Re: Fwd: Header files amd64 etch

[Marko Randjelovic]

Francesco Pietra wrote:
> Probably I have found the way with
> 
> apt-cache search (show) xorg-dev
> 
> V 1:7.1.0-16
> 
> fp
> 

I don't think so. Try this:


[EMAIL PROTECTED]:~$ apt-file search X11/IntrinsicP.h
libxt-dev: usr/include/X11/IntrinsicP.h
[EMAIL PROTECTED]:~$ apt-file search X11/ShellP.h
libxt-dev: usr/include/X11/ShellP.h

So, you need package libxt-dev.


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

Re: Header files amd64 etch

[Kushal Kumaran]

On Fri, Jun 29, 2007 at 11:00:52PM -0700, Francesco Pietra wrote:
> A mixed compilation (ifort/gcc 4.1.2) of Amber9 on amd64 etch
> dual-core-opterons (Linux deb32 2.6.18-3-k7 #1 SMP Mon Dec 4 17:23:11 UTC 2006
> i686 GNU/Linux)
> requests the following header files;
> 
> WcActCB.c:22:28: error: X11/IntrinsicP.h: No such file or directory
> WcActCB.c:29:24: error: X11/ShellP.h: No such file or directory
> 
> 
> as the single errors out of an enormous number of modules that passed "make".
> 
> As a beginner in compilations, I would appreciate being instructed how to find
> the appropriate (64bit if they differ from 32bit) headers. I have seen 
> packages
> that provide all headers, though above etch, with a more recent kernel. In
> other words, I don't know the appropriate console command to trace the package
> to install. And, for the moment I refrain from dist upgrade.
> 

You need the apt-file package.  Install apt-file:
aptitude install apt-file

Update its lists:
apt-file update

Search away:
apt-file search X11/ShellP.h

libxt-dev turns up.

-- 
Kushal Kumaran   [EMAIL PROTECTED]


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

Re: Fwd: Header files amd64 etch

[Francesco Pietra]

Marko:
Thanks. The installation of xorg-dev includes libxt-dev.

Though, "make" failed again. I have been instructed to first carry out a serial
installation, and parallel only after serial is OK. If not else because
parallel does not compile all modules.

However, the configure command line 

 ./configure -static ifort_x86_64

implies that gcc is taken from the system. If I understand, gcc is not prepared
to furnish static libraries (not checked). Why things are not set up for a
homogeneous compilation (intel provides *.a libraries) is not clear to me. The
parallel compilation will be homogenous, with intels for fortran, C and C++ for
bothe amber and openmpi.

Cheers

francesco


--- Marko Randjelovic <[EMAIL PROTECTED]> wrote:

> Francesco Pietra wrote:
> > Probably I have found the way with
> > 
> > apt-cache search (show) xorg-dev
> > 
> > V 1:7.1.0-16
> > 
> > fp
> > 
> 
> I don't think so. Try this:
> 
> 
> [EMAIL PROTECTED]:~$ apt-file search X11/IntrinsicP.h
> libxt-dev: usr/include/X11/IntrinsicP.h
> [EMAIL PROTECTED]:~$ apt-file search X11/ShellP.h
> libxt-dev: usr/include/X11/ShellP.h
> 
> So, you need package libxt-dev.
> 
> 
> -- 
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 
> 



 

8:00? 8:25? 8:40? Find a flick in no time 
with the Yahoo! Search movie showtime shortcut.
http://tools.search.yahoo.com/shortcuts/#news


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

Re: Updating without Apt

[William Pursell]

Andrew Gray wrote:

My name is Andrew, and I use Debian Lenny on an x86 machine.

I love Debian. I love apt. The Debian/Apt combination has allowed me to 
use software I would have never before dreamed to be possible to use. 
I'm quite content with my system at the moment. My problem comes when a 
piece of software that I use with frequency is updated by the developers 
in a major way. The Debian repositories do not reflect this change, and 
likely won't for some time.


The software in question is emacs22. 


I would be surprised if a major update to emacs weren't in
sid within a few days, if not hours.


Currently I have it compiled in my 
home directory at ~/source/emacs/, where I run it directly out of the 
src directory. I dare not "make install", or my apt install will be 
permanently harmed and/or it won't work and I will be faced with a 
crippled, non-running version of emacs on my system. Not only that, but 
with this compilation I was able to enable GTK for the X11 run of emacs. 
I loved having that kind of flexibility with my compilation.


Is there anything I can do to ensure that if I did a make install it 
wouldn't affect my current install or something like that?


As was suggested else-thread, you can probably do:
configure --prefix=/usr/local && make && make install.
(I'm not an emacs user, so the build may be more complicated
than that, but this will very likely work.)  In fact, /usr/local
is the defaul prefix, so you don't even need that.  In other
words, you can make install, since it will install to /usr/local/bin,
rather than /usr/bin and not interfere with the version installed
via apt at all.  If you want to test it more carefully, try

make DESTDIR=/tmp/foo install

and examine the resulting tree in /tmp/foo.  You should only get
executables in /usr/local/bin.

However, you might want to look into apt pinning.  This allows
you to pull emacs from sid without requiring you to update your
entire system to sid, and will give you the updates more
quickly.  Or, you can simply download the .deb files from
sid and install them with dpkg.  If the upstream is truly
not showing up in sid for too long, that won't work.  If
that is the case, you can grab the .dsc from sid and make
the modifications yourself to create a newer deb.


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

running iceweasel on remote host seems wrong

[Miernik]

Strange problem. I have two debian machines, one local at home, and one
remote somewhere. At home I am connected behind a NAT, and the remote
machine has a public IP.

I want to run iceweasel on the remote machine, and let it display on my
local machine. I have iceweasels installed on both machines.

So from the local machine I do:

ssh -Cv [EMAIL PROTECTED] -X

After a while I get a prompt of the remote machine, looks OK, ls gives
me listing of files on that machine.

env shows DISPLAY=localhost:10.0

So I run 'iceweasel' there. And after some time I get it displayed,
but... it seems that it runs it somehow on my local machine, for example
file:///
displays the root directory of my local machine, not the remote one, and
when I browse, with tcpdump I see HTTP traffic, and not SSH traffic as I
expected.

I don't understand it. How I am supposed to run iceweasel on the remote
machine (using only the local display), so it would work as it would be
sitting at the remote machine?

-- 
Miernik
http://miernik.name/


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

Re: No Sound After Playing XMMS

[Chris Lale]

TW wrote:
> Karl E. Jorgensen wrote:
>> On Thu, Jun 28, 2007 at 11:51:29AM -0600, TW wrote:
>>  
>>> Hi,
>>>
>>>After playing music with the XMMS application, I seem to be unable
>>> to hear sound after closing the XMMS app and going to watch videos at
>>> Google Video (or anywhere else on the web).  I've looked over the net
>>> and couldn't find the answer to this.  The sound is fine, but, after
>>> listening to XMMS audio, I have to log out (and back in) in order to
>>> hear sound over the internet.  The reverse isn't a problem, though,
>>> meaning that I can listen to Google Video and then play XMMS without
>>> problems.  What am I missing?  Thanks in advance.  ~Telly
>>> 
>>
>> esd getting in the way?
>>
>> How is your XMMS configured? ALSA? Esd?
>>
>>   
> It was configured with OSS.  ~Telly
> 
> 

XMMS should still work with Alsa in Etch even when it is configured for OSS (the
default). Other apps may be affected by the underlying problem eg Audacity.

A similar problem was resolved in the recent thread "Some ALSA apps stopped
working" [1].

