susbscribe

[Nick Jennings]

subscribe [EMAIL PROTECTED]

Re: modversions.h ?

[Gary Hennigan]

Pollywog <[EMAIL PROTECTED]> writes:
> On 20-Jan-2000 aphro wrote:
> > if the program you are compiling requires 2.2.x i suggest re linking
> > /usr/include/linux to /usr/src/linux/include/linux (i think thats
> > right) .. rename /usr/include/linux
> 
> That is the action I was considering, but I did not think it should be done
> with 2.2.x kernels because the README in the Linux tarball no longer mentions
> it.
> 
> > 
> > the program is probably looking for something 2.1/2.2/2.3 specific and
> > debian's default headers in /usr/include dont cut it for that(i had the
> > similar problems trying to compile smbmount from samba 3)

In general, it's not a good thing to do because your libc was compiled
against what's in there. However, I have had an occasion where I
moved the existing directories to different names, eg., 

mv /usr/include/asm /usr/include/asm.orig 

created the links to the kernel source tree, eg.,

ln -s /usr/src/kernel-2.2.14/include/asm-i386 /usr/include/asm

done the compile and copied them back, eg.

rm /usr/include/asm
mv /usr/include/asm.orig /usr/include/asm

Of course I don't ever recall having to do this for the kernel
source. I think I even had a 2.2.x kernel running on slink, when I was
running slink, and don't recall doing this. I certainly haven't had to
do it under potato.

Gary

Re: slink -> potato

[Phil Brutsche]

A long time ago, in a galaxy far, far way, someone said...

> Hi, 
> 
> My box running slink here would like to do the
> potato upgrade but I'm concerned that it might run 
> out of disk space:
> 
> Filesystem 1024-blocks  Used Available Capacity Mounted on
> /dev/hda1 296633  24881332500 88%   /
> /dev/hda3 675175  343296   297003 54%   /home
> 
> Safe to upgrade?

No it's not safe for you to upgrade - not enough disk space.  When I
downloaded everything it was 155MB.  You need to delete quite a few files
from your root file system.

One way to get around that problem would be to do this:

rmdir /var/lib/dpkg/methods//debian
mkdir /home/debian
ln -s /home/debian /var/lib/dpkg/methods//debian

before you start upgrading.  That will alleviate your disk space problem.

-- 
--
Phil Brutsche   [EMAIL PROTECTED]

"There are two things that are infinite; Human stupidity and the
universe. And I'm not sure about the universe." - Albert Einstein

Re: PS/2 Mouse Port

[Mike Werner]

On Fri, Jan 21, 2000 at 07:55:50AM +0930, Terri and Glen Kondos wrote:
> I'm having trouble figuring out what device I should use for my PS/2 port
> mouse. I have a Microsoft PS/2 mouse and notice on boot up the Linux 
> recognizes that there is a PS/2 mouse port. When configuring /etc/gpm.conf, I
> don't know what device to use for the mouse.

/dev/psaux
-- 
Mike Werner  KA8YSD   |  "Where do you want to go today?"
ICQ# 12934898 |  "As far from Redmond as possible!"
'91 GS500E|
Morgantown WV |  Only dead fish go with the flow.

Re: PS/2 Mouse Port

[aphro]

try /dev/psaux

that is the 'standard' port to use for 99% of ps/2 mice.

nate

On Fri, 21 Jan 2000, Terri and Glen Kondos wrote:

gkondo >I'm having trouble figuring out what device I should use for my PS/2 
port
gkondo >mouse. I have a Microsoft PS/2 mouse and notice on boot up the Linux 
gkondo >recognizes that there is a PS/2 mouse port. When configuring 
/etc/gpm.conf, I 
gkondo >don't know what device to use for the mouse.
gkondo >
gkondo >Couldn't find a HOWTO on the topic and I didn't see an appropriate 
device driver to
gkondo >load during the installation phase. I've run out options on other 
resources to use.
gkondo >I'm just moving to Debian GNU/Linux from Redhat in which the mouse just
gkondo >automagically worked.
gkondo >
gkondo >Appreciate any support
gkondo >
gkondo >Glen Kondos, Alice Springs, Australia
gkondo >
gkondo >
gkondo >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
2:50pm up 154 days, 2:54, 2 users, load average: 1.51, 1.45, 1.50

Re: modversions.h ?

[Pollywog]

On 20-Jan-2000 aphro wrote:
> if the program you are compiling requires 2.2.x i suggest re linking
> /usr/include/linux to /usr/src/linux/include/linux (i think thats
> right) .. rename /usr/include/linux

That is the action I was considering, but I did not think it should be done
with 2.2.x kernels because the README in the Linux tarball no longer mentions
it.

> 
> the program is probably looking for something 2.1/2.2/2.3 specific and
> debian's default headers in /usr/include dont cut it for that(i had the
> similar problems trying to compile smbmount from samba 3)

Thanks

--
Andrew

Re: apt changes mount status of /usr ?!?!!

[Alexander List]

On Wed, 19 Jan 2000, Jason Gunthorpe wrote:

> > Obviously, apt seems to fiddle around with mount during the upgrade
> > procedure, /usr is mounted ro after the upgrade!!!
> 
> rm /etc/apt/apt.conf

Well, must have been an attack of stupidity when I installed the example
apt.conf without checking it thoroughly ;-((. 

Of course that was a classical species of "Hit user to continue".

Thanx for showing me the way out of the tunnel...

Alex

Re: making a pine deb?

[Colin Watson]

[EMAIL PROTECTED] (Ron Farrer) wrote:
>Does anyone have a diff that I could use to make a deb of pine 4.21?
>I've got the source tarball all unpacked and ready to go, but I'd rather
>not have to create all of the debian files needed to make a deb.
>You can skip the usual, mutt's better try it instead bit as I DO use
>mutt. It's just that some users want pine...

You can get pine-4.21 debs (and source packages) from
.

-- 
Colin Watson   [EMAIL PROTECTED]

Re: Question: how to use an LDAP backend for logins

[Joseph A. Martin]

On Thu, 20 Jan 2000, Konrad Mierendorff wrote:

> Could you please send me the HOWTO as well. (Or an URL if it is online)
> I'm planning to do a similar setup soon, for twenty machines though.
> If you know any webpages on the topic you could also send them to me.

I will collect some links this evening and mail them to you. The HOWTO
will be posted online, on my own site if nowhere else, though I do
have hopes of making the HOWTO part of the LDP. And, no, I don't have
a website yet, but I have been planning on doing one for a while now
and this provides a good excuse.

Also, I have other good news. I just talked to my boss and he has
agreed to let me bill the company for the hours I spend writing
the HOWTO. This will allow me to do the work much quicker than would
otherwise be possible.

--
the "LaterDude"
ICQ: 52640402
[EMAIL PROTECTED]

All opinions expressed are my own and not necessarily those of
my employer unless otherwise noted.

PS/2 Mouse Port

[Terri and Glen Kondos]

I'm having trouble figuring out what device I 
should use for my PS/2 portmouse. I have a Microsoft PS/2 mouse and notice 
on boot up the Linux 
recognizes that there is a PS/2 mouse port. When 
configuring /etc/gpm.conf, I 
don't know what device to use for the 
mouse.
Couldn't find a HOWTO on the topic and I 
didn't see an appropriate device driver to
load during the installation phase. I've 
run out options on other resources to use.I'm just moving to Debian 
GNU/Linux from Redhat in which the mouse justautomagically 
worked.Appreciate any supportGlen Kondos, Alice Springs, 
Australia

Re: don't want x as default startup

[eric k. wolven]

-BEGIN PGP SIGNED MESSAGE-

mpx <[EMAIL PROTECTED]> writes:

mpx:

Got this from the list:

type: more /etc/inittab
(look for line that says "id:N:initdefault)  ("N" is a number)
Remember #
type: cd /etc/rcN.d
type: ls
(look for file "S999xdm")
type: rm S99xdm

This will remove the default X start-up.
I use a text editor to edit the file--easier for me.

eric wolven



> dear debian-users,
> 
> yestarday i installed debian for the first time, and finally i am
> through this dselect-jungle;-)
> now i try since a few hours to eliminate this
> x-as-default-startup-thing.
> i would like to logon my machine in the console.
> so i did like i did on my redhat-box, changing in /etc/inittab, but in
> debian this seems not to work.
> in xf86config there's no option to change this.
> 
> anyone else who had to handle this as well?
> 
> mek
> 
> 
> -- 
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null

-BEGIN PGP SIGNATURE-
Version: 2.6.3a
Charset: noconv
Comment: Processed by Mailcrypt 3.5.5, an Emacs/PGP interface

iQB1AwUBOIeLL8aVl6CLtWIhAQEwvQMAmdtb8xVoWMgzIROY5Osl6UjI9ZkikZSf
v+rYnNZH9XpIlTbeN2TrDOSN+u9Q3wDYF66XRMRNAnIMjHVU1XpmtrWiC5C3v1yx
NOMNG8n9+y0idtFP+HADcU68ReIiX91a
=Uc8D
-END PGP SIGNATURE-

slink -> frozen upgrade .. problems encountered

[Alex Perry]

Following the freeze, I changed the sources.list
on one of my machines from stable to frozen and
did a dist-upgrade to see how it would get on.

The trick with the interim perl version didn't work.
Apt managed to leave me with three base package
versions installed but the program itself missing.
-> Suggest performing the perl upgrade manually
before letting anything else get changed automatically.

As a result of the perl hassle, a bunch of things
decided that they couldn't be installed as an upgrade
although they could be removed without trouble.
As a result, a large number of packages evaporated.

kbd insisted on being held back, even on dist-upgrade,
until I forcibly removed it and reinstalled it.

gconv-modules was not available by http but I needed
it and it discouraged a bunch of packages from installing.

The version of libncurses5-dev as listed in the website
distribution doesn't match the version in the ftp archive.
As a result, it fails to install with no warning I noticed.

Some other things that are listed in the web distribution
(and available by ftp) do not appear in the http index.

"nan.h" now lives in a different place.  Sigh.

The upgrade of ssh does correctly assert that existing
connections are not affected.  However, I found that
was unable to create any _new_ connections until I
had closed all the existing connections and restarted
sshd itself.  Not a problem, but disconcerting.

I'll try another machine in a couple of days; maybe this time
I won't accidentally delete my typescript before I'm done.

Alex.

Re: modversions.h ?

[aphro]

if the program you are compiling requires 2.2.x i suggest re linking
/usr/include/linux to /usr/src/linux/include/linux (i think thats
right) .. rename /usr/include/linux

the program is probably looking for something 2.1/2.2/2.3 specific and
debian's default headers in /usr/include dont cut it for that(i had the
similar problems trying to compile smbmount from samba 3)

nate

On Thu, 20 Jan 2000, Pollywog wrote:

pollyw >What is the story behind modversions.h ?
pollyw >
pollyw >I am using kernel 2.2.14 and I don't have this file, but I see it is 
listed
pollyw >for kernel-headers package 2.2.13.  I built a custom kernel and I am 
not using
pollyw >the default Debian kernel image.
pollyw >
pollyw >gcc  -L../lib -I../include/ -Wall -O6-c -o bcreencrypt.o 
bcreencrypt.c
pollyw >bcreencrypt.c:258: warning: `#ifdef' argument starts with a digit
pollyw >gcc  -L../lib -I../include/ -Wall -O6-c -o bcfsck.o bcfsck.c
pollyw >gcc  -L../lib -I../include/ -Wall -O6   -o bctool bctool.c alg.o 
config.o
pollyw >error.o misc.o bcmount.o bcpasswd.o bcinfo.o  bccreate.o bcformat.o
pollyw >bcreencrypt.o bcfsck.o -lkgsha
pollyw >cp -f bctool ../bin/
pollyw >make[1]: Leaving directory `/usr/src/redhat/BUILD/bcrypt-0.3b/src'
pollyw >make[1]: Entering directory `/usr/src/redhat/BUILD/bcrypt-0.3b/mod'
pollyw >gcc -o mkver mkver.c
pollyw >touch bc_dev.ver ;
pollyw >gcc -E bc_dev.c -I../include -D__GENKSYMS__ | /sbin/genksyms  -k 2.2.13 
>
pollyw >bc_dev.ver
pollyw >gcc -O6 -Wall -fno-strength-reduce -D__KERNEL__ -DMODULE -I../include   
  -c
pollyw >-o bc_dev.o bc_dev.c
pollyw >In file included from bc_dev.c:30:
pollyw >bc_cfg.h:39: linux/modversions.h: No such file or directory
pollyw >make[1]: *** [bc_dev.o] Error 1
pollyw >make[1]: Leaving directory `/usr/src/redhat/BUILD/bcrypt-0.3b/mod'
pollyw >make: *** [subdirs] Error 2
pollyw >Bad exit status from /var/tmp/rpm-tmp.99181 (%build)
pollyw >
pollyw >
pollyw >BTW I was trying to build an rpm so that I could use Alien to make a 
Debian
pollyw >package from it.
pollyw >
pollyw >
pollyw >
pollyw >
pollyw >thanks
pollyw >
pollyw >--
pollyw >Andrew
pollyw >
pollyw >-
pollyw >GnuPG Public KeyID: 0x48109681
pollyw >
pollyw >
pollyw >
pollyw >-- 
pollyw >Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
pollyw >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
2:02pm up 154 days, 2:06, 2 users, load average: 1.99, 1.67, 1.62

Re: Ethernet card dies unexpectedly

[Konrad Mierendorff]

José Luis Gómez Dans wrote:
> 
> On Thu, Jan 20, 2000 at 07:42:18AM -0800, aphro wrote:
> > what kind of machine is it on? i had a similar problem on an old P100 IBM
> > machine it wouldnt work with any PCI NIC for longer then 10-15 minutes,
> 
> This is an Athlon 550 MHz. I have downloaded the newest driver
> version (1.08 as opposed to 1.06 in the kernel) and will try with that.
> I will also try to compile the kernel with just -O to see if it survives
> :).

