Re: apt-get online package repository configuration on Debian

[Andrei POPESCU]

On Sb, 18 mai 13, 19:01:38, Steve Faleiro wrote:
> 
> 1. Add the following line to the file etc/apt/sources.list :
> 
>deb http://ftp.debian.org/debian wheezy main

Depending on where you live, using the country specific mirror (or a 
mirror next to you[1]) might improve download speeds significantly[2].

[1] http://www.debian.org/mirror/list
[2] which is why one should let the installer choose a mirror for you 
based on the country you selected

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

Re: FQDN via DHCP, then used in exim4

[Klaus Doering]

Stan,

Thank you for taking the time to explain your perspective. Maybe it is
the tone of your teachings that tickle me, maybe it's just that I'm no
big fan of sweeping statements a la "Don't do it, ever".  As I described
in my initial post, this thread concerns a small domestic setting. There
is very little redundancy in such a setup, and the
router+firewall+switch+ap is one of the single point of failures that
potentially bring down most of the installation. And that's quite
independent of whether or not it also acts a a dhcp server to hand out
ip info to -- say -- an email server. My way of dealing with this
inherent weakness is that I have an old router that can be swapped in
and configured to do most of the jobs, until the original one is
repaired or replaced. Good enough for me; your guess is as good as mine
about how many typical domestic installations even have a backup
router...

I agree that in a different setting, where there are many users,
hundreds of emails per minute and other mission-critical stuff is going
on, one needs to design the infrastructure a lot more carefully.

So, the initial question remains unanswered: what happens to the
information provided by the dhcp server as reported in the lease file in
/var/lib/dhcp, how is it accessed, and is there a way to make exim4 use
it?

Klaus

On 17/05/13 19:06, Stan Hoeppner wrote:

It's actually very clear.  Has been forever.  The problem is that the
experts who know why -not- to use DHCP here aren't writing about it.
It's common sense, rule of thumb, etc.  Everyone knows it.  Everyone
except the few folks such as yourself who ask "why not", then writing
something up showing how it 'might' work.  Simply put, you're reading
things written by folks who don't know what they're doing, who don't
have long experience.  Some food for thought:

1.  When your DHCP server goes down or is unreachable this can wreak
havoc.  When IPs and hostnames are configured in local UNIX files this
is never a problem.  So you must create a failover DHCP server.  Getting
this to work seamlessly is notoriously difficult.  It's one thing in an
ISP setting when DHCP failover has a lengthy delay.  It's quite another
when you can't receive email for you 10,000 person organization for many
minutes while the DHCP servers sink up and the MX MTA can't get its IP
and hostname again.

2.  When you swap a defective server NIC udev rules can keep a local
static IP consistent on eth0, or require little reconfiguration after
the NIC swap.  Using your DHCP scenario you must write down the MAC
address from the sticker on the card before inserting it, update the
record in the DHCP server with the new MAC, then power up the box.  The
ethernet industry has been working for many years to -eliminate- the
need for direct anipulation of MAC addresses.  In your scenario you're
relying on it.  Etc, etc, etc.

While it may be possible, in most environments it is simply impractical,
and creates far more problems than it solves.  If this was a great idea,
Rackspace et al would be doing it.  And none of them are.





--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5198913f.5070...@gmail.com

Libreoffice4 in wheezy? Several questions

[Hans-J. Ullrich]

Hello all,

I wanted to install LO4 in wheezy. I wondered, why it is in wheezy-backports 
but not in testing. Maybe the packages in wheezy-backports are too young for 
testing? 

However, I managed it by using aptitude and made sure, all dependencies are 
correct.

But I guess, there is an easier way, to update my ONLY my installed packages 
out of wheezy. In the wiki I read, you can use apt-get -t wheezy-backports 
install $package, which worked well.

And of course, i can edit the apt pinning system to give wheezy-backports a 
higher ranking. But wouldn't this affect other packages, too?

I suppose, there might be a way just to update certain packages from wheezy-
backports using the pinning system, but I fear, I can use this only for meta-
packages or directly named packages. Please correct me, if I am wrong.

And last question: If Libreoffice4 will be put into testing repo, will those 
packages then be newer than those in wheezy-backports or is this a matter of 
chance?

Thanks for your answers! This will help me to help my friends.

Best regards

Hans






-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201305191139.31038.hans.ullr...@loop.de

libdigest-sha1-perl

[Veljko]

Hi,

I'm using Wheezy, but needed ddclient from Sid, as that version supports
freedns protocol. I know there are other programs/scripts, but I use ddclient
on Slackware and just wanted to have same setup. As it turns out, that version
needs libdigest-sha1-perl which, as I understood, is dropped in Wheezy in favor
of libdigest-sha-perl. Don't know why is it in Sid then... Anyhow, to solve
this problem I just edited part of /usr/sbin/ddclient that checks for version
and loads SHA1 module and replaced it with Digest::SHA module.

It works just fine, but as I do not know perl or Debian that well, I don't know
if this was right way to do it. Any thougts?

Regards,
Veljko


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519102137.ga6...@angelina.example.com

Re: Libreoffice4 in wheezy? Several questions

[theartloy]

Hi Hans,

If you want to see why a certain package version has not yet made it 
from unstable into testing, you can go to:


   http://packages.qa.debian.org/

replacing  with the name of package in question.

Looking at the page for libreoffice[1], under "testing migration" you 
can see the pointer:

 * Too young, only 8 of 10 days old
At this point the package migration system stopped considering this 
package for migration, it will not consider it for another 2 days.
However, this does not necessarily mean that after two days the package 
will migrate; there very well might be other issues with this package, I 
did not study the QA page very long, I don't know.


I don't believe you should give wheezy-backports a higher ranking, as 
the backports webpage[2] itself suggests to only update select packages.


As a different question, could anyone please tell me why 
wheezy-backports is not showing up on packages.debian.org?


Regards,
theartloy

[1] http://packages.qa.debian.org/libr/libreoffice.html
[2] http://backports.debian.org/


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5198a9da.5060...@zoho.com

Re: apt-get online package repository configuration on Debian

[Brian]

On Sun 19 May 2013 at 11:37:45 +0300, Andrei POPESCU wrote:

> On Sb, 18 mai 13, 19:01:38, Steve Faleiro wrote:
> > 
> > 1. Add the following line to the file etc/apt/sources.list :
> > 
> >deb http://ftp.debian.org/debian wheezy main
> 
> Depending on where you live, using the country specific mirror (or a 
> mirror next to you[1]) might improve download speeds significantly[2].
> 
> [1] http://www.debian.org/mirror/list
> [2] which is why one should let the installer choose a mirror for you 
> based on the country you selected

A mention of the redirector

   http://http.debian.net/ 

might not be out of place. It saves having to think about using a
particular country mirror and can be entered manually in d-i,


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519105402.GA23758@desktop

Re: Error install postgresql-9.1

[Selim T. Erdogan]

Fernando ff77, 17.05.2013:
> hello,
> i update my server from sarge to wheezy (stable). (dist-upgrade)

Are you really trying to go from *sarge* to wheezy?  Sounds like asking 
for trouble.

> All package is ok... thanks aptitude !!!
> 
> The problem is postgresql-9.1, during the installation i read this error...
> 
> Configurazione di postgresql-9.1 (9.1.9-1)...
> [] Starting PostgreSQL 9.1 database server: main[] Use of
> uninitialized value $logsize in numeric gt (>) at /usr/bin/pg_ctlcluster
> line 215. Use of uninitialized value $logsize in numeric gt (>) at
> /usr/bin/pg_ctlcluster line 215. Use of uninitialized value $logsize in
> numeric gt (>) at /usr/bin/pg_ctlcluster line 215. Use of uninitialized
> value $logsize in numeric gt (>) at...
> 
> invoke-rc.d: initscript postgresql, action "start" failed.
> dpkg: errore nell'elaborare postgresql-9.1 (--configure):
>  il sottoprocesso installato script di post-installation ha restituito lo
> stato di errore 1
> Si sono verificati degli errori nell'elaborazione:
>  postgresql-9.1

Can you try removing the package and reinstalling it?  Maybe the new 
version's post-install script is looking for something that doesn't 
exist in your old version from sarge.

> I searched on Google but have not found a solution.

A very quick google search on your error message showed up some 
cases when a file didn't exist, but I'm not sure how similar the 
situations were to yours.  Do you know where $logsize is supposed to 
come from?


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519112205.gc29...@cs.utexas.edu

Re: How to investigate system hangs?

[Selim T. Erdogan]

Dawid Toton, 16.05.2013:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
> 
> I am running wheezy. It worked perfectly for a year or two.
> But yesterday (and today again) my system got deadlocked. I'm not sure
> where to look for causes.
> Both events went like this:
> * I open the lid, the system resumes
> * I enter the password, I get applications apparently working
> * after short time, like few mouse clicks, everything becomes unresponsive
> * the mouse cursor can move
> * the system is still able to log USB mouse plug event, however the
> plugged mouse is apparently not listened to
> * ctrl+alt+f1 doesn't bring me a terminal (even after SysRq+R)
> * SysRq commands are logged and executed, so I can reboot with it

I wonder if it's Xorg that's having problems.  Can you go stop X and 
go to a virtual console before suspending?  If resuming from that works, 
that would mean it's Xorg.

Also, if you can use another computer to connect to the laptop after 
it freezes, you could try killing Xorg.

Another thing you can test is to just use ctrl+alt+f1 to go to a 
terminal as soon as you resume.  Then wait a while to see if the freeze 
happens even when you're sitting in the console.

> I have looked at nearly everything at /var/log, but I see nothing
> suspiscious for my untrained eye.
> 
> May I ask some advice? Do you know of some recent major changes that
> would e.g. affect some WiFi operation or something else that activates
> few seconds after system resume?
> How do I check what package upgrades preceded the hangs?

You could try /var/log/dpkg.log, /var/log/dpkg.log.1, etc.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519110125.gb29...@cs.utexas.edu

Re: FQDN via DHCP, then used in exim4

[Arun Khan]

On Sun, May 19, 2013 at 2:15 PM, Klaus Doering
 wrote:
>
> I agree that in a different setting, where there are many users,
> hundreds of emails per minute and other mission-critical stuff is going
> on, one needs to design the infrastructure a lot more carefully.
>

As a thumb rule, any system providing a service (e.g. email) ought to
be configured with static IP.  With email, you also need a DNS with a
MX resolving to the hostname/IP number of your email server.

>
> So, the initial question remains unanswered: what happens to the
> information provided by the dhcp server as reported in the lease file in
> /var/lib/dhcp, how is it accessed, and is there a way to make exim4 use
> it?
>

I use postfix, so cannot comment about exim.  In postfix, to the best
of my understanding, there is no provision to get network parameter
information from dhclient files.  I suspect the same is the case with
exim.