I suggest that you check that /dev/dsp exists:

# file /dev/dsp
/dev/dsp: ERROR: cannot open `/dev/dsp' (No such file or directory)

If you get an error, re-create it and check again:

# modprobe snd_pcm_oss

# file /dev/dsp
/dev/dsp: character special (14/3)

Check that XMMS now works even when configured fo OSS.

[1] http://lists.debian.org/debian-user/2007/06/msg02712.html

-- 
Chris.


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

Apache: The requested URL / was not found on this server.

[Michael Pobega]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Thursday afternoon I reinstalled my i386 Debian installation as amd64
(To compliment my Core 2 Duo), and everything's been working pretty good
so far (Minus 3d rendering, but that's another story), but I'm having
one serious problem in Apache.

I reinstalled Apache2 and libapache2-mod-php5, and restored my old
/etc/apache2 configuration files, but I'm getting the error

> The requested URL / was not found on this server.

If you need any of my configuration files just ask, the only one that I
figure would effect anything would be
/etc/apache2/sites-available/default, so here is my config:


/etc/apache2/sites-available/default
> NameVirtualHost *
> 
> ServerAdmin [EMAIL PROTECTED]
> 
>   DocumentRoot /var/www/
>   
>   Options FollowSymLinks
>   AllowOverride
>   None
>   
>   
>   Options Indexes FollowSymLinks MultiViews
>   AllowOverride None
>   Order
>   allow,deny
>   allow from all
>   
>   [...] (There's more here, but it's all the default, I
>   only changed the above)


- -- 
http://digital-haze.net/~pobega/ - My Website
If programmers deserve to be rewarded for creating innovative
programs, by the same token they deserve to be punished if they
restrict the use of these programs. 
 - Richard Stallman
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFGhk9Og6qL2BGnx4QRAlARAJ95yWrPNL1CCPXOf+yDjfHhyuvrlgCfe1AE
ZMHfFVHOBk3u+LPI1tfVSDs=
=Vn1o
-END PGP SIGNATURE-


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

Re: running iceweasel on remote host seems wrong

[Liam O'Toole]

On Sat, 30 Jun 2007 13:15:49 +0200
Miernik <[EMAIL PROTECTED]> wrote:

> Strange problem. I have two debian machines, one local at home, and
> one remote somewhere. At home I am connected behind a NAT, and the
> remote machine has a public IP.
> 
> I want to run iceweasel on the remote machine, and let it display on
> my local machine. I have iceweasels installed on both machines.
> 
> So from the local machine I do:
> 
> ssh -Cv [EMAIL PROTECTED] -X
> 
> After a while I get a prompt of the remote machine, looks OK, ls gives
> me listing of files on that machine.
> 
> env shows DISPLAY=localhost:10.0
> 
> So I run 'iceweasel' there. And after some time I get it displayed,
> but... it seems that it runs it somehow on my local machine, for
> example file:///
> displays the root directory of my local machine, not the remote one,
> and when I browse, with tcpdump I see HTTP traffic, and not SSH
> traffic as I expected.
> 
> I don't understand it. How I am supposed to run iceweasel on the
> remote machine (using only the local display), so it would work as it
> would be sitting at the remote machine?
> 

Is iceweasel already running on the local machine when you start
iceweasel on the remote machine? If so, then the remote instance opens
a new window in the local instance, then exits. You can prevent that
behaviour by closing the local instance first, or by launching the
remote instance with the '-no-remote' option.

-- 

Liam


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

Re: How to move the master boot record?

[Rodolfo Medina]

Rodolfo Medina <[EMAIL PROTECTED]> wrote:

>> On my PC, besides the swap partition, I have one partition, hda1, for MS
>> Windows and another five for Linux: hda6, had7, hda8, hda9, hda10.
>>
>> At the moment the `boot partition' is hda6 and I want it to be, say, hda9.
>> Sorry if I can'y use the right words.  Maybe I should say that the hda6
>> Grub boot loader is now installed to the master boot record of my hard
>> drive whereas the hda9 boot loader is installed to the /dev/hda9 partition?



Joe Hart <[EMAIL PROTECTED]> writes:

> There are more than one way to do this, but this will (should) work:
>
> #grub
> grub>root (hd0,8)
> grub>setup (hd0)
> grub>quit
>
> hd0,8 = /dev/hda9, so you should be alright with those command.  Note that
> the # and the grub> are the prompts.



I tested it and it seems to work fine.  Only, I can't now restore the previous
situation: I do:

 # grub
 grub > root (hd0,5)
 grub > setup (hd0)
 grub > quit

, then reboot but the system can't get into hda6.  It's strange, I find no
reason why it works with hda9 and not with hda6.  Please, any ideas?

Thanks,
Rodolfo


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

Re: How to move the master boot record?

[Joe Hart]

On Saturday 30 June 2007 15:24:08 Rodolfo Medina wrote:
> Rodolfo Medina <[EMAIL PROTECTED]> wrote:
> >> On my PC, besides the swap partition, I have one partition, hda1, for MS
> >> Windows and another five for Linux: hda6, had7, hda8, hda9, hda10.
> >>
> >> At the moment the `boot partition' is hda6 and I want it to be, say,
> >> hda9. Sorry if I can'y use the right words.  Maybe I should say that the
> >> hda6 Grub boot loader is now installed to the master boot record of my
> >> hard drive whereas the hda9 boot loader is installed to the /dev/hda9
> >> partition?
>
> Joe Hart <[EMAIL PROTECTED]> writes:
> > There are more than one way to do this, but this will (should) work:
> >
> > #grub
> > grub>root (hd0,8)
> > grub>setup (hd0)
> > grub>quit
> >
> > hd0,8 = /dev/hda9, so you should be alright with those command.  Note
> > that the # and the grub> are the prompts.
>
> I tested it and it seems to work fine.  Only, I can't now restore the
> previous situation: I do:
>
>  # grub
>  grub > root (hd0,5)
>  grub > setup (hd0)
>  grub > quit
>
> , then reboot but the system can't get into hda6.  It's strange, I find no
> reason why it works with hda9 and not with hda6.  Please, any ideas?
>
> Thanks,
> Rodolfo

No idea.  If it works for /dev/hda9. then (providing the right files are on 
the partition) then /dev/hda6 should work fine, and you're correct in calling 
it (hd0,5).

The only thing I can think of is that the files that grub needs are missing, 
but to be honest, it really doesn't matter where grub is sitting, as long as 
you can tell it where to find all the different OS's on your machine, then 
you're good to go.

So, in short, I don't quite understand why you're trying to do all this in the 
first place.  As already mentioned, installing Etch will automatically find 
the other systems and put them in the grub menu.

Joe


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

libktable-dev and tktable-dev

[Michael Fothergill]

Dear Debianists,

What is the difference between libtktable-dev in Ubuntu and tktable-dev in 
Debian?  Are they pretty much the same package but with a name change or is 
their something more subtle going on here?


I can't seem to find libtktable-dev in the Debian package world.

Regards

Michael Fothergill

_
The next generation of Hotmail is here!  http://www.newhotmail.co.uk


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

Re: libktable-dev and tktable-dev

[Shams Fantar]

Michael Fothergill wrote:

Dear Debianists,

What is the difference between libtktable-dev in Ubuntu and 
tktable-dev in Debian?  Are they pretty much the same package but with 
a name change or is their something more subtle going on here?


I can't seem to find libtktable-dev in the Debian package world.

Regards

Michael Fothergill

__


You can use aptitude show  to have more information for a package.