Once I could trace a similar error back to a bad cable.

- Konrad Mierendorff

RE: slink -> potato

[Lewis, James M.]

> Hi, 
> 
> My box running slink here would like to do the
> potato upgrade but I'm concerned that it might run 
> out of disk space:
> 
> Filesystem 1024-blocks  Used Available Capacity Mounted on
> /dev/hda1 296633  24881332500 88%   /
> /dev/hda3 675175  343296   297003 54%   /home
> 
> Safe to upgrade?
> 
Your likely to have to put the archives directory on /home.
The downloaded .deb's go into /var/cache/apt/archives.  You can
get stuff in small chunks and clean in between apt-get's.  Or, you
can use a sym link to put the archives tree over on /home so
you have more download space.  At the end, "apt-get clean" and
put the original archives dir back in place.

jim

> tia,
> 
> -Mark
> 
> -
> "I know everything we've done is absolutely right and proper"
> --Microsoft CEO Steve Ballmer on MSNBC, 01/13/00
> -
> 
> 
> 
> -- 
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] <
> /dev/null
> 

Re: Question: how to use an LDAP backend for logins

[Konrad Mierendorff]

"Joseph A. Martin" wrote:
> Very possible. Again, I will have a HOWTO in the works within a week
> or two. I will send you info as I work on the sections.

Could you please send me the HOWTO as well. (Or an URL if it is online)
I'm planning to do a similar setup soon, for twenty machines though.
If you know any webpages on the topic you could also send them to me.

Thanks in advance!

- Konrad Mierendorff

Using the mirror package

[Timm Gleason]

I have been using the mirror package for some time now, but I am finding
that recently, it has not been removing the older packages under the potato
distribution. I am now the proud holder of many multiple copies of many
different packages. Is there a way to remove those extraneous packages since
mirror did not seem to affect them.

And yes I have what I believe to be the correct config lines in my mirror
file


# Don't delete what is mirrored by the other mirror package below
# NB no / at end of directory names to save directories from
deletion
# also save private/project directory (which contains Incoming)
#
delete_excl=(local|debian-non-US)(/|$)
#
# Do not delete if more than 20% of all files would vanish
do_deletes=true
max_delete_files=30%

Timm Gleason


Timm Gleason, Hardware Development Engineer, Sheepish Lord of Chaos
[EMAIL PROTECTED] - http://n2h2.com/ - N2H2, Creators of Bess
900 Fourth Avenue, 34th Floor--Seattle, WA 98164

January 20, 2000
Any philosophy that can be put in a nutshell belongs there

Re: how many users per apache proc?

[Konrad Mierendorff]

aphro wrote:
> 
> thanks for all the replies ..i have a better idea ..sorta :) earlier i
> caught my server with 18 active connections running on only 2 processes(18
   ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^  ^ 
^
> idle servers) running Apache/1.3.9 (Unix) mod_perl/1.21 AuthMySQL/2.20
> FrontPage/4.0.4.3 ..i set it to start 20 servers, max 250 connections, and
> ulimit is set to 125 processes.

This is wrong! Take a look a this quotation from
http://www.apache.org/docs/misc/perf-tuning.html:

"""Apache (on Unix) is a pre-forking model server. The parent process is
responsible only for forking child processes, it does not serve any
requests or service any
network sockets. The child processes actually process connections, they
serve multiple connections (one at a time) before dying. The parent
spawns new or kills off
old children in response to changes in the load on the server (it does
so by monitoring a scoreboard which the children keep up to date)."""

You might want to look at the docu at www.apache.org, it is very good!

Re: Yamaha YMF 724

[Leszek Gerwatowski]

On Thu, Jan 20, 2000 at 05:41:35PM +0100, Andreas Sliwka wrote:
> Hi,   
>   
>   I got a new board with an Yamaha YMF 724 on board. How can I get this
>   chip to play sound?

I'm afraid you cannot without money. Yamaha has decided to make NDA with
OSS/4Tech and only commercial drivers are available for this chip. For
details go to www.alsa-project.org and to black list.

-- 
o--o  ___
|Leszek Gerwatowski|_/_|_\
o--o   (o\__/o)=)
"Don't fix it if it isn't broken"

RE: modversions.h ?

[Pollywog]

On 20-Jan-2000 Pollywog wrote:
> What is the story behind modversions.h ?
> 
> I am using kernel 2.2.14 and I don't have this file, but I see it is listed
> for kernel-headers package 2.2.13.  I built a custom kernel and I am not
> using
> the default Debian kernel image.

oops it does appear that I have this file after all:

/usr/src/kernel-source-2.2.14/include/linux/modversions.h
/usr/src/kernel-source-2.2.14/include/config/modversions.h
/usr/i486-linuxlibc1/include/linux/modversions.h

A symlink would seem to be the solution, but not all these copies of the file
modversions.h are the same (they are different sizes).

Any ideas?

thanks

--
Andrew

slink -> potato

[Mark Symonds]

Hi, 

My box running slink here would like to do the
potato upgrade but I'm concerned that it might run 
out of disk space:

Filesystem 1024-blocks  Used Available Capacity Mounted on
/dev/hda1 296633  24881332500 88%   /
/dev/hda3 675175  343296   297003 54%   /home

Safe to upgrade?

tia,

-Mark

-
"I know everything we've done is absolutely right and proper"
--Microsoft CEO Steve Ballmer on MSNBC, 01/13/00
-

HELP: dhcp client probs

[Albert Hurd]

I am running Debian 1.3.1, soon to be updated to 2.1.  I am having
problems with dhcpcd which I would
like to resolve before I upgrade. On @home the cable connection is
rather routinely dropped.  Running
dhcpcd as root doesn't help, the only remedy I have found is to close
down and reboot into Win95,
which seems not to have any probs with dhcpcd (things like this are
probably why they don't officially
support Linux).
On the net I found a document "Setting up DHCP under Corel for @Home
Cable" (Dec. 23/99) posetd
by 6plpy, where the problem is acknowledged, so it seems to persist
in the more recent releases.  The
fix suggested is to make changes to /etc/inet.d/dhcpcd.  This
file is not present on my system (there is
the executable dhcpcd in /usr/sbin). I found a note th debian-user
on dhcp client bt Paul van Tilburg
where a file /etc/dhclient,conf is mentioned, but that file is also
not on my system. Can anyone tell me
what to do, short of installing 2.1?
-- 
Albert Hurd
 
--- Begin Message ---


I am running Debian 1.3.1, soon to be updated to 2.1.  I am having
problems with dhcpcd which I would like to resolve before I upgrade. On
@home the cable connection is rather routinely dropped.  Running dhcpcd
as root doesn't help, the only remedy I have found is to close down and
reboot into Win95, which seems not to have any probs with dhcpcd (things
like this are probably why they don't officially support Linux).
On the net I found a document "Setting up DHCP under Corel for @Home
Cable" (Dec. 23/99) posetd by 6plpy, where the problem is acknowledged,
so it seems to persist in the more recent releases.  The fix suggested
is to make changes to /etc/inet.d/dhcpcd.  This file is not present
on my system (there is the executable dhcpcd in /usr/sbin). I found a note
th debian-user on dhcp client bt Paul van Tilburg where a file /etc/dhclient,conf
is mentioned, but that file is also not on my system. Can anyone tell me
what to do, short of installing 2.1?
-- 
Albert Hurd
 
--- End Message ---

modversions.h ?

[Pollywog]

What is the story behind modversions.h ?

I am using kernel 2.2.14 and I don't have this file, but I see it is listed
for kernel-headers package 2.2.13.  I built a custom kernel and I am not using
the default Debian kernel image.

gcc  -L../lib -I../include/ -Wall -O6-c -o bcreencrypt.o bcreencrypt.c
bcreencrypt.c:258: warning: `#ifdef' argument starts with a digit
gcc  -L../lib -I../include/ -Wall -O6-c -o bcfsck.o bcfsck.c
gcc  -L../lib -I../include/ -Wall -O6   -o bctool bctool.c alg.o config.o
error.o misc.o bcmount.o bcpasswd.o bcinfo.o  bccreate.o bcformat.o
bcreencrypt.o bcfsck.o -lkgsha
cp -f bctool ../bin/
make[1]: Leaving directory `/usr/src/redhat/BUILD/bcrypt-0.3b/src'
make[1]: Entering directory `/usr/src/redhat/BUILD/bcrypt-0.3b/mod'
gcc -o mkver mkver.c
touch bc_dev.ver ;
gcc -E bc_dev.c -I../include -D__GENKSYMS__ | /sbin/genksyms  -k 2.2.13 >
bc_dev.ver
gcc -O6 -Wall -fno-strength-reduce -D__KERNEL__ -DMODULE -I../include -c
-o bc_dev.o bc_dev.c
In file included from bc_dev.c:30:
bc_cfg.h:39: linux/modversions.h: No such file or directory
make[1]: *** [bc_dev.o] Error 1
make[1]: Leaving directory `/usr/src/redhat/BUILD/bcrypt-0.3b/mod'
make: *** [subdirs] Error 2
Bad exit status from /var/tmp/rpm-tmp.99181 (%build)


BTW I was trying to build an rpm so that I could use Alien to make a Debian
package from it.




thanks

--
Andrew

-
GnuPG Public KeyID: 0x48109681

Re: max processes

[Konrad Mierendorff]

aphro wrote:
>
> mieren >My common sense tells me that that depends on your hardware.
> 
> dual p2-233 128MB ..

Well, my common sense was a bit lazy yesterday, but now it is perfectly
aware of the fact that it depends on the processes as well :-)

But as my common sense produces rather stupid answers I thought about it
myself.

I would suggest that you push the number of processes to the limit step
by step, i.e. you increase until some processes get swapped out. Be
carefull that you still have enough disk-cache to ensure that your
webservers don't have to wait for the hdds. You should also keep an eye
on the number of file descriptors/inodes (/proc/sys/fs/* see
/usr/src/linux/Documentation/proc.txt for an explanation).

To test the setup I would generate a lot of request for some
(avarage-sized) files which will increase the number of processes. If
you get a lot of disk activity you did something wrong.

Hope it helps!

- Konrad Mierendorff

making a pine deb?

[Ron Farrer]

Hello all;

Does anyone have a diff that I could use to make a deb of pine 4.21?
I've got the source tarball all unpacked and ready to go, but I'd rather
not have to create all of the debian files needed to make a deb.
You can skip the usual, mutt's better try it instead bit as I DO use
mutt. It's just that some users want pine...
 

TIA,

Ron 
-- 
===
=  Ronald Burnett Farrer  =
===
=  =
=  =
===
|||
|Alpha Linux Powered! |
|  |
||| 

Re: Lilo

[Konrad Mierendorff]

Mats Johansson wrote:
> And the problem is that the LILO prompt comes up on the on screen and 
> the boot prompt also comes up becouse i use the prompt option but i 
> can't typ anything it is very strange and when scroll look is down it 
> come up alot  of *interrupted* messages as it should but i cant'stil 
> type anything 

Solution: Don't press scroll lock :-)

> , and when the timout has readched my linux boots as 
> usual.I have installed Linux with lilo a lto of times before and never 
> got this problem.


