is disabled here by default
# because it doesn't always work.
#
# compact
# Installs the specified file as the new boot sector
# You have the choice between: bmp, compat, menu and text
# Look in /boot/ and in lilo.conf(5) manpage for details
#
install=/boot/boot.b
# Specifies the location of the map file
#
map=/boot/map
# You can set a password here, and uncomment the `restricted' lines
# in the image definitions below to make it so that a password must
# be typed to boot anything but a default configuration.  If a
# command line is given, other than one specified by an `append'
# statement in `lilo.conf', the password will be required, but a
# standard default boot will not require one.
#
# This will, for instance, prevent anyone with access to the
# console from booting with something like `Linux init=/bin/sh',
# and thus becoming `root' without proper authorization.
#
# Note that if you really need this type of security, you will
# Note that if you really need this type of security, you will
# likely also want to use `install-mbr' to reconfigure the MBR
# program, as well as set up your BIOS to disallow booting from
# removable disk or CD-ROM, then put a password on getting into the
# BIOS configuration as well.  Please RTFM `install-mbr(8)'.
#
# password=tatercounter2000
# Specifies the number of deciseconds (0.1 seconds) LILO should
# wait before booting the first image.
#
delay=100
# You can put a customized boot message up if you like.  If you use
# `prompt', and this computer may need to reboot unattended, you
# must specify a `timeout', or it will sit there forever waiting
# for a keypress.  `single-key' goes with the `alias' lines in the
# `image' configurations below.  eg: You can press `1' to boot
# `Linux', `2' to boot `LinuxOLD', if you uncomment the `alias'.
#
# message=/boot/bootmess.txt
prompt
#   single-key
#   single-key
#   delay=100
#   timeout=100
# Specifies the VGA text mode at boot time. (normal, extended, ask, )
#
# vga=ask
# vga=9
#
vga=normal
# 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="ide2=0xd400,0xd002 ide3=0xcc00,0xc802 hdb=ide-scsi hdc=ide-scsi"
# Boot up Linux by default.
#
default=Linux
image=/boot/vmlinuz-2.4.20-x
label=Linux
read-only
#   restricted
#   alias=1
image=/vmlinuz
label=OldLinux
read-only
#   restricted
#   alias=2
# If you have another OS on this machine to boot, you can uncomment the
# 

Re: problems with networking on boot

[Andrea Tasso]

On Tue, Dec 09, 2003 at 02:02:49PM -0500 or thereabouts, Lou Losee wrote:
> I am having problems getting networking to start up on boot.  I can get
> it running manually, but it seems like there is an interface problem
> between ifup/down and ifconfig.

what do you have in /etc/network/interfaces ?

-- 
[EMAIL PROTECTED]


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

RE: Virtual PC 5.2

[Michael Martinell]

It appears that xfree86 does not support the base S3 card after 3.3.6.  It
only supports the savage series.

Is there a recommended downgrade path that works?

Or in other words, what would the apt-get command for this be?

-Original Message-
From: Michael Martinell [mailto:[EMAIL PROTECTED] 
Sent: Tuesday, December 09, 2003 1:05 PM
To: [EMAIL PROTECTED]
Subject: RE: Virtual PC 5.2

I have tried the vga driver with no success.

How do you get the busID?  I have tried searching google but have came up
empty.

-Original Message-
From: Derrick 'dman' Hudson [mailto:[EMAIL PROTECTED] 
Sent: Tuesday, December 09, 2003 1:01 PM
To: [EMAIL PROTECTED]
Subject: Re: Virtual PC 5.2

On Tue, Dec 09, 2003 at 11:53:29AM -0600, Michael Martinell wrote:
| It lists it as a VGA Compatible controller S3 Inc 86c764/765
[Trio32/64/64+]

Try the 'vga' driver.  It might not be as good as the s3 driver on
real hardware, but it should at least be operational and give you
something to start with.

| The x error is 
| Assigning device section with no busID to primary device
| (EE) No devices detected.

The driver you tried was unable to detect the device.  Perhaps
specifying the busID will solve that.

-- 
One OS to rule them all, one OS to find them,
One OS to bring them all and in the darkness bind them,
In the Land of Redmond, where the Shadows lie.
 
www: http://dman13.dyndns.org/~dman/jabber:
[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]

Kde3.2beta for Sarge/testing: any apt repositories?

[Alexander Fitterling]

Where could I get binaries from Kde3.2beta for Sarge/testing? I'd wish to take 
mirrors from as I use a recursive wget and prepare iso images out of it. This 
actually worked great so far. But I can't find a valid apt-repository for 
Sarge and Kde3.2. By the way will Kde3.2 be part of Sarge if becoming stable?