-- 
Arun Khan
Sent from my non-iphone/non-android device


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAHhM8gC0cf3d-f724is5XHJiyr9QRMZ=4-trqrw2sxefaot...@mail.gmail.com

No date in "uname -a" anymore

[Markus Schönhaber]

Hi all,

if I do this

~# uname -a
Linux squeeze 2.6.32-5-xen-amd64 #1 SMP Mon Feb 25 02:51:39 UTC 2013
x86_64 GNU/Linux
~# ls -l /boot/vm*
-rw-r--r-- 1 root root 2482528 May 10 15:32 /boot/vmlinuz-2.6.32-5-xen-amd64
~#

on a machine running squeeze, it is trivially easy to see that it
doesn't run the installed kernel, just by comparing the dates. The
machine hasn't been rebooted since the last update of the kernel package.

On wheezy this is a different matter:

~# uname -a
Linux wheezy 3.2.0-4-amd64 #1 SMP Debian 3.2.41-2 x86_64 GNU/Linux
~# ls -l /boot/vm*
-rw-r--r-- 1 root root 2833376 May 15 23:58 /boot/vmlinuz-3.2.0-4-amd64
~#

There is no date in the uname output anymore, which I could compare to
the kernel image file's timestamp.

Can someone please explain what exactly the "improvement" is in removing
information from uname's output?

If I'd like to have taken away from me stuff of which others have
decided that I don't need it, I'd still use GNOME.


And yes, I've seen
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=694565#10
but
| The package version identifies what source was used and is thus a lot
| more useful than the build date.
doesn't sound convincing to me. As an administrator, I couldn't care
less what source was used to build the kernel package installed.

Oh, and what is nowadays the recommended method to determine whether or
not the machine is running the kernel from the latest package update?

-- 
Regards
  mks


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5198c146.5070...@list-post.mks-mail.de

Odd Network Problem

[george cox]

I have a piece of equipment that when conntected to my cablebox/TV allows me to 
view my home TV on any computer from anywhere on the internet. This device only 
has a hardwired internet connection so I use a wireless print-server (it has a 
4 ethernet ports that it bridges to the wireless) to connect the device to my 
home wireless network. Friday I changed my internet provider, and needed to 
reconfigure the print-server to use the new wireless network.

The problem was I could not get my laptop to connect to the printer-server (via 
a hardwired cable). I tried letting my laptop auto configure, I tried setting 
the address manually (ifconfig). None of this worked I could not get connected 
to the print-server. I turned off the wireless network on my laptop so the 
hardwired ethernet was the only NIC available. When plugging in the ethernet 
cable, the network manager would grid away for 3 or 4 minutes but the 
connection would always fail.

I bought a my laptop last week and installed wheezy on it. I still have the old 
one and have not updated, so it still runs squeeze. I pulled out my old laptop, 
plugged-in the ethernet cable to the print-server and entered the same ifconfig 
commands to set the IP address manually and viola, I could connect to the 
print-server and modified its settings with no problem.

Any idea why one laptop worked while the other didn't.

As a side note, in the network manager on the old laptop with squeeze, if you 
left-clicked one of the options listed there is "auto eth0", this doesn't seem 
to exist on the network manager on the new wheezy laptop. Is there an 
equivalent function in the wheezy version of the GUI? I have used "auto eth0" 
several times to connect to various hardwired equipment and it comes in handy. 
Does anyone know exactly what auto eth0 does to the ethernet port?

Thanks.

Re: Grub2 gives error :File not found and goes into rescue mode

[Sebastian Canagaratna]

Hi Igor:
The Computer was bootoing off the mbr, though I think I made
the vista partition /dev/sda1 the debian stable partition /de v/sda2 as
well as the debian unsable partition  /dev/sdb2 bootable. This happened
when I was upgrading from 3.8.10-slh-aptosid-686 to the the 3.9-2.slh.1
verdion. The latter must be OK because when I boot from the suoer grub disk
everything is OK. Either the MBR is damaged ( when I grub install into
/dev/sda eveything is sussessful and no error message is reported or there
must be a bug in the version of grub which comes with the latest
distribtion.

As I said earlier, I blanked out the first 512 sectors of /dev/sda and  no
error is reported whenI grub-install.There is no /etc/grub in debian
aptosid. all the files reside is /boot/grub. grub.gfg is here. I di d a
grub-update and I verified that the UUID,s were fine, Grub is not finding a
file, because the path has presumably got messed up. Afre ther any log
files to llok at?

Thanks

Sebastian









On Sat, May 18, 2013 at 8:37 PM, Igor Cicimov  wrote:

>
> On 19/05/2013 3:12 AM, "Sebastian Canagaratna" 
> wrote:
> >
> > Hi : I have a hard disk /dev/sda and a removable disk /dev/sdb The 1st
> partition
> > is a vista partition, but this has been inoperable for some time
> /dev/sda2 has a linux
> > stable partition /dev/sdb1 has linux unstable (aptosid) on it, Yesterday
> I was upgrading to Linux 3.9.2  and thereafter I got this error. The
> updrade did nt\ot go to
> > completion. I have a super frub rescue disk with which I can start
> 3.9.2, but I am unable to repair the grub with update grub and grub-install
> /dev/sda.
>
> So which partition is marked as bootable? Whats your boot order in BIOS?
> Which linux did you upgrade? Which grub are you trying to install and
> where? What do you have in /etc/grub/grub.cfg?
>
> I have blanked out the first 512 sectors of dev/sda but that didn't help.
> Presumably grub is unabke
> > to find grub.cfg.Trying to run grub-install with other versions of Linux
> did not help. s this a bug in grub  or is there something I can do to
> overcome this. I cab see in
> > Google thaqt others have had this file not found error. I have tried
> every likely thing ,but I am aksways left in rescue mode when booting, I'd
> appreciate any suggestion
> > Are there any logd which I can look at?
> >
> > Sebastian
>

Re: FQDN via DHCP, then used in exim4

[Klaus Doering]

On 19/05/13 12:48, Arun Khan wrote:

On Sun, May 19, 2013 at 2:15 PM, Klaus Doering
 wrote:

I agree that in a different setting, where there are many users,
hundreds of emails per minute and other mission-critical stuff is going
on, one needs to design the infrastructure a lot more carefully.


As a thumb rule, any system providing a service (e.g. email) ought to
be configured with static IP.  With email, you also need a DNS with a
MX resolving to the hostname/IP number of your email server.


So, the initial question remains unanswered: what happens to the
information provided by the dhcp server as reported in the lease file in
/var/lib/dhcp, how is it accessed, and is there a way to make exim4 use
it?


I use postfix, so cannot comment about exim.  In postfix, to the best
of my understanding, there is no provision to get network parameter
information from dhclient files.  I suspect the same is the case with
exim.


Thanks Arun,

This email server is not directly connected to the 'net, it sits behind
a router. Thus, there is one external IP for which I've registered an "A"
record and an "MX" record on a public DNS server, and then there is an
internal IP server on my LAN. The later one is fixed, but is not defined
in a config file on the server, but is "reserved" in the DHCP server and
assigned to the server. Same for the domain name, which internally
(i.e. on my LAN) is distributed by the DHCP server. In "dnsmasq" this
would be the "domain=" option. And it is this option that is not
acknowledged by exim.

Klaus


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5198cab5.9080...@gmail.com

Re: libdigest-sha1-perl

[Marcus Karlsson]

> Hi,
>
> I'm using Wheezy, but needed ddclient from Sid, as that version supports
> freedns protocol. I know there are other programs/scripts, but I use ddclient
> on Slackware and just wanted to have same setup. As it turns out, that version
> needs libdigest-sha1-perl which, as I understood, is dropped in Wheezy in 
> favor
> of libdigest-sha-perl.

That's a bit odd, Digest::SHA is bundled as part of Perl, and when looking
at that package [1] there's no mention of libdigest-sha1-perl. It sounds
like a bug if the package actually requires Digest::SHA1.

> Don't know why is it in Sid then... Anyhow, to solve
> this problem I just edited part of /usr/sbin/ddclient that checks for version
> and loads SHA1 module and replaced it with Digest::SHA module.
>
> It works just fine, but as I do not know perl or Debian that well, I don't 
> know
> if this was right way to do it. Any thougts?

If you installed /usr/sbin/ddclient as part of package then things might
break if the package is updated. I would do a separate install (eg. copy
the binary and purge the package), or use dpkg-diverge to override the
packaged version with your modified copy.


[1] http://packages.debian.org/sid/ddclienty


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/874ndzf9rr@acc.umu.se

Re: No date in "uname -a" anymore

[Claudius Hubig]

Dear Markus,

Markus Schönhaber wrote:
> ~# uname -a
> Linux wheezy 3.2.0-4-amd64 #1 SMP Debian 3.2.41-2 x86_64 GNU/Linux
> ~# ls -l /boot/vm*
> -rw-r--r-- 1 root root 2833376 May 15 23:58 /boot/vmlinuz-3.2.0-4-amd64
> ~#
> 
> There is no date in the uname output anymore, which I could compare to
> the kernel image file's timestamp.

You are now supposed to compare the Debian package version reported
by uname (3.2.41-2 in the example above) against the one currently
installed (e.g. using dpkg -l). /proc/version still reports the build
time for me, though.

> doesn't sound convincing to me. As an administrator, I couldn't care
> less what source was used to build the kernel package installed.

As the source package used to build the kernel uniquely identifies the
kernel, you should only care about the source package version?
 
Best,

Claudius
-- 
Please don’t CC me.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519134126.46c7a...@ares.home.chubig.net

Re: libdigest-sha1-perl

[Celejar]

On Sun, 19 May 2013 12:21:37 +0200
Veljko  wrote:

> 
> Hi,
> 
> I'm using Wheezy, but needed ddclient from Sid, as that version supports
> freedns protocol. I know there are other programs/scripts, but I use ddclient
> on Slackware and just wanted to have same setup. As it turns out, that version
> needs libdigest-sha1-perl which, as I understood, is dropped in Wheezy in 
> favor

I don't see a direct dependency upon any libdigest*; what's the dependency
chain?

> of libdigest-sha-perl. Don't know why is it in Sid then... Anyhow, to solve

In Sid, it seems to be available only for m68k, and not for most
architectures.

> this problem I just edited part of /usr/sbin/ddclient that checks for version
> and loads SHA1 module and replaced it with Digest::SHA module.
> 
> It works just fine, but as I do not know perl or Debian that well, I don't 
> know
> if this was right way to do it. Any thougts?
> 
> Regards,
> Veljko

Celejar


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20130519085751.9e35c62d4a8fdef6378a3...@gmail.com

Re: Odd Network Problem

[Klaus Doering]

On 19/05/13 13:30, george cox wrote:
I have a piece of equipment that when conntected to my cablebox/TV 
allows me to view my home TV on any computer from anywhere on the 
internet. This device only has a hardwired internet connection so I 
use a wireless print-server (it has a 4 ethernet ports that it bridges 
to the wireless) to connect the device to my home wireless network. 
Friday I changed my internet provider, and needed to reconfigure the 
print-server to use the new wireless network.