So, you will have the difference between theses packages.
//

--
Shams Fantar (http://snurf.info)


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

USB Stick mount

[Benjamí Villoslada]

Lately I receive this error (in KDE, Konqueror, Debian Sid) when I 
select «open in new window» after USB stick plugging:

«mount: wrong fs type, bad option, bad superblock on /dev/sdd1, missing 
codepage or other error In some cases useful info is found in syslog - try 
dmesg | tail or so»

dmesg:

---
scsi6 : SCSI emulation for USB Mass Storage devices
usb-storage: device found at 6
usb-storage: waiting for device to settle before scanning
  Vendor: JetFlash  Model: TS128MJF2ARev: 1.00
  Type:   Direct-Access  ANSI SCSI revision: 02
SCSI device sdd: 250800 512-byte hdwr sectors (128 MB)
sdd: Write Protect is off
sdd: Mode Sense: 0b 00 00 08
sdd: assuming drive cache: write through
SCSI device sdd: 250800 512-byte hdwr sectors (128 MB)
sdd: Write Protect is off
sdd: Mode Sense: 0b 00 00 08
sdd: assuming drive cache: write through
 sdd: sdd1
sd 6:0:0:0: Attached scsi removable disk sdd
usb-storage: device scan complete
FAT: Unrecognized mount option "flush" or missing value
---

Then I manually mount sdd1 with pmount, and work fine.

Maybe is a hal error?


-- 
Benjamí
http://blog.bitassa.cat



.

Re: Aptitude can't upgrade Apt!

[Florian Kulzer]

[ Please keep the discussion on the list. There is a "reply-to-list"
  extension available for Thunderbird, or you can use "reply-to-all" and
  then remove the other email addresses (except debian-user). ]

On Fri, Jun 29, 2007 at 13:26:26 -0500, Richard Thomas wrote:
> [ citing an old message: ]
>>
>> You probably upgraded to apt version 0.7.2 recently; check with
>> "apt-cache policy apt". This new version provides the virtual package
>> libapt-pkg-libc6.5-6-4.4. Aptitude still depends on
>> libapt-pkg-libc6.3-6-3.11 which was provided by the previous version of
>> apt (0.6.46.4-0.1). You have to wait until a new version of aptitude
>> becomes available or you have to downgrade both apt and synaptic again.
>> (The newest version of synaptic already depends on the newest version of
>> apt.)
>
>
> My installation of aptitude disappeared during an update issue and this 
> problem is preventing me from reinstalling. Is there a correct downgrade 
> procedure listed anywhere?

You might still have the older versions of apt and aptitude in your
package cache. Check the output of: 

ls /var/cache/apt/archives/apt_0.6.46.4-0.1*
ls /var/cache/apt/archives/aptitude_0.4.4-4*

If you still have the two .deb files then you can install them with:

dpkg -i /var/cache/apt/archives/apt_0.6.46.4-0.1*
dpkg -i /var/cache/apt/archives/aptitude_0.4.4-4*

Another possibility is to add a line for "testing" to your
/etc/apt/sources.list, run "apt-get update" and then do

apt-get install apt/testing aptitude/testing

Officially, downgrades are "not supported", but I think in this case it
should work out.

-- 
Regards,| http://users.icfo.es/Florian.Kulzer
  Florian   |


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

Re: USB Stick mount

[Elimar Riesebieter]

On Sat, 30 Jun 2007 the mental interface of
Benjamí Villoslada told:

[...]
> FAT: Unrecognized mount option "flush" or missing value
> ---
> 
> Then I manually mount sdd1 with pmount, and work fine.
> 
> Maybe is a hal error?

$ grep sd /etc/fstab
?

Elimar

-- 
  Do you smell something burning or ist it me?

Re: pb with locales

[Florian Kulzer]

On Fri, Jun 29, 2007 at 17:32:46 +, abdelkader belahcene wrote:
> Hi,
> When I try to install a package a receive the following  warning. I am
> using the french language and the qwerty keyboard.
>
> bela# dpkg-reconfigure locales
> perl: warning: Setting locale failed.
> perl: warning: Please check that your locale settings:
>LANGUAGE = (unset),
>LC_ALL = (unset),
>LANG = "fr_FR.UTF-8"
>are supported and installed on your system.
> perl: warning: Falling back to the standard locale ("C").
> locale: Ne peut initialiser LC_ALL �� la locale par d��faut: Aucun
> fichier ( means NO FILE FOUND) ou r��pertoire de ce type
>
> where to change and fix the pb

Try this (as root):

LC_ALL="C" dpkg-reconfigure locales

If this allows you to run the command then you can make sure that
"fr_FR.UTF-8" is created and that it is set as the default. (If you do
not want to set a system-wide default then you have to configure the
relevant environmental variables of your $SHELL yourself.)

If you still cannot run the "dpkg-reconfigure locales" command then you
should post the output of

locale -a

grep '^[^#]' /etc/locale.gen

-- 
Regards,| http://users.icfo.es/Florian.Kulzer
  Florian   |

Re: Apache: The requested URL / was not found on this server.

[Nick Adie]

Michael

I had a deal of trouble ut here is my solution:


   ServerName webalizer.brokenmould.com
   ServerAlias monaco webalizer webalizer.brokenmould.com
   DocumentRoot /var/www/monaco
   ServerRoot /var/www/monaco

   AllowOverride All

   ScriptAlias /cgi-bin/ /var/www/~cgi-bin/monaco/
   ErrorLog  /var/www/~log/monaco/error.log
   CustomLog /var/www/~log/monaco/access.log combined
   ErrorDocument 404 /404.htm



   ServerName mail.brokenmould.com
   ServerAlias mail
   DocumentRoot /var/www/monaco/mail/
   ServerRoot /var/www/monaco/mail/

   AllowOverride All

   ScriptAlias /cgi-bin/ /var/www/~cgi-bin/monaco/
SSLEngine on
SSLCertificateFile /etc/apache2/apache.pem
   ErrorLog  /var/www/~log/monaco/error.log
   CustomLog /var/www/~log/monaco/access.log combined
   ErrorDocument 404 /404.htm


But remember you need to link 'available' to enabled, here's part of the
script that I used.

Regards
Nick


#!/bin/sh
cd /etc/apache2/sites-enabled
ln -s /etc/apache2/sites-available/000-monaco 000-monaco



On 6/30/07, Michael Pobega <[EMAIL PROTECTED]> wrote:


-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Thursday afternoon I reinstalled my i386 Debian installation as amd64
(To compliment my Core 2 Duo), and everything's been working pretty good
so far (Minus 3d rendering, but that's another story), but I'm having
one serious problem in Apache.

I reinstalled Apache2 and libapache2-mod-php5, and restored my old
/etc/apache2 configuration files, but I'm getting the error

> The requested URL / was not found on this server.

If you need any of my configuration files just ask, the only one that I
figure would effect anything would be
/etc/apache2/sites-available/default, so here is my config:


/etc/apache2/sites-available/default
> NameVirtualHost *
> 
> ServerAdmin [EMAIL PROTECTED]
>
>   DocumentRoot /var/www/
>   
>   Options FollowSymLinks
>   AllowOverride
>   None
>   
>   
>   Options Indexes FollowSymLinks MultiViews
>   AllowOverride None
>   Order
>   allow,deny
>   allow from all
>   
>   [...] (There's more here, but it's all the default, I
>   only changed the above)


- --
http://digital-haze.net/~pobega/ - My Website
If programmers deserve to be rewarded for creating innovative
programs, by the same token they deserve to be punished if they
restrict the use of these programs.
- Richard Stallman
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFGhk9Og6qL2BGnx4QRAlARAJ95yWrPNL1CCPXOf+yDjfHhyuvrlgCfe1AE
ZMHfFVHOBk3u+LPI1tfVSDs=
=Vn1o
-END PGP SIGNATURE-


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

Re: FileSystem Question

[Manon Metten]

Hi Douglas,

On 6/29/07, Douglas Allan Tutty <[EMAIL PROTECTED]> wrote:

On Fri, Jun 29, 2007 at 06:45:03PM +, Manon Metten wrote:


I've never used it but you can probably use a CVS repository for this
more conveniently.