> delay=20
  

You might try skipping this one as a delay is rather senseless when
prompting for input.

> timeout=20
> 
> compact
> prompt

- Konrad

Re: non-US

[Colin Watson]

[EMAIL PROTECTED] (Konstantin Kivi) wrote:
>ëÁË ÎÁÐÉÓÁÔØ ÓÔÒÏÞËÕ × sources.lst
>ÄÌÑ potato ÐÒÏ non-us?
>ôÁÍ ËÁËÏÅ ÔÏ ÓÅÊÞÁÓ ÎÅÔÒÉ×ÉÁÌØÎÏÅ ÒÁÚÂÉÅÎÉÅ

It would have probably been more useful in English (as that's the
primary language of most of the debian-* lists), but I assume you're
looking for the following lines:

deb ftp://non-us.debian.org/debian-non-US potato non-US/main non-US/contrib 
non-US/non-free
deb-src ftp://non-us.debian.org/debian-non-US potato non-US/main non-US/contrib 
non-US/non-free

-- 
Colin Watson   [EMAIL PROTECTED]

Re: non-US

[Oleg Krivosheev]

On Thu, 20 Jan 2000, Konstantin Kivi wrote:

> ???  ??? ? sources.lst
> ??? potato ??? non-us?
> ??? ? ?? ?? ? ?

deb ftp://non-us.debian.org/debian-non-US unstable non-US/main non-US/contrib 
non-US/non-free


> 
> -- 
> Sincerely yours, Konstantin Kivi <[EMAIL PROTECTED]>
> 

OK

Re: Boot

[Dave Blears]

On Thu, Jan 20, 2000 at 07:34:40AM -0800, aphro wrote:

> the problem is the kernel on the debian cd is too old to run on athlon,
> its probably running linux 2.2.12, i hear 2.0.36 works but if you use
> 2.2.x i believe you need 2.2.13 or newer.


slink/2.0.36 locks at detecting the ide controllers when i try an install on
an athlon 600 based system with an msi mainboard.

both freebsd 3.3 and sco openserver 5.0.5 install fine... sigh

> 
> only thing I can suggest is install to another machine and move the HD
> over..you can try to make your own boot/root disk but i've never tried it.
> 

i will be trying mkrboot today to see if i can get it going with 2.2.14


-- 

Best Regards

Dave

---
Closer To Home Systems * Shop 6, 44 King Street * Caboolture, Qld, 4510
Ph 07 5499 3000 * Fax 07 5499 1822 * Mob 0416 173 522 * [EMAIL PROTECTED]
ICQ 1955906 * We Specialise in Computing Solutions and Personal Support
New Systems * Upgrades * Repairs * Small Business Networking * Software
---

non-US

[Konstantin Kivi]

Как написать строчку в sources.lst
для potato про non-us?
Там какое то сейчас нетривиальное разбиение

-- 
Sincerely yours, Konstantin Kivi <[EMAIL PROTECTED]>

gs/magicfilter on a hp lj3

[Lewis, James M.]

I need to make magicfilter (ljet3-filter) or gs leave a bit more
left margin.  The lj3 has about .25 in that can't be written on
and the output gets cut a little.  Does anyone know how to make
gs and/or magicfilter leave more left margin?

At the moment, I can use a2ps with the --margin option but I 
would like to fix it so I don't have to change a2ps options
when changing printers.

tia
jim

Re: ncp (novel) wrongly locked

[aphro]

if you want lsof (its useful) id suggest grabbing it and recompiling it:

lsof 4.43 (latest revision at ftp://vic.cc.purdue.edu/pub/tools/unix/lsof)

you probably need the kernel source for the kernel you are running to
compile it.

nate


On Thu, 20 Jan 2000 [EMAIL PROTECTED] wrote:

hawk >
hawk >aphro added,
hawk >> did u check lsof? it would show if something is using that partition..i
hawk >> just checked one of my machines and saw that one partition was in use,
hawk >> just because i started a program on which forked to the
hawk >> backgroupd(vpnd) it is not"using" the partition at all but lsof shows it
hawk >> as having the mount point as an open file
hawk >
hawk >I just tried it, and get 
hawk >
hawk >hawkins:~# lsof /dev/hda2
hawk >lsof: can't determine system map file path
hawk >
hawk >The only thing in the man page that seems to be vaguely relevantis
hawk >
hawk >Linux 2.0.3[234] and 2.1.8[89] for Intel-based systems
hawk >
hawk >--I have 2.0.36
hawk >
hawk >I get the same result as user and as root.
hawk >
hawk >rick
hawk >
hawk >-- 
hawk >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
12:47pm up 154 days, 52 min, 1 user, load average: 1.34, 1.56, 1.62

Re: potato - where's lp

[John Lapeyre]

> On Wed, Jan 19, 2000 at 06:44:25AM -0800, Kenneth Scharf wrote:
> > One last question is potato now linked to frozen? 
> > Should I change my sources.list to point to frozen or
> > potato instead of unstable now?

 As a rule of thumb, you should keep using potato, as it is more urgent
to get it into release shape.  You may notice a bug that you just
have to report, or an incompatibility  may arise that affects your package.
But the release-police won't kick in your door if you run an unstable
distribution.


-- 
John Lapeyre <[EMAIL PROTECTED]>,  [EMAIL PROTECTED]
Tucson,AZ http://www.physics.arizona.edu/~lapeyre

RE: char-major-6

[Lewis, James M.]

> What is char-major-6?  Where do I find this information so I don't have
> to bother clueful people about it?
> 
> TIA,
> 
Try /usr/src/linux/Documentation/devices.txt

jim
> --
> Nathan Norman
> MidcoNet  410 South Phillips Avenue  Sioux Falls, SD
> mailto:[EMAIL PROTECTED]   http://www.midco.net
> finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9)
> 
> 
> 
> -- 
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] <
> /dev/null
> 

dropping libc 2.1 into stable for some development work possible?

[Britton]

I have been told that I probably need the features of glibc 2.1 for some
of the development I am doing, however, I don't want to risk an upgrade to
unstable at the moment.  

Is it possible to get the upstream library, install it in /usr/local, and
then compile certain packages against it (static or whatever is fine for
the moment) without disturbing the rest of the system?  Will I see full
benefits of newer conformance symbols and everything?

How much work am I looking at to do this?  I have never done it before and
if it's going to be really confusing or require knowledge of many strange
compilation options or both I will probably just brave unstable.

Thanks for any advice you can offer me.

Britton Kerin
__
GNU GPL: "The Source will be with you... always."

Re: Remote xterm trouble

[Gary Hennigan]

"Gary Hennigan" <[EMAIL PROTECTED]> writes:
> Is anyone else having remote xterm trouble? If I log in to my Debian
> laptop, running potato, from my SGI, running IRIX 6.5, via an xterm
> the terminal is almost unusable. It seems to lose track of the
> cursor. For example, if I fire up dselect and I start moving down the
> list of choices when I get to the bottom, [Q]uit, and keep scrolling
> [Q]uit appears to move to the top of the screen. A similiar phenomena
> happens when I "vi" a file in the terminal or do a "man
> ". Debian seems to get confused about how to handle the
> terminal. In the case of man I get almost continuous underlining and
> reverse-video in areas of the man page that shouldn't have this type
> of highlighting.
[snip]

For those of you interested I found a bug report dealing with
this. It's #54833 and is filed against libncurses4. I still don't know
how it was determined to be a libncurses4 problem, but at least it's
known and in the Debian system. Hopefully that means it'll be fixed.

Gary

char-major-6

[Nathan E Norman]

What is char-major-6?  Where do I find this information so I don't have
to bother clueful people about it?

TIA,

--
Nathan Norman
MidcoNet  410 South Phillips Avenue  Sioux Falls, SD
mailto:[EMAIL PROTECTED]   http://www.midco.net
finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9)

Re: ncp (novel) wrongly locked

[hawk]

aphro added,
> did u check lsof? it would show if something is using that partition..i
> just checked one of my machines and saw that one partition was in use,
> just because i started a program on which forked to the
> backgroupd(vpnd) it is not"using" the partition at all but lsof shows it
> as having the mount point as an open file

I just tried it, and get 

hawkins:~# lsof /dev/hda2
lsof: can't determine system map file path

The only thing in the man page that seems to be vaguely relevantis

Linux 2.0.3[234] and 2.1.8[89] for Intel-based systems

--I have 2.0.36

I get the same result as user and as root.

rick

-- 

Re: mail confusion

[aphro]

fetchmail sends mail by resending it through the local SMTP server, try to
telnet to your smtp server (telnet localhost 25) chances are you dont have
a SMTP server running, or it is firewalled

nate

On Thu, 20 Jan 2000, Paul wrote:

skin >It`s a long story but in order to install fidogate I replaced my
skin >perfectly working exim with sendmail. Now when I try to collect my mail
skin >with fetchmail I get:
skin >
skin >reading message 1 of 10 (1598 header octets)
skin >fetchmail: SMTP connect to localhost failed
skin >fetchmail: IMAP> A0007 LOGOUT
skin >fetchmail: IMAP< )
skin >fetchmail: IMAP< A0006 OK Completed
skin >fetchmail: IMAP< * BYE LOGOUT received
skin >fetchmail: IMAP< A0007 OK Completed
skin >fetchmail: SMTP transaction error while fetching from pop.dial.pipex.com
skin >fetchmail: Query status=10
skin >fetchmail: normal termination, status 10 
skin >
skin >
skin >If I test sendmail with  "sendmail -v" it says :
skin >
skin > Recipient names must be specified
skin >
skin >I would imagine that this is my problem but I`ve read loads of Docs. and
skin >still can`t work out what`s wrong.
skin >Easy with the advice please  - my head hurts!
skin >
skin >Paul
skin >
skin >-- 
skin >Paul Walton * Powered by * 
skin >Cambridge* Debian GNU Linux *   
skin >U.K.   * http://www.debian.org *
skin >  * PGP Public Key: http://pgp5.ai.mit.edu/ *
skin >
skin >
skin >-- 
skin >Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
skin >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
11:24am up 153 days, 23:29, 1 user, load average: 1.40, 1.60, 1.61

Re: ppp/chat: BLACKLISTED after (many) tries

[Wojciech Zabolotny]

On Wed, Jan 19, 2000 at 12:26:19AM +0100, Konrad Mierendorff wrote:
> Henrique M Holschuh wrote:
> > Some modems can be configured to believe they're in the wrong country and
> > ignore the blacklisting thing. 
> 
> ALMOST ALL modems can be switched off and on again :-)
> And this is legal!
> 
> A german user who has made good experiences with switching off/on his
> modem before redialing ;-}
> 
If you have an external modem, you can even build a simple hardware device,
controlled by one of unused RS-232 lines, (or by one pin in the printer port),
which will switch off/on your modem under the software controll.
(Switching modem on and off after each two trials, when you have to try
100 times before you get connected is really tiring :-)).
-- 
Wojciech Zabolotny
http://www.ise.pw.edu.pl/~wzab

http://www.freedos.org  Free DOS for free people!

Re: mail confusion

[Gary Hennigan]

Paul <[EMAIL PROTECTED]> writes:
> It`s a long story but in order to install fidogate I replaced my
> perfectly working exim with sendmail. Now when I try to collect my mail
> with fetchmail I get:
> 
> reading message 1 of 10 (1598 header octets)
> fetchmail: SMTP connect to localhost failed
The above is the problem. Sendmail is not allowing fetchmail to
connect to your local host. You can test this with a simple:
  telnet localhost 25

If everything's working fine you should connect and get a prompt from
sendmail. If it fails, so will fetchmail.

[snip] 
> 
> If I test sendmail with  "sendmail -v" it says :
> 
>  Recipient names must be specified
> 
> I would imagine that this is my problem but I`ve read loads of Docs. and
> still can`t work out what`s wrong.
> Easy with the advice please  - my head hurts!

No, you need to tell sendmail what to use as an address when you test
it. Like:

sendmail -v [EMAIL PROTECTED]
This is a test.
.

That's how you get it a test run working. 

As I stated though, I don't believe this is the problem. It's that
sendmail isn't accepting connections from localhost. I'm not that
familiar with sendmail but I suspect you just need to add something in
the sendmail.cf file to tell it that localhost is allowed to use
sendmail as a relay.

