Re: Can't reboot or halt from gdm

[Oki DZ]

Todd Myhre wrote:

I've set up gdm on a new laptop but the System menu does not
seem to work correctly.  When I try to halt the system, it
simply reloads gdm and the following error messages appear
in my syslog:

  gdm_slave_xioerror_handler: Fatal X error - Restarting :0
  gdm_child_action: Reboot or Halt request when there is no
 system menu from display :0


Which gdm version are you using?
It happened to me once, when I upgraded gdm; I don't remember correctly, 
 whether it was potato -> potato/woody or woody -> woody/sid. I found 
out that the newer gdm set the "gdm" username to not having the 
privilege to shutdown the system. So, I ended up utilizing sudo so that 
gdm can do the shutdown. In /etc/gdm/gdm.conf:

RebootCommand=/usr/bin/sudo /sbin/reboot
HaltCommand=/usr/bin/sudo /sbin/poweroff

And, of course, you have to add the gdm username to the sudo entries 
file using visudo.


Oki

Re: apt-get source vs. binary

[Paul Miller]

Modify the debian/changelog in your debian source tree so that the version
is 2:1.0.6-3, etc... the number in front of the colon will not show up in
dselect, etc.  This change will keep your version "current."
Unfortunately, dselect doesn't provide an indication when a newer debian
version is available.  You can still get the latest source version with apt-
get source.
-Paul

> When I install via apt-get source and compile the package, then
> install.  apt-get -u upgrade wants to replace my compiled version with
> the binary package from debian servers, of the same version.  Ever time
> I install a package from apt-get source, apt-get wants to replace it
> with the binary version.  I compiled nautilus (1.0.6-3), then
> $ apt-get -s install nautilus
> Reading Package Lists... Done
> Building Dependency Tree... Done
> 1 packages upgraded, 0 newly installed, 0 to remove and 0  not
> upgraded. Inst nautilus (1.0.6-3 Debian:unstable)
> Conf nautilus (1.0.6-3 Debian:unstable)
>
> It should says that nautilus is the latest version, but it thinks that
> my optimize nautilus should be replaced with the i386 package.  I have
> been optimizing the compiled packages for my athlon using
> pentium-builder.
>
> any thoughts?
> james
>
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact
> [EMAIL PROTECTED]

unix programing question

[a]

i use popen to run a command and read output from the command,
but the command outputs error message to something else than std out.

how can i read the error message?

i'll leave the list soon, please reply to [EMAIL PROTECTED]

Re: Change from SuSE to Debian

[Osamu Aoki]

On Mon, Mar 11, 2002 at 07:36:05PM +, pete atkinson wrote:
> I would appreciate some tips...
> 
> I have been running SuSE in various releases for the last 3 years and like(d) 
> to think I had a good grasp of what was going on.
> 
> In an effort to expand my knowledge, I have commandeered a spare box that I 
> have and have chosen 'Potato' as my first Debian distribution. (It is a 400 
> Celeron, 8Gb Disk, 512Mb RAM, Nvidia TNT2 graphics, Realtek NIC (no sound 
> card) and an old IBM SVGA monitor - which will be replaced when funds allow)
> 
> Boy ! what a difference - I am plodding through various HOWTO's and have 
> configured my display and network card successfully (big deal some might say 
> !) but the thing that is getting me is the whole culture change between the 
> relatively spoon fed SuSE and the somewhat (to me) esoteric Debian.
> 
> Are there any hints/tips/watch-out-fors that you could offer, principally, I 
> am a bit confused over the non-RPMness of packages and the lack of the config 
> suites such as YAST/YAST2 that SusE employs.
> 
> I hope you can help and apologise if this seems a woolly request
> 
I started Linux with Slackware but immidiately moved to Redhat when I
founf out V5 came to my notice.  Anyway, Debian is tougher than RPM
Linux system.

I suggest you to read some basic debian documentations including
developer ones, release notes and install manuals.

  http://www.debian.org/doc
  http://www.debian.org/doc/ddp

Oh, I hope mu web page below may help too.

Cheers :)

-- 
~\^o^/~~~ ~\^.^/~~~ ~\^*^/~~~ ~\^_^/~~~ ~\^+^/~~~ ~\^:^/~~~ ~\^v^/~~~ +
Osamu Aoki <[EMAIL PROTECTED]>, GnuPG-key: 1024D/D5DE453D
Visit Debian reference http://www.debian.org/doc/manuals/quick-reference/
There are 6 files: index.{en|fr|it}.html quick-reference.{en|fr|it}.txt
I welcome your constructive criticisms and corrections.

when compile Kernel2.4.18 need Ncursor Library,how to add this library?

[debian2002]

HI:
same as title~
anyone can help me,,thanks 

From:
Ming Wu

Re: Still looking to replace Eudora

[O Polite]

On Fri, 2002-03-08 at 17:30, Brian Stults wrote:

> Did you install all the dependencies (e.g. aspell, pspell) and a
> dictionary?

Thanks. That did the trick.

op

-- 
http://plusseven.com/gpg/

Re: unix programing question

[Sean 'Shaleh' Perry]

On 12-Mar-2002 a wrote:
> i use popen to run a command and read output from the command,
> but the command outputs error message to something else than std out.
> 
> how can i read the error message?
> 
> i'll leave the list soon, please reply to [EMAIL PROTECTED]
> 

Advanced Programming in the UNIX Environment, W. Richard Stevens.  If you write
UNIX software, you should own it.

Re: Samba and CUPS

[O Polite]

On Sun, 2002-03-10 at 23:21, Bill Wohler wrote:
>   This has been a painful day. I've got printing working nicely now.
>   However, Samba is another story.
> 
>   My Win2k clients could see the new printer during the create printer
>   phase. However, I'm getting the message "Access denied, unable to
>   connect." Some messages from a Google search said that one can still
>   print, but that's not the case here.
> 
> 
Funny. I had this same problem just a few days ago. I googled around
until I came up with this smb.conf

I think that 
hosts allow = 192.168.0. 
hosts deny = ALL

and 
guest ok = Yes  

are were the lines that did it. 

Notice that the printer hpviatilda is set up in CUPS with that same
name.

good luck.

op

--
[global]
workgroup = WORKGROUP
server string = %h server (Samba %v)
security = SHARE
obey pam restrictions = Yes
passwd program = /usr/bin/passwd %u
passwd chat = *Enter\snew\sUNIX\spassword:* %n\n
*Retype\snew\sUNIX\spassword:* %n\n .
syslog = 0
max log size = 1000
socket options = IPTOS_LOWDELAY TCP_NODELAY SO_SNDBUF=4096
SO_RCVBUF=4096
dns proxy = No
invalid users = root
hosts allow = 192.168.0. 
hosts deny = ALL

[homes]
comment = Home Directories
create mask = 0700
directory mask = 0700
browseable = Yes

[printers]
comment = All Printers
path = /tmp
create mask = 0700
printable = Yes
browseable = Yes

[hpviatilda]
path = /tmp
guest ok = Yes  
read only = No
create mask = 0700
printable = Yes
postscript = Yes
printing = cups
print command = lpr -P%p %s
printer name = hpviatilda
oplocks = No
share modes = No
--

-- 
http://plusseven.com/gpg/

Re: when compile Kernel2.4.18 need Ncursor Library,how to add this library?

[Scott Henson]

On Tue, 2002-03-12 at 01:48, debian2002 wrote:
> HI:
> same as title~
> anyone can help me,,thanks 
> 

apt-get install libncurses5-dev
-- 
-Scott Henson

[EMAIL PROTECTED]

 "Eternal vigilance is the price of liberty: power is ever stealing from
the many to the few.  The manna of popular liberty must be gathered each
day, or it is rotten... The hand entrusted with power becomes, either
from human depravity or esprit de corps, the necessary enemy of the
people.  Only by continual oversight can the democrat in office be
prevented from hardening into a despot: only by unintermitted agitation
can a people be kept sufficiently awake to principle not to let liberty
be smothered in material prosperity... Never look, for an age when the
people can be quiet and safe.  At such times despotism, like a shrouding
mist, steals over the mirror of Freedom"
- Wendell Phillips

Re: unix programing question

[Crispin Wellington]

On Tue, 2002-03-12 at 14:43, a wrote:
> i use popen to run a command and read output from the command,
> but the command outputs error message to something else than std out.
> 
> how can i read the error message?

popen is not really your command. use pipe() to create pipe pairs. One
for stdout, stdin and stderr. fork() and in the child close the reading
end of stdout and in the parent close the writing end (and visa versa
for stdin). In the child close() stdin, stdout and stderr. Then use
dup2() to tie these (stdin/out/err) to your pipes (the pipe() ends).
then close your duped pipes (you've now attached them to standard
pipes). Then exec() your task.

In the parent write into the stdin pipe and read from the others. You'll
need to select() them.

You may also be able to popen("process 2>&1") or something like that.
Unsure if the shell is used with popen().

Sorry about the brevity, but this isn't the list for this.

Crispin

opening a directory for samba, netatalk, www all at once.

[O Polite]

I'm running this distro http://www.e-smith.org/ on a box for my
colleagues. The only feature that I use is the "Ibay", this lets me
create a directory that is accessible through samba (windows), netatalk
(apple), www and ftp. It will also let me treat user home directories as
Ibays. 

Any package that will do this for debian, check that all the needed
kernel modules are in place etc?

op

-- 
http://plusseven.com/gpg/

Re: lite applications

[Karsten M. Self]

on Sun, Mar 10, 2002, Joe ([EMAIL PROTECTED]) wrote:
> On Sat, Mar 09, 2002 at 11:52:43PM -0800, Karsten M. Self wrote:
> > on Fri, Mar 08, 2002, Nicholas Imfeld ([EMAIL PROTECTED]) wrote:
> > > I have set up an old 486 laptop with debian.  I am looking for an x
> > > email client (preferably graphical as opposed to test based) and a web
> > > browser that don't require a lot of space, memory, etc.  
> > 
> > I'd ***STRONGLY*** encourage you to stick with the text alternatives,
> > unless you're using the 486 _just_ as an X11 terminal, you might be able
> > to run more advanced apps on it.  Otherwise, I'd stick with mutt, w3m,
> > lynx, and similar text-mode apps.
> > 
> > If X is an absolute requirement, look to BrowseX as a browser.  It's
> > Tk/Tcl based and pretty light.  Nothing else I'm aware of comes close.
> > Dillo's small of itself, but the gtk requirements are large.
> 
> If you install KDE 1.x 

This is simply not an option on a 486.

Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What part of "Gestalt" don't you understand?   There is no K5 cabal
  http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org



pgptE5fxyLfFp.pgp
Description: PGP signature

Re: portsentry: port 162 attack

[Jeff]

Thomas Shemanske, 2002-Mar-11 16:46 -0500:
> I have a sid system and installed portsentry on it (and several other 
> woody machines in the department).
> 
> I left it in log-only mode, but immediately after starting it up, I 
> discovered that a machine of a colleague of mine is
> banging away (every three minutes exactly) on port 162 (snmp-trap) on 
> all the machines in our subnet, but on no machines outside the subnet.
> 
> I also have a debian potato box in a different subnet doing the same 
> thing to all the machines in its subnet.
> 
> While my colleague has perhaps been a little slack in maintaining 
> upgrades, the potato machine is always upgraded as soon as a security 
> upgrade appears.  So while it is possible both machines have been 
> hacked, I am leaning against that opinion for the moment, largely 
> because the activity seems restricted to the subnet.
> 
> I have no idea how to track this down.  I can see no process he is 
> running which would indicate the activity, and certainly there is 
> nothing in the logs.  The password and shadow files have the same 
> timestamp and haven't been changed (apparently) since October.
> 
> On my machine netstat -a | grep snmp returns
> udp0  0 *:snmp  *:*
> udp0  0 *:snmp-trap *:*
> 
> On his machine, it returns nothing.
> 
> Any ideas on whether this is a real concern and/or how to track it down 
> would greatly be appreciated.
> 
> Thanks
> 
> Tom

It appears to me that your colleagues machine has an snmp agent
running and has your machine configured as a trap reciever.  It's
also apparent that you have an snmp trap server running, since
the ports are listening.  Have you looked at the snmp trap log to
see what the trap is?

Your colleagues machine won't have listening ports since it's the
agent/client.  Check the processes running and look for the
agent.  If the agent must run, check the config to tune it.

hope this helps,
jc

-- 
Jeff CoppockSystems Engineer
Diggin' Debian  Admin and User

Re: Still looking to replace Eudora

[Karsten M. Self]

on Mon, Mar 11, 2002, csj ([EMAIL PROTECTED]) wrote:
> On Sat, 9 Mar 2002 23:28:31 -0800
> "Karsten M. Self"  wrote:
> 
> >  In general, I find the MDI paradigm to be frustrating to the extreme,
> >  and the alternatives provided by GNU/Linux are far superior (IMVAO).
> 
> Then why do you continuously plug galeon? ;-)

Winks aside, there's a distinction.

MDI is often applied to applications in which there is significant
authoring activity.  Examples most users would be familiar with would
include MS Word, MS Excel, etc.  One particular instance I'm familiar
with is SAS, which includes an "IDE" type getup -- there's a program
editor, a log window (programming, error, debugging output), and a
"list" window (reports output).  Under Unix, the windows are free-form
and may be placed at any place on the desktop.  Under MS Legacy MS
Windows, they're all stuck in this stupid _box_ you can't get rid of.
At the same time, there's a lot of work involving reading or copying out
of one window and writing into another.

You're also likely dealing primarially with a small number of documents
or equivalents.  The framing is (IMVAO) not necessary for organization.

Galeon is (mostly) a _browsing_ application.  While some websites allow
for input (and I'm using web-based tools more frequently these days),
the principle actions are:

  - Search for a website.
  - Open a website.
  - Open links within a page.
  - Pursue some other vein of thought.

What galeon offers is choice:  you can stack a bunch of pages together
in a single window (not contained loosely within it in the MDI style,
but tabbed within a frame).  Or you can arrange them as fully separate
windows.  And you can switch a given page between the two choices,
detaching it from a tabbed view, or adding it to another window's
tab(s).  While it's possible to focus on a single document when browsing
the Web, I find I'm far more likely to be scanning a larger number --
five pages would be a small browsing session, 20-40 typical, and 100 or
more pages (5-8 windows of 12-20 tabs) is not unheard of.