Thanks.
Alex


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

New Debian installer?

[Alexander Fitterling]

Everyone.

I would like to know where I could get a new Debian installer - that one I 
heard wispering about. Interesting to me seems the question if it comes 
already with Sarge/testing, if downloading using the jigdo utility. Or must I 
take a look some place else at?

Appreciate any hints ...
Best Regards,
Alex


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

making a floppy for essential modules

[New Disorder Records]

Hi,
  I am trying to install debian on a dell poweredge, but I need drivers 
for the hard drives and ethernet card.  I have found the drivers, but I 
have no idea how to put them on a floppy disk so that debian will see them 
when I put the floppy in.  I don't have other debian machines in the 
office, so I have to make the floppies on a windows machine.  Any advice 
about how to put the drivers on a floppy so that debian will see them 
would be greatly appreciated.

Ernst Schoen-Rene,
HBL Industries.


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

RE: File system error (power outage) SOLVED

[Francisco Castellon]

Thank-you guys!

Got it working... I just ran

# e2fsck /dev/hda3

and just answered 'y' to all of the questions. Everything seems to be
working ok. Thanx again!

Francisco


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

Re: Info Problem

[Arnt Karlsen]

On Tue, 9 Dec 2003 17:59:09 +, 
Colin Watson <[EMAIL PROTECTED]> wrote in message 
<[EMAIL PROTECTED]>:

> On Tue, Dec 09, 2003 at 05:33:46PM +0100, Arnt Karlsen wrote:
> > On Tue, 9 Dec 2003 13:58:15 +, 
> > Colin Watson <[EMAIL PROTECTED]> wrote in message 
> > <[EMAIL PROTECTED]>:
> > > On Mon, Dec 08, 2003 at 04:58:26PM -0800, Karsten M. Self wrote:
> > > > On a more practical note:  copyright would pass to the estate
> > > > and heirs. It's possible (not necessarially easy) to get
> > > > assignments in this case.
> > > 
> > > Yeah. The thought has occurred to me, but it seems rather
> > > tasteless to go and ask. In practice my assessment is that it's
> > > vanishingly unlikely to become an actual problem (and if it does,
> > > I'll just remove all copyrightable pieces of code he wrote), so
> > > again I've gone for the easy road.
> > 
> > ..I disagree: It _is_ appropriate to approach the heirs and ask,
> > informing them of the 2 alternatives, removal of his code (just
> > like we're gonna do with any SCO code), and the heirs giving 
> > their permission to let the dead man's code work survive under 
> > the GPL and copyright law.  Chances are very good they will 
> > consider what he would have wanted them to do. 
> > 
> > .._not_ asking, I find _less_ appropriate, because chances are, 
> > the heirs may feel we take liberties with _their_ copyright.
> 
> Eh? The code is GPL. The person in question was perfectly aware of
> this and licensed his code the same way. *No* liberties are being
> taken! 

.._we_ know that.  His _heirs_, does not neccessarily know, understand,
or appreciate, this fact.  Which again _may_ cause such bad feelings. 

..myself, I prefer to err on the side of caution, such bad feelings
_can_ be prevented, both by removal of dead mens code, and 
by asking the heirs.  Removal and rewriting, is of course the 
safest approach, but forces "re-inventing the wheel", I like to see
re-innovation happen on technological merits alone, not just 
because someone dies.

> The discussion with Karsten is about the problem of getting copyright
> assignments to the FSF for the purpose of making it a GNU project,
> which is purely hypothetical; it's something I have no intention of
> actually doing.

..nevertheless we oughtta think thru this, we will all die, so we wanna 
secure our copyrights so our work _remains_ available, and ours,
FSF is a good tool to enforce our deads copyright.

> Please read the thread a little more carefully.

.. ;-)

-- 
..med vennlig hilsen = with Kind Regards from Arnt... ;-)
...with a number of polar bear hunters in his ancestry...
  Scenarios always come in sets of three: 
  best case, worst case, and just in case.


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

RE: Virtual PC 5.2

[Michael Martinell]

I have tried the vga driver with no success.

How do you get the busID?  I have tried searching google but have came up
empty.

-Original Message-
From: Derrick 'dman' Hudson [mailto:[EMAIL PROTECTED] 
Sent: Tuesday, December 09, 2003 1:01 PM
To: [EMAIL PROTECTED]
Subject: Re: Virtual PC 5.2

On Tue, Dec 09, 2003 at 11:53:29AM -0600, Michael Martinell wrote:
| It lists it as a VGA Compatible controller S3 Inc 86c764/765
[Trio32/64/64+]

Try the 'vga' driver.  It might not be as good as the s3 driver on
real hardware, but it should at least be operational and give you
something to start with.

| The x error is 
| Assigning device section with no busID to primary device
| (EE) No devices detected.