I don't know nothing about CVS repositories. Can you explain a little
more, please?

Manon.

entering data into a pdf file

[Alan Ianson]

Hello List!

I need to enter data into a pdf file and print it. I can view the pdf 
using 
evince but I can't seem to enter any data into the fields. How can I do this?


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

Re: Updating without Apt

[Bob Proulx]

William Pursell wrote:
> I would be surprised if a major update to emacs weren't in
> sid within a few days, if not hours.

You missed the very long thread in this list last month with the
subject "I am ANGRY with Debian" where this topic was discussed.

  http://lists.debian.org/debian-user/2007/05/msg04442.html

In summary, no, recent emacs is not available in Sid and won't be for
some time.  This is a result of the GNU GFDL not being DFSG-free.

The emacs packages have been orphaned:

  http://lists.debian.org/debian-emacsen/2007/03/msg00012.html

Fortunately Romain Francoise has set up an unofficial emacs-snapshot
that may provide relief for emacs users desiring something newer than
emacs21.

  http://lists.debian.org/debian-emacsen/2007/04/msg2.html

Bob


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

Re: Updating without Apt

[Andrew Gray]

William Pursell wrote:

Andrew Gray wrote:

My name is Andrew, and I use Debian Lenny on an x86 machine.

I love Debian. I love apt. The Debian/Apt combination has allowed me 
to use software I would have never before dreamed to be possible to 
use. I'm quite content with my system at the moment. My problem comes 
when a piece of software that I use with frequency is updated by the 
developers in a major way. The Debian repositories do not reflect 
this change, and likely won't for some time.


The software in question is emacs22. 


I would be surprised if a major update to emacs weren't in
sid within a few days, if not hours.


Currently I have it compiled in my home directory at ~/source/emacs/, 
where I run it directly out of the src directory. I dare not "make 
install", or my apt install will be permanently harmed and/or it 
won't work and I will be faced with a crippled, non-running version 
of emacs on my system. Not only that, but with this compilation I was 
able to enable GTK for the X11 run of emacs. I loved having that kind 
of flexibility with my compilation.


Is there anything I can do to ensure that if I did a make install it 
wouldn't affect my current install or something like that?


As was suggested else-thread, you can probably do:
configure --prefix=/usr/local && make && make install.
(I'm not an emacs user, so the build may be more complicated
than that, but this will very likely work.)  In fact, /usr/local
is the defaul prefix, so you don't even need that.  In other
words, you can make install, since it will install to /usr/local/bin,
rather than /usr/bin and not interfere with the version installed
via apt at all.  If you want to test it more carefully, try

make DESTDIR=/tmp/foo install

and examine the resulting tree in /tmp/foo.  You should only get
executables in /usr/local/bin.

However, you might want to look into apt pinning.  This allows
you to pull emacs from sid without requiring you to update your
entire system to sid, and will give you the updates more
quickly.  Or, you can simply download the .deb files from
sid and install them with dpkg.  If the upstream is truly
not showing up in sid for too long, that won't work.  If
that is the case, you can grab the .dsc from sid and make
the modifications yourself to create a newer deb.


I was going to to a configure --prefix=/usr/local as you suggested, but 
I made the mistake of looking through the configure file and seeing that 
perhaps exec_prefix was the variable I wanted to change. I then figured 
that hey, this is what people do for all the other packages, it's likely 
to work for emacs, and gave it a shot with prefix.


It's worked beautifully! Installation went flawlessly, and now I'll just 
install slime and a few other things and get runnin'. Thanks so much Will!


Andrew Gray
Seattle,WA


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

Re: FileSystem Question

[Manon Metten]

Hi Sam,

On 6/30/07, Sam Leon <[EMAIL PROTECTED]> wrote:

ext3cow does this but it is not in debian repos for some reason


http://linux.slashdot.org/article.pl?sid=07/05/02/0413253




Thanks for the link. I checked out ext3cow, but taking snapshots is
not exactly what I meant. SFS takes no snapshots. It sort of backs
up all files I save. IE: it does not overwrite an existing file, but creates

a new file every time I save it. The old file will be renamed and moved
to the .recycled directory. So this is much less overhead than
creating snapshots from time to time.

1. First save of ~/my_doc
2. Second save of ~/my_doc
   The existing ~/my_doc is moved to ~/.recycled/my_doc
   ~/my_doc is saved to another part of the hd
3. Third save of ~/my_doc
   The existing ~/my_doc is moved/as to ~/.recycled/my_doc$AAA
   ~/my_doc is saved to another part of the hd
4. Fourth save of ~/my_doc
   The existing ~/my_doc is moved/as to ~/.recycled/my_doc$AAB
   ~/my_doc is saved to yet another part of the hd
etc.

At this point there exist four copies of my_doc:

three in ~/.recycled
my_doc
my_doc$AAA
my_doc$AAB

and one in ~
my_doc

All this only occurs when I save some file. No snapshots.
Besides that, .recycled only allows file read access and
deletion, no modification of files. SFS takes care of all this.
I don't have to think of it at all. It only comes in handy if I
want to recover some data.

Manon.

Re: entering data into a pdf file

[Andrew Gray]

Alan Ianson wrote:


Hello List!

	I need to enter data into a pdf file and print it. I can view the pdf using 
evince but I can't seem to enter any data into the fields. How can I do this?



  
I remember from my dad's experience at work, where he does this very act 
on an extremely regular basis, that you need the professional version of 
Adobe Acrobat in order to enter data into fields; not even the regular 
Acrobat Reader would do the job. Evince won't work because it's only a 
viewer, similar to Acrobat Reader. I don't know about free software 
alternatives.




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

Re: entering data into a pdf file

[Wackojacko]

Andrew Gray wrote:

Alan Ianson wrote:


Hello List!

I need to enter data into a pdf file and print it. I can view the 
pdf using evince but I can't seem to enter any data into the fields. 
How can I do this?



  
I remember from my dad's experience at work, where he does this very act 
on an extremely regular basis, that you need the professional version of 
Adobe Acrobat in order to enter data into fields; not even the regular 
Acrobat Reader would do the job. Evince won't work because it's only a 
viewer, similar to Acrobat Reader. I don't know about free software 
alternatives.
 


Acrobat reader is available from the unofficial debian repository at 
www.debian-multimedia.com. With the additional plugins available from 
this site, it does allow filling of forms in pdf format.


HTH

Wackojacko




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

Re: resolv.conf wrongly gets Belkin router's address

[Bob Proulx]

Kent West wrote:
> I replaced a dead router the other day with a new Belkin F5D7230-4 
> wireless/4-port unit.

First let me say that I am not familiar with that particular piece of
hardware.

> When networking starts on my Etch box, /etc/resolv.conf gets rewritten 
> like so:
> 
> search hydroplatenet
> nameserver 192.168.2.1
> nameserver 208.180.42.68
> nameserver 208.180.42.100

I assume this interface is configured using DHCP?  Are you using
'resolvconf'?  What is in your /etc/network/interfaces file for that
interface?

> Web browsing (and other name-dependent Intarweb Pipes stuff) is very 
> slow with these settings, but if I comment/remove the first nameserver 
> line, which is the address of the Belkin router, things are fine.

Some routers proxy DNS information through themselves.  On the boxes I
have seen that do this it is configurable and can be turned off.  It
appears to me from reading your report that the Belkin box has offered
its own address as a DNS server in addition to some upstream DNS
server.