FWIW, I also like and use 'screen' extensively, a terminal / console
multiplexer.  It (generally) follows a similar mode of restricting the
user to viewing one session at a time, but still can be extremely
useful.  I find it indespensible.  One application I've got is based on
a set of shell wrappers to launch xterm with screen, with w3m as a
shell, wrapped to default to opening to a bookmarks page (it's a lot
easier to use than it is to describe).  The result is, approximately, a
tabbed, text-mode browser.  Damned useful stuff.

Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What part of "Gestalt" don't you understand?   There is no K5 cabal
  http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org



pgpNNS4V1fvqh.pgp
Description: PGP signature

Re: the quest for a *stable* browser

[Karsten M. Self]

on Fri, Mar 08, 2002, timothy bauscher ([EMAIL PROTECTED]) wrote:
> I downloaded Mozilla via apt-get. It crashes often. I would like a
> stable open source alternative to Mozilla.

What version of Debian?  I'd recommend Testing (Woody) for a desktop,
and Galeon as the browser.  It simply kicks royal ass, then starts to
*really* rock.

Browser uptimes of days to weeks.

Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What part of "Gestalt" don't you understand?   There is no K5 cabal
  http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org



pgpRO9s8GsFsa.pgp
Description: PGP signature

Re: unix programing question

[Colin Watson]

On Tue, Mar 12, 2002 at 02:43:14PM +0800, a wrote:
> i use popen to run a command and read output from the command,
> but the command outputs error message to something else than std out.
> 
> how can i read the error message?

Another homework question? Please don't. There's plenty of documentation
installed on any reasonable Unix system which should help you figure it
out.

-- 
Colin Watson  [EMAIL PROTECTED]

Re: woody upgrade bugs

[Colin Watson]

On Tue, Mar 12, 2002 at 12:25:13PM +1100, John Griffiths wrote:
> i've got couple of things that while fixable, were certainly suboptimal...
> 
> should I submit them as bugs?

Please do.

> not being familiar with bug submission, is anyone interested in this stuff?

Check http://bugs.debian.org/ first to avoid duplicates,
then use 'reportbug' to submit.

Thanks,

-- 
Colin Watson  [EMAIL PROTECTED]

Nvidia Debian Kernel

[Kai Hendry]

On the user list you mentioned building a kernel the debian way with
nvidia stuff.

I apt-get source kernel-source-2.4.18

however building the kernel I got the error:
dpkg-gencontrol: error: package kernel-image-2.4.18 not in control info
make[1]: *** [real_stamp_image] Error 29
make[1]: Leaving directory `/usr/src/kernel-source-2.4.18-2.4.18'
bilbo:/usr/src/kernel-source-2.4.18-2.4.18# make-kpkg kernel_image

Have I forgotten to do something?

Also I tried building a modules image for the nvidia driver:

bilbo:/usr/src/kernel-source-2.4.18-2.4.18# make-kpkg modules_image
find: /usr/src/modules: No such file or directory
test -f stamp-debian || /usr/bin/make -f /usr/share/kernel-package/rules
debian
test -f .config  || /usr/bin/make -f /usr/share/kernel-package/rules
.config
test ! -f stamp-configure \
  && /usr/bin/make\
 ARCH=i386 oldconfig \
  && /usr/bin/make\
 ARCH=i386  dep \
  && /usr/bin/make\
 ARCH=i386 clean && \
  touch stamp-configure
make: [stamp-configure] Error 1 (ignored)
for module in  ; do\
  if test -d  $module; then\
(cd $module;   \
  if ./debian/rules KVERS="2.4.18"
KSRC="/usr/src/kernel-source-2.4.18-2.4.18" \
 KMAINT="Unknown Kernel Package Maintainer"
KEMAIL="[EMAIL PROTECTED]"
  \
 KDREV="2.4.18-2" kdist_image; then\
  echo "Module $module processed fine";\
  else  \
   echo "Module $module failed.";  \
   echo "Hit return to Continue";   \
   read ans;\
  fi;   \
 ); \
  fi;   \
done
bilbo:/usr/src/kernel-source-2.4.18-2.4.18# cd ..
bilbo:/usr/src# ls -al
total 30704
drwxrwsr-x5 root src  4096 Mar 12 00:27 .
drwxr-xr-x   13 root root 4096 Feb  9 13:12 ..
-rw-r--r--1 root src 22403 Mar 11 20:09 .config
drwxrwxr-x2 502  502  4096 Nov 27 22:39
NVIDIA_kernel-1.0-2313
drwx--   15 hendry   hendry   4096 Mar 12 01:10
kernel-source-2.4.18-2.4.18
-rw-r--r--1 root src 20509 Mar  7 12:32
kernel-source-2.4.18_2.4.18-2.diff.gz
-rw-r--r--1 root src   764 Mar  7 12:32
kernel-source-2.4.18_2.4.18-2.dsc
-rw-r--r--1 root src  29818323 Feb 28 15:02
kernel-source-2.4.18_2.4.18.orig.tar.gz
drwxr-xr-x4 root src  4096 Mar 11 20:07
nvidia-glx-1.0.2802
-rw-r--r--1 root src 29082 Mar 11 20:07
nvidia-glx-dev_1.0.2802-1_i386.deb
-rw-r--r--1 root src   1474220 Mar 11 20:07
nvidia-glx_1.0.2802-1_i386.deb
bilbo:/usr/src#

Just a comment. Isn't it silly to apt-source then kernel and get also a
diff and orig.tar.gz you don't need?!

Even though I was JUST trying to build a new kernel last night, when I
booted up this morning I now get error with my old trusted 2.4.17 error
while starting X:
(II) NVIDIA(0): AGP 4X successfully initialized
(EE) NVIDIA(0): Failed to allocate a DMA channel
(EE) NVIDIA(0):  *** Aborting ***
(EE) NVIDIA(0): Failed to allocate DMA push buffer
(EE) NVIDIA(0):  *** Aborting ***

Fatal server error:
AddScreen/ScreenInit failed for driver 0

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

XIO:  fatal IO error 104 (Connection reset by peer) on X server ":0.0"
  after 0 requests (0 known processed) with 0 events remaining.


As you can imagine I am quite frustrated, and I am stuck in console.
Maybe it was the glx upgrade that did it:
bilbo:/usr/src# dpkg -l nvidia*
No packages found matching nvidia-glx-1.0.2802.
No packages found matching nvidia-glx-dev_1.0.2802-1_i386.deb.
No packages found matching nvidia-glx_1.0.2802-1_i386.deb.

Do I need to compile my own kernel with this nvidia source module? Can't I
use a debian 2.4.18 kernel and in the module somehow?

Maybe someone can point me to a Debian specific guide to building a kernel
with nvidia shit from start to finish?

Regards
-Kai Hendry

best way to install non debian package

[François]

Hello, 


What's the best way to install package which are not(yet) in debian
package format or in the current distribution ?

For example, I need pilot-link > 0.10.0 to use with palm m125
but have no deb package.

In which directory may I install it ?
What's the Debian philosophy ?


Thanks  

François 

RE: Nvidia Debian Kernel

[Wayne Brown]

Hi Kai,
I have just compiled this kernel with the Nvidia modules, I have a geforceII
mx and got there after a bit of fiddling.

> Cc: debian-user@lists.debian.org
> Subject: Nvidia Debian Kernel
> 
> 
> 
> On the user list you mentioned building a kernel the debian way with
> nvidia stuff.
> 
> I apt-get source kernel-source-2.4.18
> 
> however building the kernel I got the error:
> dpkg-gencontrol: error: package kernel-image-2.4.18 not in 
> control info

Dont know what this error is, but did you make-kpkg clean before starting.
You created a .config file with make xconfig or similiar yes?
I am not at my Linux box at the moment but I remember I needed another
package installed, something like kpkg-dev?
Instructions in /usr/share/doc/kernel...

> make[1]: *** [real_stamp_image] Error 29
> make[1]: Leaving directory `/usr/src/kernel-source-2.4.18-2.4.18'
> bilbo:/usr/src/kernel-source-2.4.18-2.4.18# make-kpkg kernel_image
> 
> Have I forgotten to do something?
> 
> Also I tried building a modules image for the nvidia driver:
> 
> bilbo:/usr/src/kernel-source-2.4.18-2.4.18# make-kpkg modules_image
> find: /usr/src/modules: No such file or directory

>From the 'ls' below, I see you do not have this directory. 
You should install the debian src packages as a pair, glx and nvidia kernel.
(I wouldnt try and use the source from the NVIDIA site as I dont know if
this will work and will not create the necessary Dir.)
On doing so the modules dir. will be created. When you run make-kpkg
modules_image, this is where it looks.

HTH
Wayne.
(Newbie)
 

Re: best way to install non debian package

[Karl E. Jorgensen]

On Tue, Mar 12, 2002 at 11:01:01AM +0100, François Chenais wrote:
> Hello, 
> 
> 
>   What's the best way to install package which are not(yet) in debian
>   package format or in the current distribution ?
> 
>   For example, I need pilot-link > 0.10.0 to use with palm m125
>   but have no deb package.
> 
>   In which directory may I install it ?
>   What's the Debian philosophy ?

The debian package manage system will keep its hands off /usr/local, so
that is a good choice.

Have you checked whether your desired version of pilot-link is available
in sid? It's probably quicker to get it from there.


HTH
-- 
Karl E. Jørgensen
[EMAIL PROTECTED]
www.karl.jorgensen.com
 Today's fortune:
Real Men don't make backups.  They upload it via ftp and let the world mirror 
it.
-- Linus Torvalds


pgpueboQQxRDl.pgp
Description: PGP signature

Acroread not displaying menu text 2nd try

[james]

Pat

Thanks for the e-mail. Unfortunately yours was the only responses I
received. In fact I am using this reply as a rather lame excuse to
post this question a second time.

Needless to say I would be grateful for any pointer to a solution for
the acroread/Acrobat problem described below.

On 11 March 2002 12:57:04 Patrick Colbeck <[EMAIL PROTECTED]> wrote
>Hi
>
>Did you get a reply to this question ? I am having exactly the same issue and 
>guess waht I am in th UK as well. If you got a solution I would love to know 
>what it is.
>
>Thanks 
>
>Pat
>
>On Sunday 10 March 2002 11:42 pm, [EMAIL PROTECTED] wrote:
>> Hello
>>
>> I am experiencing a problem with acroread (Acobat reader)v.4.05-5 on
>> what I believe is a typical Debian GNU/Linux 3 (Woody/Testing) system
>> running on a Compaq Professional Workstation 5000 (P-Pro SMP).
>>
>> Acroread does not display any menu text at all (just gray line boxes
>> where the text should be) and when it is started from an xterm the
>> following error message is displayed:
>>
>> Warning: charset of fontList (ISO10646-1) does not match locale
>> (ISO8859-1).
>>
>> I suspect this has come about because recently I had to change the
>> user and root locales to ISO8859-1 (using dpkg-reconfigure locales)to
>> enable recent releases of multi-gnome-terminal (currently v1.3.13) to
>> run on my system.
>>
>> Reluctantly (because ISO8859-1 seems to be ideal for a UK based Debian
>> box) I tried to set both the user and root locales to ISO10646-1. But
>> this option is not available using dpkg-reconfigure locales. My system
>> does have the following relevant looking file:
>>
>> /usr/share/i18n/charmaps/ISO_10646.gz
>>
>> But so far I have been unable to discover if it can be used to change
>> the locales.
>>
>> I would be grateful for any suggestions on how to sort this problem

>> out, since I need to use acroread to read work related docs.
>>
>> James



--
Personalised email by http://another.com

Re: kernel compile problem -- unresolved symbols

[Alan James]

On Mon, Mar 11, 2002 at 04:53:52PM -0600, Christopher M. Jones wrote:
> I'm compiling for a K7/Athlon/Duron system. As far as I can tell, -all- 
> modules come up with unresolved symbols. In other words, any module I try to 
> load comes up with unresolved symbols, whether part of the kernel, or 
> standalone. Clearly, I am doing something very wrong, though I haven't 
> changed the methods for compiling, that have worked successfully for me since 
> about 2.4.13. I have made sure (to the best of my knowledge) that I have 
> sufficiently new versions of the important packages.

This may be a daft question but when you load modules are you using inmod or 
modprobe ? Try running "depmod -ae" too to see if that makes any difference.

Post the error messages here too, that might help.

Re: kernel compile problem -- unresolved symbols

[Alan James]

On Tue, Mar 12, 2002 at 11:24:22AM +, Alan James wrote:

> This may be a daft question but when you load modules are you using inmod or 

Sorry I meant "insmod" there of course. Still modprobe is the one to use.

RE: The future of Debian install??

[Craig Sampson]

On Mon, 11 Mar 2002 11:37:08 -0800, Bedford, Donald T. wrote:

>a few years back. Yes, my first install on a x86 box as 
>anything but easy as
>I build my own box. But, I now know more about my system than 
>I ever did w/
>RH. This is why I chose the Debian path instead of re-
>installing RH on the
>new system.
>
>Auto-detect would be nice but I sure learned a lot when it 
>didn't...


I may have missed something (sure hope so), but what I'd find 
immensely useful is a way of being able to choose, at install 
(or other) time what packages I want then save this selection 
'list' to a file so that when I next install Debian on another 
box I can just tell it to use the previously made selections.

SuSE has this capability and its fantastic for installing a 
sane and similar OS onto vastly differing server hardware.

I imagine I could make a new 'task' to achieve this, but I've 
yet to see any way, at install time, of importing something 
that's not already on the install CD.  

Something like 'would you like to import package selections 
from floppy (or other) disk' would be just great.

Any pointers anyone?

As for the install - well, I -hate- dselect with a passion, it 
gets the job done but gawd!  Does it have to be so time 
consuming/difficult?  

I'm deeply distrustful of GUI installers and never use em.  
Quite liked the SuSE YAST1 (text only) except for the use of 
RPM which gives you -big- inherent problems later in your 
servers life.

Cheers,
Craig

Moving /var to another drive

[Andrew Stephen]

Hi

My /var and /usr partitions have just run out of space and I was wondering
what is the best way to copy them to a new drive that has just been
installed.

Any suggestions would be greatly appreciated.

Regards,
Andrew

ssh-client & gui

[Ines Rieger]

Hi all,
is there any graphical ssh-client for Unix?