Gary

mail confusion

[Paul]

It`s a long story but in order to install fidogate I replaced my
perfectly working exim with sendmail. Now when I try to collect my mail
with fetchmail I get:

reading message 1 of 10 (1598 header octets)
fetchmail: SMTP connect to localhost failed
fetchmail: IMAP> A0007 LOGOUT
fetchmail: IMAP< )
fetchmail: IMAP< A0006 OK Completed
fetchmail: IMAP< * BYE LOGOUT received
fetchmail: IMAP< A0007 OK Completed
fetchmail: SMTP transaction error while fetching from pop.dial.pipex.com
fetchmail: Query status=10
fetchmail: normal termination, status 10 


If I test sendmail with  "sendmail -v" it says :

 Recipient names must be specified

I would imagine that this is my problem but I`ve read loads of Docs. and
still can`t work out what`s wrong.
Easy with the advice please  - my head hurts!

Paul

-- 
Paul Walton * Powered by * 
Cambridge* Debian GNU Linux *   
U.K.   * http://www.debian.org *
  * PGP Public Key: http://pgp5.ai.mit.edu/ *

Re: Epplet question

[John Miskinis]

Thanks for your reply. I will try it :)
The logout thing might be interesting. I usually
just do "init 0" as root to turn of my machine.


The CTRL-ALT-DELETE is basically the same as selecting the
"Logout" option from the mouse button 1 menu.  I wanted to add
another note, that someone on #E suggested.  They suggested
that I might be OK logging out the bad ways (CTRL-ALT-BACKSPACE
or "init 0" as you did) as long as no NEW epplets were started.
I tried this for a while, making sure to logout correctly whenever
I tried a NEW epplet, and although it greatly reduced the frequency
of epplet problems, now and then they would come back duplicated.

It appears to be caused by E's "session management" capabilities,
but that's about all I know...

I wish there was a way to turn that stuff off completely, and start
epplets from .xsession instead.  I start my wmaker dock apps this
way, as it's easy to comment them out, when I need to run color
intensive apps.

At some point I may grab the E sources, and see what's going on,
but for now I switch to wmaker when I get too frustrated, but
I really love E, and always come back to it.

For what it is worth ...

John
__
Get Your Private, Free Email at http://www.hotmail.com

Re: stable innd is segfaulting

[Colin Watson]

[EMAIL PROTECTED] (Miquel van Smoorenburg) wrote:
>In article <[EMAIL PROTECTED]>,
>Colin Watson  <[EMAIL PROTECTED]> wrote:
>>Eventually I traced the problem to innd itself dying when innxmit tries
>>to submit articles to it. The tail of an strace looks like this
>>(apologies for the long lines):
>
>There is a bug in innd with a certain kind of headers - if an
>article contains "Distribution: ," innd will fall over and die.
>Ofcourse it didn't accept the article yet, so after a restart it
>will most likely receive the same article again, and keep on dying

That sounds very plausible; I now wish I'd kept the article that caused
this to happen, but at the time I had got myself fixated on the idea
that it must be something in the system rather than a particular article
(as we'd been fiddling with the INN configuration immediately before it
happened).

Various workarounds in suck suggest themselves; I'll hack something
together. Thanks!

>The From: and Reply-To: addresses are internal news2mail gateway addresses.
>Reply to the list or to [EMAIL PROTECTED] (Miquel van Smoorenburg)

Are they? I see that address in your From: header, and no Reply-To:
header (though the latter could be an artefact of my own mail2news
setup). Does this .sig only apply internally?

-- 
Colin Watson   [EMAIL PROTECTED]

rsync:"@ERROR: chroot failed"

[Gerhard Kroder]

i run into problems using rsync to [EMAIL PROTECTED]: chroot failedr debians
ftp archive, which worked for a while. but now all i get is this error
message: "@ERROR: chroot failed". i don't even know who makes this
error. i tried to do it manually, as follows:

> sh-2.03$ LOGFILE=rsync.debian.log
> sh-2.03$ TO=/mnt/mirror/debian
> sh-2.03$ #MIRRORTOP=/mnt/hydra/mirror
> sh-2.03$ MIRRORTOP=/mnt/mirror
> sh-2.03$ RSYNC_HOST=ftp.de.debian.org
> sh-2.03$ RSYNC_DIR=debian/
> sh-2.03$ LOCK="${TO}/Archive-Update-in-Progress-mail.data-sciences.de"
> sh-2.03$ cd $MIRRORTOP
> sh-2.03$ pwd
> /mnt/mirror
> sh-2.03$ umask 002
> sh-2.03$ if lockfile -! -l 43200 -r 0 "$LOCK"; then
> >   echo `hostname` is unable to start rsync, lock file exists
> >   exit 1
> > fi
> sh-2.03$ echo $?
> 0
> sh-2.03$ trap "rm -f $LOCK > /dev/null 2>&1" exit
> sh-2.03$ echo $?
> 0
> sh-2.03$ set +e
> sh-2.03$ rsync -v -rltvz \
> >  $EXCLUDE \
> >  $RSYNC_HOST::$RSYNC_DIR $TO
> @ERROR: chroot failed
> sh-2.03$ rsync -v -v -rltvz  $EXCLUDE  $RSYNC_HOST::$RSYNC_DIR $TO
> @ERROR: chroot failed
> sh-2.03$ rsync -v -v -v -rltvz  $EXCLUDE  $RSYNC_HOST::$RSYNC_DIR $TO
> @ERROR: chroot failed
> sh-2.03$ rsync -v -v -v -v -rltvz  $EXCLUDE  $RSYNC_HOST::$RSYNC_DIR 
> $TO
> @ERROR: chroot failed
> sh-2.03$ 




any idea what's going on? maybe some conflict with nis, which is running
here, too?

some more infos to my versions:

> Subject: rsync: @ERROR: chroot failed
> Version: 2.3.2-1.1
> Severity: normal
> 
> -- System Information
> Debian Release: potato
> Kernel Version: Linux stardust 2.2.13 #1 Sat Nov 20 12:44:19 EST 1999 i686 
> unknown
> 
> Versions of the packages rsync depends on:
> ii  libc62.1.2-11 GNU C Library: 
> Shared libraries and Timezone data

gerhard

Re: ncp (novel) wrongly locked

[aphro]

did u check lsof? it would show if something is using that partition..i
just checked one of my machines and saw that one partition was in use,
just because i started a program on which forked to the
backgroupd(vpnd) it is not"using" the partition at all but lsof shows it
as having the mount point as an open file

nate

On Thu, 20 Jan 2000 [EMAIL PROTECTED] wrote:

hawk >
hawk >I'm using the ncpfs to use space on networked novell drives.  It works 
hawk >most of the time with most applications, but it has a tendency to 
hawk >halfway dismount itself on errors.  Some I can reproduce, but they all 
hawk >involve manually dismounting then remounting.
hawk >
hawk >This time, though, it wrongly concluded that the file system was still 
hawk >in use, and refused to dismount.  I could find no way to remount other 
hawk >than to reboot!  I searched for lock files in the usual places in /var, 
hawk >but found nothing.
hawk >
hawk >Is there some other way to deal with this situation?
hawk >-- 
hawk >
hawk >
hawk >
hawk >-- 
hawk >Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
hawk >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
10:11am up 153 days, 22:15, 1 user, load average: 1.64, 1.53, 1.56

Re: Athlon kernel compilation blues

[José Luis Gómez Dans]

I have been trying to put everybody's suggestions on compiling
the kernel. Now, I have traced the problem to the initialization of
modutils in in rcS. Basically, depmod -a > /dev/null segfaults and then
a variety of kernel not being able to use NULL pointers pop up,
eventually halting the system.

If I boot with the old kernel, and I try to the depmod line, it
works fine. In fact, if I issue now 
# depmod -a 2.2.14 -F /boot/System.map-2.2.14, I get

/lib/modules/2.2.14/fs/nls_iso8859-1.o:

/lib/modules/2.2.14/fs/fat.o:

/lib/modules/2.2.14/fs/vfat.o:  /lib/modules/2.2.14/fs/fat.o

/lib/modules/2.2.14/fs/nls_iso8859-15.o:

/lib/modules/2.2.14/net/rtl8139.o:

/lib/modules/2.2.14/net/dummy.o:

/lib/modules/2.2.14/scsi/scsi_mod.o:

/lib/modules/2.2.14/scsi/sd_mod.o:  /lib/modules/2.2.14/scsi/scsi_mod.o

/lib/modules/2.2.14/scsi/ppa.o: /lib/modules/2.2.14/misc/parport.o \
/lib/modules/2.2.14/scsi/scsi_mod.o

/lib/modules/2.2.14/block/ide-cd.o: /lib/modules/2.2.14/cdrom/cdrom.o

/lib/modules/2.2.14/cdrom/cdrom.o:

/lib/modules/2.2.14/misc/parport_pc.o:  /lib/modules/2.2.14/misc/parport.o

/lib/modules/2.2.14/misc/esssolo1.o:/lib/modules/2.2.14/misc/soundcore.o

/lib/modules/2.2.14/misc/lp.o:  /lib/modules/2.2.14/misc/parport.o

/lib/modules/2.2.14/misc/parport.o:

/lib/modules/2.2.14/misc/soundcore.o:

which suggests that the module dependencies have been calculated without
any problem whatsoever. My /etc/modules file is

# /etc/modules: kernel modules to load at boot time.
noauto
vfat
esssolo1
serial
parport
parport_pc
rtl8139

so all modules here are present on the above file. I could really skip
the noauto stuff, but it *shouldn't* harm. As for the kernel compilation
line, this goes as follows:

-D__KERNEL__ -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -pipe
-fno-streght-reduce -mpreferred-stack-boundary=2 -m486 -malign-loop=2
-malign-jumps=2 -malign-functions=2 -DCPU=586

The CPU is an AMD Athlon at 550 MHz, so I *guess* I have the
right DCPU option. I do a make dep ; make clean ; make zImage, copy the
resultant zImage to /boot/vmlinuz-2.2.14 and then run lilo
appropriately.

And it doesn't work! I'm getting somewhat frustrated by all
this, it used to work first time on the other computers!

Thanks for your help
José

-- 
José L Gómez Dans   PhD student
Radar & Communications Group
Department of Electronic Engineering
University of Sheffield UK

help using 3COM PCMCI Ethernet card 3CCE589ET

[Frank Mattes]

Hi,

I have installed debian 2.1 new on my thinkpad. PCMCIA ethernet 
support worked
with out any modification when I used a Dayna CommuniCard. However 
installing debian with
a 3COM Megahertz Ethernet Card Model 3CCe569ET did't work.

Does anybody can help me how to set up debian that I can use this card 
instead of the
Dayne Card.

Many thanks Framl

ncp (novel) wrongly locked

[hawk]

I'm using the ncpfs to use space on networked novell drives.  It works 
most of the time with most applications, but it has a tendency to 
halfway dismount itself on errors.  Some I can reproduce, but they all 
involve manually dismounting then remounting.

This time, though, it wrongly concluded that the file system was still 
in use, and refused to dismount.  I could find no way to remount other 
than to reboot!  I searched for lock files in the usual places in /var, 
but found nothing.

Is there some other way to deal with this situation?
-- 

Re: apt changes mount status of /usr ?!?!!

[Clyde Wilson]

Check out /etc/apt/apt.conf.  If you remove the lines that
remount /usr ( 2 I think ) everything should work fine.

On Thu, 20 Jan 2000, Alexander List wrote:

> 
> Obviously, apt seems to fiddle around with mount during the upgrade
> procedure, /usr is mounted ro after the upgrade!!!
> 
> On another system using slink, I noticed that apt doesn't work correctly
> if there's no separate mount point for /usr, eg when you put everything in
> one big partition. Did anyone else encounter similar problems?
> 

system

[Hutchinson, Floyd]

I just install Debian Linux on a Dell 133Mhz with 16MB of ram.  However
after I install debian and reboot when it ask me to reboot it stop booting
at the message line Partition check the came back with a error message.
Irq timeout: status=0x50 {driveready seek complete} and hangs again.

Re: Unidentified subject!

[Jan Ludewig]

On Thu, Jan 20, 2000 at 09:03:22AM -0500, Jan Ludewig wrote:
> unsubscribe [EMAIL PROTECTED]
sorry, i should not use mailing-list-handling scripts in
alpha-state.

Jan

Unidentified subject!