Inspect the Belkin client DNS configuration and see if that address
has been added to the configuration or if there is a way to remove
that configuration.  On consumer routers I have worked with it has
always been possible to specially add in a set of DNS addresses to
provide to DHCP clients, by default empty.  Hopefully in your case
these can be removed specifically.  Although it is a terrible thing
that it came out-of-the-box with a bad configuration.

> I found a couple of other people with similar problems 
> (http://ubuntuforums.org/showthread.php?t=297834 and 
> http://www.mepis.org/node/3221), one of which provided the kludgey fix 
> of pre-pending the good addresses via /etc/dhcp3/dhclient.conf to force 
> duplicates of the good address to the top of the file, but that's just ugly.

And you could also completely override the dns addresses provided by
dhcp or you could move to static addresses.  In both cases you are not
getting good use out of the box.

> I've sent this problem to Belkin, and their tech support responded that 
> "The Belkin Router only provides a Gateway address but is not a DNS 
> server" 

Then why did it offer itself as one through DHCP?  That could be a
confirmation of a bug.

> (http://www.belkin.com/contact/milan/default1.aspx?id=5B5B5956095F560D0D5C5F0E5B0A5D56305F5E).

Note: That web url has expired.

> They indicated that I should call them on the phone (I hate dealing with 
> stuff over the phone...), but I wanted to get all my ducks in a row 
> before I do so.

If they hear Debian GNU/Linux they will probably drop your support
call immediately too.  Sigh.  It would be useful to crosscheck this
with tests from other operating system clients.  If they all behave
the same then you have confirmation.  If they behave different then
that is also useful information.

> Is Debian somehow responsible for incorrectly seeing this "Gateway 
> address" as a name server address (and if so, how do I fix it), or is 
> the Belkin tech supporter incorrect about his product (or some other 
> option)?

When you say Debian here you really should say dhcp3 client or pump or
other specific dhcp client that you have installed.  If you don't know
then you are probably using dhcp3.  Debian just bundles it up for you.

I believe that the Belkin DHCP server is offering itself as a DNS
server inappropriately.  Yours is not the only report of the problem.
It may be possible correct the bad configuration by explicit
configuration of the box to avoid this.

Commodity routers such as those are inexpensive.  Return yours and buy
a different brand that works correctly and avoid the problem.

Bob


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

Re: USB Stick mount

[Aenn Seidhe Priest]

On 30.06.2007 at 18:43 Elimar Riesebieter wrote:

>On Sat, 30 Jun 2007 the mental interface of
>Benjamí Villoslada told:
>
>[...]
>> FAT: Unrecognized mount option "flush" or missing value
>> ---
>> 
>> Then I manually mount sdd1 with pmount, and work fine.
>> 
>> Maybe is a hal error?
>
>$ grep sd /etc/fstab
>?
>
>Elimar
>
>-- 
>  Do you smell something burning or ist it me?

Maybe it's cacheless :-)



 Not far from here, by a white sun, behind a green star, lived the 
Steelypips, illustrious, industrious, and they hadn't a care: no spats 
in their vats, no rules, no schools, no gloom, no evil influence of the 
moon, no trouble from matter or antimatter -- for they had a 
machine, a dream of a machine, with springs and gears and perfect 
in every respect. And they lived with it, and on it, and under it, and 
inside it, for it was all they had -- first they saved up all their atoms, 
then they put them all together, and if one didn't fit, why they 
chipped at it a bit, and everything was just fine... 

-- Stanislaw Lem, "Cyberiad"


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

Re: entering data into a pdf file

[Brad Rogers]

On Sat, 30 Jun 2007 11:04:09 -0700
Andrew Gray <[EMAIL PROTECTED]> wrote:

Hello Andrew,

> I remember from my dad's experience at work, where he does this very
> act on an extremely regular basis, that you need the professional
> version of Adobe Acrobat in order to enter data into fields; not even

You can enter data using Acroread.  You can even print out the
completed form.  There are, of course, caveats.  First, the document has
to be set up to allow it at the design stage (i.e. in Adobe Designer),
and second, you can't save, from Acroread, the document with the data
included.

Details are included in the Acroread help files in the section "Filling
in Adobe PDF forms".

-- 
 Regards  _
 / )   "The blindingly obvious is
/ _)radnever immediately apparent"

Every single one of us
Devil Inside - INXS


signature.asc
Description: PGP signature

Re: entering data into a pdf file

[Alan Ianson]

On Sat June 30 2007 11:06:48 am Wackojacko wrote:
> Andrew Gray wrote:
> > Alan Ianson wrote:
> >> Hello List!
> >>
> >> I need to enter data into a pdf file and print it. I can view the
> >> pdf using evince but I can't seem to enter any data into the fields.
> >> How can I do this?
> >
> > I remember from my dad's experience at work, where he does this very act
> > on an extremely regular basis, that you need the professional version of
> > Adobe Acrobat in order to enter data into fields; not even the regular
> > Acrobat Reader would do the job. Evince won't work because it's only a
> > viewer, similar to Acrobat Reader. I don't know about free software
> > alternatives.
>
> Acrobat reader is available from the unofficial debian repository at
> www.debian-multimedia.com. With the additional plugins available from
> this site, it does allow filling of forms in pdf format.
>
> HTH

It sure does. For a minute there I though I hit a show stopper. I have it 
installed on my laptop and it works as needed so I can get it installed when 
I get back to work and we'll be good to go.

I don't see any acroread package on the amd64 box but that's not critical (at 
this point).

Thanks for the replies.. :)


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

Re: USB Stick mount

[Benjamí Villoslada]

El Dissabte 30 Juny 2007, Elimar Riesebieter va escriure:
> $ grep sd /etc/fstab

$ grep sd /etc/fstab
/dev/sda5   /   ext3defaults,errors=remount-ro 0   1
/dev/sda7   /home   ext3defaults0   2
/dev/sda6   noneswapsw  0   0

Without entry for sdd.

Thanks!

-- 
Benjamí
http://blog.bitassa.cat



.

Mounting USB HDs the same every time.

[Eric A. Bonney]

   I have 4 USB hard drives on my server in my basement.  The power 
went out the other night for about 30 seconds and when the server came 
back online, most of my Samba shares where messed up.  I figured out 
what happened was that the drives were not initialized in the same order 
as they were when I setup the shares.  Now a drive that used to be 
/media/usb0 was say /media/usb1.


So I have to go back and reconfigure my samba.conf file, not a huge 
deal, but still a pia.  Is there anyway to make it so that the drives 
will always be named correctly upon rebooting of the server and 
mounted?  I think it needs to be done in fstab is that correct?  The 
other issue I have is that I always have to actually click on each drive 
before it is able to be used by anyone else.  Not sure why, but I don't 
think the drives are getting mounted at bootup.


Thanks for the help.

-Eric


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

k3b mesure of dir size across nfs

[pol]

Hi all, 

I use k3b to burn backup cd  (dvd-r 4.7 gb) of directories on ext3 file
system across nfs connection. 

K3b has just reported the selected remote directory to be more than 2 gb
large. Yet the same mounted directory, as measured by 'du -s' on the local
station, where k3b is running, returs about 0.5 gb only . That is the
probable correct value, being the same that is found by 'du -s' on the
remote machine. 
How to fix that strange k3b behaviour?

thank you 

--
Pol 


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

Re: FileSystem Question

[Sam Leon]

Manon Metten wrote:

Hi Sam,

On 6/30/07, *Sam Leon* < [EMAIL PROTECTED] 
> wrote:


ext3cow does this but it is not in debian repos for some reason

http://linux.slashdot.org/article.pl?sid=07/05/02/0413253




Thanks for the link. I checked out ext3cow, but taking snapshots is
not exactly what I meant. SFS takes no snapshots. It sort of backs
up all files I save. IE: it does not overwrite an existing file, but 
creates

a new file every time I save it. The old file will be renamed and moved
to the .recycled directory. So this is much less overhead than
creating snapshots from time to time.