THX
Ines

--
Es ist nichts sicher, ausser dass nichts sicher ist, und nicht einmal das.

Re: Squid - Had to "recreate" the cache...

[Pedro Zorzenon Neto]

On Sat, Feb 09, 2002 at 06:23:09PM +0100, Johannes Franken wrote:
> On Sat, Feb 09, 2002 at 02:36:23PM -0200, Pedro Zorzenon Neto wrote:
> >Did squid cache got corrupted? why? what happened?
> 
> check /var/log/squid/squid.out*

I found the following error in squid's logs:

FATAL: You've run out of swap file numbers.

The solution is at the FAQ:

http://www.squid-cache.org/Doc/FAQ/FAQ-11.html#ss11.26

Re: Change from SuSE to Debian

[Craig Sampson]

Hi Pete,

On Mon, 11 Mar 2002 19:36:05 +, pete atkinson wrote:

>I have been running SuSE in various releases for the last 3 
>years and like(d) 
>to think I had a good grasp of what was going on.

I'm a fellow SuSEer.  Finally got tired of RPM hell so I'm here 
trying on a Debian coat.  So far its been rough, plenty of 
difference between Debian and most of the other RPM based 
distros.


>Boy ! what a difference - I am plodding through various 
>HOWTO's and have 
>configured my display and network card successfully (big deal 
>some might say 
>!) but the thing that is getting me is the whole culture 
>change between the 
>relatively spoon fed SuSE and the somewhat (to me) esoteric 
>Debian.

Yep.  Its a blast isn't it?  I never found SuSE to be too big 
on the spoon feeding myself, but then I never -ever- used YAST2 
(the GUI thing).

What I found and liked about SuSE was that it gave a really 
good amount of granularity on the install package selection, 
yet didnt bog you down too much in the tiny and (after a while) 
annoying basic stuff.

I usually just installed a minimum system (about 80MB), then 
built on that for the purpose the server was for.  Better yet, 
when building server farms I'd make liberal use of the 
'save/load config' option and just build all the boxes with 
exactly the same software even across widely different 
hardware.

However, sysadmining many SuSE (or any RPM) box's is hell.  Its 
easy, but immensely time consuming.  YAST1 falls down badly in 
this area.  I never tried YOU, the online update thing.

I'm not so much a refugee from SuSE as I am from RPM.

Having said that though, I've been tremendously unhappy with 
the stability and production quality of all the SuSE distro's 
since V6.4

>Are there any hints/tips/watch-out-fors that you could offer, 
>principally, I 
>am a bit confused over the non-RPMness of packages and the 
>lack of the config 
>suites such as YAST/YAST2 that SusE employs.

Can't help you here as I'm stuck in the same quagmire.  Looking 
for admin tools that just are not there - thats fine though, I 
always hated them but I'm lazy :))

After jumping in with both feet and no idea I've now come to 
the conclusion that the best thing for a new Debianite to do is 
to learn -all- about dselect, apt and dpkg.  I don't mean skim 
over it, learn everything.  The whole Debian distro seems to be 
intimately linked to the packaging system (as they all are).  

The .deb packaging system and APT in particular would appear to 
be reason enough on their own to make the switch to this 
distro, but using them effectively is non trivial and takes a 
committment I never had to come up with before when using SuSE.

Best regards,
Craig

Re: What's worng with ALSA???

[Romuald DELAVERGNE]

The onboard SiS 7018 Audio chip is also supported by the Linux kernel
(modprobe trident)
I don't test it but others says that's it's works well.

(Google is your friend)

Le 2002.03.12 02:21, Max Koszela a écrit :

I've got the SiS 7018 audio-chip, supposedly supported both under the
kernel and in alsa with the module trident. The reason for starting to
play with alsa in the first place was the fact that the kernel (2.4.17)
OSS driver doesn't seem to support anything else than 48Khz, correct me
here if i'm wrong!

Re: ssh-client & gui

[Ross Burton]

On Tue, 2002-03-12 at 12:16, Ines Rieger wrote:
> Hi all,
> is there any graphical ssh-client for Unix?

Any terminal emulator, from xterm upwards, can run ssh.

Ross
-- 
Ross Burton mail: [EMAIL PROTECTED]
   jabber: [EMAIL PROTECTED]
 PGP Fingerprint: 1A21 F5B0 D8D0 CFE3 81D4 E25A 2D09 E447 D0B4 33DF

Re: ssh-client & gui

[Pietro Cagnoni]

> Hi all,
> is there any graphical ssh-client for Unix?
> 
> THX
> Ines

xterm -e ssh -X 

:-)

really, it's hard to me to understand what exactly should do a gui for
ssh.

maybe you need a way to launch gui apps via ssh? if so, you just need
the -X switch.

pietro.

Re: What's worng with ALSA???

[Mark Janssen]

On Tue, 2002-03-12 at 13:30, Romuald DELAVERGNE wrote:
> The onboard SiS 7018 Audio chip is also supported by the Linux kernel
> (modprobe trident)
> I don't test it but others says that's it's works well.
> 
> (Google is your friend)
> 
> Le 2002.03.12 02:21, Max Koszela a écrit :
> > I've got the SiS 7018 audio-chip, supposedly supported both under the
> > kernel and in alsa with the module trident. The reason for starting to
> > play with alsa in the first place was the fact that the kernel (2.4.17)
> > OSS driver doesn't seem to support anything else than 48Khz, correct me
> > here if i'm wrong!

Yep... it works like a charm... Only thing is... don't boot into linux
after windows (without shutdown) or sound won't work. The windows
drivers screw up some registers /me thinks :)

AFAIK 44Khz (and others) work just as well... I've been playing mpegs /
mp3's and qt's for ages, never had problems with it.

Mark Janssen


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

Re: ssh-client & gui

[Ines Rieger]

Hi,

> On Tue, 2002-03-12 at 12:16, Ines Rieger wrote:
> > Hi all,
> > is there any graphical ssh-client for Unix?
>
> Any terminal emulator, from xterm upwards, can run ssh.
>
yes, I know that (: . I am searching for a graphical user interface to run
ssh. An equivalent program to PuTTY, TeraTerm etc but for Unix.

THX
Ines

--
Es ist nichts sicher, ausser dass nichts sicher ist, und nicht einmal das.

Re: ssh-client & gui

[Timo Benk]

On Tue, Mar 12, 2002 at 01:16:45PM +0100, Ines Rieger wrote:
> Hi all,
> is there any graphical ssh-client for Unix?
I like mindterm. You can even use it from any Windows machine, it is
a java applett.

-timo

-- 
gpg key fingerprint = 6832 C8EC D823 4059 0CD1  6FBF 9383 7DBD 109E 98DC



pgpPTULWFXeO5.pgp
Description: PGP signature

RE: The future of Debian install??

[Tony Crawford]

Craig Sampson wrote (on 12 Mar 2002 at 19:57):

> I may have missed something (sure hope so), but what I'd find
> immensely useful is a way of being able to choose, at install (or
> other) time what packages I want then save this selection 'list'
> to a file so that when I next install Debian on another box I can
> just tell it to use the previously made selections.

Maybe you can adapt this:
===
From:   Eric Richardson <[EMAIL PROTECTED]>
To: "Sean 'Shaleh' Perry" <[EMAIL PROTECTED]>
Subject:Re: Debian Reinstall
Date sent:  Wed, 23 May 2001 08:56:08 -0700
Copies to:  "debian-user@lists.debian.org" 
Forwarded by:   debian-user@lists.debian.org 
Date forwarded: Wed, 23 May 2001 18:07:00 +0200 (MET DST)
Organization:   MilagroSoft Inc.

Sean 'Shaleh' Perry wrote:
> 
> On 09-May-2001 Ronan O'Sullivan wrote:
> > Hi there,
> >I am wondering is there anyway to save your current installed
> >packages information and when you reinstall for apt or dselect to
> >know what packages to install or remove to restore your system to
> >the previous state?
> >
> 
> 2 steps:
> 
> a)
> # dpkg --get-selections|sed -e 's/deinstall$/purge/'|dpkg
> # --set-selections apt-get dselect-upgrade # this removes any lingering
> # packages in your list
> 
> b)
> # dpkg --get-selections|sed -e 's/hold$/install/' > package_list
> # copy package_install /somewhere/safe
> 
> Then, after the install:
> 
> # dpkg --set-selections < package_list
> # apt-get dselect-upgrade
> 
> The sed call is to ensure that even packages on hold get stored
> properly.

This is very clever and effective. I tried it today and it 
worked like a
charm to get a good package list for backup and of course purge
deinstalled packages as per a).

BTW, How can I find why a package is on hold?

Thanks again,
Eric
=





-- 
-- Tony Crawford
-- [EMAIL PROTECTED]
-- +49-3341-30 99 99
-- 

Re: Using Alt in xterm

[Vincent Lefevre]

On Mon, Mar 11, 2002 at 14:29:40 -0600, Mark J. Tilford wrote:
> One program I use (frotz) runs in an xterm and uses Alt-(various
> keys).  The program ran fine on my old redhat system, but on this setup,
> Alt-U (or other key) simply registers as U.  It works fine in character
> mode, though.  Any ideas how to get it to work in X?
> 
> (from my ~/.Xdefaults file)
> XTerm*metaSendsEscape:  true
> XTerm*eightBitInput:true

OK.

> (from my ~/.xmodmap)
> keycode 113 = Mode_switch
> keycode 115 = Meta_L 
> clear mod1
> add mod1 = Meta_L
> clear mod3
> add mod3 = Mode_switch

OK (make sure that these are the right keycodes for Alt and Meta,
you can check with xev).

Is the remaining of your .xmodmap OK?

For instance, I have:

keycode  38 = a A ae AE

Thus Alt-a gives æ and Shift-Alt-a gives Æ.

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA

Re: ? about C++

[Anthony DeRobertis]

On Thursday, March 7, 2002, at 03:57 PM, Dimitri Maziuk wrote:

Anyhow, my point was, name 4 problem areas in C.



You're lucky with 'none of the above'. It could be...


1. No array bounds checking (Fix: use vector<> or equivalent)


Real-world idiot fix: My name is 7 characters; give it twenty, 
we'll never need more than that! Then use that twenty inline in 
the code, as well as simplifying the expressions a little so we 
get the related constants 21 (w/ null), 19 (can't count, 
paranoid), 8, 9, 10 (unicode), and infinity (core dump).



2. Pointers (Fix: use references, iterators etc.).


Real-world idiot fix: Pass everything by value. Who needs 
pointers, references, etc.? Use indices with magic constants as 
the bounds (see above) to iterate arrays.


BTW: I disagree _strongly_ that pointers are a misfeature of C.
They are a very useful tool. However, misuse of pointers
is a misfeature of some programmers.


3. Forgetting to free() malloc()'ed memory (Fix: use auto_ptr<>,
   destructors etc.)


Real-world idiot fix: Buy more memory. If that proves 
impossible, increase amount of disk used for swap. In the event 
of hitting architectural limits, go 64-bit. Or just make the 
program crash more often. Then it won't leak as much memory. 
Tell the user to save often.



4. Using pointers to (hopefully array) (hopefully allocated) (hopefully
   null-terminated) of char as string data type (see also 1, 2, & 3)
   (Fix: use std::string)


Real-world idiot fix: Add 'hopefully not overflowed' to the 
list. Fix allocation problems by allocating on the stack. If 
worried about array not being large enough, increase static 
size. See #1. If worried about excess stack usage, see #3. And, 
for the hell of it, see #4 ;-)

Re: Moving /var to another drive

[George Karaolides]

Hi Andrew,

You need to be root on your machine to do this.

Make two partitions on the new drive using

cfdisk /dev/

Then make a filesystem on each partition.  Stick to the tried-and-tested
ext2, or go for one of the new journalling ones like ext3 or reiserfs, if
you're running a kernel recent enough to support them or can compile one
that does.

E.g. mke2fs /dev/

Then mount the first partition e.g. under /mnt

mount /dev/ /mnt

And transfer the data using tar:

tar cplf - -C / var | tar xvf - -C /mnt

Unmount the partition

umount /mnt

Mount the other one and do the same thing for /usr:

mount /dev/
tar cplf - -C / usr | tar xvf - -C /mnt
umount /mnt

Make sure you type the tar commands exactly as above, you don't want to go
dropping the entire copnmtents of /var or /usr (or even / if you v=get it
wrong) in the wrong place!

Edit /etc/fstab and either add lines for /usr and /var, or change existing
ones, to mount your new filesystems instead of the old ones.  Here's an
example:

--- fstab excerpt ---

/dev/sdb6 /usr ext2 rw  0   2
/dev/sdb7 /var ext2 rw  0   2

--- end fstab excerpt ---

Then go to single user mode:

telinit S

Unmount /usr and /var:

umount /usr
umount /var

Mount all partitions using your new fstab:

mount -a

Check that everything is mounted where it should be:

mount

And go back to your usual runlevel (usually 2 on Debian):

telinit 2

That's it.  Note that if /usr and /var were originally part of your root
filesystem, the data will still be there but the new filesystem will be
mounted on the top level directory so you won't see it.  After you've
successfully transferred /usr and /var out of the root filesystem, you can
reclaim the space by going to single user mode, unmounting /usr and /var,
and doing the following:

rm -rvf /var
rm -rvf /usr
mkdir /var
mkdir /usr

Be VERY careful that /usr and /var are NOT mounted when you do this!
Also note, a space or a slash in the wrong place in either of the above
commands can wreck your system completely!

Good luck,

|   George Karaolides  Linustech Advanced Solutions,  |
|   tel:   +357 22 55 61 29  86 Ifigenias Street, 3rd Floor,  |
|   web:   www.linustech.com.cy  Strovolos, Nicosia CY 2003,  |
|   email: [EMAIL PROTECTED]   Republic  of Cyprus.  |

On Tue, 12 Mar 2002, Andrew Stephen wrote:

> Hi
>
> My /var and /usr partitions have just run out of space and I was wondering
> what is the best way to copy them to a new drive that has just been
> installed.
>
> Any suggestions would be greatly appreciated.
>
> Regards,
> Andrew
>
>
>
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
>
>
>

Re: Using Alt in xterm

[Andrew M. Davenport]

Set eightBitInput to false.  You might want to avoid xmodmap, too, if you can.