[Jan Ludewig]

unsubscribe [EMAIL PROTECTED]

Re: Athlon kernel compilation blues

[Dänzer]

--- José_Luis_Gómez_Dans <[EMAIL PROTECTED]> wrote:
> On Thu, Jan 20, 2000 at 07:56:43AM -0800, Michel Dänzer wrote:
> > This is not needed anymore with 2.95.2
> > 
> > On the contrary, if you want to build apps with strict aliasing, you have
> to
> > add -fstrict-aliasing.
> 
>   So what flags should I use for compiling the kernel then? I
> don't really know what this strict-aliasing is ;)

It can optimize better, but only works with "correct C programs", which the
kernel is not, just like it's not an app.

In summary: Don't add it :)


Michel


=
"Software is like sex; it's better when it's free"
 -- Linus Torvalds

"If you continue running Windows, your system may become unstable."
 -- Windows 95 BSOD
__
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

Yamaha YMF 724

[Andreas Sliwka]

Hi, 
  I got a new board with an Yamaha YMF 724 on board. How can I get this 
chip to play sound?   

mfg 
-- 
Andreas Sliwka   | [EMAIL PROTECTED] | ICQ goff:13961062

Re: Athlon kernel compilation blues

[José Luis Gómez Dans]

On Thu, Jan 20, 2000 at 07:56:43AM -0800, Michel Dänzer wrote:
> This is not needed anymore with 2.95.2
> 
> On the contrary, if you want to build apps with strict aliasing, you have to
> add -fstrict-aliasing.

So what flags should I use for compiling the kernel then? I
don't really know what this strict-aliasing is ;)

> I deleted your original post, but do you run /sbin/lilo after you have built
> the kernel? What target do you build (zImage, bzImage, ...)?

I have run lilo, and as for the target, I tend to run the kernel
package (whichever that one makes).

-- 
José L Gómez Dans   PhD student
Radar & Communications Group
Department of Electronic Engineering
University of Sheffield UK

Re: Remote xterm trouble

[Gary Hennigan]

Roger Pittman <[EMAIL PROTECTED]> writes:
> On Tue, Jan 18, 2000 at 02:16:36PM -0700, Gary Hennigan wrote:
> > Is anyone else having remote xterm trouble? ...
> 
>   I am.  Remote boxes are Ultra 1's, AXi's, and AXMP's, running
> Solaris 2.6 or 7.  Worked fine until a recent apt-get run.  If I
> telnet in via an xterm, well, it's hosed.  Eg in vi, the ':q' to
> quit displays on the top line, trying to move the cursor may
> appear to do nothing, may jump around by several rows and/or
> columns at a time, may disappear completely, who knows.  $TERM,
> $LINES, and $COLUMNS are all correct.  Remote access via a terminal
> window (dtterm) still works ok, so getting by for now.

Exactly the same problem! I'd fire off a bug report but so much was
upgraded during my apt-get run that I don't know where to start.

Anyone have any suggestions on how to go about finding the offending
piece of software? My best guess is that the xterm definition got
hosed up somehow, but an xterm works fine if you log in locally.

Gary

Re: Ethernet card dies unexpectedly

[José Luis Gómez Dans]

On Thu, Jan 20, 2000 at 07:42:18AM -0800, aphro wrote:
> what kind of machine is it on? i had a similar problem on an old P100 IBM
> machine it wouldnt work with any PCI NIC for longer then 10-15 minutes,

This is an Athlon 550 MHz. I have downloaded the newest driver
version (1.08 as opposed to 1.06 in the kernel) and will try with that.
I will also try to compile the kernel with just -O to see if it survives
:).

Thanks for your help!
José
-- 
José L Gómez Dans   PhD student
Radar & Communications Group
Department of Electronic Engineering
University of Sheffield UK

Re: Athlon kernel compilation blues

[Dänzer]

--- José_Luis_Gómez_Dans <[EMAIL PROTECTED]> wrote:

> Reading specs from /usr/lib/gcc-lib/i386-linux/2.95.2/specs
> gcc version 2.95.2 2116 (Debian GNU/Linux)
> 
>   Optimization is set to O2. However, I also issue the
> -fno-strict-aliasing option, as recommended by
> /usr/doc/gcc/README.Debian.gz.

This is not needed anymore with 2.95.2

On the contrary, if you want to build apps with strict aliasing, you have to
add -fstrict-aliasing.


>   I suppose I could try compiling it without optimization. It's
> not a massive piece of kernel anyway :).
> 
>   However, I do get a lot of NULL pointer dereference errors.

I deleted your original post, but do you run /sbin/lilo after you have built
the kernel? What target do you build (zImage, bzImage, ...)?


Michel


=
"Software is like sex; it's better when it's free"
 -- Linus Torvalds

"If you continue running Windows, your system may become unstable."
 -- Windows 95 BSOD
__
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

Re: Athlon kernel compilation blues

[José Luis Gómez Dans]

--- Begin Message ---
I'm using standard stuff found in potato. gcc -v returns:

Reading specs from /usr/lib/gcc-lib/i386-linux/2.95.2/specs
gcc version 2.95.2 2116 (Debian GNU/Linux)

Optimization is set to O2. However, I also issue the
-fno-strict-aliasing option, as recommended by
/usr/doc/gcc/README.Debian.gz.

I suppose I could try compiling it without optimization. It's
not a massive piece of kernel anyway :).

However, I do get a lot of NULL pointer dereference errors.

Thanks,
José
-- 
José L Gómez Dans   PhD student
Radar & Communications Group
Department of Electronic Engineering
University of Sheffield UK
--- End Message ---

Re: slink mirror

[Nathan E Norman]

On Thu, 20 Jan 2000, Timothy C. Phan wrote:

 : Hi Nathan,
 : 
 :   Many thanks for you support.  I guess you are right about
 :   large sites.
 : 
 :   I finally try to mirror the debian.crosslink.net and got
 :   the complete slink (i386) minus the source mirrored.
 : 
 :   So I'll try to upgrade my 1.3.1 to slink this weekend.

I forgot to mention that ftp.debian.org is usually swamped ... if you
find a mirror that works for you, cool :)  Feel free to try
debian.midco.net; we should be fairly fast to most places.

Good luck!!

--
Nathan Norman
MidcoNet  410 South Phillips Avenue  Sioux Falls, SD
mailto:[EMAIL PROTECTED]   http://www.midco.net
finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9)

Re: Ethernet card dies unexpectedly

[aphro]

what kind of machine is it on? i had a similar problem on an old P100 IBM
machine it wouldnt work with any PCI NIC for longer then 10-15 minutes,
the network would die as well..my solution was to just use an ISA card.

you could also try upgrading the driver.. check NASA's site-- i forget the
URL go to http://yahoo.aphroland.org and search for nasa i got a link to
it. Even if you have the latest kernel chances are the NIC drivers are
still very old.

nate

On Thu, 20 Jan 2000, [iso-8859-1] Jos? Luis G?mez Dans wrote:

j.l.go >Hi!
j.l.go >I'm having some trouble with my ethernet card (a PCI fast
j.l.go >ethernet RealTek 8139). Basically, this card sits in 0xe800, IRQ 11 and
j.l.go >is perfectly detected by the rt8139 driver. It is configured to work in
j.l.go >half duplex mode.
j.l.go >
j.l.go >When the system boots up, it all works fine, it is
j.l.go >connected to our local network (10Mb Ethernet), and normal operation
j.l.go >starts. However, if I try to download a large file off the network, the
j.l.go >collisions LED in the hub starts flashing. When I had another computer
j.l.go >here with an ISA card, none of this happened.
j.l.go >
j.l.go >Moreover, I have been experiencing problems with the downloaded 
j.l.go >files: it turns out that sometimes the archives are corrupted. I have
j.l.go >downloaded Debian packages and have had probmes as they cannot be
j.l.go >decompressed.
j.l.go >
j.l.go >In the end, and after a while of normal use (only as a client
j.l.go >for internet uses, such as e-mail, web, usenet, ftp and telnet; no fancy
j.l.go >networked filesystems or stuff like that), the network becomes invisible
j.l.go >to my machine. The ifconfig configuration stays there, with proper IP
j.l.go >address and all that. However, route and netstat seem to freeze up when
j.l.go >showing my gateway table. In the end, the default gateway never shows
j.l.go >up. The only way around this has been to reboot.
j.l.go >
j.l.go >Does anyone know what could be going wrong here?
j.l.go >Thanks for your help,
j.l.go >Jos?
j.l.go >
j.l.go >-- 
j.l.go >Jos? L G?mez Dans   PhD student
j.l.go >Radar & Communications Group
j.l.go >Department of Electronic 
Engineering
j.l.go >University of Sheffield UK
j.l.go >
j.l.go >
j.l.go >-- 
j.l.go >Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
j.l.go >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
7:28am up 153 days, 19:33, 1 user, load average: 1.65, 1.66, 1.59

Re: slink mirror

[Timothy C. Phan]

Hi Nathan,

  Many thanks for you support.  I guess you are right about
  large sites.

  I finally try to mirror the debian.crosslink.net and got
  the complete slink (i386) minus the source mirrored.

  So I'll try to upgrade my 1.3.1 to slink this weekend.

  Again, thank you!

Nathan E Norman wrote:
> 
> On Mon, 17 Jan 2000, Timothy C. Phan wrote:
> 
>  : Hi Nathan,
>  :
>  :   Thanks for the information.  I, however, have problem using rsync because
>  :   I'm still using 1.3.1 and rsync required libc6.  Is there another way 
> around
>  :   this.  Does 'mirror' is no longer used for mirror site?
> 
> mirror has trouble with large sites ... I would guess that many mirrors
> still use it though.  I don't use it except for small sites, so I don't
> know if I can answer your question or not :( (what was the original
> problem?)
> 
> You could download the rsync source package from potato or slink and try
> compiling it locally.  Is upgrading to libc6 (this would take you from
> hamm to slink?) out of the question?
> 
> --
> Nathan Norman
> MidcoNet  410 South Phillips Avenue  Sioux Falls, SD
> mailto:[EMAIL PROTECTED]   http://www.midco.net
> finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9)
> 
> --
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null

Re: Athlon kernel compilation blues

[aphro]

what compiler are you using? what optimization do you have the kernel set
to?

nate

On Thu, 20 Jan 2000, [iso-8859-1] Jos? Luis G?mez Dans wrote:

j.l.go >Dear all,
j.l.go >I have an Athlon 550 MHz system in which I installed Potato.
j.l.go >It's all good with the installation boot disks and all that. The only
j.l.go >problem comes when trying to compile a new kernel other than the one
j.l.go >that came with the distro. So I downloaded the sources for 2.2.14 and
j.l.go >configured it (I can provide you with the Makefile, I won't send it to
j.l.go >the list so as not to clutter it). It's a pretty easy system: normal
j.l.go >EIDE hard drive, CD-ROM, serial mouse, keyboard and an ESS Solo1
j.l.go >soundcard. I don't even need networked filesystems (at least not for the
j.l.go >time being), so all there is is some network support (I use this as a
j.l.go >workstatio from which I read mail and news, and browse the web). Most of
j.l.go >the stuff is located as modules, with kmod support and all that. Oh, and
j.l.go >there's also a RealTek 8139 PCI network card.
j.l.go >
j.l.go >After compiling a number of kernels (both using and not using
j.l.go >kernel-package), I always have to restore to a bootdisk I created with
j.l.go >the installation's default kernel. Basically, I get loads of different
j.l.go >errors: from CRC errors when decompressing the kernel (!) to dodgy
j.l.go >pointr assignments and stuff that outputs tons of debugging information.
j.l.go >I get kernel panics and stuff like that.
j.l.go >
j.l.go >Obviously, I am missing some kernel option here which is causing
j.l.go >all these problems. I don't really know where the problem lies, and I
j.l.go >don't know where to look in order to sort it out. Does anyone have any
j.l.go >suggestions? I suppose I could go on using the default kernel, but I
j.l.go >find it annoying not knowing what's going on here!
j.l.go >
j.l.go >Thanks,
j.l.go >Jos?
j.l.go >-- 
j.l.go >Jos? L G?mez Dans   PhD student
j.l.go >Radar & Communications Group
j.l.go >Department of Electronic 
Engineering
j.l.go >University of Sheffield UK
j.l.go >
j.l.go >
j.l.go >-- 
j.l.go >Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
j.l.go >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
7:28am up 153 days, 19:33, 1 user, load average: 1.65, 1.66, 1.59

Re: Boot

[aphro]

the problem is the kernel on the debian cd is too old to run on athlon,
its probably running linux 2.2.12, i hear 2.0.36 works but if you use
2.2.x i believe you need 2.2.13 or newer.

only thing I can suggest is install to another machine and move the HD
over..you can try to make your own boot/root disk but i've never tried it.

nate

On Thu, 20 Jan 2000, Nicola Tessari wrote:

nitess >Hello there,
nitess >I'm here because my official debian cd (bought at linuxcentral) doesn't
nitess >boot...
nitess >After few seconds the booting process in my machine stops at the line:
nitess >md driver 0.36.3 MAX_MD_DEV=4, MAXREL=8
nitess >Anyone could tell me which is the problem?
nitess >I've choosen the hardware checking out if it was supported (k7 500MHz
nitess >with asus k7m, cd creative 48x).
nitess >Slackware 7.0 boots well (from linux central too)
nitess >Thanks a lot in advance,
nitess >Nicola
nitess >
nitess >
nitess >-- 
nitess >Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
nitess >

[mailto:[EMAIL PROTECTED] ]--
   Vice President Network Operations   http://www.firetrail.com/
  Firetrail Internet Services Limited  http://www.aphroland.org/
   Everett, WA 425-348-7336http://www.linuxpowered.net/
Powered By:http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMPhttp://yahoo.aphroland.org/
-[mailto:[EMAIL PROTECTED] ]--
7:28am up 153 days, 19:33, 1 user, load average: 1.65, 1.66, 1.59

Ethernet card dies unexpectedly

[José Luis Gómez Dans]

Hi!
I'm having some trouble with my ethernet card (a PCI fast
ethernet RealTek 8139). Basically, this card sits in 0xe800, IRQ 11 and
is perfectly detected by the rt8139 driver. It is configured to work in
half duplex mode.