1. First save of ~/my_doc 
2. Second save of ~/my_doc

The existing ~/my_doc is moved to ~/.recycled/my_doc
~/my_doc is saved to another part of the hd
3. Third save of ~/my_doc
The existing ~/my_doc is moved/as to ~/.recycled/my_doc$AAA
~/my_doc is saved to another part of the hd
4. Fourth save of ~/my_doc
The existing ~/my_doc is moved/as to ~/.recycled/my_doc$AAB
~/my_doc is saved to yet another part of the hd
etc.

At this point there exist four copies of my_doc:

three in ~/.recycled
my_doc
my_doc$AAA
my_doc$AAB

and one in ~
my_doc

All this only occurs when I save some file. No snapshots.
Besides that, .recycled only allows file read access and
deletion, no modification of files. SFS takes care of all this.
I don't have to think of it at all. It only comes in handy if I
want to recover some data.

Manon.



I swear I have heard of something like that for ext3.  I though it was 
ext3cow.  Took me like 30 minutes to find it on google, lol.  hmm


Sam

Re: FileSystem Question

[Douglas Allan Tutty]

On Sat, Jun 30, 2007 at 05:29:58PM +, Manon Metten wrote:
> On 6/29/07, Douglas Allan Tutty <[EMAIL PROTECTED]> wrote:
> On Fri, Jun 29, 2007 at 06:45:03PM +, Manon Metten wrote:
> >
> >I've never used it but you can probably use a CVS repository for this
> >more conveniently.
> 
> I don't know nothing about CVS repositories. Can you explain a little
> more, please?

Hi Manon,

See the apt descriptions for subversion and cvs.  From the description
of cvs:
CVS is a version control system, which allows you to keep old
versions of files (usually source code), keep a log of who,
when, and why changes occurred, etc.

As I understand it, a user will 'check out' a file from the cvs to work
on it.  While it is 'checked out', others can read that version of the
file but they can't 'check it out' to edit it.  The person who has it
checked out can then check the file back in with whatever changes they
made.  The cvs then keeps diffs and checkpoints so that the state of the
repository at any point in time can be recreated.  I.e. the question
"what did that file look like on June 15?" is a valid question that the
cvs can answer.

The whole cvs sits on top of a regular file system.  There's nothing
preventing root from directly editing a file owned by cvs.

I think that there are web-based interfaces to the cvs so that files can
be retreived over https as well.

This is all I know about cvs and it may not be accurate.  

Doug.


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

Re: Mounting USB HDs the same every time.

[Douglas Allan Tutty]

On Sat, Jun 30, 2007 at 02:45:59PM -0400, Eric A. Bonney wrote:
>I have 4 USB hard drives on my server in my basement.  The power 
> went out the other night for about 30 seconds and when the server came 
> back online, most of my Samba shares where messed up.  I figured out 
> what happened was that the drives were not initialized in the same order 
> as they were when I setup the shares.  Now a drive that used to be 
> /media/usb0 was say /media/usb1.
> 
> So I have to go back and reconfigure my samba.conf file, not a huge 
> deal, but still a pia.  Is there anyway to make it so that the drives 
> will always be named correctly upon rebooting of the server and 
> mounted?  I think it needs to be done in fstab is that correct?  The 
> other issue I have is that I always have to actually click on each drive 
> before it is able to be used by anyone else.  Not sure why, but I don't 
> think the drives are getting mounted at bootup.
> 

Use volume label names.  Name the volume with the tools for the specific
filesystem (ext2, JFS, etc) then use LABEL= in fstab and with the mount
command.  The the respective man pages.

Doug.


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

rock solid

[will trillich]

typical debian server--

i logged in, connected to an old, neglected SCREEN session, and this
was still on the screen:

# uptime
20:30:17 up 15 days,  6:11,  2 users,  load average: 0.76, 0.24, 0.08
[EMAIL PROTECTED]:/etc
  Fri Jan 05 20:30:17

and then just for symmetry i added:

# uptime
15:42:12 up 190 days, 23:42,  2 users,  load average: 1.05, 1.05, 1.00
[EMAIL PROTECTED]:/etc
  Sat Jun 30 15:42:12
#

ho hum, serves files, backs up, yada yada, all in a year's work.

--
will trillich
"Gratitude is riches. Complaint is poverty." -- Doris Von Kappelhoff


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

Re: FileSystem Question

[Manon Metten]

Hi Sam,

On 6/30/07, Sam Leon <[EMAIL PROTECTED]> wrote:

I swear I have heard of something like that for ext3.  I though it was

ext3cow.  Took me like 30 minutes to find it on google, lol.  hmm




Oh my! Well, thanks again. Although ext3cow is not what I am looking for,
your effort is much appreciated. You might have hit the bull's eye.

Greetings, Manon.

Re: Apache: The requested URL / was not found on this server.

[Michael Pobega]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Sat, Jun 30, 2007 at 06:17:30PM +0100, Nick Adie wrote:
> On 6/30/07, Michael Pobega <[EMAIL PROTECTED]> wrote:
> >
> >-BEGIN PGP SIGNED MESSAGE-
> >Hash: SHA1
> >
> >Thursday afternoon I reinstalled my i386 Debian installation as amd64
> >(To compliment my Core 2 Duo), and everything's been working pretty good
> >so far (Minus 3d rendering, but that's another story), but I'm having
> >one serious problem in Apache.
> >
> >I reinstalled Apache2 and libapache2-mod-php5, and restored my old
> >/etc/apache2 configuration files, but I'm getting the error
> >
> >> The requested URL / was not found on this server.
> >
> >If you need any of my configuration files just ask, the only one that I
> >figure would effect anything would be
> >/etc/apache2/sites-available/default, so here is my config:
> >
> > [cut config file]
> >
>
> Michael,
> 
> I had a deal of trouble ut here is my solution:
> 
> [cut config file]
> 
> But remember you need to link 'available' to enabled, here's part of the
> script that I used.
> 
> Regards
> Nick
> 

That was it, I removed sites-enabled and symbolically linked
sites-available to sites-enabled, then restarted Apache and it began to
work. I guess maybe this is a repercussion from moving my /etc from my
i386 install to my amd64 install (Not that the architecture makes any
difference, but I guess maybe moving the files broke the symbolic link).

Thanks for the help and the quick response!

- -- 
http://digital-haze.net/~pobega/ - My Website
If programmers deserve to be rewarded for creating innovative
programs, by the same token they deserve to be punished if they
restrict the use of these programs. 
 - Richard Stallman
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFGhrQCg6qL2BGnx4QRAo7KAJ4w8J3zpKB+Vdis9/hwW7Jaayo8OwCfSjtK
1MyIRLPzSrrfXZMzSUpXzy0=
=XaMt
-END PGP SIGNATURE-


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

Re: FileSystem Question

[Manon Metten]

Hi Douglas,

On 6/30/07, Douglas Allan Tutty <[EMAIL PROTECTED]> wrote:

See the apt descriptions for subversion and cvs.  From the description

of cvs:
CVS is a version control system, which allows you to keep old
versions of files (usually source code), keep a log of who,
when, and why changes occurred, etc.



The whole cvs sits on top of a regular file system.  There's nothing
preventing root from directly editing a file owned by cvs.




Thanks for explaining Doug. I ain't got the time right now to check this
out in detail, but at first glance it surely looks useful to me.

Greetings, Manon.