-Andrew

On Mon, Mar 11, 2002 at 02:29:40PM -0600, Mark J. Tilford wrote:
> One program I use (frotz) runs in an xterm and uses Alt-(various
> keys).  The program ran fine on my old redhat system, but on this setup,
> Alt-U (or other key) simply registers as U.  It works fine in character
> mode, though.  Any ideas how to get it to work in X?
> 
> (from my ~/.Xdefaults file)
> *eightBitInput: true
> *metaSendsEscape:   true
> XTerm*metaSendsEscape:  true
> XTerm*eightBitInput:true
> 
> 
> 
> (from my ~/.xmodmap)
> keycode 113 = Mode_switch
> keycode 115 = Meta_L 
> clear mod1
> add mod1 = Meta_L
> clear mod3
> add mod3 = Mode_switch
> 
> 
> 
> Does anybody know how to get this to work correctly?
> 
> 
> Mark   Jeffrey   Tilford
> [EMAIL PROTECTED]
-- 
Andrew Davenport   | All good gifts around us
[EMAIL PROTECTED] | are sent from Heaven above.
http://www.tropnevad.org/~adavenpo | Then thank the Lord! Thank the Lord
103 Cluff Crossing Road, Apt. X-8  | for all his love!
Salem, NH  03079   | (Chorus of All Good Gifts, from GodSpell)

Re: ssh-client & gui

[Vincent Lefevre]

On Tue, Mar 12, 2002 at 13:41:29 +0100, Ines Rieger wrote:
> > On Tue, 2002-03-12 at 12:16, Ines Rieger wrote:
> > > is there any graphical ssh-client for Unix?
> >
> > Any terminal emulator, from xterm upwards, can run ssh.
> >
> yes, I know that (: . I am searching for a graphical user interface to run
> ssh. An equivalent program to PuTTY, TeraTerm etc but for Unix.

What's the difference with running ssh in a xterm?

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA

Debian as server?

[R. van der Slot]

Dear support,

I was wandering on the internet in search for the perfect server(s) 
distribution of Linux. My eye felt on Debian, only it's not completely clear 
if I can use Debian as PROXY, File/print, mail server (with Exchange 
coorporation) and/or Oracle server.


I want to setup these servers for a big project. Only I want to know which 
servers I can use Debian for and or what problems I can encounter with i.e. 
exchange?


Thanks in advance,

R. van der Slot

_
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com

AltGr dont work !?!

[Martin A. Hansen]

hi

after an upgrade of xf3 to xf4 my AltGr key dont work in x. it do work in 
console thou.

i have a dell latitude laptop.

i have no idea what to do?


help needed :)

martin
-- 

Re: Using Alt in xterm

[Mark J. Tilford]

On Tue, 12 Mar 2002, Vincent Lefevre wrote:

> On Mon, Mar 11, 2002 at 14:29:40 -0600, Mark J. Tilford wrote:
> > One program I use (frotz) runs in an xterm and uses Alt-(various
> > keys).  The program ran fine on my old redhat system, but on this setup,
> > Alt-U (or other key) simply registers as U.  It works fine in character
> > mode, though.  Any ideas how to get it to work in X?
> > 
> > (from my ~/.Xdefaults file)
> > XTerm*metaSendsEscape:  true
> > XTerm*eightBitInput:true
> 
> OK.
> 
> > (from my ~/.xmodmap)
> > keycode 113 = Mode_switch
> > keycode 115 = Meta_L 
> > clear mod1
> > add mod1 = Meta_L
> > clear mod3
> > add mod3 = Mode_switch
> 
> OK (make sure that these are the right keycodes for Alt and Meta,
> you can check with xev).
> 
> Is the remaining of your .xmodmap OK?

Actually, that's my entire .xmodmap; I had never used it before and
somebody suggested setting up the file like that.

> For instance, I have:
> 
> keycode  38 = a A ae AE
> 
> Thus Alt-a gives ? and Shift-Alt-a gives ?.
> 


Mark   Jeffrey   Tilford
[EMAIL PROTECTED]

Re: ssh-client & gui

[Ines Rieger]

Hi,

> > yes, I know that (: . I am searching for a graphical user interface to run
> > ssh. An equivalent program to PuTTY, TeraTerm etc but for Unix.
>
> What's the difference with running ssh in a xterm?

the buttons and drag an drop felling and so on. My users want to select a
file in a file browser an copy by ssh to another machine. They do not want
to use an xterm an a keyboard. They only want to use their mice. (:

cu
Ines

--
Es ist nichts sicher, ausser dass nichts sicher ist, und nicht einmal das.

Re: ssh-client & gui

[Craig Dickson]

begin  Vincent Lefevre  quotation:

> On Tue, Mar 12, 2002 at 13:41:29 +0100, Ines Rieger wrote:
> > > On Tue, 2002-03-12 at 12:16, Ines Rieger wrote:
> > > > is there any graphical ssh-client for Unix?
> > >
> > > Any terminal emulator, from xterm upwards, can run ssh.
> > >
> > yes, I know that (: . I am searching for a graphical user interface to run
> > ssh. An equivalent program to PuTTY, TeraTerm etc but for Unix.
> 
> What's the difference with running ssh in a xterm?

Not much. Configuration dialog boxes, mostly. I use PuTTY on Win32
systems, but would drop it if only the cygwin bash window behaved like a
proper xterm.

Craig


pgpGun1a1Fdls.pgp
Description: PGP signature

Re: The future of Debian install??

[Sridhar M.A.]

On Tue, Mar 12, 2002 at 07:57:31PM +0800, Craig Sampson wrote:
   > 
   > Something like 'would you like to import package selections 
   > from floppy (or other) disk' would be just great.
   > 
It is there. Recently installed woody on about 20 machines with the
following method:

  1. On some machine that I had already installed woody with a bunch of
 packages did
 
   dpkg --get-selections > inst.list
   
  2. Copied inst.list on to a floppy.

  3. Installed the base system on a new machine.

  4. Copied the inst.list from the floppy on to the new machine
 (went the mount floppy route; no mtools yet)

  5. Did 
   dpkg --set-selections < inst.list

  6. Pop in the woody cd (or the appropriate sources) and did

apt-get dselect-upgrade

Now, you have an installation which is exactly what  you had one some
other machine.
  
The process was so smooth, it did not fail even once. Hats off to the
debian team for the apt-get and friends.

-- 
Sridhar M.A.mas at uomphysics dot net

This fortune was brought to you by the people at Hewlett-Packard.

Re: Debian as server?

[Juhan Kundla]

Ühel ilusal päeval [12.03.2002] kirjutas R. van der Slot <[EMAIL PROTECTED]>:
> Dear support,
> 
> I was wandering on the internet in search for the perfect server(s) 
> distribution of Linux. My eye felt on Debian, only it's not completely 
> clear if I can use Debian as PROXY, File/print, mail server (with Exchange 
> coorporation) and/or Oracle server.
> 
> I want to setup these servers for a big project. Only I want to know which 
> servers I can use Debian for and or what problems I can encounter with i.e. 
> exchange?

Heips!

Note, that this is not official Debian support, this is Debian users'
mailing list.

Anyway, i am quite convinced, that you can use Debian for those purposes
you described in your posting. I am no expert on mail/file/print
servers, but i have been experimenting with Oracle on Debian a little.

If you are planning to run Oracle8i server on Linux platform, then i
truely recommend Debian Potato. Oracle
install process has been historically major PITA on the linux platform.
However, on Debian Potato it went smoothly, except one little glitch --
the root.sh script didn't find awk, so the script needed to be edited.
Oracle on Debian seems very stable to me, I haven't had any problems
with it. Oh, and when installing Oracle on Debian Potato, one does not
have to mess with this scary downgrading glibc stuff, this means, that
install on Debian is much cleaner than install on RedHat.

Namarie!
Juhan
(Ugh! I am not a native english speaker, as you have probably noticed.)

-- 
In the early morning hour, when the pub was closing, my grandpa
emptied his tankard, stood up and said his famous words:
  Save Your Breath ... You'll need it to blow up your date!

http://www.hot.ee/oppalilli/

Re: ssh-client & gui

[Ross Burton]

On Tue, 2002-03-12 at 14:24, Craig Dickson wrote:
> > What's the difference with running ssh in a xterm?
> Not much. Configuration dialog boxes, mostly. I use PuTTY on Win32
> systems, but would drop it if only the cygwin bash window behaved like a
> proper xterm.

Have you noticed that Cygwin are shipping rxvt as a package?  That makes
everything far, far saner.

Ross
-- 
Ross Burton mail: [EMAIL PROTECTED]
   jabber: [EMAIL PROTECTED]
 PGP Fingerprint: 1A21 F5B0 D8D0 CFE3 81D4 E25A 2D09 E447 D0B4 33DF

Re: ssh-client & gui

[Brett Parker]

On Tue, Mar 12, 2002 at 01:16:45PM +0100, Ines Rieger wrote:
> Hi all,
> is there any graphical ssh-client for Unix?

Testing onwards has secpanel, which I've used a couple of times
and at a guess would do what you're after.

Thanks,

Brett Parker

Solved: AltGr dont work !?!

[Martin A. Hansen]

it worked when i hashed out

Option"XkbVariant"  "dk"

in /etc/X11/XF86Config-4


:/

--m

On Tue, Mar 12, 2002 at 03:08:17PM +0100, Martin A. Hansen wrote:
> hi
> 
> after an upgrade of xf3 to xf4 my AltGr key dont work in x. it do work in 
> console thou.
> 
> i have a dell latitude laptop.
> 
> i have no idea what to do?
> 
> 
> help needed :)
> 
> martin
> -- 
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

-- 

Re: new installation from /var/cache/apt/archives

[Craig Sampson]

>New installation from /var/cache/apt/archives
>
>> May I do that ?
>>
>huh, what do you mean?

I think what is meant here is that if you have a (fairly 
massive) bunch of .deb files in your apt cache could you do a 
completely fresh install of debian using this cache as the 
source?

My guess would be no, probably not.  I think the best way to 
preserve your hard gotten downloads (hard gotten for those of 
us with modems) would be to apt-move the cache first so that it 
at least resembles a partial mirror.

There would, I assume, be at least a few missing files that an 
initial install would be looking for using this method as well.

Perhaps the best way of attacking the problem would be to do 
the base install from local media like a CD, then do an 
immediate upgrade using the files you've stored via apt-move 
from /var/cache/apt


Cheers,
Craig

Re: Debian as server?

[Sebastiaan]

High,

On Tue, 12 Mar 2002, R. van der Slot wrote:

> Dear support,
>
> I was wandering on the internet in search for the perfect server(s)
> distribution of Linux. My eye felt on Debian, only it's not completely clear
> if I can use Debian as PROXY, File/print, mail server (with Exchange
> coorporation) and/or Oracle server.
>
Debian fits perfectly well. I have a Debian box running as
proxy/mail/firewall server and another one as file/print server (for Linux
printers, that is. I have no experience with SAMBA to share windoze
printers, but it should work fine too).


Greetz,
Sebastiaan

> I want to setup these servers for a big project. Only I want to know which
> servers I can use Debian for and or what problems I can encounter with i.e.
> exchange?
>



--
  NT is the OS of the future. The main engine is the 16-bit Subsystem
  (also called MS-DOS Subsystem). Above that, there is the windoze 95/98
  16-bit Subsystem. Anyone can see that 16+16=32, so windoze NT is a
  *real* 32-bit system.

Re: mozilla-psm installed but not https

[Roberto Pereyra]

You could  fix it, download the mozilla binaries from

www.mozilla.org

The last version support ssl

bye
 
 
Roberto Pereyra
Gualeguaychu
Argentina
GnuPG keyID: BB43E337
http://pgp.mit.edu

On Fri, 8 Mar 2002, Bill Moseley wrote:

> Did I miss a step or is a dependency missing?
> 
> I had mozilla installed on Sid.  But couldn't got to a SSL site.  Installed 
> mozilla-psm, but still same situation.
> 
> Granted all I did was apt-get install mozilla-psm, and poke around google.  
> Still seems like that should have got my https working.
> 
> Thanks,
> 
> 
> -- 
> Bill Moseley
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 
> 

FYI: Installing the Intel C++ compiler on Debian SID

[Douglas Eck]

Hello All,

I wanted to play around with the free (as in beer) Intel C++ compiler.
It was a bit of work to get it going on Sid. I thought I'd post
my experiences to debian-user. These instructions might be invalid
in a week. I'm not a glibc pro so I can't say

PS Please don't flame me about using the Intel compiler. I'm not
advocating it, and I think the GNU compiler is great. I just
wanted to play around with it.

Anyway, here goes.

1) Go to http://www.intel.com/software/products/compilers/c50/linux/
and create a Premier Support Account. It's free.

2) Log in.

3) Download the latest version of the compiler
for RedHat 7.1 As of the writing of this email, the file is
cc_rh71_020201.tar.

4) Download the file LinuxOtherOSSupport5_0_1.tar.

5) Untar cc_rh71_020201.tar

6) Run alien --to-tgz on all .rpms  (making .debs failed for me)

7) Run tar -xzf on all resulting .tgz files.
This yields an "/opt" directory

8) The install is self-contained and you can put it where you want (that is,
it writes nothing to system dirs like /usr/include. I like /usr/local for stuff
like this (when there is no .deb file) so I did this:

cp -a opt/intel /usr/local

9) Edit /usr/local/intel/compiler50/ia32/bin/iccvars.sh and iccvars.csh, 
replacing
 with /usr/local/intel. I don't have a 64 bit machine so I ignored
the ia64 directory.

10)Untar LinuxOtherOSSupport_5_0_1.tar

11) Follow the instructions in LinuxOtherOSSupport5_0_1.htm for installing the 
replacement
libs.

12) Do *not* (stress *not*) install the new substitute_header files... only the 
new libs.
This differs from the instructions for SuSE or Caldera.

All done. As far as I can tell, this works great.

Cheers,
Doug
--
Dr. Douglas Eck, http://www.idsia.ch/~doug
Istituto Dalle Molle di Studi sull'Intelligenza Artificiale (IDSIA)
Neural Networks, Rhythm Perception and Production, Dynamical Systems

kernel-2.4.17

[Gerard Robin]

hello,
I have a problem with the kernel 2.4.17:
My machine: HDD1 in hda
HDD2 in hdb

The machine boots with lilo on hda.
I boot with a disket 1.44 Mb for hdb.