When the system boots up, it all works fine, it is
connected to our local network (10Mb Ethernet), and normal operation
starts. However, if I try to download a large file off the network, the
collisions LED in the hub starts flashing. When I had another computer
here with an ISA card, none of this happened.

Moreover, I have been experiencing problems with the downloaded 
files: it turns out that sometimes the archives are corrupted. I have
downloaded Debian packages and have had probmes as they cannot be
decompressed.

In the end, and after a while of normal use (only as a client
for internet uses, such as e-mail, web, usenet, ftp and telnet; no fancy
networked filesystems or stuff like that), the network becomes invisible
to my machine. The ifconfig configuration stays there, with proper IP
address and all that. However, route and netstat seem to freeze up when
showing my gateway table. In the end, the default gateway never shows
up. The only way around this has been to reboot.

Does anyone know what could be going wrong here?
Thanks for your help,
José

-- 
José L Gómez Dans   PhD student
Radar & Communications Group
Department of Electronic Engineering
University of Sheffield UK

Re: Strange LILO question

[Ron Rademaker]

> Is there then realy no way to boot the floppy also WITH a DOS filesystem?
> Probably not :-(

I think there is a way, just not using LILO, but loadlin. You just take a
DOS disk boot DOS and make some kind of menu in you're autoexec.bat file
with the choice between dos and linux. (For linux you just do a dos
command loadlin with some arguments.

Ron

Re: slink mirror

[Nathan E Norman]

On Mon, 17 Jan 2000, Timothy C. Phan wrote:

 : Hi Nathan,
 : 
 :   Thanks for the information.  I, however, have problem using rsync because
 :   I'm still using 1.3.1 and rsync required libc6.  Is there another way 
around
 :   this.  Does 'mirror' is no longer used for mirror site?

mirror has trouble with large sites ... I would guess that many mirrors
still use it though.  I don't use it except for small sites, so I don't
know if I can answer your question or not :( (what was the original
problem?)

You could download the rsync source package from potato or slink and try
compiling it locally.  Is upgrading to libc6 (this would take you from
hamm to slink?) out of the question?

--
Nathan Norman
MidcoNet  410 South Phillips Avenue  Sioux Falls, SD
mailto:[EMAIL PROTECTED]   http://www.midco.net
finger [EMAIL PROTECTED] for PGP Key: (0xA33B86E9)

Enlightenment CVS questions

[Joakim Svensson]

Hi all,

I would like to try out imlib2 and efm for enlightenment.
As a relative newbie how should I do this, or should I do
it at all ?
Do I need to use CVS and compile myself or is there some
sort of debian CVS (my search so far indicates this) that
should be used (this means I will compile myself but with
some sort of debianized stuff so I wont mess up).
I would like to try these things but wouldn't want to mess
up my debian packages.

Any hints are of interest.

Best regards
Joakim Svensson

RE: Cannot mount cdrom in new potato install

[Irving Frederick]

Thanks for all the replies. The problem was as
described and subsequently solve.

FRED

___
To get your own FREE ZDNet onebox - FREE voicemail, email, and fax,
all in one place - sign up today at http://www.zdnetonebox.com

CD's and such for a CD handout?

[Robert L. Harris]

  I'm trying to organize a CD handout on 2/17 at the local CompUSA.
RedHat is going to send me some stuff to hand out and a local vendor
is going to bring some things like bumper sticker's, some "tux" dolls,
etc.  Does anyone know any contacts at debian.org that can send me 
a cd or so of slink and/or potato that can be handed out and/or dupped
if needed?

Robert


:wq!
---
Robert L. Harris|  Low quality in a product happens.
Senior System Engineer  |That doesn't mean it's right and
  at RnD Consulting.|  definitely doesn't mean it should
\_   be accepted.  Require quality.

http://www.rnd-consulting.com/~nomad

DISCLAIMER:
  These are MY OPINIONS ALONE.  I speak for no-one else.

FYI:
 perl -e 'print $i=pack(c5,(41*2),sqrt(7056),(unpack(c,H)-2),oct(115),10);'

bootable CD on non cd booting machine?

[David Teague]

List folks 
and Richard --

The subject line almost says it all:

How can I make a bootable CD boot on hardware that does not support
booting from a CD? 

In particular, have the rescue CD from the Atlanta Linux Showcase,
but the machine I really want to use it on does not boot from CD.

I could put syslinux on floppy and point things to the CD or I could
dd the kernel from my system to a floppy and point things into the
cd.  The structure of that disk is a little strange to me.  Do you
have a suggestion on how to set things up for that disk?

If there is a more appropriate forum, a pointer will be appreciated. 
Meanwhile, I'll be hacking at this.

--David
David Teague, [EMAIL PROTECTED]
Debian GNU/Linux Because software support is free, timely,
 useful, technically accurate, and friendly.

Re: wrong CD!!!

[David Wright]

Quoting dkphoto ([EMAIL PROTECTED]):
> Found a great looking book at the bookstore that had the added bonus of 
> containing a Debian CD so I wouldn't have to spend the next 10 years 
> downloading. Only one problem... it only has platform specific files on 
> it for Wintel boxes. I have a 68K Mac.
> Now I assume that I can just use a handful of downloaded files along with 
> this CD for an install (I hope), but need to know which files to put 
> where to start, and if the setup will pull the files it wants off the CD.

The installation procedure would pull the right files off the CD except
they won't be there because the architecture is wrong. The few files
that would work are those which contain just documentation as they are
architecture-independant (they will be links in the m68k tree on the ftp
site).

Cheers,

-- 
Email:  [EMAIL PROTECTED]   Tel: +44 1908 653 739  Fax: +44 1908 655 151
Snail:  David Wright, Earth Science Dept., Milton Keynes, England, MK7 6AA
Disclaimer:   These addresses are only for reaching me, and do not signify
official stationery. Views expressed here are either my own or plagiarised.

Athlon kernel compilation blues

[José Luis Gómez Dans]

Dear all,
I have an Athlon 550 MHz system in which I installed Potato.
It's all good with the installation boot disks and all that. The only
problem comes when trying to compile a new kernel other than the one
that came with the distro. So I downloaded the sources for 2.2.14 and
configured it (I can provide you with the Makefile, I won't send it to
the list so as not to clutter it). It's a pretty easy system: normal
EIDE hard drive, CD-ROM, serial mouse, keyboard and an ESS Solo1
soundcard. I don't even need networked filesystems (at least not for the
time being), so all there is is some network support (I use this as a
workstatio from which I read mail and news, and browse the web). Most of
the stuff is located as modules, with kmod support and all that. Oh, and
there's also a RealTek 8139 PCI network card.

After compiling a number of kernels (both using and not using
kernel-package), I always have to restore to a bootdisk I created with
the installation's default kernel. Basically, I get loads of different
errors: from CRC errors when decompressing the kernel (!) to dodgy
pointr assignments and stuff that outputs tons of debugging information.
I get kernel panics and stuff like that.

Obviously, I am missing some kernel option here which is causing
all these problems. I don't really know where the problem lies, and I
don't know where to look in order to sort it out. Does anyone have any
suggestions? I suppose I could go on using the default kernel, but I
find it annoying not knowing what's going on here!

Thanks,
José
-- 
José L Gómez Dans   PhD student
Radar & Communications Group
Department of Electronic Engineering
University of Sheffield UK

PPP problem - HELP!!!

[A . Zurek]

Hi!

I'm pretty new on this list, but I think I my mails wount be too lame for 
you...
(sorry for my bad english)

Ok, let's start:

I've a problem, and I don't know how to fix this! (nearly every1 has on 
this list ;))

I have a Debian 2.1 aka "Slink" changed slightely to Potato (all packages 
related
to kernel and networking and a few other are from Potato).
I use kernel 2.2.14 and I have a PROBLEM with ppp - I want to connect to 
my ISP
(actually this isn't my real ISP - do you know what TP S.A. is? :)) and 
everything is
fine except... when the connection is established, I cannot ping to my 
internet gateway,
and everywhere in the internet.
Ah, I forgot to tell you that I don't have permanent connection to 
internet, and I have a small
LAN (two PCs connected each other with ethernet). My LAN address on eth0 
is 192.168.1.1.

Now I want to show you my scripts - created automagically by pppconfig:
(I use CHAT to connect)

file /etc/chatscripts/provider
-
# This chatfile was generated by pppconfig 2.0.1.
# Please do not delete any of the comments.  Pppconfig needs them.
# 
# ispauth chat
# abortstring
ABORT BUSY ABORT 'NO CARRIER' ABORT VOICE ABORT 'NO DIALTONE' ABORT 'NO 
DIAL TONE' ABORT 'NO ANSWER'
# modeminit
'' ATZ
# ispnumber
OK-AT-OK ATDT0202122
# ispconnect
CONNECT ''
# prelogin
# ispname
sername: ppp
# isppassword
ssword: \qppp
# postlogin
'' \d\c
# end of pppconfig stuff
--

and options file /etc/ppp/peers/provider:
---
# This optionfile was generated by pppconfig 2.0.1. 
# 
#
hide-password 
noauth
connect "/usr/sbin/chat -v -f /etc/chatscripts/provider"
debug
/dev/ttyS1
115200
defaultroute
noipdefault 
user ppp
remotename provider
ipparam provider
---

Everything seems to be OK - I've set static nameservers - as I know their 
addreses and
my /etc/resolv.conf file looks like this:
---
domain lodz.tpnet.pl
search lodz.tpnet.pl tpnet.pl
nameserver 194.204.152.34
nameserver 194.204.159.1
-

And now my logs - how it all works (actually DON'T!):