The problem was I could not get my laptop to connect to the 
printer-server (via a hardwired cable). I tried letting my laptop auto 
configure, I tried setting the address manually (ifconfig). None of 
this worked I could not get connected to the print-server. I turned 
off the wireless network on my laptop so the hardwired ethernet was 
the only NIC available. When plugging in the ethernet cable, the 
network manager would grid away for 3 or 4 minutes but the connection 
would always fail.


I bought a my laptop last week and installed wheezy on it. I still 
have the old one and have not updated, so it still runs squeeze. I 
pulled out my old laptop, plugged-in the ethernet cable to the 
print-server and entered the same ifconfig commands to set the IP 
address manually and viola, I could connect to the print-server and 
modified its settings with no problem.


Any idea why one laptop worked while the other didn't.

As a side note, in the network manager on the old laptop with squeeze, 
if you left-clicked one of the options listed there is "auto eth0", 
this doesn't seem to exist on the network manager on the new wheezy 
laptop. Is there an equivalent function in the wheezy version of the 
GUI? I have used "auto eth0" several times to connect to various 
hardwired equipment and it comes in handy. Does anyone know exactly 
what auto eth0 does to the ethernet port?


Thanks.


George

This could still be a network config issue. An easy way around might
be to connect (wired) both, the print server and the new laptop, to your
router (assuming that the router also acts as the dhcp server, and that both
clients are configured to get their ip address through dhcp). In the
router's  log you can then see the connected devices and their
respective IP addresses. Can you connect from laptop to print server now?

Klaus


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5198d2de.3040...@gmail.com

Re: No date in "uname -a" anymore

[Markus Schönhaber]

Thanks for your answer, Claudius!

19.05.2013 14:41, Claudius Hubig:

> Markus Schönhaber wrote:
>> ~# uname -a
>> Linux wheezy 3.2.0-4-amd64 #1 SMP Debian 3.2.41-2 x86_64 GNU/Linux
>> ~# ls -l /boot/vm*
>> -rw-r--r-- 1 root root 2833376 May 15 23:58 /boot/vmlinuz-3.2.0-4-amd64
>> ~#
>>
>> There is no date in the uname output anymore, which I could compare to
>> the kernel image file's timestamp.
> 
> You are now supposed to compare the Debian package version reported
> by uname (3.2.41-2 in the example above) against the one currently
> installed (e.g. using dpkg -l).

OK, that is a way. To me, this seems to be a very awkward way, though.

> /proc/version still reports the build
> time for me, though.

Alas, for me it doesn't.

>> doesn't sound convincing to me. As an administrator, I couldn't care
>> less what source was used to build the kernel package installed.
> 
> As the source package used to build the kernel uniquely identifies the
> kernel, you should only care about the source package version?

Someone seems to think I should care. They think I should care on wheezy
- but not on squid or any other Linux distribution I have to cope with.
Well, I'll continue to not care (mostly, at least)...

Anyway, it is still way beyond me how this change can be thought of as
an improvement.
I could well live with it if the oh-so-important source package version
was *added* to uname's output instead of being put in the place of the
info I really *do* care for. But as it is, someone has made a change
that gains me nothing but makes my life a tiny bit harder.

-- 
Regards
  mks


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5198dd3a.2070...@list-post.mks-mail.de

Re: FQDN via DHCP, then used in exim4

[Andrei POPESCU]

On Du, 19 mai 13, 13:51:01, Klaus Doering wrote:
> 
> This email server is not directly connected to the 'net, it sits behind
> a router. Thus, there is one external IP for which I've registered an "A"
> record and an "MX" record on a public DNS server, and then there is an
> internal IP server on my LAN. The later one is fixed, but is not defined
> in a config file on the server, but is "reserved" in the DHCP server and
> assigned to the server. Same for the domain name, which internally
> (i.e. on my LAN) is distributed by the DHCP server. In "dnsmasq" this
> would be the "domain=" option. And it is this option that is not
> acknowledged by exim.

Let me try to put it differently: because server software (including if 
not especially mail servers) are usually on static IPs server software 
does not need to reconfigure itself based on information from a DHCP 
server. You would have to write your own scripts to achieve that.

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

wheezy-backports not on p.d.o? [was: Re: Libreoffice4 in wheezy? Several questions]

[Andrei POPESCU]

On Du, 19 mai 13, 11:30:50, theart...@zoho.com wrote:
> 
> As a different question, could anyone please tell me why
> wheezy-backports is not showing up on packages.debian.org?

No, but we can ask the people taking care of it ;)

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

Re: Libreoffice4 in wheezy? Several questions

[Andrei POPESCU]

On Du, 19 mai 13, 11:39:30, Hans-J. Ullrich wrote:
> Hello all,
> 
> I wanted to install LO4 in wheezy. I wondered, why it is in wheezy-backports 
> but not in testing. Maybe the packages in wheezy-backports are too young for 
> testing? 
> 
> However, I managed it by using aptitude and made sure, all dependencies are 
> correct.
> 
> But I guess, there is an easier way, to update my ONLY my installed packages 
> out of wheezy. In the wiki I read, you can use apt-get -t wheezy-backports 
> install $package, which worked well.
 
Not sure what you mean here, could you please rephrase/elaborate?

> And of course, i can edit the apt pinning system to give wheezy-backports a 
> higher ranking. But wouldn't this affect other packages, too?

Yes it would and it isn't necessary, the defaults should work fine.

> I suppose, there might be a way just to update certain packages from wheezy-
> backports using the pinning system, but I fear, I can use this only for meta-
> packages or directly named packages. Please correct me, if I am wrong.

In the default configuration packages that you choose to install from 
backports (via -t) will be kept updated, but no other packages from 
backports will be installed. Does this answer your question?

> And last question: If Libreoffice4 will be put into testing repo, will those 
> packages then be newer than those in wheezy-backports or is this a matter of 
> chance?

Normally a package has to be in testing first and only then backported 
(to ensure some testing). The package in backports has a special version 
so that apt/dpkg always consider the package in testing to be newer 
(even if the software itself is at the same version).

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

Re: Libreoffice4 in wheezy? Several questions

[Hans-J. Ullrich]

Hi Andrei, 
> > But I guess, there is an easier way, to update my ONLY my installed
> > packages out of wheezy. In the wiki I read, you can use apt-get -t
> > wheezy-backports install $package, which worked well.
> 
> Not sure what you mean here, could you please rephrase/elaborate?
> 

No, not quite. I imagined something like "apt-get -t update wheezy-backports" 
and then "apt-get upgrade libreoffice-* | grep installed" or some similar 
command. Just to make sure, to update ONLY the libreoffice-packages (and libs) 
out of wheezy-backports. I suppose, you know, what I mean

 
> 
> In the default configuration packages that you choose to install from
> backports (via -t) will be kept updated, but no other packages from
> backports will be installed. Does this answer your question?
> 
> > And last question: If Libreoffice4 will be put into testing repo, will
> > those packages then be newer than those in wheezy-backports or is this a
> > matter of chance?
> 
> Normally a package has to be in testing first and only then backported
> (to ensure some testing). The package in backports has a special version
> so that apt/dpkg always consider the package in testing to be newer
> (even if the software itself is at the same version).

Yes, this is , what I thought, too. But I learned, that this is not the case. 
In testing the libreoffice version is 3.5.4, whilst in wheezy-backports already 
is 4.0! IMO this is strange, as testing IMO should be higher and newer 
versions than in stable. But in real, libreoffice4 is in unstable(!), and from 
unstable ported to backports. I do not quite understand the policy in this 
behaviour, sorry. :)

Best regards

Hans
> 
> Kind regards,
> Andrei



-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201305191701.52429.hans.ullr...@loop.de

Why can ubuntu dist upgrade dist from within an X session?

[Πρεκατές Αλέξανδρος]

In release notes it says:

"You should not upgrade using telnet, rlogin, rsh, or from an X session 
managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is 
because each of those services may well be terminated during the upgrade, 
which can result in an inaccessible system that is only half-upgraded. Use of 
the GNOME application update-manager is strongly discouraged for upgrades to 
new releases, as this tool relies on the desktop session remaining active. "|

So i wonder why can ubuntu dist upgrade dist from within an X session?


aprekates



-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201305191811.56837.apreka...@gmail.com

Re: No date in "uname -a" anymore

[Claudius Hubig]

Dear Markus,

Markus Schönhaber wrote:
> Thanks for your answer, Claudius!

Welcome :)

> 19.05.2013 14:41, Claudius Hubig:
> > You are now supposed to compare the Debian package version reported
> > by uname (3.2.41-2 in the example above) against the one currently
> > installed (e.g. using dpkg -l).
> 
> OK, that is a way. To me, this seems to be a very awkward way, though.

It doesn’t break if you move files around, and at least to me
comparing version numbers is per se more reasonable than comparing
build dates, but YMMV, of course.

> > /proc/version still reports the build
> > time for me, though.
> 
> Alas, for me it doesn't.

Strange. I am running a self-compiled 3.7 as follows:

$ cat /proc/version 
Linux version 3.7.1.a2017.3 (root@ares) (gcc version 4.7.2 (Debian 4.7.2-4) ) 
#3 SMP Sun Jan 13 16:24:52 GMT 2013

Maybe there is an option somewhere in the kernel compile-time
configuration?

> I could well live with it if the oh-so-important source package version
> was *added* to uname's output instead of being put in the place of the
> info I really *do* care for. But as it is, someone has made a change
> that gains me nothing but makes my life a tiny bit harder.

If I recall correctly, there is some length-limit imposed on this
string (for whatever reason), so merely adding it was not an option.

Best,

Claudius
-- 
Please don’t CC me.


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519160707.47bd0...@ares.home.chubig.net

Re: Odd Network Problem

[george cox]

This could still be a network config issue. An easy way around might be to 
connect (wired) both, the print server and the new laptop, to your router 
(assuming that the router also acts as the dhcp server, and that both clients 
are configured to get their ip address through dhcp). In the router's log you 
can then see the connected devices and their respective IP addresses. Can you 
connect from laptop to print server now? Klaus I don't think that can work. The 
print-server doesn't have the ability to connect it to the router through a 
wired port, it just has the wireless. The ethernet ports are only used to 
bridge other non-wireless equipment through the printer-server to whatever 
wireless network the printer-server is configured to use. The rub is the only 
way to configure what wireless network the print-server uses is to put a 
computer on one of those ethernet ports, hold the reset button on the 
print-server resetting it to factory defaults, then configure the wireless via 
a 
 web-page generated by the print-server. What seems weird to me, is using 
ifconfig command to set an IP worked on the squeeze laptop, but didn't on the 
wheezy laptop. Both systems seem to take the ip and the output of running 
ifconfig -a looked the same on both. It was just that on one box I could 
connect to the print-servers webpage and the other I couldn't. On both I took 
care to disconnect from all other networks, so I wasn't a routing issue.