After the compile, my machine booted correctly with the
kernel 2.4.17 on the disket 1.44 Mb, but I forgot to 
include a module in the kernel and I recompiled the 
kernel 2.4.17 and now when I boot the machine with the 
boot disket I get :

NET4:unix domain socket 1.0/SMP for LINUX NET4.0.
request_module[ide-disk]:Rot fs not mounted
hdb:driver not present
VFS:Cannot open root device "343" or 03:43
Please apend a correct "root=" boot option
Kernel panic:VFS:Unable to mount root fs on 03:43

Can someone help me to solve this problem, please.
TIA
-- 
Gerard

Re: Moving /var to another drive

[François]

And what happens if the /var/log and /var/run dirs that can change during
the tar ?

François


On Tue, 12 Mar 2002 15:33:55 +0200 (EET)
George Karaolides <[EMAIL PROTECTED]> wrote:

> 
> Hi Andrew,
> 
> You need to be root on your machine to do this.
> 
> Make two partitions on the new drive using
> 
> cfdisk /dev/
> 
> Then make a filesystem on each partition.  Stick to the tried-and-tested
> ext2, or go for one of the new journalling ones like ext3 or reiserfs, if
> you're running a kernel recent enough to support them or can compile one
> that does.
> 
> E.g. mke2fs /dev/
> 
> Then mount the first partition e.g. under /mnt
> 
> mount /dev/ /mnt
> 
> And transfer the data using tar:
> 
> tar cplf - -C / var | tar xvf - -C /mnt
> 
> Unmount the partition
> 
> umount /mnt
> 
> Mount the other one and do the same thing for /usr:
> 
> mount /dev/
> tar cplf - -C / usr | tar xvf - -C /mnt
> umount /mnt
> 
> Make sure you type the tar commands exactly as above, you don't want to go
> dropping the entire copnmtents of /var or /usr (or even / if you v=get it
> wrong) in the wrong place!
> 
> Edit /etc/fstab and either add lines for /usr and /var, or change existing
> ones, to mount your new filesystems instead of the old ones.  Here's an
> example:
> 
> --- fstab excerpt ---
> 
> /dev/sdb6 /usr ext2 rw0   2
> /dev/sdb7 /var ext2 rw0   2
> 
> --- end fstab excerpt ---
> 
> Then go to single user mode:
> 
> telinit S
> 
> Unmount /usr and /var:
> 
> umount /usr
> umount /var
> 
> Mount all partitions using your new fstab:
> 
> mount -a
> 
> Check that everything is mounted where it should be:
> 
> mount
> 
> And go back to your usual runlevel (usually 2 on Debian):
> 
> telinit 2
> 
> That's it.  Note that if /usr and /var were originally part of your root
> filesystem, the data will still be there but the new filesystem will be
> mounted on the top level directory so you won't see it.  After you've
> successfully transferred /usr and /var out of the root filesystem, you can
> reclaim the space by going to single user mode, unmounting /usr and /var,
> and doing the following:
> 
> rm -rvf /var
> rm -rvf /usr
> mkdir /var
> mkdir /usr
> 
> Be VERY careful that /usr and /var are NOT mounted when you do this!
> Also note, a space or a slash in the wrong place in either of the above
> commands can wreck your system completely!
> 
> Good luck,
> 
> |   George Karaolides  Linustech Advanced Solutions,  |
> |   tel:   +357 22 55 61 29  86 Ifigenias Street, 3rd Floor,  |
> |   web:   www.linustech.com.cy  Strovolos, Nicosia CY 2003,  |
> |   email: [EMAIL PROTECTED]   Republic  of Cyprus.  |
> 
> On Tue, 12 Mar 2002, Andrew Stephen wrote:
> 
> > Hi
> >
> > My /var and /usr partitions have just run out of space and I was wondering
> > what is the best way to copy them to a new drive that has just been
> > installed.
> >
> > Any suggestions would be greatly appreciated.
> >
> > Regards,
> > Andrew
> >
> >
> >
> > --
> > To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> >
> >
> >
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 
> 

making my own dist. ? ?

[Adam craig]

Hello,

I've 3 servers to install - not the same hardware but the software needs to be 
the same.
I have installed the first server and got everything working the way I want it 
to.
My problem is that I need to use interbase with php4 and I can't get it to work 
with php4.1.x - the most recent version it will work with is php4.0.6.
Now using apt-move get;apt-move move, I have a copy on hard drive of all the 
packages I need to install the other 2 servers.
My idea was to update the first server direct from the internet, test, then 
(apt-move get/move) update the 2 other servers from the first server. So how 
can I disable any updates relating to php4 ?
Or should I (can I ?) create my own php4.deb and change some config. file (that 
I haven't yet found) so that my version will always be the most recent for me ?
I'm sure this must be a fairly common situation - any suggestions would be 
greatly appreciated.

Adam CRAIG 

Re: ssh-client & gui

[Dale Hair]

gftp has the option of using SSH & SSH2, haven't tried it.  

On Tue, 2002-03-12 at 08:16, Ines Rieger wrote:
> Hi,
> 
> > > yes, I know that (: . I am searching for a graphical user interface to run
> > > ssh. An equivalent program to PuTTY, TeraTerm etc but for Unix.
> >
> > What's the difference with running ssh in a xterm?
> 
> the buttons and drag an drop felling and so on. My users want to select a
> file in a file browser an copy by ssh to another machine. They do not want
> to use an xterm an a keyboard. They only want to use their mice. (:
> 
> cu
> Ines
> 
> --
> Es ist nichts sicher, ausser dass nichts sicher ist, und nicht einmal das.
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 

Re: Nvidia Debian Kernel

[Jonathan Ard]

Hi, Kai,
> 
> I apt-get source kernel-source-2.4.18

Actually, I think you want to use "apt-get install
kernel-source-2.4.18".  You are downloading the source to the kernel
source package, which, as you say later, gives you files you don't
need.  If you just apt-get install the kernel-source, it will put the
kernel source tarred and zipped into /usr/src.  Bunzip2 and untar it. 
Now you will have a kernel-source-2.4.18 directory.  (I always move it
to /usr/src/linux; but I don't know if you are suppose to or not).  
Then cd to the source directory and type make config, make menuconfig,
or make xconfig to configure the kernel.  
When you are done, type "make-kpkg --rev [whatever you want]
kernel-image".
Now you will have a .deb of your kernel image in /usr/src.  Then follow
the directions in /usr/share/doc/nvidia-glx-src/README.Debian and
/usr/share/doc/nvidia-kernel-src/README.Debian.  

Hope that helps,

Jonathan 





_
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com

Re: ? about C++

[Dimitri Maziuk]

* Anthony DeRobertis ([EMAIL PROTECTED]) spake thusly:
> 
> On Thursday, March 7, 2002, at 03:57 PM, Dimitri Maziuk wrote:
> >Anyhow, my point was, name 4 problem areas in C.
> >
> 
> You're lucky with 'none of the above'. It could be...

Luck has nothing to do with it

> >1. No array bounds checking (Fix: use vector<> or equivalent)
> 
> Real-world idiot fix: My name is 7 characters; give it twenty, 
> we'll never need more than that! Then use that twenty inline in 
> the code, as well as simplifying the expressions a little so we 
> get the related constants 21 (w/ null), 19 (can't count, 
> paranoid), 8, 9, 10 (unicode), and infinity (core dump).
>
> >2. Pointers (Fix: use references, iterators etc.).
> 
> Real-world idiot fix: Pass everything by value. Who needs 
> pointers, references, etc.? Use indices with magic constants as 
> the bounds (see above) to iterate arrays.
> 
> BTW: I disagree _strongly_ that pointers are a misfeature of C.
>   They are a very useful tool. However, misuse of pointers
>   is a misfeature of some programmers.

Real-world troubleshooting technique: put ulimit on stack size.
If program segfaults on int main( int argc, char **argv ),
- segfault here ---
it has both of the above.

[ snip ]

Yeah. BTDT, got more t-shirts than I'll ever need.

Dima
-- 
The wombat is a mixture of chalk and clay used for respiration.  -- MegaHal

Re: new installation from /var/cache/apt/archives

[François]

On Tue, 12 Mar 2002 23:09:24 +0800
"Craig Sampson" <[EMAIL PROTECTED]> wrote:

> >New installation from /var/cache/apt/archives
> >
> >> May I do that ?
> >>
> >huh, what do you mean?
> 
> I think what is meant here is that if you have a (fairly 
> massive) bunch of .deb files in your apt cache could you do a 
> completely fresh install of debian using this cache as the 
> source?
> 
Yes ;-)


> My guess would be no, probably not.  I think the best way to 
> preserve your hard gotten downloads (hard gotten for those of 
> us with modems) would be to apt-move the cache first so that it 
> at least resembles a partial mirror.
> 
> There would, I assume, be at least a few missing files that an 
> initial install would be looking for using this method as well.
> 
> Perhaps the best way of attacking the problem would be to do 
> the base install from local media like a CD, then do an 
> immediate upgrade using the files you've stored via apt-move 
> from /var/cache/apt
> 
In fact, I have to change my laptop disk from 8Go to 20Go. But I would
like to duplicat my system without re-installing it !

another debian-user thread talk about moving /var to another drive using tar
but I wonder if it forks for entire system !

The real pb is that I can't make an iso from my devices because they are too 
large
and I can't put 2 hard drive on my laptop !
I can put one of both on a desktop and transfert datas via ftp/nfs/rsync...
But I can't stop the laptop ! Is tar or rsync enougth to transfert an entire 
system
while this one running ?


Thanks 4 responses

François

Re: best way to install non debian package

[Sridhar M.A.]

On Tue, Mar 12, 2002 at 11:02:15AM +, Karl E. Jorgensen wrote:
   > On Tue, Mar 12, 2002 at 11:01:01AM +0100, Fran?ois Chenais wrote:
   > > 
   > >  What's the best way to install package which are not(yet) in debian
   > >  package format or in the current distribution ?
   > > 
   > >  For example, I need pilot-link > 0.10.0 to use with palm m125
   > >  but have no deb package.
   > > 
   > 
   > The debian package manage system will keep its hands off /usr/local, so
   > that is a good choice.
   > 
   > Have you checked whether your desired version of pilot-link is available
   > in sid? It's probably quicker to get it from there.
   > 

If you want install non-debian software, i.e., after compiling a
compressed tarball, I would suggest that you give checkinstall a try. It
creates a deb package, which installs the binaries in /usr/local.
Installation/Removal is thru dpkg -i/-r. Bonus is that you can install
the resulating deb on another machine with out going thru the hassle of
compiling gain.

Checkinstall debs are available at:

http://asic-linux.com.mx/~izto/checkinstall/

Regards,

-- 
Sridhar M.A.mas at uomphysics dot net

You are the only person to ever get this message.

Re: GnuCash vs MoneyDance

[James D Strandboge]

On Sat, 2002-03-09 at 18:47, David Roundy wrote:
> On Sat, Mar 09, 2002 at 02:51:30PM -0600, Alex Malinovich wrote:
> >
> > Personally, I would very much prefer to keep my system as free as
> > possible. I'd much prefer to use gnucash over some other alternative,
> > but so far I've found it severely lacking in one area, and that's
> > importing bank statements via QIF. I use my Discover and American
> > Express cards quite a bit, and entering each of those transactions into
> > gnucash by hand would get very tedious very quickly. Other than that, I
> > love gnucash. Unfortunately, for me at least, that's reason enough to
> > start looking elsewhere.
> 
> I have a question about this.  I've just started using gnucash, so I
> haven't gotten to the point of importing my bank statements (since it
> hasn't been a month yet), but it seems like it has a working import
> function (I tried it on an old month, but didn't actually do it, since it
> would have confused things).
> 
> I've heard before that gnucash has problems when you import your monthly
> statements, but never understood what that problem is.  Can you explain
> this for me? I'd rather know what I'm up against when I get my bank
> statements next month...
> -- 

I am probably not the best person to answer this since I haven't
actually done it myself, but I know people who are able to import
quicken files.  They just need to be in the right format, or something. 
Thry got all he needed from the gnucash site:

http://www.gnucash.org/

Jamie

-- 
Email:[EMAIL PROTECTED]
GPG/PGP ID:   26384A3A
Fingerprint:  D9FF DF4A 2D46 A353 A289  E8F5 AA75 DCBE 2638 4A3A


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

Re: Using Alt in xterm

[Vincent Lefevre]

On Tue, Mar 12, 2002 at 08:10:02 -0600, Mark J. Tilford wrote:
> Actually, that's my entire .xmodmap; I had never used it before and
> somebody suggested setting up the file like that.

I did, because this is my config on my Powerbook and it works very well.
But this is the configuration of the Alt and Meta keys only. You have to
configure the other keys too, but this depends on your personal choices.

Note that setting XTerm*eightBitInput to false (in your .Xdefaults)
will not work on some Linux/PPC kernels.

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA

Re: Debian as server?

[Sean 'Shaleh' Perry]

On 12-Mar-2002 R. van der Slot wrote:
> Dear support,
> 
> I was wandering on the internet in search for the perfect server(s) 
> distribution of Linux. My eye felt on Debian, only it's not completely clear 
> if I can use Debian as PROXY, File/print, mail server (with Exchange 
> coorporation) and/or Oracle server.
> 
> I want to setup these servers for a big project. Only I want to know which 
> servers I can use Debian for and or what problems I can encounter with i.e. 
> exchange?
> 

look at the debian mailing list home page.  One of the lists is 'debian-isp --
Debian at work in Internet Service Providers'.  If they are willing to run
Debian, I do not see why you would not be.

Exchange can be a bear to interoperate with from any OS other than Windows. 
Beyond that everything should be clear sailing.

Re: fatal server error: no screens found (SOLVED)

[timothy bauscher]

Nermind, problem solved.

Thank you for your help,
(==timothy==)

=

I copied the file to the /etc/X11/
directory, renamed it, and made gdm my
default display manager by running:

dpkg-reconfigure gdm

Now, X starts via gdm. I had xdm as
the default dm but it kept booting kde.

Small problem: X is running at 640x480,
I can't CNTRL+ALT+[+], and the
/etc/X11/XF86Config-4 file doesn't
mention any resolutions I can change,
only color depths.

I am guessing there is another configuration
tool I should use, or the same one with
different options?

Thank you Patrick and Alan for your help.