/var/log/ppp.log:
-
Jan 19 23:04:55 guru pppd[2741]: pppd 2.3.10 started by root, uid 0
Jan 19 23:04:56 guru chat[2743]: abort on (BUSY)
Jan 19 23:04:56 guru chat[2743]: abort on (NO CARRIER)
Jan 19 23:04:56 guru chat[2743]: abort on (VOICE)
Jan 19 23:04:56 guru chat[2743]: abort on (NO DIALTONE)
Jan 19 23:04:56 guru chat[2743]: abort on (NO DIAL TONE)
Jan 19 23:04:56 guru chat[2743]: abort on (NO ANSWER)
Jan 19 23:04:56 guru chat[2743]: send (ATZ^M)
Jan 19 23:04:56 guru chat[2743]: expect (OK)
Jan 19 23:04:56 guru chat[2743]: ATZ^M^M
Jan 19 23:04:56 guru chat[2743]: OK
Jan 19 23:04:56 guru chat[2743]:  -- got it 
Jan 19 23:04:56 guru chat[2743]: send (ATDT0202122^M)
Jan 19 23:04:57 guru chat[2743]: expect (CONNECT)
Jan 19 23:04:57 guru chat[2743]: ^M
Jan 19 23:05:20 guru chat[2743]: ATDT0202122^M^M
Jan 19 23:05:20 guru chat[2743]: CONNECT
Jan 19 23:05:20 guru chat[2743]:  -- got it 
Jan 19 23:05:20 guru chat[2743]: send (^M)
Jan 19 23:05:20 guru chat[2743]: expect (sername:)
Jan 19 23:05:20 guru chat[2743]:  115200^M
Jan 19 23:05:23 guru chat[2743]: ^M
Jan 19 23:05:23 guru chat[2743]:   WITAMY  W  INTERNECIE (Lodz)^M
Jan 19 23:05:23 guru chat[2743]: ^M
Jan 19 23:05:23 guru chat[2743]: ^M
Jan 19 23:05:23 guru chat[2743]: http://www.tpnet.pl^M
Jan 19 23:05:23 guru chat[2743]: http://www.tpsa.pl^M
Jan 19 23:05:23 guru chat[2743]: ^M
Jan 19 23:05:23 guru chat[2743]: Name server:  194.204.152.34, 
194.204.159.1^M
Jan 19 23:05:23 guru chat[2743]: ^M
Jan 19 23:05:23 guru chat[2743]:   Prosze wpisac:^M
Jan 19 23:05:23 guru chat[2743]: ^M
Jan 19 23:05:23 guru chat[2743]:   Username:
Jan 19 23:05:23 guru chat[2743]:  -- got it 
Jan 19 23:05:23 guru chat[2743]: send (ppp^M)
Jan 19 23:05:23 guru chat[2743]: expect (ssword:)
Jan 19 23:05:23 guru chat[2743]: ppp  Password:
Jan 19 23:05:23 guru chat[2743]:  -- got it 
Jan 19 23:05:23 guru chat[2743]: send (??)
Jan 19 23:05:23 guru chat[2743]: send (\d)
Jan 19 23:05:24 guru pppd[2741]: Serial connection established.
Jan 19 23:05:24 guru pppd[2741]: Using interface ppp0
Jan 19 23:05:24 guru pppd[2741]: Connect: ppp0 <--> /dev/ttyS1
Jan 19 23:05:25 guru pppd[2741]: sent [LCP ConfReq id=0x1  
   ]
Jan 19 23:05:26 guru pppd[2741]: rcvd [LCP ConfReq id=0x1  
   ]
Jan 19 23:05:26 guru pppd[2741]: sent [LCP ConfAck id=0x1  
   ]
Jan 19 23:05:28 guru pppd[2741]: sent [LCP ConfReq id=0x1  
   ]
Jan 19 23:05:28 guru pppd[2741]: rcvd [LCP ConfNak id=0x1 ]
Jan 19 23:05:28 guru pppd[2741]: sent [LCP ConfReq id=0x2  
  ]
Jan 19 23:05:29 guru pppd[2741]: rcvd [LCP ConfAck id=0x2  
  ]
Jan 19 23:05:29 guru pppd[2741]: sent [LCP EchoReq id=0x0 
magic=0x4c8198cd]
Jan 19 23:05:29 guru pppd[2741]: sent [IPCP ConfReq id=0x1  
]
Jan 19 23:05:29 guru 

Vodoo 2

[erasmo perez]

hello people

i am in a bit of buying a Vodoo 2 video card, just the last minute question?
there seems to be problems with this card, working under X11 ?
this has 12 MB
thanks a lot

erasmo

Re: Strange LILO question

[Ethan Benson]

On 20/1/2000 Onno Ebbinge wrote:


Yes, that is what I ment, maybe I should put LILO on the
hard disk and then set the default boot to the hard disk
with an option to boot from A: ???


perhaps, i have never tried it on a floppy (did with a cd and it 
didnt work) you could add:


other=/dev/fd0
label=floppy

to lilo.conf and run lilo WITH the floppy in the drive (otherwise it 
won't work)



 >you can however install lilo on a floppy and have it boot the hard disk.

Is there then realy no way to boot the floppy also WITH a DOS filesystem?
Probably not :-(


not with lilo, maybe with syslinux or something.


--
Ethan Benson
To obtain my PGP key: http://www.alaska.net/~erbenson/pgp/

Re: 3c575_cb driver

[M. Tavasti]

"Kjohn Sasitorn" <[EMAIL PROTECTED]> writes:

> I have a sony n505ve and purchased the above ethernet card. I've spent more
> than enough hours trying to get debian to work. It works flawlessly in
> redhat. I've tried to get slink to recognize eth0. I did upgrade pcmcia-cs
> to the newest version, and it will recognize eth0, but no network functions
> work. It's quite frustrating. Anyone have any ideas where to begin with

Might be same problem as I had. Remove ipmasq.

-- 
 (Edeltava teksti ei tainnut sisaltaa skandeja, oikeinkirjoitusta tai asiaa)
M. Tavasti  | Puh. | Mahdolliset mielipiteet varastettu,
[EMAIL PROTECTED]| 040-5078254  | lainattu, muitten tai jopa omia!
(Poista tupla-x mail-osoitteesta. Remove double-x from mail address)

Re: Strange LILO question

[Onno Ebbinge]

At 01:43 PM 1/20/00 +, Ethan Benson wrote:
>On 20/1/2000 Onno Ebbinge wrote:
>
>>Is it possible to set LILO on a DOS floppy disk
>>with the default boot for the hard disk and an
>>alternate boot for the DOS floppy itself?
>
>if you mean install the lilo boot sector on the DOS floppy then no, 
>dosfs does not leave any room for a bootblock, so installing lilo on 
>it would destroy the dos filesystem, which would not be problem 
>except you say you want to boot from it.

Yes, that is what I ment, maybe I should put LILO on the
hard disk and then set the default boot to the hard disk
with an option to boot from A: ???

>you can however install lilo on a floppy and have it boot the hard disk.

Is there then realy no way to boot the floppy also WITH a DOS filesystem?
Probably not :-(

Thanks,

Onno

RE: Help gettin' X to work...

[Rob Hensley]

Rob Hensley <[EMAIL PROTECTED]> writes:

}Hi, i recently ran 'apt-get install xf86setup xserver-svga' and installed
}the svga x server as my default. After configuring it and everything, I
}tried to run 'startx', but it didn't exist. So then I went into
}/usr/X11R6/bin/ and ran './startx'. That's where I got the error:
}
}X: X: cannot execute binary file
}
}Then it just sits there. Is there anyway somone could give me
}step-by-step directions on installing a working x windows system with
}gnome and enlightenment? I would greatly appreciate it. Thanks a
lot. Bye.

> Does the file exist? Do you have the xfreecommon package installed?
>-- 
>  {exile} - http://www.freespeech.org/apophysis {v6.1} - ICQ# 47439354
>
>   "It is god's will that I inflame."
>   {exile}


If you're speaking of the xserver-common packages, yes it is installed. Is
there any way you could give me step-by-step directions on setting it
up. Including which packages to get, and what commands to run. Thanks a
lot, Be!

Re: IBM PS/1 Consultant problem

[Bradley Pursley]

Actually, I tried that and it worked just fine. 
My problem was why it couldn't/wouldn't boot from a
floppy.

> If you've got a DOS partition, why not boot from it
> using
> loadlin. All that goes on the DOS partition is a
> copy of
> the kernel, one 32KB file and you may want a .bat
> file to
> save remembering the command. Then shutdown to a DOS
> prompt 
> and up comes linux.
> 
> Cheers,
> 
> -- 
> Email:  [EMAIL PROTECTED]   Tel: +44 1908 653 739 
> Fax: +44 1908 655 151
> Snail:  David Wright, Earth Science Dept., Milton
> Keynes, England, MK7 6AA
> Disclaimer:   These addresses are only for reaching
> me, and do not signify
> official stationery. Views expressed here are either
> my own or plagiarised.
> 
> 
> -- 
> Unsubscribe?  mail -s unsubscribe
> [EMAIL PROTECTED] < /dev/null
> 
> 
__
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

Re: Strange LILO question

[Ethan Benson]

On 20/1/2000 Onno Ebbinge wrote:


Is it possible to set LILO on a DOS floppy disk
with the default boot for the hard disk and an
alternate boot for the DOS floppy itself?


if you mean install the lilo boot sector on the DOS floppy then no, 
dosfs does not leave any room for a bootblock, so installing lilo on 
it would destroy the dos filesystem, which would not be problem 
except you say you want to boot from it.


you can however install lilo on a floppy and have it boot the hard disk.


--
Ethan Benson
To obtain my PGP key: http://www.alaska.net/~erbenson/pgp/

Re: Question: how to use an LDAP backend for logins

[Joseph A. Martin]

>   Well, I'm not answering your question but asking you: How to use
> an LDAP backend for logins?

I can answer this question, but not super quickly. I am planning on
writing a Debian specific HOWTO with example config files,
explanations, etc. Of course the explanations will require that I go
back and understand what it is I did...

>   What I have: one server with shadow.
>   What I want: to put 2 workstations with central authentication
> and exported home directory?

That is exactly the setup I have, except that I only have one
workstation. So, my advice should help a lot. Though after I give you
my config files, etc you should talk to others about real security. My
setup was done for my home network where quick usability is more
important that real security. I suspect your setup may be different.

>   I have tried (not hardly) nis but I couldn't get it to work.
> Mainly because the howto says you need a static hosts files for name
> resolution and I already have a dns running with dynamic dhcp updates.

I read that to, but I don't have a static hosts file and it still
appears to be working. I don't know why you would need static files,
but my advice would be to try it. I was able to get NIS working in a
little over 2 hours. Again, I can't speak for security.

>   Is it possible to use LDAP in this situation?

Very possible. Again, I will have a HOWTO in the works within a week
or two. I will send you info as I work on the sections.

>   Thanks,

You're welcome,
later,
joseph
--
the "LaterDude"
ICQ: 52640402
[EMAIL PROTECTED]

All opinions expressed are my own and not necessarily those of
my employer unless otherwise noted.

Strange LILO question

[Onno Ebbinge]

Is it possible to set LILO on a DOS floppy disk
with the default boot for the hard disk and an
alternate boot for the DOS floppy itself?

Regards,

Onno

One step further (WAS: Re: Advanced hard disk mirroring!)

[Onno Ebbinge]

The high quality replies I received (especially from 
Jens B. Jorgensen) solved my problem. 
(see http://www.debian.org/Lists-Archives/debian-user-0001/msg02027.html)

It also gave me an idea: 

Wouldn't it be great if there was ONE Linux boot-floppy 
that would mount (SMB or NFS?) a complete filesystem after 
the initial boot over a network (and complete the boot 
sequence?).

I'm probable not the first who came up with this idea, so 
does anyone know If such a thing is already pulled off?

Regards,

Onno
 

Re: Is potatoe's exim 3.12-1 OK?

[Mark]

Hi, 

Wouter Hanegraaff wrote:
> 
> I had similar problems when I didn't use an smtp while sending mail to
> certain adresses, only I use postfix. My error was:
> 
> (conversation with columbo.hvu.nl[145.89.50.200] timed out while sending
> end of data -- message may be sent more than once)
> 
> However, this turned out to be a router problem, not a Postfix
[]
> If you have this problem, you can't solve it yourself because it's
> caused by some router fragmenting packets it should not fragment.

Thanks for the info. I think you are correct as my ISP was out of action
for 1 day - they possibly changed something around. I have also noticed
a drop in ftp speed - 3.3k to 2.3k. Also, I bypass thier mta and use the 
main unsw one where my mail account is (access permissions are ok
though).
Obviously I will have to change this until the problem is fixed.

The exim smpt transfer agent does seem to transfer mail to a second
debian
box I have running exim so I guess exim's smtp tranfer is OK.

Thanks,
Mark.

Re: IBM PS/1 Consultant problem

[David Wright]

Quoting Bradley Pursley ([EMAIL PROTECTED]):
> Actually, it wasn't so much that my system was too
> small that was causing the problem, but just that it
> can't boot Linux from the floppy drive.  I could have
> theoretically gotten it to work on the drive I had,
> but my wife was getting tired of my playing with the
> Win95 settings on the DOS partition.  

If you've got a DOS partition, why not boot from it using
loadlin. All that goes on the DOS partition is a copy of
the kernel, one 32KB file and you may want a .bat file to
save remembering the command. Then shutdown to a DOS prompt 
and up comes linux.

Cheers,

-- 
Email:  [EMAIL PROTECTED]   Tel: +44 1908 653 739  Fax: +44 1908 655 151
Snail:  David Wright, Earth Science Dept., Milton Keynes, England, MK7 6AA
Disclaimer:   These addresses are only for reaching me, and do not signify
official stationery. Views expressed here are either my own or plagiarised.