Re: Why can ubuntu dist upgrade dist from within an X session?

[Ralf Mardorf]

On Sun, 2013-05-19 at 17:51 +0200, Ralf Mardorf wrote:
> On Sun, 2013-05-19 at 18:11 +0300, Πρεκατές Αλέξανδρος wrote:
> > In release notes it says:
> > 
> > "You should not upgrade using telnet, rlogin, rsh, or from an X session 
> > managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is 
> > because each of those services may well be terminated during the upgrade, 
> > which can result in an inaccessible system that is only half-upgraded. Use 
> > of 
> > the GNOME application update-manager is strongly discouraged for upgrades 
> > to 
> > new releases, as this tool relies on the desktop session remaining active. 
> > "|
> > 
> > So i wonder why can ubuntu dist upgrade dist from within an X session?
> 
> I don't know what's different, but in general you for example can't
> backup a complete Linux, while it's running, a way to do it anyway, is
> to make a snapshot first and to backup the snapshot, independent from
> the system. In general you need to reboot to switch to a new kernel, but
> there's a tool that enables to switch the kernel without the need to
> restart the computer. IOW, they perhaps have programmed something to
> enable this and FWIW, it might be related or not, they don't use
Oops, Ubuntu does use upstart ;), a typo.
> upstart, so a basic component for the startup process already is
> completely different to Debian. Perhaps files for services such as
> display managers will be stored temporary and copied to the correct
> location during a startup and again, upstart is a completely newer
> process for startup than Debian's init. How many distros nowadays stay
> with this kind of init? I suspect nearly every distro does use upstart
> or systemd. You can't compare something old-fashioned like Debian, with
> distros that are more modern, closer to upstream. OTOH something e.g.
> running in the RAM can be updated and there e.g. is no need to stop X to
> update a display manager for a regular backup. What's the source of your
> quotation?
> 
> Regards,
> Ralf



-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1368978816.4892.26.camel@archlinux

Re: Why can ubuntu dist upgrade dist from within an X session?

[Ralf Mardorf]

On Sun, 2013-05-19 at 18:11 +0300, Πρεκατές Αλέξανδρος wrote:
> In release notes it says:
> 
> "You should not upgrade using telnet, rlogin, rsh, or from an X session 
> managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is 
> because each of those services may well be terminated during the upgrade, 
> which can result in an inaccessible system that is only half-upgraded. Use of 
> the GNOME application update-manager is strongly discouraged for upgrades to 
> new releases, as this tool relies on the desktop session remaining active. "|
> 
> So i wonder why can ubuntu dist upgrade dist from within an X session?

I don't know what's different, but in general you for example can't
backup a complete Linux, while it's running, a way to do it anyway, is
to make a snapshot first and to backup the snapshot, independent from
the system. In general you need to reboot to switch to a new kernel, but
there's a tool that enables to switch the kernel without the need to
restart the computer. IOW, they perhaps have programmed something to
enable this and FWIW, it might be related or not, they don't use
upstart, so a basic component for the startup process already is
completely different to Debian. Perhaps files for services such as
display managers will be stored temporary and copied to the correct
location during a startup and again, upstart is a completely newer
process for startup than Debian's init. How many distros nowadays stay
with this kind of init? I suspect nearly every distro does use upstart
or systemd. You can't compare something old-fashioned like Debian, with
distros that are more modern, closer to upstream. OTOH something e.g.
running in the RAM can be updated and there e.g. is no need to stop X to
update a display manager for a regular backup. What's the source of your
quotation?

Regards,
Ralf


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1368978695.4892.25.camel@archlinux

Re: Libreoffice4 in wheezy? Several questions

[Lisi Reisz]

On Sunday 19 May 2013 16:01:51 Hans-J. Ullrich wrote:
> > > But I guess, there is an easier way, to update my ONLY my installed
> > > packages out of wheezy. In the wiki I read, you can use apt-get -t
> > > wheezy-backports install $package, which worked well.
> >
> > Not sure what you mean here, could you please rephrase/elaborate?
>
> No, not quite. I imagined something like "apt-get -t update
> wheezy-backports" and then "apt-get upgrade libreoffice-* | grep installed"
> or some similar command. Just to make sure, to update ONLY the
> libreoffice-packages (and libs) out of wheezy-backports. I suppose, you
> know, what I mean

No, you do not need to do any such thing.  Leave the settings as they are by 
default.  Don't change any preferences or pinning from what was set 
automatically.  Your backported packages, _and_ _only_  your backported 
packages will be updated from wheezy-backports.

If you can just accept the system and not fuss about it, it will work fine.  
Sid and Jessie are in a certain amount of chaos at the moment because Wheezy 
has only just been released as Stable.  They will settle down, and everything 
will be fine, as will your installation, provided that you do not try to 
fiddle with it.

Lisi


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201305191705.40418.lisi.re...@gmail.com

Re: Why can ubuntu dist upgrade dist from within an X session?

[Lisi Reisz]

On Sunday 19 May 2013 16:11:56 Πρεκατές Αλέξανδρος wrote:
> In release notes it says:
>
> "You should not upgrade using telnet, rlogin, rsh, or from an X session
> managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is
> because each of those services may well be terminated during the upgrade,
> which can result in an inaccessible system that is only half-upgraded. Use
> of the GNOME application update-manager is strongly discouraged for
> upgrades to new releases, as this tool relies on the desktop session
> remaining active. "|
>
> So i wonder why can ubuntu dist upgrade dist from within an X session?

What has what Ubuntu does got to do with Debian?  

Ubuntu seems to be deliberately moving away from the CLI and towards the GUI.  
Debian is not.

Lisi


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201305191708.40479.lisi.re...@gmail.com

Re: Odd Network Problem

[Lisi Reisz]

On Sunday 19 May 2013 16:43:31 george cox wrote:
> This could still be a network config issue. An easy way around might be to
> connect (wired) both, the print server and the new laptop, to your router
> (assuming that the router also acts as the dhcp server, and that both
> clients are configured to get their ip address through dhcp). In the
> router's log you can then see the connected devices and their respective IP
> addresses. Can you connect from laptop to print server now? Klaus I don't
> think that can work. The print-server doesn't have the ability to connect
> it to the router through a wired port, it just has the wireless. The
> ethernet ports are only used to bridge other non-wireless equipment through
> the printer-server to whatever wireless network the printer-server is
> configured to use. The rub is the only way to configure what wireless
> network the print-server uses is to put a computer on one of those ethernet
> ports, hold the reset button on the print-server resetting it to factory
> defaults, then configure the wireless via a web-page generated by the
> print-server. What seems weird to me, is using ifconfig command to set an
> IP worked on the squeeze laptop, but didn't on the wheezy laptop. Both
> systems seem to take the ip and the output of running ifconfig -a looked
> the same on both. It was just that on one box I could connect to the
> print-servers webpage and the other I couldn't. On both I took care to
> disconnect from all other networks, so I wasn't a routing issue.

I am interested in your problem and might even be able to help, but I cannot 
cope with this.  You have not quoted what you are replying to, and there is 
no air here at all.  I could, of course, reformat it, but I would still not 
have the quotation.

You will of course get help from those who are not bothered by such things. 
But you might get more answers if you made yourself more accessible.

Lisi


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201305191712.54992.lisi.re...@gmail.com

Re: Libreoffice4 in wheezy? Several questions

[Lisi Reisz]

On Sunday 19 May 2013 17:05:40 Lisi Reisz wrote:
> Your backported packages, _and_ _only_  your backported
> packages will be updated from wheezy-backports.

Sorry, upgraded.

Lisi


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201305191716.24764.lisi.re...@gmail.com

Re: FQDN via DHCP, then used in exim4

[Arun Khan]

On Sun, May 19, 2013 at 6:21 PM, Klaus Doering
 wrote:
>
> On 19/05/13 12:48, Arun Khan wrote:
>>
>> On Sun, May 19, 2013 at 2:15 PM, Klaus Doering
>>  wrote:
>>>
>>> I agree that in a different setting, where there are many users,
>>> hundreds of emails per minute and other mission-critical stuff is going
>>> on, one needs to design the infrastructure a lot more carefully.
>>>
>> As a thumb rule, any system providing a service (e.g. email) ought to
>> be configured with static IP.  With email, you also need a DNS with a
>> MX resolving to the hostname/IP number of your email server.
>>
>>> So, the initial question remains unanswered: what happens to the
>>> information provided by the dhcp server as reported in the lease file in
>>> /var/lib/dhcp, how is it accessed, and is there a way to make exim4 use
>>> it?
>>>
>> I use postfix, so cannot comment about exim.  In postfix, to the best
>> of my understanding, there is no provision to get network parameter
>> information from dhclient files.  I suspect the same is the case with
>> exim.
>>
> Thanks Arun,
>
> This email server is not directly connected to the 'net, it sits behind
> a router. Thus, there is one external IP for which I've registered an "A"
> record and an "MX" record on a public DNS server, and then there is an
> internal IP server on my LAN. The later one is fixed, but is not defined
> in a config file on the server, but is "reserved" in the DHCP server and
> assigned to the server. Same for the domain name, which internally
> (i.e. on my LAN) is distributed by the DHCP server. In "dnsmasq" this
> would be the "domain=" option. And it is this option that is not
> acknowledged by exim.
>

Whether a server sits on the WAN or an internal LAN (with no
visibility on the WAN),  the  principles of configuration are the
same.  Instead of public route-able IPs you would be dealing with
private IPs.

The FQDN is extracted from /etc/hosts entry and hostname from /etc/hostname [a]

[a] 


IMO, you would be better off configuring the system with static ip and
FQDN in /etc/hosts.

As others have pointed out, this is the standard practice for server setup.

Alternately, you could query on the exim mailing list to resolve your
requirement.

-- 
Arun Khan
Sent from my non-iphone/non-android device


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/cahhm8gaokwkodocucs_f2qwgbwq8c+861eqxucqcx0wh1ti...@mail.gmail.com

Preseed with wheezy overwritting interfaces file

[ML mail]

  
Hello,

I
 am installing wheezy automatically by using preseed and am using its 
late_command feature to run a script at the end of the installation. 
This script is actually modifying the /etc/network/interfaces file to 
setup bonding and VLAN access. This all used to work fine on squeeze but
 now with wheezy it looks like the /etc/network/interfaces file gets 
rewritten after the late_command of preseed.

Does anyone know more about this or have any tips?