(==timothy==)

===

On Mon, Mar 11, 2002 at 01:23:31PM -0500, timothy bauscher wrote:
>x-window-system and gnome-session are installed. xdm is the default 
display manager. I am using the nv driver for my RIVA TNT2, and have 
just completed a fresh install of unstable.
>
>When xdm (or gdm) are launched, they display a black screen 3 times 
and then fail. When startx is launched it also fails. The error I 
receive is:
>
>(EE) Screen(s) found, but none have a usable configuration.
>Fatal server error:
>no screens found
>

login as root
run XFree86 -configure

A working XF86Config file should now be in the root directory for you 
to either use as is if everything has worked perfectly or forward to us 
for help if you have problems.



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


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

Re: ssh-client & gui

[Vincent Lefevre]

On Tue, Mar 12, 2002 at 15:16:41 +0100, Ines Rieger wrote:
> > > yes, I know that (: . I am searching for a graphical user
> > > interface to run ssh. An equivalent program to PuTTY, TeraTerm
> > > etc but for Unix.
> >
> > What's the difference with running ssh in a xterm?
> 
> the buttons and drag an drop felling and so on. My users want to select a
> file in a file browser an copy by ssh to another machine. They do not want
> to use an xterm an a keyboard. They only want to use their mice. (:

Then, you want a GUI for scp, not for ssh, I suppose. As the manpage
says, ssh is a remote login program, not a file transfer program.

-- 
Vincent Lefèvre <[EMAIL PROTECTED]> - Web:  - 100%
validated (X)HTML - Acorn Risc PC, Yellow Pig 17, Championnat International
des Jeux Mathématiques et Logiques, TETRHEX, etc.
Work: CR INRIA - computer arithmetic / SPACES project at LORIA

Re: TrueType

[Simon Hepburn]

Oki DZ wrote:

> It seems that TrueType fonts are always displayed more jagged compared
> to X fonts. Is that the case?

Not normally. What do you mean by X fonts - postscript ? Where did you get 
your truetype fonts from ? There are 3 methods of displaying truetype in X4

-freetype backend
-xtt backend
-xfs-xtt font server

Which one are you using ?

-- 
Simon Hepburn.

RE: Re Root symlink to /boot/vmlinuz

[Brooks R. Robinson]

| > me thinks you would like make-kpkg (kernel-package)
| >
|
| IIRC, kernel-package asks if you want the symlinks.  I may be wrong, it's
| been a while since I've needed a new kernel :)
|

Last I ran it (a couple of days ago) out of woody, it didn't ask about
symlinks.  It just does them as if the kernel you are installing is the new
default kernel.  I always have to go back and update my symlinks, but tehn
again I typically have at least three different kernels ready to go at a
given moment.

HTH,

Brooks

Pump and Dhclient

[hanasaki]

I have both installed.
- is there a way to tell ifup/down which to use?
dhclient seems to need entries for each interface, that will get a 
lease, in /etc/dhclient.conf.  Pump is nice in that it only requireds on 
place to specifiy that the interface will be dhcp

(/etc/network/interfaces)
Is what are the pro's cons of using these differnt tools?
Thank you.
--
=
= [EMAIL PROTECTED]  =
= Spam : Unhealthy and High in Sodium and Cholesterol   =
=

RE: Debian as server?

[Panuganty, Ramesh]

> I was wandering on the internet in search for the perfect server(s) 
> distribution of Linux. My eye felt on Debian, only it's not completely
clear 
> I can use Debian as PROXY, File/print, mail server (with Exchange 
> orporation) and/or Oracle server.

I had earlier setup a Debian Potato as a proxy & mail server (with squid
& qmail) and the system once ran for a 200 days uptime with nearly 150
users. Quite stable!

RE: Missing nsupdate

[Panuganty, Ramesh]

I guess that file is moved to dnsutils in the newer versions. I can see
this in bind package in the earlier releases. Next time onwards, you can
search for the contents of packages in
http://www.debian.org/distrib/packages. Select option 1, which allows to
search for a file across all packages.

-Ramesh

-Original Message-
From: Jerome Acks Jr [mailto:[EMAIL PROTECTED] 
Sent: Saturday, March 09, 2002 2:11 PM
To: debian-user@lists.debian.org
Subject: Re: Missing nsupdate


On Tue, Mar 05, 2002 at 12:51:44PM -0500, David Priban wrote:
> I'm probably missing somethig here but does anybody know what happend 
> to nsupdate utility from bind package? It was included in bind 8.2.4 
> but is not in 8.3.1 .deb file but it is part of original sources from 
> ISC. Same thing applies for bind 9.2.0 (didn't check 9.1). Is it part 
> of different package or is there something else to be used for dynamic

> zone updates?
/usr/bin/nsupdate is in dnsutils
-- 
Jerome


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

Re: ssh-client & gui

[Craig Dickson]

begin  Ross Burton  quotation:

> Have you noticed that Cygwin are shipping rxvt as a package?  That makes
> everything far, far saner.

I knew it was there, but I thought it required XFree86/Cygwin. I am
pleasantly surprised to find that it does not, and simply appears in
its own top-level Win32 frame if X is not running. Nice. Now I have to
figure out how to configure it (~/.Xresources, I assume?). At the very
least, I want the scroll bar on the right, the colors set to white text
on black, a larger window, and a nicer font. "man rxvt", I suppose.

Thanks.

Craig


pgpsUFlrirCKv.pgp
Description: PGP signature

Re: the quest for a *stable* browser

[James D Strandboge]

On Sat, 2002-03-09 at 09:24, stan wrote:
> On Fri, Mar 08, 2002 at 11:15:01PM -0500, timothy bauscher wrote:
> > I downloaded Mozilla via apt-get. It crashes
> > often. I would like a stable open source
> > alternative to Mozilla.
> >=20
> I've become a BIG fan of galeon, myselg.
>=20
Me too, and if you are into this sort of thing, galeon handles flash,
realplayer and java well in my experience.

Jamie Strandboge




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

Re: the quest for a *stable* browser

[Eamon Roque]

Am Die, 2002-03-12 um 09.12 schrieb Karsten M. Self:
> on Fri, Mar 08, 2002, timothy bauscher ([EMAIL PROTECTED]) wrote:
> > I downloaded Mozilla via apt-get. It crashes often. I would like a
> > stable open source alternative to Mozilla.
> 
> What version of Debian?  I'd recommend Testing (Woody) for a desktop,
> and Galeon as the browser.  It simply kicks royal ass, then starts to
> *really* rock.
> 
> Browser uptimes of days to weeks.

Well, I must say that you're quite lucky; I'm happy when I can get a few
pages seen before it says goodbye... I haven't found anything really
stable yet.

I'm running an up-to-date Woody.

Cheers.

Eamon Roque.

problem with dialup connection

[Siward de Groot]

  Hello,
 
  I have a problem getting my dial-up internet connection to work.
  Your help would be appreciated.

  I cant get net connection from netscape, when using Debian,
due to 'host unreachable'.
  Running netscape from windows shows that provider is ok.

  The ppp link through my external modem seems ok,
but pings i send to my provider (from xterm+bash)
dont seem to make it back to bash,
although lights on modem seem to indicate that every ping is answered
promptly,
and /var/log/debug says that too.
 
  after executing pon, as user siward, connection is made,
and ifconfig (executed as root) says :
 
 Si:/my# ifconfig
 loLink encap:Local Loopback
   inet addr:127.0.0.1  Mask:255.0.0.0
   inet6 addr: ::1/128 Scope:Host
   UP LOOPBACK RUNNING  MTU:3924  Metric:1
   RX packets:34 errors:0 dropped:0 overruns:0 frame:0
   TX packets:34 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:0
 
 ppp0  Link encap:Point-to-Point Protocol
   inet addr:62.251.55.135  P-t-P:193.172.249.23  Mask:255.255.255.255
   UP POINTOPOINT RUNNING NOARP MULTICAST  MTU:1524  Metric:1
   RX packets:11 errors:0 dropped:0 overruns:0 frame:0
   TX packets:12 errors:0 dropped:0 overruns:0 carrier:0
   collisions:0 txqueuelen:10

 
  route seems to be missing an entry for loopback device,
but i dont know wether it needs that :
 
 Si:/my# route -n
 Kernel IP routing table
 Destination Gateway Genmask Flags Metric RefUse Iface
 193.172.249.23  0.0.0.0 255.255.255.255 UH0  00 ppp0
 0.0.0.0 193.172.249.23  0.0.0.0 UG0  00 ppp0
 
  I can ping localhost, but pinging 193.172.249.23 is not successfull :
 
 [EMAIL PROTECTED]:/my$ ping localhost
 PING localhost (127.0.0.1): 56 data bytes
 64 bytes from 127.0.0.1: icmp_seq=0 ttl=255 time=1.2 ms
 64 bytes from 127.0.0.1: icmp_seq=1 ttl=255 time=0.8 ms
 64 bytes from 127.0.0.1: icmp_seq=2 ttl=255 time=0.8 ms
 
 --- localhost ping statistics ---
 3 packets transmitted, 3 packets received, 0% packet loss
 round-trip min/avg/max = 0.8/0.9/1.2 ms
 
 [EMAIL PROTECTED]:/my$ ping 193.172.249.23
 PING 193.172.249.23 (193.172.249.23): 56 data bytes
 
 --- 193.172.249.23 ping statistics ---
 14 packets transmitted, 0 packets received, 100% packet loss
 
  Although ppp0 seems to be set up correctly, netstat says i dont have a
connection :
 
 Si:/my# netstat -n
 Active Internet connections (w/o servers)
 Proto Recv-Q Send-Q Local Address   Foreign Address State
 Active UNIX domain sockets (w/o servers)
 Proto RefCnt Flags   Type   State I-Node Path
 unix  1  [ ] STREAM CONNECTED 2446   @001b
 unix  1  [ ] STREAM CONNECTED 2667   @002d
 unix  1  [ ] STREAM CONNECTED 2600   @002c
 unix  1  [ ] STREAM CONNECTED 2468   @0022
 unix  1  [ ] STREAM CONNECTED 66 @0001
 unix  1  [ ] STREAM CONNECTED 2497   @0025
 unix  1  [ ] STREAM CONNECTED 2482   @0024
 unix  1  [ ] STREAM CONNECTED 205@0006
 unix  1  [ ] STREAM CONNECTED 2473   @0023
 unix  1  [ ] STREAM CONNECTED 159@0005
 unix  1  [ ] STREAM CONNECTED 2668   /dev/log
 unix  1  [ ] STREAM CONNECTED 2601   /tmp/.X11-unix/X0
 unix  1  [ ] STREAM CONNECTED 2498   /tmp/.X11-unix/X0
 unix  1  [ ] STREAM CONNECTED 2483   /tmp/.X11-unix/X0
 unix  1  [ ] STREAM CONNECTED 2474   /tmp/.X11-unix/X0
 unix  1  [ ] STREAM CONNECTED 2469   /tmp/.X11-unix/X0
 unix  1  [ ] STREAM CONNECTED 2449   /tmp/.X11-unix/X0
 unix  1  [ ] STREAM CONNECTED 206/dev/log
 unix  1  [ ] STREAM CONNECTED 160/dev/log
 unix  1  [ ] STREAM CONNECTED 67 /dev/log

 
  Meanwhile, /var/log/debug has recorded this :
 
 Mar 12 16:56:46 Si pppd[867]: sent [LCP ConfReq id=0x1]
 Mar 12 16:56:46 Si pppd[867]: rcvd [LCP ConfReq id=0x1  < 11 04 05 f4> < 13 05 01 55 54>]
 Mar 12 16:56:46 Si pppd[867]: sent [LCP ConfRej id=0x1 < 11 04 05 f4> < 13 05
01 55 54>]
 Mar 12 16:56:46 Si pppd[867]: rcvd [LCP ConfAck id=0x1]
 Mar 12 16:56:46 Si pppd[867]: rcvd [LCP ConfReq id=0x2 ]
 Mar 12 16:56:46 Si pppd[867]: sent [LCP ConfNak id=0x2 ]
 Mar 12 16:56:46 Si pppd[867]: rcvd [LCP ConfReq id=0x3 ]
 Mar 12 16:56:46 Si pppd[867]: sent [LCP ConfAck id=0x3 ]
 Mar 12 16:56:46 Si pppd[867]: sent [LCP EchoReq id=0x0 magic=0xab9ee4f6]
 Mar 12 16:56:46 Si pppd[867]: sent [PAP AuthReq id=0x1 user="siward.de.groot"
password=]
 Mar 12 16:56:46 Si pppd[867]: rcvd [LCP EchoRep id=0x0 magic=0x0]
 Mar 12 16:

Re: Moving /var to another drive

[Scott Henson]

On Tue, 2002-03-12 at 08:33, George Karaolides wrote:

> And transfer the data using tar:
> 
> tar cplf - -C / var | tar xvf - -C /mnt
I would use rsync for this.  It is faster and just all around
better(IMHO).


> Unmount the partition
> 
> umount /mnt
> 
> Mount the other one and do the same thing for /usr:
> 
> mount /dev/
> tar cplf - -C / usr | tar xvf - -C /mnt
> umount /mnt

-- 
-Peace kid
  Scott Henson  [EMAIL PROTECTED]

"God's the ultimate playa, so naturally He's going to have some haters,"
rapper Ice Cube said. "But these haters need to realize that  if you
mess with the man upstairs, you will get your ass smote. True dat."

Re: CUPS auto-banner

[Angus D Madden]

David Wright, Sun, Mar 10, 2002 at 06:20:34PM -0800: 
> 
> Forgive my density, but the CUPS manuals are singularly impenetratable.
> 
> How to I confiure a CUPS server to automatically print a banner with 
> every job?
> 
> Just to be clear, I don't want my users to have to type
>   -o job-sheets=standard
> each time.
> 
> 

I think this can be done from the web interface
http://localhost:631/printers . Select "Configure Printer" for the
printer in question and there will be a "Banners" section at the bottom
of the page.

The existing banners can be found in /usr/share/cups/banners

g



-- 
Brought to you by Debian 3.0
Linux took 2.4.16 #1 SMP Sat Jan 5 12:52:24 EST 2002 i686 unknown


pgpBYeQqddnNm.pgp
Description: PGP signature

Re: Massive issues with WMaker and X

[Karsten M. Self]

on Fri, Mar 08, 2002, Jonathan Ard ([EMAIL PROTECTED]) wrote:
> 
> Ok, so I am having some major issues with Window Maker.  When I start it
> up, it tells me it can't load the Applications Menu, and that I should
> check the output in the terminal to see what the problem is. 