Re: Updating without Apt

[Tyler Smith]

On 2007-06-30, Andrew Gray <[EMAIL PROTECTED]> wrote:
>>
>>
> I was going to to a configure --prefix=/usr/local as you suggested, but 
> I made the mistake of looking through the configure file and seeing that 
> perhaps exec_prefix was the variable I wanted to change. I then figured 
> that hey, this is what people do for all the other packages, it's likely 
> to work for emacs, and gave it a shot with prefix.
>
> It's worked beautifully! Installation went flawlessly, and now I'll just 
> install slime and a few other things and get runnin'. Thanks so much Will!
>

A little late now, but I went through the same thing last week. I
wasn't sure what the best approach would be. From some very limited
experimenting it looks like you can have the apt emacs21 and a
compiled emacs22 on the same system without problem. But since stuff
is so easy to install with apt if I change my mind, to be safe I just
purged everything related to emacs (except my .emacs, of course). With
a clean setup, I then installed emacs22 from source using all the
standard settings. Took me about ten minutes. And another couple of
hours to install Auctex, ESS, and slime, but just because I was poking
around a lot figuring out how the load-path stuff works. No problems
at all so far.

Cheers,

Tyler


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

Re: How to move the master boot record?

[Felix Karpfen]

On Fri, 29 Jun 2007 15:33:09 +0200, Rodolfo Medina wrote:

On Fri, 29 Jun 2007 15:33:09 +0200, Rodolfo Medina wrote:


> I want to do so beacuse: now I'm still using Debian Sarge, which is
> installed in hda6; I want to install Debian Etch in hda9; then when I'm
> sure that everything is all right with Etch I want to boot from hda9, so
> hda6 can be formatted again.
> 
Has anyone done this successfully?

I faced a similar problem, wrote to "linux.debian.user" for advice and
scored a zero response.

In my case I went ahead, made a backup of Sarge to a newly-created
partition, checked that I could boot into it and then ran a "dist-upgrade"
on my main Sarge partition.

Result:

- Etch boots and works perfectly (fortunately);

- Sarge boots (despite some "fatal" notifications during the boot);
  all the tested programs still work; the mouse does not. Running
  "dpkg-reconfigure xserver-xfree86" achieved nothing.

For the record, I use LILO.

Felix Karpfen



-- 
Felix Karpfen
Public Key 72FDF9DF (DH/DSA)


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

Re: k3b mesure of dir size across nfs - no symb links

[pol]

pol wrote:
> K3b has just reported the selected remote directory to be more than 2 gb
> large. Yet the same mounted directory, as measured by 'du -s' on the local
> station, where k3b is running, returs about 0.5 gb only . That is the
> probable correct value, being the same that is found by 'du -s' on the
> remote machine.

Add to this that no instruction to follow symbolic links has been given 
(I hope k3b does not follow symbolic links by default)

thanks fot your hints
--
Pol


> How to fix that strange k3b behaviour?
> 
> thank you
> 
> --
> Pol



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

Re: How to move the master boot record?

[Nigel Henry]

On Saturday 30 June 2007 22:41, Felix Karpfen wrote:
> On Fri, 29 Jun 2007 15:33:09 +0200, Rodolfo Medina wrote:
>
> On Fri, 29 Jun 2007 15:33:09 +0200, Rodolfo Medina wrote:
> > I want to do so beacuse: now I'm still using Debian Sarge, which is
> > installed in hda6; I want to install Debian Etch in hda9; then when I'm
> > sure that everything is all right with Etch I want to boot from hda9, so
> > hda6 can be formatted again.
>
> Has anyone done this successfully?
>
> I faced a similar problem, wrote to "linux.debian.user" for advice and
> scored a zero response.
>
> In my case I went ahead, made a backup of Sarge to a newly-created
> partition, checked that I could boot into it and then ran a "dist-upgrade"
> on my main Sarge partition.
>
> Result:
>
> - Etch boots and works perfectly (fortunately);
>
> - Sarge boots (despite some "fatal" notifications during the boot);
>   all the tested programs still work; the mouse does not. Running
>   "dpkg-reconfigure xserver-xfree86" achieved nothing.
>
> For the record, I use LILO.
>
> Felix Karpfen

This is going back a bit, but all my Debian installs started off as Woody 
3.0r2, and have been constantly upgraded. I had no mouse problems (ps2 mouse) 
with the 2.4.27 kernel, but moving to a 2.6.8 kernel  caused problems.

I had to add to /etc/modules a couple of modules. I can't remember in which 
order, but at first I had no mouse pointer showing at all. I modprobed one 
module, then had the mouse pointer, but couldn't move it, then modprobed the 
second module, and all was working ok. See the 2 modules below.

mousedev
psmouse

Don't know if this helps, but it fixed my problems at the time.

Nigel.


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

Re: FileSystem Question

[William Pursell]

Douglas Allan Tutty wrote:

On Sat, Jun 30, 2007 at 05:29:58PM +, Manon Metten wrote:

On 6/29/07, Douglas Allan Tutty <[EMAIL PROTECTED]> wrote:
On Fri, Jun 29, 2007 at 06:45:03PM +, Manon Metten wrote:

I've never used it but you can probably use a CVS repository for this
more conveniently.

I don't know nothing about CVS repositories. Can you explain a little
more, please?


Hi Manon,

See the apt descriptions for subversion and cvs.  From the description
of cvs:
CVS is a version control system, which allows you to keep old
versions of files (usually source code), keep a log of who,
when, and why changes occurred, etc.

As I understand it, a user will 'check out' a file from the cvs to work
on it.  While it is 'checked out', others can read that version of the
file but they can't 'check it out' to edit it.  The person who has it
checked out can then check the file back in with whatever changes they
made.  The cvs then keeps diffs and checkpoints so that the state of the
repository at any point in time can be recreated.  I.e. the question
"what did that file look like on June 15?" is a valid question that the
cvs can answer.

The whole cvs sits on top of a regular file system.  There's nothing
preventing root from directly editing a file owned by cvs.

I think that there are web-based interfaces to the cvs so that files can
be retreived over https as well.

This is all I know about cvs and it may not be accurate.  


Keep in mind that CVS is extremely old, and entirely obsolete.
Subversion was a new implementation of the same idea, and
did in fact address many of CVS's shortcomings.  However,
if you are going to look into using a VCS (Version Control
System) for doing backups like this, look into git.
http://git.or.cz



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

Re: FileSystem Question

[Glen Pfeiffer]

On 06/30/2007 03:00 PM, William Pursell wrote:
> Keep in mind that CVS is extremely old, and entirely obsolete.
> Subversion was a new implementation of the same idea, and did
> in fact address many of CVS's shortcomings.

I second that.

> However, if you are going to look into using a VCS (Version
> Control System) for doing backups like this, look into git.
> http://git.or.cz

Care to explain why you suggest using Git. I cannot find a reason
to use it unless one needs good decentralized version control.

-- 
Glen


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

Triple/Quad screen options

[Hans du Plooy]

Hi guys,

I'm Looking for your experiences using three or more monitors under 
Linux?  We use two to four screens at work (depending on our exact 
requirements).  Using two screens on a dual-head card works mostly OK, 
at least for my legitimate work needs (some things like fullscreen 
mplayer don't behave so nicely, but that's not essential), but as soon 
as you run three or more screens and have to use two graphics card, 
things get a little messy.  Xinerama, for one, breaks hardware 3D, and 
some things behave weirdly


I've looked online at cards like the Matrox G450 Quad, and wondered if 
anyone has experience running these under linux.  How well does things 
like 3D and hardwar everlay (as in MPlayer fullscreen with xv without 
trying to fullscreen over all the screens) work?