Regards.
John


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1368983570.43112.yahoomail...@web164603.mail.gq1.yahoo.com

Re: Why can ubuntu dist upgrade dist from within an X session?

[Brian]

On Sun 19 May 2013 at 18:11:56 +0300, Πρεκατές Αλέξανδρος wrote:

> In release notes it says:
> 
> "You should not upgrade using telnet, rlogin, rsh, or from an X session 
> managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is 
> because each of those services may well be terminated during the upgrade, 
> which can result in an inaccessible system that is only half-upgraded. Use of 
> the GNOME application update-manager is strongly discouraged for upgrades to 
> new releases, as this tool relies on the desktop session remaining active. "|
> 
> So i wonder why can ubuntu dist upgrade dist from within an X session?

One imagines killing X during an upgrade is avoided by not restarting
gdm (for example). Looking at the postinst, postrm and prerm files and
what is run in debian/rules is very likely to reduce your wondering.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519172407.GB23758@desktop

Re: Libreoffice4 in wheezy? Several questions

[Andrei POPESCU]

On Du, 19 mai 13, 17:01:51, Hans-J. Ullrich wrote:
> > 
> > Normally a package has to be in testing first and only then backported
> > (to ensure some testing). The package in backports has a special version
> > so that apt/dpkg always consider the package in testing to be newer
> > (even if the software itself is at the same version).
> 
> Yes, this is , what I thought, too. But I learned, that this is not the case. 
> In testing the libreoffice version is 3.5.4, whilst in wheezy-backports 
> already 
> is 4.0! IMO this is strange, as testing IMO should be higher and newer 
> versions than in stable. But in real, libreoffice4 is in unstable(!), and 
> from 
> unstable ported to backports. I do not quite understand the policy in this 
> behaviour, sorry. :)

In special cases the backports ftpmasters do accept packages before the 
equivalent version in unstable has migrated to testing. My guess is that 
in this particular case it is because libreoffice is such an important 
package to backport and because the Maintainer has a good history of 
providing quality packages.

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

Re: Why can ubuntu dist upgrade dist from within an X session?

[Andrei POPESCU]

On Du, 19 mai 13, 18:11:56, Πρεκατές Αλέξανδρος wrote:
> In release notes it says:
> 
> "You should not upgrade using telnet, rlogin, rsh, or from an X session 
> managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is 
> because each of those services may well be terminated during the upgrade, 
> which can result in an inaccessible system that is only half-upgraded. Use of 
> the GNOME application update-manager is strongly discouraged for upgrades to 
> new releases, as this tool relies on the desktop session remaining active. "|
> 
> So i wonder why can ubuntu dist upgrade dist from within an X session?

I have no idea what Ubuntu can or can not do[1], but I can assure you it 
is safer not to do it from within an X session. Besides, do you know how 
many ways there are in Debian to start an X session? I know of at least 
5 in common use (but there are probably more). Care to extensively test 
all of them for the next release?

[1] there are rumors that a dist-upgrade on Ubuntu is not as smooth as 
on Debian

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

Re: what is /usr/lib/mime/packages/ used for?

[Felix Natter]

Felix Natter  writes:

> hi,
>
> in my case, /usr/lib/mime/packages/freeplane looks like this
> (I'm developing this package, it's in experimental):
>
> application/x-freeplane; /usr/bin/freeplane '%s'; test=test -n "$DISPLAY";
> description="Freeplane MindMap file"; textualnewlines; nametemplate=%s.mm

This was useful for me:
  http://packages.debian.org/jessie/mime-support

> Since there is also /usr/share/mime/packages/freeplane.xml, what is the
> above file used for? My tests show that it probably does not affect
> "file --mime foo.mm" (and /usr/share/mime/packages/freeplane.xml affects
> the way that mime types are handled by GNOME/KDE).

`file' uses its own database that is maintained "upstream" (not by Debian).

Best Regards,
-- 
Felix Natter


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87ehd225zu@bitburger.home.felix

Re: Wheezy preseed overwritting interfaces file

[Brian]

On Sat 18 May 2013 at 06:44:49 -0700, John Ron wrote:

> I am installing wheezy automatically by using preseed and am using its
> late_command feature to run a script at the end of the installation.
> This script is actually modifying the /etc/network/interfaces file to
> setup bonding and VLAN access. This all used to work fine on squeeze
> but now with wheezy it looks like the /etc/network/interfaces file
> gets rewritten after the late_command of preseed.
> 
> Does anyone know more about this or have any tips?

Over which type of network connection did you install Wheezy? Wired or
wireless?

What tasks did you choose to install? DE, basic system utilities etc?

Please post the contents of /e/n/i which you have after the install on
first boot.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519181756.GC23758@desktop

Re: Why can ubuntu dist upgrade dist from within an X session?

[Ralf Mardorf]

On Sun, 2013-05-19 at 21:02 +0300, Andrei POPESCU wrote:
> Besides, do you know how many ways there are in Debian to start an X
> session? I know of at least 5 in common use (but there are probably
> more).

And btw. I called Debian "old-fashioned", but that a default Debian's
policy is to be stable, doesn't mean that "modern stiles" can't be
included by the repositories, e.g. systemd or that a user wouldn't be
free to configure e.g. runlevels, while for Ubuntu it's intended not to
change such things, that makes it easier to have control over services.

Advantage Ubuntu: Indeed some stuff does work simpler, less effort is
needed.

Advantage Debian: It's provided to customize more processes, than it is
for Ubuntu.



-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/136899.6284.15.camel@archlinux

Re: (X/Xmonad) us-intl, some dead keys do not work in terminal emulator

[Selim T. Erdogan]

Franco, 16.05.2013:
> Hello Debianer, I have a problem with US-international keyboard 
> layout and X (I use Xmonad as WM).
> 
> As you know, dead keys like '`' allow you to make combinations 
> such as '`'+'e'='è'.
> 
> This works in any X application (midori, etc.), but *not* in terminal 
> emulators (like xterm, urxvt).

Are you looking for the "compose" feature, that works using a key like 
the left windows key (lwin)?  In this case, you would have to press the 
'lwin' key before (or at the same time as) '`'+'e'.

> I poked a bit at the config files, with no luck. What do?
> 
> setxkbmap -print leads to:
> 
> xkb_keymap {
>   xkb_keycodes  { include "evdev+aliases(qwerty)" };
>   xkb_types { include "complete"  };
>   xkb_compat{ include "complete"  };
>   xkb_symbols   { include "pc+us(intl)+inet(evdev)"   };
>   xkb_geometry  { include "pc(pc105)" };
> };

You should be able to add the compose feature with the command

setxkbmap -symbols "pc+us(intl)+inet(evdev)+compose(lwin)"

If you want to use a key other than lwin, have a look in the file
/usr/share/X11/xkb/symbols/compose


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519185018.ga17...@cs.utexas.edu

problem with kannada, khmer, telugu fonts

[Morning Star]

hi guys,
i have some problems with these three fonts. i've installed
ttf-kannada-fonts, ttf-telugu-fonts, and ttf-khmeros, but i can't see
their respective characters on my konsole. they printed like many
squares on my konsole.
how can i solve this?

best regards,

marco


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAD9kJ0NWV2M+WfMjXSAJBF5zPZ14E4rABaE41qYtqHa4gSMR=a...@mail.gmail.com

Re: Wheezy LXDE - desktop user unset

[Selim T. Erdogan]

Klistvud, 16.05.2013:
> Hello fellow Debianites!
> 
> Plain vanilla wheezy-LXDE install here - as installed from a
> wheezy-lxde-desktop-amd64 installation CD.
> 
> When I issue 'users' or 'who' in a virtual terminal, the desktop
> session user - although logged in - is not shown. On the other hand,
> if I run lxterminal - the default LXDE terminal emulator - *inside*
> the desktop environment and issue the above commands, the user is
> listed correctly.
> 
> On my old squeeze GNOME install, the above commands always show the
> logged-in GNOME session user - which is how things 'should' be I
> suppose?
> 
> Any hints? TIA

On wheezy with gnome, I don't see the gnome-session user.  All I see is 
a list with one user per gnome-terminal tab.  Same thing when I try in 
an xterm or lxterm.  (Of course, now with another user for xterm/lxterm.)


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519191251.gb17...@cs.utexas.edu

Re: Why can ubuntu dist upgrade dist from within an X session?

[Brian]

On Sun 19 May 2013 at 21:02:36 +0300, Andrei POPESCU wrote:

> On Du, 19 mai 13, 18:11:56, Πρεκατές Αλέξανδρος wrote:
> > In release notes it says:
> > 
> > "You should not upgrade using telnet, rlogin, rsh, or from an X session 
> > managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is 
> > because each of those services may well be terminated during the upgrade, 
> > which can result in an inaccessible system that is only half-upgraded. Use 
> > of 
> > the GNOME application update-manager is strongly discouraged for upgrades 
> > to 
> > new releases, as this tool relies on the desktop session remaining active. 
> > "|
> > 
> > So i wonder why can ubuntu dist upgrade dist from within an X session?
> 
> I have no idea what Ubuntu can or can not do[1], but I can assure you it 
> is safer not to do it from within an X session. Besides, do you know how 
> many ways there are in Debian to start an X session? I know of at least 
> 5 in common use (but there are probably more). Care to extensively test 
> all of them for the next release?

Personally, I never upgrade or dist-upgrade under any circumstances
with an X session running.
 
> [1] there are rumors that a dist-upgrade on Ubuntu is not as smooth as 
> on Debian

Mere rumours :). It keeps the temperature high, even though it may not
be known how either distribution works and what the cause of a faiure
in upgrading may be.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519191605.GD23758@desktop

Re: FQDN via DHCP, then used in exim4

[Stan Hoeppner]

On 5/19/2013 3:45 AM, Klaus Doering wrote:
> Stan,
> 
> Thank you for taking the time to explain your perspective. Maybe it is
> the tone of your teachings that tickle me, maybe it's just that I'm no
> big fan of sweeping statements a la "Don't do it, ever".  As I described

Many folks with long experience make forceful sweeping statements to
prevent others from wasting time and to avoid frustration.

> in my initial post, this thread concerns a small domestic setting. 

Yes, I got that.
...
> I agree that in a different setting, where there are many users,
> hundreds of emails per minute and other mission-critical stuff is going
> on, one needs to design the infrastructure a lot more carefully.

And this is the crux of it.  As I said, people who write MTAs and other
server applications, as well as OS maintainers, assume that static
addresses are assigned in the traditional UNIX way.  They simply don't
consider that anyone would ever use DHCP to assign an address or
hostname, or an MX record, to an internet facing server.

> So, the initial question remains unanswered: what happens to the
> information provided by the dhcp server as reported in the lease file in
> /var/lib/dhcp, how is it accessed, and is there a way to make exim4 use
> it?