Re: S3 Trio3D

[Robert Varga]

On 19 Jan 2000, David Z. Maze wrote:

> erasmo perez <[EMAIL PROTECTED]> writes:
> EP> more or less one week ago, i see in this lists the report of a S3
> EP> Trio3D video card that refuses to work in SVGA, just accepting VGA
> EP> 
> EP> and now i face the same problem, is there some way to overcome this
> EP> failure ?
> 
> I've had some luck using the XF86_SVGA server from XFree86 3.3.4, and
> specifically using the binary from the tarball from xfree86.org.  I
> have had no luck using any of the Debian-packaged X servers, nor with
> any more recent X server.

I am using an Appolo-named (noname) Trio3D with 4M of RAM and the 365
chipset. It works flawlessly with the 3.3.5-2 drivers in glibc2.1.

Previously I tried it with the 3.3.5-0slink1 packages, then it was only
working with only one resolution. 

The main problem with this card is that it is very picky on the monitor
type you set for it. Don't try to set it better than 1024x768-NI.
1024x768-HR is something it can't cope with, according to my experience.

And don't try to set 32bit color depth. That is the other thing it is
incapable of doing in X.

And of course it needs trying. On some machines you must not enter the
Card menu of the XF86Setup program. (that was with the slink package).

Good luck with it.

Robert Varga

> 
> The only thing is, this server seems to have about a 50% chance of
> failure on restart, where "failure" means "takes down entire system
> with it".  So now I'm using the XF86_FBDev server, which is actually
> giving more or less equivalent performance at the cost of a much worse
> refresh rate (43 Hz interlaced vs. 60Hz noninterlaced).  You might try
> reading the kernel documentation for 2.2.* kernels or look at the
> Framebuffer-HOWTO.

You can avoid system breakdowns if you recognize it early enough that it
won't work, and press alt-ctrl-backspace. It occurs upon resolution
changes, when it can't properly set the video mode. After a couple of
tries you can guess from the state of the leds on your monitor, what will
happen.

It only occurs with videomodes that cannot be displayed properly, but
leave the card in some state, that it hangs the system. 

My monitor tends to behave differently for the modes that hang the
machine, and those that don't, so probably the output of the card's sync
lines is different in the two states. 

I haven't tried the 3.3.6 version yet. I expect it will work correctly as
well. 

However I have seen another machine with a Trio3d as well, which I was not
able to make work at all in modes above 640x480. However it had a very old
monitor. Maybe that was the cause.

Robert Varga




> 
> -- 
> David Maze [EMAIL PROTECTED]  http://donut.mit.edu/dmaze/
> "Theoretical politics is interesting.  Politicking should be illegal."
>   -- Abra Mitchell
> 
> 
> -- 
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
> 
> 

Re: ipchains firewall rules

[Ethan Benson]

On 20/1/2000 dyer wrote:



Are the deny's perhaps UDP packets?


no, they are tcp, I have not yet tried to do anything about udp packets.

looking closer at the way portmapper seems to do things, it appears 
to me that any rpc service that is registered is given a random port, 
not necessarly privileged. that would make firewalling off statd 
lockd et al very difficult.


is there any way to mount an nfs share without opening up all these 
holes?  how much of a risk do they pose? (given they run as root, im 
less then thrilled with them being open)


Ethan

Re: don't want x as default startup

[John Gould]

Just remove the sym link in the default run level directory in /etc
i.e. if the default runlevel is 2, then remove the sym link to S89xdm in
the /etc/rc2.d directory. If there's a better way I'm sure others will
tell you. You'll find the default run level from /etc/inittab.

Hope this helps...

Regards JohnG

_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
_/  John Gould  _/
_/  Senior Systems Support Engineer _/
_/  Power Innovations Limited   _/
_/  Manton Lane, Bedford. MK41 7BJ, England _/
_/  PGP - Public key available  _/ 
_/  email: [EMAIL PROTECTED], [EMAIL PROTECTED]  _/
_/  Tel:   +44 1234 223002  Fax: +44 1234 223000_/
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/


On Thu, 20 Jan 2000, mpx wrote:

> dear debian-users,
> 
> yestarday i installed debian for the first time, and finally i am
> through this dselect-jungle;-)
> now i try since a few hours to eliminate this
> x-as-default-startup-thing.
> i would like to logon my machine in the console.
> so i did like i did on my redhat-box, changing in /etc/inittab, but in
> debian this seems not to work.
> in xf86config there's no option to change this.
> 
> anyone else who had to handle this as well?
> 
> mek
> 
> 
> -- 
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
> 
> 

Re: ipchains firewall rules

[dyer]

Ethan Benson wrote:

> hi,
>
> 
>
> what i tried was adding 3 new rules to the very beginning of the input chain
>
> ipchains -I input 1 -p tcp -i ppp0 -d 0.0.0.0/0 53 -j ACCEPT
> ipchains -I input 2 -p tcp -i ppp0 -d 0.0.0.0/0 113 -j ACCEPT
> ipchains -I input 3 -l -p tcp -y -i ppp0 0.0.0.0/0 :1023 -j REJECT
>
> but as i mentioned i get DENY logs for various ports below 1024 from
> IPs which i presume are from my ISP.
>
> any pointers to good books and such on firewalling would be nice too.
>
> 

Are the deny's perhaps UDP packets?

dyer

Re: Epplet question

[John Miskinis]

So How do I make this epplet that gives me the error to not
be started ?


Hello,

I have had problems with epplets coming back, sometimes many
copies of them at once.  Deleting the files under .enlightenment
doesn't get rid of them, so when I want to start "clean", I
simply rename the directory where I keep my epplets.  You can
try ALT-BUTTON3, chose "Remember" and toggle off the restart
on login parameter, but in my experiences this doesn't always
work.

I keep my epplets in /usr/enlightenment/bin_x3 at the moment, and
changing the name of the directory might be a "hack" but it has
never caused me any problems, and I can be guaranteed that no
epplets will start automatically once I rename the directory.

So far, this epplet stuff is my only complaint issue with E.

I also found that exiting E properly (CTRL-ALT-DELETE) then
return to select "Yes, Really log out" helps things.  I got
this advice on the #E IRC channel.

John
__
Get Your Private, Free Email at http://www.hotmail.com

Epplet question

[Joakim Svensson]

i list,

I have an E epplet question. Perhaps its more suitable to
post to the E lists but since I use ljlanes epplet deb I
thought I tried here first.

I happened to start an epplet (some network monitoring epplet)
but since I don't have a network It just returned an error box.
My problem now is that everytime I start X I will get this
error box.

So How do I make this epplet that gives me the error to not
be started ?

Also does anyone know if E-FancyLauncher is to be found as a
deb package ?

I could probably get all the epplets from CVS and compile
myself but since I am not very experienced In doing that
stuff I would prefer a .deb
But if anyone feels like helping me point out where to start
(pointers FAQ etc) using CVS and compiling myself that would
be very nice too.

Best regards
Joakim

Re: don't want x as default startup

[Ethan Benson]

On 20/1/2000 mpx wrote:



yestarday i installed debian for the first time, and finally i am
through this dselect-jungle;-)
now i try since a few hours to eliminate this
x-as-default-startup-thing.
i would like to logon my machine in the console.
so i did like i did on my redhat-box, changing in /etc/inittab, but in
debian this seems not to work.
in xf86config there's no option to change this.

anyone else who had to handle this as well?


rm /etc/rc2.d/S99xdm

or update-rc.d -f xdm remove

but then the next time xdm is upgraded the links will be put back so 
in that case i would then do


update-rc.d xdm start 5 . stop 0 1 2 3 4 6 .

then you have redhat like behaviour, boot to runlevel 5 -> get X 
otherwise console.


you could always remove xdm too, unless a bunch of other things 
depend on it for some reason..



--
Ethan Benson

OpenPGP encrypted mail accepted.
To obtain my PGP key: http://www.alaska.net/~erbenson/pgp/
Key FingerPrint: 371A 7416 5D39 CF2D 9366  8AF6 0139 54F5 3EBD 0FE6
RSA Key FingerPrint: DE8B 74D0 79F1 6176  9AF5 120F 47AD 9B0A

Boot

[Nicola Tessari]

Hello there,
I'm here because my official debian cd (bought at linuxcentral) doesn't
boot...
After few seconds the booting process in my machine stops at the line:
md driver 0.36.3 MAX_MD_DEV=4, MAXREL=8
Anyone could tell me which is the problem?
I've choosen the hardware checking out if it was supported (k7 500MHz
with asus k7m, cd creative 48x).
Slackware 7.0 boots well (from linux central too)
Thanks a lot in advance,
Nicola

Lilo

[Mats Johansson]

Hi!

I have newly installed Potato on my new IBM hardrive and have got any very 
strange problem with lilo.



And the problem is that the LILO prompt comes up on the on screen and the boot 
prompt also comes up becouse i use the prompt option but i can't typ anything 
it is very strange and when scroll look is down it come up alot  of 
*interrupted* messages as it should but i cant'stil type anything , and when 
the timout has readched my linux boots as usual.I have installed Linux with 
lilo a lto of times before and never got this problem.
Anyone  got a solution?



my lilo.conf look like this:

boot=/dev/hda
root=/dev/hda1
install=/boot/boot.b
map=/boot/map

delay=20
timeout=20

compact
prompt

vga=normal

image=/vmlinuz
label=Linux
read-only
 

//Mats

Question: how to use an LDAP backend for logins

[Mario Olimpio de Menezes]

Hi Joseph,

Well, I'm not answering your question but asking you: How to use
an LDAP backend for logins?
What I have: one server with shadow.
What I want: to put 2 workstations with central authentication
and exported home directory?
I have tried (not hardly) nis but I couldn't get it to work.
Mainly because the howto says you need a static hosts files for name
resolution and I already have a dns running with dynamic dhcp updates.
Is it possible to use LDAP in this situation?
Thanks,

[]s,
Mario O.de Menezes"Many are the plans in a man's heart, but
IPEN-CNEN/SP is the Lord's purpose that prevails"
http://curiango.ipen.br/~mario Prov. 19.21
   http://www.revistalinux.com.br

Re: how many users per apache proc?

[Mario Olimpio de Menezes]

On Wed, 19 Jan 2000, aphro wrote:

> i was wondeirng if anyone knew approx how many connection 1 apache process
> could handle? just 1? or is it more..
 
my $0.02:

According to Kabir, Mohammed J., Apache Server Administrator's
Handbook, IDG Books, pp.68-69, "Apache uses one child server for each
request..."  The MaxClients directive limits the number of simultaneous
requests that Apache can service. 
The MaxRequestsPerChild: "Apache launches a child server process
to service a request, however, a child server can process multiple
requests. The number of requests a child server can process is limited
by the MaxRequestsPerChild directive. After servicing the maximum number
of requests, the child process terminates.
If the MaxRequestsPerChild is 0 (zero) then the process will
never expire. If you suspect there are libraries on your operating
system that have memory-leaking code, you may want to set this directive
to a non-zero value. This enables you to define a life cycle for a child
process, and therefore reduces the chances of a process consuming leaked
memory and slowly eating up all available memory. It also provides you
with a small load average number for your system, since the
Apache-related load is reduced as your Web server becomes less busy".

The are other directives related with performance and resources,
but I think the above are more closely related with your question. 

[]s,
Mario O.de Menezes"Many are the plans in a man's heart, but
IPEN-CNEN/SP is the Lord's purpose that prevails"
http://curiango.ipen.br/~mario Prov. 19.21
   http://www.revistalinux.com.br

don't want x as default startup

[mpx]

dear debian-users,

yestarday i installed debian for the first time, and finally i am
through this dselect-jungle;-)
now i try since a few hours to eliminate this
x-as-default-startup-thing.
i would like to logon my machine in the console.
so i did like i did on my redhat-box, changing in /etc/inittab, but in
debian this seems not to work.
in xf86config there's no option to change this.

anyone else who had to handle this as well?

mek

Re: ftp errors

[Ethan Benson]

On 20/1/2000 John Leget wrote:


ive been trying to use gftp and im getting the following error
"Warning: Cannot parse listing /bin/ls: can't open cache
'/etc/ld.so.cache'"
and of course remote listings fail ??.
but what has /etc/ld.so.cache got to do with this

what does this mean ??.


this looks like a problem on the server, anonymous ftp is usually 
dropped into a chroot jail with a copy of /bin/ls which of course 
needs a lib or two which are also copied, sometimes a ld.so.cache is 
created there as well, in this case it seems to be wrong or corrupted.


im not sure why the ld.so.cache would be needed really, i don't have 
one on my ftp server and ls works fine.


Ethan