Thanks


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

Mailing list problems with Thunderbird

[Sam Leon]

Ok, I have had this problem for awhile and I can't find any info on it.  
I am using thunderbird 1.5 in testing.  All the other mailing lists that 
I have subscribed to work fine.  When I see a post that I want to 
respond to, I simply click the Reply button and it will bring up a new 
box with the message and the "To" field will be set to the correct 
mailing list. 

However with all the debian lists, if I click reply, it will open a new 
box and the "To" field will be set to the user that sent the message and 
not the debian list.  I will have to delete the users name, and then 
manually type in the correct list.  Sometimes using "Reply to all" will 
kinda work.  The original user will be in the "To" field and the list 
will be in the "CC" field, sometimes but not always. 


I don't know.  Anyone know what I am talking about?  Is this just "normal"?


Sam


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

Re: resolv.conf wrongly gets Belkin router's address

[Kent West]

Bob Proulx wrote:
> Kent West wrote:
>   
>> I replaced a dead router the other day with a new Belkin F5D7230-4 
>> wireless/4-port unit.
>> 
>
> First let me say that I am not familiar with that particular piece of
> hardware.
>   
>From Wal-Mart; about $40. Wireless G Router with 4 wired ports.


>> When networking starts on my Etch box, /etc/resolv.conf gets rewritten 
>> like so:
>>
>> search hydroplatenet
>> nameserver 192.168.2.1
>> nameserver 208.180.42.68
>> nameserver 208.180.42.100
>> 
>
> I assume this interface is configured using DHCP?  Are you using
> 'resolvconf'?  What is in your /etc/network/interfaces file for that
> interface?
>   
Yes, DHCP.

> [EMAIL PROTECTED]:/home/westk> cat /etc/network/interfaces
> # /etc/network/interfaces -- configuration file for ifup(8), ifdown(8)
>
> # The loopback interface
> auto lo
> iface lo inet loopback
>
> # The first network card - this entry was created during the Debian
> installation
> auto eth0
> iface eth0 inet dhcp

(Actually, this particular interfaces file is from a totally different
machine, but it also has the same symptoms. Also, when I booted this
machine up with a Knoppix LiveCD, it too go the bogus Belkin DNS entry
in resolv.conf. So I now know it's not machine- or Debian-specific,
although it might be Debian-derived-specific.)

>> Web browsing (and other name-dependent Intarweb Pipes stuff) is very 
>> slow with these settings, but if I comment/remove the first nameserver 
>> line, which is the address of the Belkin router, things are fine.
>> 
>
> Some routers proxy DNS information through themselves.  On the boxes I
> have seen that do this it is configurable and can be turned off.  It
> appears to me from reading your report that the Belkin box has offered
> its own address as a DNS server in addition to some upstream DNS
> server.
>   
That's the way it seems to me also, but when Belkin support told me that
their router does NOT offer its own address as a DNS server, I figured I
better double-check with other Debianistas before concluding that Belkin
support doesn't know what they're talking about.

> Inspect the Belkin client DNS configuration and see if that address
> has been added to the configuration or if there is a way to remove
> that configuration.
I don't find any such settings; here's a snapshot of the only relevant
page on the Belkin setup that I can find: 
http://www.acu.edu/~westk/belkin.jpg

> If they hear Debian GNU/Linux they will probably drop your support
> call immediately too.  Sigh.  It would be useful to crosscheck this
> with tests from other operating system clients.
Ag! I'm so dense. I'm so accustomed to being a Debian-only house that I
forgot my work laptop is dual-boot with Vista. I'll give that a shot.
Thanks!

> When you say Debian here you really should say dhcp3 client or pump or
> other specific dhcp client that you have installed.  If you don't know
> then you are probably using dhcp3.
>   
Yes, dhcp3.

> Commodity routers such as those are inexpensive.  Return yours and buy
> a different brand that works correctly and avoid the problem.
>   
Thanks for the response! I appreciate it!

-- 
Kent


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

Re: Mailing list problems with Thunderbird

[Kamaraju S Kusumanchi]

Sam Leon wrote:

> Ok, I have had this problem for awhile and I can't find any info on it.
> I am using thunderbird 1.5 in testing.  All the other mailing lists that
> I have subscribed to work fine.  When I see a post that I want to
> respond to, I simply click the Reply button and it will bring up a new
> box with the message and the "To" field will be set to the correct
> mailing list.
> 
> However with all the debian lists, if I click reply, it will open a new
> box and the "To" field will be set to the user that sent the message and
> not the debian list.  I will have to delete the users name, and then
> manually type in the correct list.  Sometimes using "Reply to all" will
> kinda work.  The original user will be in the "To" field and the list
> will be in the "CC" field, sometimes but not always.
> 

According to
http://packages.debian.org/changelogs/pool/main/i/icedove/icedove_2.0.0.4-1/changelog
the replytolist feature is introduced since icedove 1.5.0.5-1 onwards. What
is your version of thunderbird or icedove?


> I don't know.  Anyone know what I am talking about?  Is this just
> "normal"?

This question comes up pretty frequently. Your best bet is to use
thunderbird version which has replytolist feature or install the extension
which provides this feature or use an email client which knows how to deal
with mailing lists (Ex :- kmail, mutt etc.,) or use a news reader (Ex :-
knode) to read the list via a newsgroup.

You might also want to read
https://bugzilla.mozilla.org/show_bug.cgi?id=45715 which tells that this
bug has not been fixed upstream in around 7 years!

hth
raju


-- 
Kamaraju S Kusumanchi
http://www.people.cornell.edu/pages/kk288/
http://malayamaarutham.blogspot.com/


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

Re: FileSystem Question

[William Pursell]

Glen Pfeiffer wrote:

On 06/30/2007 03:00 PM, William Pursell wrote:

Keep in mind that CVS is extremely old, and entirely obsolete.
Subversion was a new implementation of the same idea, and did
in fact address many of CVS's shortcomings.


I second that.


However, if you are going to look into using a VCS (Version
Control System) for doing backups like this, look into git.
http://git.or.cz


Care to explain why you suggest using Git. I cannot find a reason
to use it unless one needs good decentralized version control.



I was (and I suppose still am) a huge fan of subversion.  I
loved RCS, hated CVS, and found subversion to be a God-send.
I started using git about 3 weeks ago, and in that short
time I've found that its flexibility has already changed
my work flow (for the better).  In particular, it is actually
*easy* to merge. The decentralized nature of it is
really nice for (at least) one big reason--it strongly
encourages 'commit-early, commit often',  since you can
munge your own repository as much as you want, and then only
push clean versions into the central repository.

Another point: consider the case in which you spend an hour
working, and then realize that you do not want to incorporate
those changes into the trunk.  In subversion, if you didn't
make a branch an hour ago, you've got a little bit of
work to do: make a branch now and switch your working copy
to it.  (I know, it doesn't seem like much work, but it
is a kludge.)  With git, you simply declare the current
working copy to be a new branch.  This is perhaps not a good
description of the issue, and undoubtedly will not convince
anyone of anything, but I think it's indicative of why
the subsequent merge is so much cleaner.

I was reluctant to investigate git for reasons which are
unclear to me, and was frankly somewhat offended by Linus'
comments that "subversion is the most pointless open-source
project ever started" along with remarks like "subversion
says it is 'CVS done right'...which says a lot about
why it is lousy [massively paraphrased due to memory loss]".
However, Linus generally knows what he's talking about,
and I respect him enough to give some credence to his
opinions.  I still like subversion, but I suspect it is
only because I am required to use it in some situations.

Maybe I'm overstating things, but it's possible that
Linus will end up being remembered for creating git
rather than for creating the Linux kernel.  It really is
that good.


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