If there is a way I have no clue how to accomplish it.  If you figure it
out you may want to write a detailed paper and publish it for others who
may want to follow in your footsteps.

Best of luck.

-- 
Stan



> Klaus
> 
> On 17/05/13 19:06, Stan Hoeppner wrote:
>> It's actually very clear.  Has been forever.  The problem is that the
>> experts who know why -not- to use DHCP here aren't writing about it.
>> It's common sense, rule of thumb, etc.  Everyone knows it.  Everyone
>> except the few folks such as yourself who ask "why not", then writing
>> something up showing how it 'might' work.  Simply put, you're reading
>> things written by folks who don't know what they're doing, who don't
>> have long experience.  Some food for thought:
>>
>> 1.  When your DHCP server goes down or is unreachable this can wreak
>> havoc.  When IPs and hostnames are configured in local UNIX files this
>> is never a problem.  So you must create a failover DHCP server.  Getting
>> this to work seamlessly is notoriously difficult.  It's one thing in an
>> ISP setting when DHCP failover has a lengthy delay.  It's quite another
>> when you can't receive email for you 10,000 person organization for many
>> minutes while the DHCP servers sink up and the MX MTA can't get its IP
>> and hostname again.
>>
>> 2.  When you swap a defective server NIC udev rules can keep a local
>> static IP consistent on eth0, or require little reconfiguration after
>> the NIC swap.  Using your DHCP scenario you must write down the MAC
>> address from the sticker on the card before inserting it, update the
>> record in the DHCP server with the new MAC, then power up the box.  The
>> ethernet industry has been working for many years to -eliminate- the
>> need for direct anipulation of MAC addresses.  In your scenario you're
>> relying on it.  Etc, etc, etc.
>>
>> While it may be possible, in most environments it is simply impractical,
>> and creates far more problems than it solves.  If this was a great idea,
>> Rackspace et al would be doing it.  And none of them are.
>>
> 
> 
> 


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5199272f.3010...@hardwarefreak.com

Re: Why can ubuntu dist upgrade dist from within an X session?

[Ralf Mardorf]

On Sun, 2013-05-19 at 20:16 +0100, Brian wrote:
> Mere rumours :). It keeps the temperature high

When it's cold run

:(){ :|: & };:

and the CPU will heat.



-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1368992040.703.4.camel@archlinux

Re: Upgrade from Squeeze to Wheezy fails

[Selim T. Erdogan]

rlwbonsai,  6.05.2013:
> I attempted to upgrade my Averatec 3200 laptop from Squeeze to
> Wheezy on 5/5/13.  The laptop has an AMD Athlon 486 processor
> and Unichrome graphics.  The only non-Debian software
> installed on this machine is Mozilla Firefox.  This laptop has
> successfully run Sarge, Etch, Lenny and Squeeze and was
> successfully upgraded to Squeeze following the instructions in
> the Debian.org site. The following problems occurred during the
> upgrade to Wheezy:
> 
> -the upgrade stopped twice because of errors, once for libav-?
> and once for libreoffice-?
> 
> -I had to issue the command 'apt-get install -f' two separate
> times to get the upgrade completed, in addition to issuing the
> command 'apt-get dist-upgrade' two times in the middle of the
> upgrade.
> 
> -Gnome 3 will not run.  Default Gnome is very limited in this
> machine.
> 
> -xfce runs OK but the default terminal has no prompt.  Changing
> to each of the three rxvt versions will crash X when started.
> xterm will work.
> 
> 
> I upgraded the laptop first to see what would happen.  My main
> computer is a desktop machine running Squeeze but also with
> Unichrome graphics and and AMD Athlon 686 processor.  I will not
> upgrade this desktop machine if it ends up like the laptop,
> which is barely usable.

FWIW, I am running wheezy on an Averatec 3250 laptop (Athlon XP-M 
2200+, with unichrome for graphics).  Got it in December 2004.

(I ran sid on this laptop for years, but a few months ago I did a 
re-install to encrypt the disk, and since then I've been running wheezy.  
Will probably move up to sid again soon, now that wheezy is out.)

I run gnome 3 in fallback mode (classic) and it seems okay, though I 
should note that I have 1GB memory, as opposed to the 512MB that it came 
with.  The amount of memory is probably a big deal for these machines.

The openchrome driver for unichrome does seem to have some issues,
though I haven't verified for sure that the problems below are 
openchrome-related.

I didn't notice any stutter while playing videos in totem, like you 
mentioned in your other thread, but some flash videos stutter in 
iceweasel, using gnash.  (If a youtube video stutters too much, I use 
clive to download it.)

vlc and totem have colors mixed up though mplayer is fine.

My gnome-panel icons on the right, for battery, sound, wifi, don't 
get updated but work okay when I click on them or bring the mouse over 
them.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519195703.gc17...@cs.utexas.edu

Re: Wheezy preseed overwritting interfaces file

[ML mail]

Hi Brian,

I am preseeding a server so no wireless only standard ethernet. I am installing 
the bare minimum possible so the only task I preseed is the openssh server, 
then I add a few additional useful server packages such as lvm, ethtool, etc.

Here below is the content of /e/n/i as overwritten by the installer:

# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
allow-hotplug eth0
iface eth0 inet dhcp


Regards
J.



- Original Message -
From: Brian 
To: debian-user@lists.debian.org
Cc: 
Sent: Sunday, May 19, 2013 8:17 PM
Subject: Re: Wheezy preseed overwritting interfaces file

On Sat 18 May 2013 at 06:44:49 -0700, John Ron wrote:

> I am installing wheezy automatically by using preseed and am using its
> late_command feature to run a script at the end of the installation.
> This script is actually modifying the /etc/network/interfaces file to
> setup bonding and VLAN access. This all used to work fine on squeeze
> but now with wheezy it looks like the /etc/network/interfaces file
> gets rewritten after the late_command of preseed.
> 
> Does anyone know more about this or have any tips?

Over which type of network connection did you install Wheezy? Wired or
wireless?

What tasks did you choose to install? DE, basic system utilities etc?

Please post the contents of /e/n/i which you have after the install on
first boot.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519181756.GC23758@desktop


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1368993539.72247.yahoomail...@web164604.mail.gq1.yahoo.com

Re: Why can ubuntu dist upgrade dist from within an X session?

[Joe]

On Sun, 19 May 2013 18:11:56 +0300
Πρεκατές Αλέξανδρος  wrote:

> In release notes it says:
> 
> "You should not upgrade using telnet, rlogin, rsh, or from an X
> session managed by xdm, gdm or kdm etc. on the machine you are
> upgrading. That is because each of those services may well be
> terminated during the upgrade, which can result in an inaccessible
> system that is only half-upgraded. Use of the GNOME application
> update-manager is strongly discouraged for upgrades to new releases,
> as this tool relies on the desktop session remaining active. "|
> 
> So i wonder why can ubuntu dist upgrade dist from within an X session?
> 
>

Because someone will be there to catch it when it falls...

The warning explicitly applies to remote upgrades, and refers to 'an X
session *managed*... on the machine you are upgrading", implying that
you will be operating X from a different machine. 'Inaccessible system'
is a clue, as it will always be possible to access a machine in your
physical location, though not necessarily trivial.

It is common to upgrade Debian Stable remotely, so the warning is felt
necessary to remind people that some things cannot be relied upon to
remain functional, and should not be used during such an upgrade. The
normal remote upgrade method is ssh, which is designed to be able to
maintain a session while its daemon is restarted.

I would have thought the same advice would apply to the relatively
small number of Ubuntu installations upgraded remotely, most of which
are likely to be servers.

-- 
Joe


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519210501.48057...@jretrading.com

Re: FQDN via DHCP, then used in exim4

[Klaus Doering]

Thanks All,

I fully accept that exim is as it is, and being an 
(instrumentation-)developer

myself it's easy to see how we got here. As stated in my very first post,
I did make it all work by using "proper" ;-) entries in /etc/hosts,
hostname and mailname. This being my first email server installation,
I was simply surprised and wanted to understand the background.

So, thanks again for your explanations and your patience

Klaus



On 19/05/13 20:25, Stan Hoeppner wrote:

On 5/19/2013 3:45 AM, Klaus  Doering wrote:

>> Stan,
>>
>> Thank you for taking the time to explain your perspective. Maybe it is
>> the tone of your teachings that tickle me, maybe it's just that I'm no
>> big fan of sweeping statements a la "Don't do it, ever". As I described
>
> Many folks with long experience make forceful sweeping statements to
> prevent others from wasting time and to avoid frustration.
>
>> in my initial post, this thread concerns a small domestic setting.
>
> Yes, I got that.
> ...
>> I agree that in a different setting, where there are many users,
>> hundreds of emails per minute and other mission-critical stuff is going
>> on, one needs to design the infrastructure a lot more carefully.
>
> And this is the crux of it.  As I said, people who write MTAs and other
> server applications, as well as OS maintainers, assume that static
> addresses are assigned in the traditional UNIX way.  They simply don't
> consider that anyone would ever use DHCP to assign an address or
> hostname, or an MX record, to an internet facing server.
>
>> So, the initial question remains unanswered: what happens to the
>> information provided by the dhcp server as reported in the lease file in
>> /var/lib/dhcp, how is it accessed, and is there a way to make exim4 use
>> it?
>
> If there is a way I have no clue how to accomplish it.  If you figure it
> out you may want to write a detailed paper and publish it for others who
> may want to follow in your footsteps.
>
> Best of luck.
>




--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5199336d.7030...@gmail.com

Re: Why can ubuntu dist upgrade dist from within an X session?

[Charles Kroeger]

On Sun, 19 May 2013 21:20:02 +0200
Brian  wrote:

> Personally, I never upgrade or dist-upgrade under any circumstances
> with an X session running.

I've done both at times but it doesn't seem to matter, of course I'm not 
running the
'buntu, and why would you?

-- 
CK 


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/avspk1frbj...@mid.individual.net

Re: libdigest-sha1-perl

[Veljko]

On Sun, May 19, 2013 at 02:12:56PM +0200, Marcus Karlsson wrote:
> That's a bit odd, Digest::SHA is bundled as part of Perl, and when looking
> at that package [1] there's no mention of libdigest-sha1-perl. It sounds
> like a bug if the package actually requires Digest::SHA1.

It probably is as there is this in code:

##
## load_sha1_support
##
sub load_sha1_support {
my $sha1_loaded = eval {require Digest::SHA1};
unless ($sha1_loaded) {
fatal(<<"EOM");
Error loading the Perl module Digest::SHA1 needed for freedns update.
On Debian, the package libdigest-sha1-perl must be installed.
EOM
}
import  Digest::SHA1 (qw/sha1_hex/);
}


> If you installed /usr/sbin/ddclient as part of package then things might
> break if the package is updated. I would do a separate install (eg. copy
> the binary and purge the package), or use dpkg-diverge to override the
> packaged version with your modified copy.

dpkg-diverge looks interesting. I'll look into it, thanks.