The driver you tried was unable to detect the device.  Perhaps
specifying the busID will solve that.

-- 
One OS to rule them all, one OS to find them,
One OS to bring them all and in the darkness bind them,
In the Land of Redmond, where the Shadows lie.
 
www: http://dman13.dyndns.org/~dman/jabber:
[EMAIL PROTECTED]


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

problems with networking on boot

[Lou Losee]

I am having problems getting networking to start up on boot.  I can get
it running manually, but it seems like there is an interface problem
between ifup/down and ifconfig.

I am running net-tools (ifconfig) 1.60.8 and ifupdown 0.6.4-4.6 but have
also tried this with net-tools 1.60.4 and ifupdown 0.5.4-4

After bootingthe system, there are no network interfaces setup.  If I
manually run ifup -av I get the following output:

Configuring interface lo=lo (inet)
run-parts /etc/network/if-pre-up.d
ifconfig lo 127.0.0.1 up
ifconfig: only one address allowed for interface `lo'
Usage: ifconfig [OPTION]... [SYSTEM OPTION]...
Try `ifconfig --help' for more information.
Configuring interface eth0=eth0 (inet)
run-parts /etc/network/if-pre-up.d
pump -i eth0  
run-parts /etc/network/if-up.d

However networking comes up fine if I issue the commands:
ifconfig lo up 127.0.0.1
pump -i eth0

Similarly, if I try ifdown -av I get:

Configuring interface lo=lo (inet)
run-parts /etc/network/if-down.d
ifconfig lo down
ifconfig: can not resolve `down': Unknown host
Configuring interface eth0=eth0 (inet)
run-parts /etc/network/if-down.d
pump -i eth0 -k
ifconfig eth0 down
ifconfig: can not resolve `down': Host name lookup failure

With regards to the boot sequence, in /etc/rcS.d I have:

lrwxrwxrwx1 root root   18 Dec  8 07:31 /etc/rcS.d/S40ifupdown -> 
../init.d/ifupdown
lrwxrwxrwx1 root root   20 Dec  8 07:31 /etc/rcS.d/S40networking -> 
../init.d/networking

I tried searching bugs and the changelogs for ifconfig and ifupdown and
the forum archives but didn't see anything relavent.

Anyone have any ideas where/what to look at next.

TIA,
Lou


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

Re: Mixing woody and sarge

[Rajkumar S]

Derrick 'dman' Hudson wrote:
The hitch is that libc6 has changed, and so installing any package in
 sarge or sid will pull in the new libc, and that in turn will force
 other packages to be upgraded.  This is just a matter of following 
the dependencies.
Yes, This is my experience also, I installed ulogd and whole libc and
perl also came along with it. so I guess the safe way is to compile the
any packages, that I need from testing, from source in woody and use
that, or get it from backport.org if available.
One of the most urgent need is samhain (a file integrity checker which
can detect kernel root kits) With debian servers getting rooted this
 has become a necessity.
I hope this clarifies things.
Yup, thanks a lot.

raj

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

Re: Virtual PC 5.2

[Derrick 'dman' Hudson]

On Tue, Dec 09, 2003 at 11:53:29AM -0600, Michael Martinell wrote:
| It lists it as a VGA Compatible controller S3 Inc 86c764/765 [Trio32/64/64+]

Try the 'vga' driver.  It might not be as good as the s3 driver on
real hardware, but it should at least be operational and give you
something to start with.

| The x error is 
| Assigning device section with no busID to primary device
| (EE) No devices detected.

The driver you tried was unable to detect the device.  Perhaps
specifying the busID will solve that.

-- 
One OS to rule them all, one OS to find them,
One OS to bring them all and in the darkness bind them,
In the Land of Redmond, where the Shadows lie.
 
www: http://dman13.dyndns.org/~dman/jabber: [EMAIL PROTECTED]


pgp0.pgp
Description: PGP signature

Re: File system error (power outage)

[Vineet Kumar]

* Alvin Oga ([EMAIL PROTECTED]) [031209 08:40]:
> - now fix /   -- Important to use the right -o options
> 
>   root# mount -o ro,remount /dev/hda1 /
>   root# e2fsck /dev/hda
  
Don't forget the 1!
'e2fsck /dev/hda' is not a good idea in this case.  You really want
'e2fsck /dev/hda1'

good times,
Vineet
-- 
http://www.doorstop.net/
-- 
http://www.sprintpcs-sucks.org/


signature.asc
Description: Digital signature