You've resolved this issue, not addressed.

> Unfortunately, I am running wdm, so I don't have a terminal to look at. 

Wrong:  unless you've changed your /etc/inittab, you're running six text
consoles in virtual terminals 1-6.  You can reach these via the 'chvt'
(change virtual terminal) command (run with root priviliges), or with
the keystroke  (that's control plus alt plus any of
the function keys one through six.

There's a short FAQ on GNU/Linux X display manager disabling you may
find useful, at:

http://kmself.home.netcom.com/Linux/FAQs/xdm-disable.html

Please address your quoting style as well.

Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What part of "Gestalt" don't you understand?   There is no K5 cabal
  http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org



pgpApRpFHVZzY.pgp
Description: PGP signature

Re: kernel-2.4.17

[Sebastiaan]

High,

On Tue, 12 Mar 2002, Gerard Robin wrote:

> hello,
> I have a problem with the kernel 2.4.17:
> My machine: HDD1 in hda
> HDD2 in hdb
>
> The machine boots with lilo on hda.
> I boot with a disket 1.44 Mb for hdb.
>
> After the compile, my machine booted correctly with the
> kernel 2.4.17 on the disket 1.44 Mb, but I forgot to
> include a module in the kernel and I recompiled the
> kernel 2.4.17 and now when I boot the machine with the
> boot disket I get :
>
> NET4:unix domain socket 1.0/SMP for LINUX NET4.0.
> request_module[ide-disk]:Rot fs not mounted
> hdb:driver not present
> VFS:Cannot open root device "343" or 03:43
> Please apend a correct "root=" boot option
> Kernel panic:VFS:Unable to mount root fs on 03:43
>
Ok, I do not really see how your system is configured. On which harddisk
is Linux installed? Why do you use a floppy to boot when you have LILO
installed too?

The problem is that the ide driver is not loaded (that's what the kernel
says). You have to recompile the kernel and test it first.

Greetz,
Sebastiaan

Re: making my own dist. ? ?

[Adam Majer]

On Tue, Mar 12, 2002 at 04:46:42PM +0100, Adam craig wrote:
> Hello,
> 
> I've 3 servers to install - not the same hardware but the software needs to 
> be the same.
> I have installed the first server and got everything working the way I want 
> it to.
> My problem is that I need to use interbase with php4 and I can't get it to 
> work with php4.1.x - the most recent version it will work with is php4.0.6.
> Now using apt-move get;apt-move move, I have a copy on hard drive of all the 
> packages I need to install the other 2 servers.
> My idea was to update the first server direct from the internet, test, then 
> (apt-move get/move) update the 2 other servers from the first server. So how 
> can I disable any updates relating to php4 ?
> Or should I (can I ?) create my own php4.deb and change some config. file 
> (that I haven't yet found) so that my version will always be the most recent 
> for me ?
> I'm sure this must be a fairly common situation - any suggestions would be 
> greatly appreciated.

Go to dselect, and put a hold on all the packages you do not want updated (use 
'=' for that).
Then they will never get updated unless the hold is released :)

- Adam

Re: Copying Partition, Using Reiser.

[Karsten M. Self]

on Sat, Mar 09, 2002, Timothy R. Butler ([EMAIL PROTECTED]) wrote:
> 
> > 1. Does your current kernel support Reiser?  If so, then you
> >can put the destination disk in the current disk and copy.
> 
>   Well in the computer that my system is currently installed, yes, I have 
> Reiser support. In the latter system however, I do not currently have an OS 
> installed, and the net_inst CD does not support Reiser.
> 
> > 2. Is your whole install (including /boot, /var, etc. in _one_
> > partition?  If so, then tar will work and not screw up links.
> 
>   Yup, it's all one big partition.

Not generally recommended.  My own partitioning guidelines:

http://kmself.home.netcom.com/Linux/FAQs/partition.html

As for the reiserfs conversion:

  - Buy, bug, steal, or build a kernel with Reiserfs support.

  - Archive any existing data you wish to retain on the partition(s) you
wish to convert to reiserfs.  I recommend tarring to an alternate
partition, computer, or tape.

  - Repartition, if necessary.

  - Create the reiserfs filesystem.

  - Restore or copy data from archives or other source(s).

  - If you've modified your boot or root partition, update your
bootloader (typically LILO or GRUB).

  - Update your /etc/fstab to reflect current reality (partitions, mount
points, filesystems).

  - Restart.

I recommend using ext2fs or ext3fs for partitions smaller than ~150 MiB 
due to the overhead of the rieser journal data (about 32 MiB, fixed
size).  This would include your /, /boot, and often /tmp partitions,
typically.

Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What part of "Gestalt" don't you understand?   There is no K5 cabal
  http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org



pgpdXTWyDbr2D.pgp
Description: PGP signature

Re: Moving /var to another drive

[Shri Shrikumar]

On Tue, 2002-03-12 at 15:42, François Chenais wrote:
> And what happens if the /var/log and /var/run dirs that can change during
> the tar ?
> 
>   François

Go into single user mode telinit 1 and then tar. I actually use cp -a
which seems to preserve all the required attributes.

I have done this process before.

%telinit 1
%mkdir mnt/var
%mkdir mnt/usr
%mount /dev/ /mnt/var
%mount /dev/ /mnt/usr
%cp -a /usr/* /mnt/var
%cp -a /var/* /mnt/usr
%umount /mnt/var
%umount /mnt/usr
%rm -r /mnt/var
%rm -r /mnt/usr
%umount /var
%umount /usr
%mv  /var /var.old
%mv /usr /usr.old
%mkdir /var
%mkdir /usr
%mkdir /dev/ /var
%mkdir /dev/ /usr

update /etc/fstab to reflect new changes

%telinit 2

to come back to multiuser

Please note that above steps are from memory - no guarantees - it worked
for me atleast three times. make backups blah blah.

Good luck


Shri

> 
> On Tue, 12 Mar 2002 15:33:55 +0200 (EET)
> George Karaolides <[EMAIL PROTECTED]> wrote:
> 
> > 
> > Hi Andrew,
> > 
> > You need to be root on your machine to do this.
> > 
> > Make two partitions on the new drive using
> > 
> > cfdisk /dev/
> > 
> > Then make a filesystem on each partition.  Stick to the tried-and-tested
> > ext2, or go for one of the new journalling ones like ext3 or reiserfs, if
> > you're running a kernel recent enough to support them or can compile one
> > that does.
> > 
> > E.g. mke2fs /dev/
> > 
> > Then mount the first partition e.g. under /mnt
> > 
> > mount /dev/ /mnt
> > 
> > And transfer the data using tar:
> > 
> > tar cplf - -C / var | tar xvf - -C /mnt
> > 
> > Unmount the partition
> > 
> > umount /mnt
> > 
> > Mount the other one and do the same thing for /usr:
> > 
> > mount /dev/
> > tar cplf - -C / usr | tar xvf - -C /mnt
> > umount /mnt
> > 
> > Make sure you type the tar commands exactly as above, you don't want to go
> > dropping the entire copnmtents of /var or /usr (or even / if you v=get it
> > wrong) in the wrong place!
> > 
> > Edit /etc/fstab and either add lines for /usr and /var, or change existing
> > ones, to mount your new filesystems instead of the old ones.  Here's an
> > example:
> > 
> > --- fstab excerpt ---
> > 
> > /dev/sdb6 /usr ext2 rw  0   2
> > /dev/sdb7 /var ext2 rw  0   2
> > 
> > --- end fstab excerpt ---
> > 
> > Then go to single user mode:
> > 
> > telinit S
> > 
> > Unmount /usr and /var:
> > 
> > umount /usr
> > umount /var
> > 
> > Mount all partitions using your new fstab:
> > 
> > mount -a
> > 
> > Check that everything is mounted where it should be:
> > 
> > mount
> > 
> > And go back to your usual runlevel (usually 2 on Debian):
> > 
> > telinit 2
> > 
> > That's it.  Note that if /usr and /var were originally part of your root
> > filesystem, the data will still be there but the new filesystem will be
> > mounted on the top level directory so you won't see it.  After you've
> > successfully transferred /usr and /var out of the root filesystem, you can
> > reclaim the space by going to single user mode, unmounting /usr and /var,
> > and doing the following:
> > 
> > rm -rvf /var
> > rm -rvf /usr
> > mkdir /var
> > mkdir /usr
> > 
> > Be VERY careful that /usr and /var are NOT mounted when you do this!
> > Also note, a space or a slash in the wrong place in either of the above
> > commands can wreck your system completely!
> > 
> > Good luck,
> > 
> > |   George Karaolides  Linustech Advanced Solutions,  |
> > |   tel:   +357 22 55 61 29  86 Ifigenias Street, 3rd Floor,  |
> > |   web:   www.linustech.com.cy  Strovolos, Nicosia CY 2003,  |
> > |   email: [EMAIL PROTECTED]   Republic  of Cyprus.  |
> > 
> > On Tue, 12 Mar 2002, Andrew Stephen wrote:
> > 
> > > Hi
> > >
> > > My /var and /usr partitions have just run out of space and I was wondering
> > > what is the best way to copy them to a new drive that has just been
> > > installed.
> > >
> > > Any suggestions would be greatly appreciated.
> > >
> > > Regards,
> > > Andrew
> > >
> > >
> > >
> > > --
> > > To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> > > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> > >
> > >
> > >
> > 
> > 
> > -- 
> > To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> > 
> > 
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 

(no subject)

[Peggysfamily]

Re: Wine

[csj]

On Tue, 12 Mar 2002 14:30:03 +1030
Tom Cook <[EMAIL PROTECTED]> wrote:

> Tom Cook wrote:
> > 
> > Hi,
> > 
> > Does anyone know how to make windoze programs I run under Wine stick to
> > one desktop?  I can figure it out, and it doesn't seem to be
> > WM-specific.  I am using IceWM.

Maybe this can help:

apt-get install winesetuptk

> Ooops!  I _can't_ figure it out!  Which is why I'm asking the
> question...

Re: making my own dist. ? ?

[Daniel Farnsworth Teichert]

And I heard Adam craig exclaim:
> My idea was to update the first server direct from the internet,
> test, then (apt-move get/move) update the 2 other servers from
> the first server. So how can I disable any updates relating to php4 ?

There is a way to put a hold on a package (which someone else will
probably tell you, I'm not really sure what it is). But we do a similar
thing to what you're describing here, and what we do is use dpkg with
the -R option (see the man page for dpkg for more on this) to install
all of the .deb files in a directory.  This won't go out to the network
or wherever and get other stuff--like more recent versions of files
(although if some of the .deb's you're trying to install are missing
dependencies, it might choke on them).

Anyway, HTH.

  --Daniel F. T.

Strange messages on all terminals

[Scott Henson]

I am having a problem with all my virtual terminals(ctr-alt-F<1-6>)  For
some reason a constant stream of information is sent to all of them.  It
seems to be something to do with my internet connection, because its
specifying MAC addresses and a couple of other things which makes me
think its displaying all incomming packets to the terminals.  Now this
is not affecting X or any of the X-terms, but its sure a bother while
working in a virtual terminal.  I am running woody and a 2.4.19 ac
pre-patch.  Anyone have an idea why this is happening.  Its been going
on for a while now and I tried re-compiling the kernel taking out all
verbose debug messages having to do with the net connection.  I am also
running the new isc dhcp client 3.0, but this was happening before with
the old 2.2 version.  Im not sure what else might be causing this.  This
is especially embaracing because Im trying to get a bunch of my friends
to switch from windows, and having all that trash there and not knowing
what is causeing it is not a major selling point.  Anyone have an idea? 
Thanks.
-- 
-Peace kid
  Scott Henson  [EMAIL PROTECTED]

"God's the ultimate playa, so naturally He's going to have some haters,"
rapper Ice Cube said. "But these haters need to realize that  if you
mess with the man upstairs, you will get your ass smote. True dat."

Re: DVD players under Linux (was: Re: 2.4 kernels with potato and Movie players)

[Jim Gettys]

> From: christophe =?iso-8859-15?Q?barb=E9?= <[EMAIL PROTECTED]>
> Resent-From: debian-user@lists.debian.org
> Date: Mon, 11 Mar 2002 16:10:29 -0500
> To: debian-user@lists.debian.org
> Subject: Re: DVD players under Linux (was: Re: 2.4 kernels with potato and
> Movie players)
> -
> On Mon, Mar 11, 2002 at 08:33:08AM -0800, Jim Gettys wrote:
> > I'm a happy camper with current Xine, on a display running the Xvideo
> > extension, on a 600mhz pentium III.  Pretty much doesn't drop frames.
> >
> > The other tip is to kill esd if it is running.  That lame audio server
> > is hopeless on being able to keep audio and video synced up.
> 
> It is a bit strong to kill esd.
> I would suggest 'man esdctl'
> 
> Christophe
> 
>

Which gets me:

"No manual entry for esdctl"

--
Jim Gettys
Cambridge Research Laboratory
Compaq Computer Corporation
[EMAIL PROTECTED]

Re: DVD players under Linux (was: Re: 2.4 kernels with potato and Movie players)

[christophe barbé]

On Tue, Mar 12, 2002 at 09:32:58AM -0800, Jim Gettys wrote:
> 
> > From: christophe =?iso-8859-15?Q?barb=E9?= <[EMAIL PROTECTED]>
> > Resent-From: debian-user@lists.debian.org
> > Date: Mon, 11 Mar 2002 16:10:29 -0500
> > To: debian-user@lists.debian.org
> > Subject: Re: DVD players under Linux (was: Re: 2.4 kernels with potato and
> > Movie players)
> > -
> > On Mon, Mar 11, 2002 at 08:33:08AM -0800, Jim Gettys wrote:
> > > I'm a happy camper with current Xine, on a display running the Xvideo
> > > extension, on a 600mhz pentium III.  Pretty much doesn't drop frames.
> > >
> > > The other tip is to kill esd if it is running.  That lame audio server
> > > is hopeless on being able to keep audio and video synced up.
> > 
> > It is a bit strong to kill esd.
> > I would suggest 'man esdctl'
> > 
> > Christophe
> > 
> >
> 
> Which gets me:
> 
> "No manual entry for esdctl"

Which is in the esound-clients package.