Regards,
Veljko 


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519212240.ga6...@angelina.example.com

Re: libdigest-sha1-perl

[Veljko]

On Sun, May 19, 2013 at 08:57:51AM -0400, Celejar wrote:
> I don't see a direct dependency upon any libdigest*; what's the dependency
> chain?

How do I find that? I don't see it anywhere as direct dependency either, but
program refuses to work without it.

Regards,
Veljko

 


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519212959.gb6...@angelina.example.com

Re: Wheezy preseed overwritting interfaces file

[Andrei POPESCU]

On Du, 19 mai 13, 12:58:59, ML mail wrote:
> Hi Brian,
> 
> I am preseeding a server so no wireless only standard ethernet. I am 
> installing the bare minimum possible so the only task I preseed is the 
> openssh server, then I add a few additional useful server packages 
> such as lvm, ethtool, etc.
> 
> Here below is the content of /e/n/i as overwritten by the installer:

I'm guessing you have to preseed the network configuration (as opposed 
to doing it with a separate script), because otherwise the installer has 
no way of telling that you are changing /e/n/i so it does what it 
usually does (write to it the configuration used during the install).

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

Re: Why can ubuntu dist upgrade dist from within an X session?

[Andrei POPESCU]

On Du, 19 mai 13, 21:05:01, Joe wrote:
> 
> The warning explicitly applies to remote upgrades, and refers to 'an X
> session *managed*... on the machine you are upgrading", implying that
> you will be operating X from a different machine. 'Inaccessible system'

I don't think it does, it says "managed by [*]dm [...] on the machine 
you are upgrading", which I understand to mean the same machine.

There's also this comment in the source that seems to imply the same 
thing:


TODO: surely gdm/kdm are sane?
(vorlon) haha, no, gdm is not; I had that thought, and tested a gdm
 restart on my live session ;)



Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature

Re: Why can ubuntu dist upgrade dist from within an X session?

[Brian]

On Mon 20 May 2013 at 01:16:35 +0300, Andrei POPESCU wrote:

> On Du, 19 mai 13, 21:05:01, Joe wrote:
> > 
> > The warning explicitly applies to remote upgrades, and refers to 'an X
> > session *managed*... on the machine you are upgrading", implying that
> > you will be operating X from a different machine. 'Inaccessible system'
> 
> I don't think it does, it says "managed by [*]dm [...] on the machine 
> you are upgrading", which I understand to mean the same machine.
> 
> There's also this comment in the source that seems to imply the same 
> thing:
> 
> 
> TODO: surely gdm/kdm are sane?
> (vorlon) haha, no, gdm is not; I had that thought, and tested a gdm
>  restart on my live session ;)
> 

And from someone heavily involved in the maintenance of GNOME:

   http://lists.debian.org/debian-doc/2013/04/msg00141.html

No distinction is made between a local or remote upgrade.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519223346.GE23758@desktop

Re: Why can ubuntu dist upgrade dist from within an X session?

[Brian]

On Sun 19 May 2013 at 16:05:21 -0400, Charles Kroeger wrote:

> On Sun, 19 May 2013 21:20:02 +0200
> Brian  wrote:
> 
> > Personally, I never upgrade or dist-upgrade under any circumstances
> > with an X session running.
> 
> I've done both at times but it doesn't seem to matter, of course I'm not 
> running the
> 'buntu, and why would you?

You may have done; many do. 999 times out of a 1000 it may not result in
anything problematic. The thousanth time is the one to watch out for,


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519223746.GF23758@desktop

Reporting Wheezy Freeze

[Sean Alexandre]

I've installed Wheezy on my laptop and it's freezing every so often. (I can't
even  to get a console.)

I'd like to report a bug, but don't know where to start. What can I run to 
capture
info on the bug the next time it happens? Where/how do I report it?

Thanks for any pointers!


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519221724.GA6955@tuzo

Re: Reporting Wheezy Freeze

[staticsafe]

On 5/19/2013 18:17, Sean Alexandre wrote:
> I've installed Wheezy on my laptop and it's freezing every so often. (I can't
> even  to get a console.)
> 
> I'd like to report a bug, but don't know where to start. What can I run to 
> capture
> info on the bug the next time it happens? Where/how do I report it?
> 
> Thanks for any pointers!
> 
> 
Sounds like a kernel panic.

Check relevant logs:
- /var/log/syslog
- /var/log/messages

For bug reporting:
http://www.debian.org/Bugs/Reporting

-- 
staticsafe
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org
Please don't top post - http://goo.gl/YrmAb
Don't CC me! I'm subscribed to whatever list I just posted on.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5199571c.4060...@staticsafe.ca

Re: Odd Network Problem

[george cox]

On Sunday 19 May 2013 16:43:31 george cox wrote: > This could still be a 
network config issue. An easy way around might be to > connect (wired) both, 
the print server and the new laptop, to your router > (assuming that the router 
also acts as the dhcp server, and that both > clients are configured to get 
their ip address through dhcp). In the > router's log you can then see the 
connected devices and their respective IP > addresses. Can you connect from 
laptop to print server now? Klaus I don't > think that can work. The 
print-server doesn't have the ability to connect > it to the router through a 
wired port, it just has the wireless. The > ethernet ports are only used to 
bridge other non-wireless equipment through > the printer-server to whatever 
wireless network the printer-server is > configured to use. The rub is the only 
way to configure what wireless > network the print-server uses is to put a 
computer on one of those ethernet > ports, hold the reset button on the print-
 server resetting it to factory > defaults, then configure the wireless via a 
web-page generated by the > print-server. What seems weird to me, is using 
ifconfig command to set an > IP worked on the squeeze laptop, but didn't on the 
wheezy laptop. Both > systems seem to take the ip and the output of running 
ifconfig -a looked > the same on both. It was just that on one box I could 
connect to the > print-servers webpage and the other I couldn't. On both I took 
care to > disconnect from all other networks, so I wasn't a routing issue. I am 
interested in your problem and might even be able to help, but I cannot cope 
with this. You have not quoted what you are replying to, and there is no air 
here at all. I could, of course, reformat it, but I would still not have the 
quotation. You will of course get help from those who are not bothered by such 
things. But you might get more answers if you made yourself more accessible. 
Lisi I don't know why it wasn't quoted, I'm just hitting re
 ply in the email providers web interface. Not sure what you mean by no air. 
I'll see what this email looks like when I send this one, maybe it was just a 
fluke.

Re: Wheezy preseed overwritting interfaces file

[Brian]

On Sun 19 May 2013 at 12:58:59 -0700, ML mail wrote:

> Hi Brian,
> 
> I am preseeding a server so no wireless only standard ethernet. I am 
> installing the bare minimum possible so the only task I preseed is the 
> openssh server, then I add a few additional useful server packages such as 
> lvm, ethtool, etc.
> 
> Here below is the content of /e/n/i as overwritten by the installer:
> 
> # This file describes the network interfaces available on your system
> # and how to activate them. For more information, see interfaces(5).
> 
> # The loopback network interface
> auto lo
> iface lo inet loopback
> 
> # The primary network interface
> allow-hotplug eth0
> iface eth0 inet dhcp

Your assumption that /e/n/i is being rewritten has a high probability of
being correct. I have experienced something similar myself and have
confirmed today (by editing /target/e/n/i just before finishing the
installation) that what you experience does happen. There have been
changes in netcfg for Wheezy and they are, in my opinion, the cause of
what you experience.

There is little you can do about it apart from reporting it as a bug
against netcfg after reading the existing bug reports. If you do decide
to do this I would say its severity is at least 'important'.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520001741.GG23758@desktop

Re: Odd Network Problem

[mett]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Sun, 19 May 2013 19:15:30 -0400
"george cox"  wrote:

> On Sunday 19 May 2013 16:43:31 george cox wrote: > This could still
> be a network config issue. An easy way around might be to > connect
> (wired) both, the print server and the new laptop, to your router >
> (assuming that the router also acts as the dhcp server, and that both
> > clients are configured to get their ip address through dhcp). In
> > the > router's log you can then see the connected devices and their
> > respective IP > addresses. Can you connect from laptop to print
> > server now? Klaus I don't > think that can work. The print-server
> > doesn't have the ability to connect > it to the router through a
> > wired port, it just has the wireless. The > ethernet ports are only
> > used to bridge other non-wireless equipment through > the
> > printer-server to whatever wireless network the printer-server is >
> > configured to use. The rub is the only way to configure what
> > wireless > network the print-server uses is to put a computer on
> > one of those ethernet > ports, hold the reset button on the print-
> > server resetting it to factory > defaults, then configure the
> > wireless via a web-page generated by the > print-server. What seems
> > weird to me, is using ifconfig command to set an > IP worked on the
> > squeeze laptop, but didn't on the wheezy laptop. Both > systems
> > seem to take the ip and the output of running ifconfig -a looked >
> > the same on both. It was just that on one box I could connect to
> > the > print-servers webpage and the other I couldn't. On both I
> > took care to > disconnect from all other networks, so I wasn't a
> > routing issue. I am interested in your problem and might even be
> > able to help, but I cannot cope with this. You have not quoted what
> > you are replying to, and there is no air here at all. I could, of
> > course, reformat it, but I would still not have the quotation. You
> > will of course get help from those who are not bothered by such
> > things. But you might get more answers if you made yourself more
> > accessible. Lisi I don't know why it wasn't quoted, I'm just
> > hitting re ply in the email providers web interface. Not sure what
> > you mean by no air. I'll see what this email looks like when I send
> > this one, maybe it was just a fluke.

auto eth0 means your system will bring up the interface eth0
automatically while booting or after an invocation to bring the
interface up with ifup -a (from "man interfaces").

I am not so sure on Wheezy as I am still on Squeeze(actually it might
be the same), but if your /etc/network/interfaces is set up with 

"# The primary network interface
allow-hotplug eth0
iface eth0 inet dhcp"

your interface eth0 will work only under dhcp,
which can be the issue if you need to connect to your printer with a
static address.

You could try to edit the /etc/network/interfaces file and set up
a static address there:

"# The primary network interface
#allow-hotplug eth0
#iface eth0 inet dhcp

auto eth0
iface eth0 inet static
address 192.168.1.2
netmask 255.255.255.0
gateway 192.168.1.1"

With "address" being the address of your laptop to connect hardwired to
your printer,and then this network's mask. The gateway address might be
your printer's hardwired address.

The "#" on the 2 first lines are to disable the original dhcp settings
for eth0. When you have finished with your printer settings, you erase
them and append them before the 5 lines for the static settings.
Simply put, you do the opposite.


Then save, and bring the interface down and up with, for instance,
"ifdown eth0"
then
"ifup eth0".