Christophe

> 
> --
> Jim Gettys
> Cambridge Research Laboratory
> Compaq Computer Corporation
> [EMAIL PROTECTED]
> 

-- 
Christophe Barbé <[EMAIL PROTECTED]>
GnuPG FingerPrint: E0F6 FADF 2A5C F072 6AF8  F67A 8F45 2F1E D72C B41E

Ce que l'on conçoit bien s'énonce clairement,
Et les mots pour le dire arrivent aisément.
   Nicolas Boileau, L'Art poétique


pgp7gkHVWyhqM.pgp
Description: PGP signature

in what config file should I place 'setserial /dev/ttyS3 IRQ 5", and pon ?

[John Kennedy]

Good morning,

I have a 56k dialup/firewall box using Debian 2.2.17

I configured the dialup connection using 'ppp-config'
and selected 'demand' dial
however after reboot doing 'netstat -nr' returns only

"192.168.1.10   0.0.0.0   255.255.255.0   eth0"

When I ping an address out on the net it won't dial out.

If I type 'pon' and then 'netstat -nr' i get
a proper routing table as below

"10.112.112.112  0.0.0.0   UH   ppp0"
192.168.1.0 0.0.0.0   Ueth0
0.0.0.010.112.112.112 UG   ppp0

and I can now ping out of my lan to an ip on the net.

Because I an using IRQ-5 on my modem I must also do a
'setserial /dev/ttyS3 IRQ 5" before dialing.

I am curious as to why the modem does not dial out
after setting 'demand' in 'ppp-config' ?
I set 'defaultroute' in ppp-config.

My most urgent question here is
in what config file should I enter the commands to
1. do 'setserial /dev/ttyS3 IRQ 5"  after boot.
2. do 'pon' after boot.

I'm a former Redhat user...
With Redhat I would put it in /etc/rc.d/rc.local
but I noticed that my Potato install has no
file /etc/rc.d/rc.local...
so what file in Debian 2.2.17 performs rc.locals functions?

Also one other question,
when doing tail -f /var/log/messages  I notice that
when I'm dialed up but the connection is just sitting there
dooing nothing for a while I get a bunch of
'---MARK---'  messages and I'm wondering what they mean.

Thanks for any help you can give me.

John








_
Chat with friends online, try MSN Messenger: http://messenger.msn.com

Re: apt-get source vs. binary

[csj]

On 11 Mar 2002 23:46:11 -0500
James Leigh <[EMAIL PROTECTED]> wrote:

> When I install via apt-get source and compile the package, then
> install.  apt-get -u upgrade wants to replace my compiled version with
> the binary package from debian servers, of the same version.  Ever time
> I install a package from apt-get source, apt-get wants to replace it
> with the binary version.  I compiled nautilus (1.0.6-3), then
> $ apt-get -s install nautilus
> Reading Package Lists... Done
> Building Dependency Tree... Done
> 1 packages upgraded, 0 newly installed, 0 to remove and 0  not upgraded.
> Inst nautilus (1.0.6-3 Debian:unstable)
> Conf nautilus (1.0.6-3 Debian:unstable)
> 
> It should says that nautilus is the latest version, but it thinks that
> my optimize nautilus should be replaced with the i386 package.  I have
> been optimizing the compiled packages for my athlon using
> pentium-builder.

Create your own debian archive. Put that in front of your http:// or
ftp:// sources, something like:

deb file:/MyArchive/ ./
deb http://non-us.debian.org/debian-non-US stable/non-US main

Re: Acroread not displaying menu text 2nd try

[Karsten M. Self]

on Tue, Mar 12, 2002, [EMAIL PROTECTED] ([EMAIL PROTECTED]) wrote:
> Pat
> 
> Thanks for the e-mail. Unfortunately yours was the only responses I
> received. In fact I am using this reply as a rather lame excuse to
> post this question a second time.
> 
> Needless to say I would be grateful for any pointer to a solution for
> the acroread/Acrobat problem described below.

Please use postfix quoting format:  your reply goes below the material
cited.  Trim your quotes appropriately and ensure your attributions are
accurate.


First:  I'd strongly recommend xdpf or gv as alternatives to the
non-free Acroread produced by a company which pursues strongly
objectionable practices of persecuting software developers for being
smart and innovative.

Second:  I suspect a font/locales issue.  You're likely seeing Adobe's
broken handling of double-byte charactersets.  This is conjecture (I
don't use Adobe products if at all possible), but somewhat matches my
own experience with other software this past fall, particularly the
locales given.  The issue was discussed at length on list.  There is no
single workaround, though there are numerous tacks to try.  My own later
posts tried to summarize some of the options, see archives.

> On 11 March 2002 12:57:04 Patrick Colbeck <[EMAIL PROTECTED]> wrote
> >Hi
> >
> >Did you get a reply to this question ? I am having exactly the same
> >issue and guess what I am in th UK as well. If you got a solution I
> >would love to know what it is.

<...>

> >On Sunday 10 March 2002 11:42 pm, [EMAIL PROTECTED] wrote:

> >> I am experiencing a problem with acroread (Acobat reader)v.4.05-5 on
> >> what I believe is a typical Debian GNU/Linux 3 (Woody/Testing) system
> >> running on a Compaq Professional Workstation 5000 (P-Pro SMP).
> >>
> >> Acroread does not display any menu text at all (just gray line boxes
> >> where the text should be) and when it is started from an xterm the
> >> following error message is displayed:
> >>
> >> Warning: charset of fontList (ISO10646-1) does not match locale
> >> (ISO8859-1).
> >>
> >> I suspect this has come about because recently I had to change the
> >> user and root locales to ISO8859-1 (using dpkg-reconfigure locales)to
> >> enable recent releases of multi-gnome-terminal (currently v1.3.13) to
> >> run on my system.
> >>
> >> Reluctantly (because ISO8859-1 seems to be ideal for a UK based Debian
> >> box) I tried to set both the user and root locales to ISO10646-1. But
> >> this option is not available using dpkg-reconfigure locales. My system
> >> does have the following relevant looking file:
> >>
> >> /usr/share/i18n/charmaps/ISO_10646.gz
> >>
> >> But so far I have been unable to discover if it can be used to change
> >> the locales.

Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What part of "Gestalt" don't you understand?   There is no K5 cabal
  http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org



pgpqQwAC9LM2a.pgp
Description: PGP signature

Re: new twist on shutting down and restricting ssh users

[Karsten M. Self]

on Wed, Mar 06, 2002, Simon Hepburn ([EMAIL PROTECTED]) wrote:
> On Wednesday 06 Mar 2002 9:11 pm, dman wrote:
> 
> > I want to allow my dad to shut down the router/gateway. 
> 
> Install ext3fs. Show him the power switch  ;-)  

Journaled filesystems exist to help recover from _filesystem_ errors in
the event of an unexpected shutdown.  You're still liable to encounter
_other_ forms of inconsistent data, and journaling ***DOES NOT***
address this issue.

Hard shutdowns are _not_ recommended as a matter of course.  You're
likely to run into trouble sooner or later.

By way of analogy, you can stop your car by running it up against
something.  Hitting the wheelstop in a parking lot at 1.5 MPH is rather
different from striking a tree at 65 MPH.  Your car is stopped in both
cases.  The condition of the contents may vary.

Peace.

-- 
Karsten M. Self http://kmself.home.netcom.com/
 What part of "Gestalt" don't you understand?   There is no K5 cabal
  http://gestalt-system.sourceforge.net/ http://www.kuro5hin.org



pgpT2FIt1LmRK.pgp
Description: PGP signature

Re: lite applications

[Craig]

On Mon, Mar 11, 2002 at 11:42:44PM -0800, Karsten M. Self wrote:
> on Sun, Mar 10, 2002, Joe ([EMAIL PROTECTED]) wrote:
> > On Sat, Mar 09, 2002 at 11:52:43PM -0800, Karsten M. Self wrote:
> > > on Fri, Mar 08, 2002, Nicholas Imfeld ([EMAIL PROTECTED]) wrote:
> > > > I have set up an old 486 laptop with debian.  I am looking for an x
> > > > email client (preferably graphical as opposed to test based) and a web
> > > > browser that don't require a lot of space, memory, etc.  
> > > 
> > > I'd ***STRONGLY*** encourage you to stick with the text alternatives,
> > > unless you're using the 486 _just_ as an X11 terminal, you might be able
> > > to run more advanced apps on it.  Otherwise, I'd stick with mutt, w3m,
> > > lynx, and similar text-mode apps.
> > > 
> > > If X is an absolute requirement, look to BrowseX as a browser.  It's
> > > Tk/Tcl based and pretty light.  Nothing else I'm aware of comes close.
> > > Dillo's small of itself, but the gtk requirements are large.
> > 
> > If you install KDE 1.x 
> 
> This is simply not an option on a 486.
> 
hmm. maybe not. I had it once on a 486 box with 24 Mb RAM, but the CPU
was upgraded to a 120 Mhz Pentium.

CraigW

OT: interface card to register button presses

[Sebastiaan]

Hi,

does anyone know a good, simple  interface card which gives you the
ability to connect a panel with about 100 buttons to your computer and
registrates which butten is pressed?

Of course it should work with Linux and hopefully an easy to program one.

I was asked to make a simple voting simulation for the upcoming elections
here. Since this is the first time all votings are electronical, some
people have trouble voting (mainly elderly people). Another issue is to
train illiterates (illiterate in Latin script) in voting for the upcoming
elections.


Thanks in advance for any hints or ideas,
Sebastiaan



--
  NT is the OS of the future. The main engine is the 16-bit Subsystem
  (also called MS-DOS Subsystem). Above that, there is the windoze 95/98
  16-bit Subsystem. Anyone can see that 16+16=32, so windoze NT is a
  *real* 32-bit system.

Re: truncate utility (not split)

[csj]

On 11 Mar 2002 13:25:06 -0800
Caleb Shay <[EMAIL PROTECTED]> wrote:

> Seems to me that you should be able to do this with dd.

You're absolutely right. Reminds me of someone's Linux motto: You want
it, you have it.

> On Mon, 2002-03-11 at 11:59, csj wrote:
> > Is there a utility to truncate files such that a single smaller piece or
> > no more than two pieces are produced? Something like:
> > 
> > (1) original_file -> small_chunk (bigger_2nd_chunk discarded)
> > (2) original_file -> (smaller_1st_chunk discarded) big_chunk
> > (3) original_file -> small_chunk + big_chunk
> > 
> > split would be suboptimal, especially in scenarios (2) and (3)::
> > 
> > original_file -> small_chunk_1 + small_chunk_2 + small_chunk_n
> > delete small_chunk_2 small_chunk_2
> > concatenate small_chunk_*
> > 
> > Hope you understand what I means. TIA

Re: Moving /var to another drive

[Dave Sherohman]

On Tue, Mar 12, 2002 at 04:42:37PM +0100, François Chenais wrote:
> And what happens if the /var/log and /var/run dirs that can change during
> the tar ?

That's my main disagreement with George's instructions.  Go to
single-user mode _before_ you copy /var and /usr to the new drive
instead of after copying them.  This will shut down all the various
daemons that might be modifying data while you work.

(The lesser disagreement is that I would use cpio instead of tar, but
that's really just a matter of taste.)

So, how I do it:

1.  Shut down machine[1]
2.  Install new drive
3.  Boot to single-user mode
4.  cfdisk /dev/hdnew
5.  mount /dev/hdnewvar /mnt
6.  cd /var
7.  find . -xdev -print0 | cpio -pvdm0 /mnt
8.  umount /mnt
9.  Repeat steps 5-8 for /usr
10. Update /etc/fstab
11. Reboot

or

11. umount /var && umount /usr && mount -a && init 2

[1]  I'm assuming that the box is not hot-swap capable, so you have
to power down to install the new drive.  If it is hot-swappable, just
plug the drive in and use `init 1` to get to single-user mode.

-- 
When we reduce our own liberties to stop terrorism, the terrorists
have already won. - reverius

Innocence is no protection when governments go bad. - Tom Swiss

Re: Moving /var to another drive

[Dave Sherohman]

On Tue, Mar 12, 2002 at 03:33:55PM +0200, George Karaolides wrote:
> That's it.  Note that if /usr and /var were originally part of your root
> filesystem, the data will still be there but the new filesystem will be
> mounted on the top level directory so you won't see it.  After you've
> successfully transferred /usr and /var out of the root filesystem, you can
> reclaim the space by going to single user mode, unmounting /usr and /var,
> and doing the following:
> 
> rm -rvf /var
> rm -rvf /usr
> mkdir /var
> mkdir /usr
> 
> Be VERY careful that /usr and /var are NOT mounted when you do this!
> Also note, a space or a slash in the wrong place in either of the above
> commands can wreck your system completely!

Good warnings.  I would also recommend doing a `ls -ld /usr /var >
/tmp/perms` before deleting the old directories, then double-checking
that the ownership and permissions on them are correct after creating
and mounting the new ones.

-- 
When we reduce our own liberties to stop terrorism, the terrorists
have already won. - reverius

Innocence is no protection when governments go bad. - Tom Swiss

pppoe and kernel 2.4.x

[Frodo Baggins]

Hi debianers,
  well, I have a debian system connected to a ISP using
pppoe. Everithing works well with the 2.2.x kernels, but for different
reasons, I need to install a 2.4.x kernel. 

  Reading the file /usr/doc/pppoe/KERNEL-MODE-PPPOE I understand that
I must recompile ppp from the tar-ball at
http://www.shoshin.uwaterloo.ca/~mostrows/ which is a patched ppp,
install it and then compiling rp-pppoe enabling plugins.

  This look strightforward but will obviously clash with the package
structure of my machine. Just for one, intrusion-detection tools
(e.g. tiger) will complain about unmatching cecksums. For another,
updating will become painful. 

  Is there another, more debian conformant, way to do it? 

  Thanks a lot
-- 
Leo TheHobbit 
IRCnet #roma2
ICQ 56656060

-BEGIN GEEK CODE BLOCK-
Version: 3.12
GED/CS d? s-:++ a+ C++ UL+++ P+++>+ E+ W++ N+ o K? !w O? M V PS+++ 
PE-- Y+ PPG+ t++ 5? X-- R+ tv+ b+++ DI? D G++ e*() h r++ y+++(*)
--END GEEK CODE BLOCK--
++