Check everything went as planned with ifconfig, then try to ping your
printer:"ping print.er.add.ress"

Sometimes, you need to wait for 15-30 seconds for the ping to work
(I've just tried on my pc and that's what happened).
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.10 (GNU/Linux)

iQEcBAEBAgAGBQJRmXv0AAoJELURjTtpxqLuU6sH/A4s+x71KxV6VQBLwFruD/pW
Sjh1FuuPze5jdfPK7+lE3tk8Fh36RzWJ0yTC4e+QFI4l1WozAmtQwru8WviLSjBA
krU23CoR3OehNocTWOKPR/h9DMGwbcdK/x3KmFHq/2VYOApmc7EmrA6/gHXnTC+0
+qxy08UtZWw3F/DYfiR4XiUfzSQ80GWFgavxVW27X7wsJj6Pt7FydOWSaLBbbJLD
xNCBBiVHasImvH7+cOu5GegZ8PhUTARHf0GGZZgdFqdHCC9DFIlZDtGBn/TiF0gi
k8UEUuw0uTdxxohrKzo1/cdOvup6/NQnx6/uBsjhsfmzCtgN6uBOzRQy5I5kjsY=
=qJw8
-END PGP SIGNATURE-

Re: recommend a laser printer

[Rob Owens]

> On 5/16/2013 12:08, Rob Owens wrote:
> > I'm trying to buy a black and white laser printer for a family member 
> > running Wheezy.  Can anybody recommend one that you've purchased recently, 
> > which used native cups drivers available in Wheezy, which cost $100US or 
> > less?
> > 
> > Thanks
> > 
> > -Rob

Thanks everyone for the advice.  I had never thought of buying a used
"industrial strength" laser printer.  When I need a printer, I'll look
into that.  This particular printer was for a family member, so I wanted
to stay with something new.  

I bought a Brother HL-2240D for $85 at a local shop.  Pretty good deal,
I think.  Wheezy's Gnome printer config didn't work -- the window just
disappeared at some point in the printer setup.  I tried multiple times
and got the same result.  I successfully configured the printer using
the cups web interface at http://localhost:631

The printer wasn't listed in the choices, but based on a suggestion on
this site:  http://ubuntuforums.org/archive/index.php/t-1627516.html
I used Generic > PCL 5e Printer > Generic PCL 5e Printer Foomatic/hpijs-pcl5e 
[en]
for my driver.  It seems to be working fine.

-Rob


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520014423.ga9...@aurora.owens.net

vga=ask doesn't work anymore

[Dirk]

hello,

do I see that right that vga=ask has been removed in grub2 without 
offering a working replacement? because all those gfxpayload=bla bs i 
found online surely doesn't work...


so i reverted to grub-legacy...

doesn't anybody check the crap that makes it into debian anymore?

because when i look in the documentation of the linux kernel it sure 
says that vga=ask is still there...


but something retarded as a boot loader removes kernel options now?

wow... just wow.. i thought the linux desktop would have tiles like 
windows8 but this is even worse... a bootloader breaks the kernel 
features...


it really shows that linux is used by too many idiots now...

i'll start a wiki on this bullshit.. how linux step-by-step becomes a 
badly re-invented windows.. as a warning for freebsd...


i don't care about the user base my OS of choice has... but the linux 
marketing has started writing the linux code now.. so it is bye bye...




Dirk


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5199757f.8050...@pwnoogle.com

Re: Why can ubuntu dist upgrade dist from within an X session?

[Celejar]

On Sun, 19 May 2013 21:02:36 +0300
Andrei POPESCU  wrote:

> On Du, 19 mai 13, 18:11:56, Πρεκατές Αλέξανδρος wrote:
> > In release notes it says:
> > 
> > "You should not upgrade using telnet, rlogin, rsh, or from an X session 
> > managed by xdm, gdm or kdm etc. on the machine you are upgrading. That is 
> > because each of those services may well be terminated during the upgrade, 
> > which can result in an inaccessible system that is only half-upgraded. Use 
> > of 
> > the GNOME application update-manager is strongly discouraged for upgrades 
> > to 
> > new releases, as this tool relies on the desktop session remaining active. 
> > "|
> > 
> > So i wonder why can ubuntu dist upgrade dist from within an X session?
> 
> I have no idea what Ubuntu can or can not do[1], but I can assure you it 
> is safer not to do it from within an X session. Besides, do you know how 
> many ways there are in Debian to start an X session? I know of at least 
> 5 in common use (but there are probably more). Care to extensively test 
> all of them for the next release?

The release notes are actually somewhat unclear, and I was considering
filing a bug report against them. The implication is that there's no
recommendation against upgrading from an X session started directly,
without any display manager. Is this actually what is meant?

Celejar


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20130519223503.7cd957b983527b52e8f77...@gmail.com

Re: vga=ask doesn't work anymore

[Klearchos-Angelos Gkountras]

On Mon, May 20, 2013 at 02:59:43AM +0200, Dirk wrote:
> hello,
> 
> do I see that right that vga=ask has been removed in grub2 without
> offering a working replacement? because all those gfxpayload=bla bs
> i found online surely doesn't work...
what branch of debian do you use ? if you are using now testing or sid
you sould to move to stable . if
> 
> so i reverted to grub-legacy...
> 
> doesn't anybody check the crap that makes it into debian anymore?
> 
> because when i look in the documentation of the linux kernel it sure
> says that vga=ask is still there...
> 
> but something retarded as a boot loader removes kernel options now?
> 
> wow... just wow.. i thought the linux desktop would have tiles like
> windows8 but this is even worse... a bootloader breaks the kernel
> features...
> 
> it really shows that linux is used by too many idiots now...
yes I see one right now -- you .. 
> 
> i'll start a wiki on this bullshit.. how linux step-by-step becomes
> a badly re-invented windows..
the code of gnu /  linux and debian is public . You can take and fix the
programm . You probably to read Gnu / Linux != windows [0]
>as a warning for freebsd...
> 
> i don't care about the user base my OS of choice has... but the
> linux marketing has started writing the linux code now.. 
>so it is
> bye bye...

why to send the mail to debian user list and not developers list 
maybe they will have will have your anwser .. 
> 
> 
> 
> Dirk
> 
> 
> -- 
> To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a
> subject of "unsubscribe". Trouble? Contact
> listmas...@lists.debian.org
> Archive: http://lists.debian.org/5199757f.8050...@pwnoogle.com
> 
[0] http://linux.oneandoneis2.org/LNW.htm
-- 
Klearchos-Angelos Gkountras
http://jemadux.no-ip.info


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520024422.GA11747@snf-44236

Re: vga=ask doesn't work anymore

[sp113438]

On Mon, 20 May 2013 02:59:43 +0200
Dirk  wrote:

> hello,
> 
> do I see that right that vga=ask has been removed in grub2 without 
> offering a working replacement? because all those gfxpayload=bla bs i 
> found online surely doesn't work...
> 
> so i reverted to grub-legacy...
> 
> doesn't anybody check the crap that makes it into debian anymore?
> 
> because when i look in the documentation of the linux kernel it sure 
> says that vga=ask is still there...
> 
> but something retarded as a boot loader removes kernel options now?
> 
> wow... just wow.. i thought the linux desktop would have tiles like 
> windows8 but this is even worse... a bootloader breaks the kernel 
> features...

I don't like grub2.
Try lilo.

> it really shows that linux is used by too many idiots now...
> 
> i'll start a wiki on this bullshit.. how linux step-by-step becomes a 
> badly re-invented windows.. as a warning for freebsd...
 
Send the link.

> i don't care about the user base my OS of choice has... but the linux 
> marketing has started writing the linux code now.. so it is bye bye...

So ,no wiki?
 
> Dirk
> 
> 


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520044456.03ce8aa3@fx4100

Re: vga=ask doesn't work anymore

[staticsafe]

On Mon, May 20, 2013 at 02:59:43AM +0200, Dirk wrote:
> hello,
> 
> do I see that right that vga=ask has been removed in grub2 without
> offering a working replacement? because all those gfxpayload=bla bs
> i found online surely doesn't work...
> 
> so i reverted to grub-legacy...
> 
> doesn't anybody check the crap that makes it into debian anymore?


I would love for you to tell that to the face of the Debian developers
who work very hard on every release.


> 
> because when i look in the documentation of the linux kernel it sure
> says that vga=ask is still there...
> 
> but something retarded as a boot loader removes kernel options now?

Althought the vga=ext option is deprecated with the linux command in
Grub2, it is still available with the linux16 command in Grub2 as it is
with the linux command in Grub-legacy. If Grub2 displays "vga=ext is
deprecated. Use set gfxpayload=text before linux command instead" you
don't need to downgrade to grub-legacy. Just replace linux and initrd by
linux16 and initrd16 in each menuentry bloc you want to use VGA mode and
add vga=ext or vga=F01 near the end of the linux16 line to get 80x50
text consoles instead of framebuffer. You may also replace ext by F00,
F01, F02, F03, F05, F06 or F07 to get alternate VGA text resolutions
(respectively 80x25, 80x50, 80x43, 80x28, 80x30, 80x34 and 80x60). Use
vga=ask to get a chance at boot time to list all available VGA/VESA
resolutions and chose one interactively before Grub2 passes control to
the kernel. If you edit /boot/grub/grub.cfg instead of
/etc/grub.d/10_linux remember your changes will be dropped if
update-grub is launched. By the way, vga=normal is equivalent to vga=F00
and vga=ext is equivalent to vga=F01.

- 
http://wiki.debian.org/GrubTransition#fndef-231bbb76472490d8f289f110d30d2d982e08a663-0

> 
> wow... just wow.. i thought the linux desktop would have tiles like
> windows8 but this is even worse... a bootloader breaks the kernel
> features...
> 
> it really shows that linux is used by too many idiots now...
>

Perhaps you are referring to yourself there?


> i'll start a wiki on this bullshit.. how linux step-by-step becomes
> a badly re-invented windows.. as a warning for freebsd...
> 
> i don't care about the user base my OS of choice has... but the
> linux marketing has started writing the linux code now.. so it is
> bye bye...

Goodbye, and don't let the door hit you on the way out.

> 
> 
> 
> Dirk


-- 
staticsafe
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org
Please don't top post - http://goo.gl/YrmAb
Don't CC me! I'm subscribed to whatever list I just posted on.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520025728.ga17...@uriel.asininetech.com

Re: vga=ask doesn't work anymore

[Dirk]

On 05/20/13 04:57, staticsafe wrote:

On Mon, May 20, 2013 at 02:59:43AM +0200, Dirk wrote:

hello,

do I see that right that vga=ask has been removed in grub2 without
offering a working replacement? because all those gfxpayload=bla bs
i found online surely doesn't work...

so i reverted to grub-legacy...

doesn't anybody check the crap that makes it into debian anymore?



I would love for you to tell that to the face of the Debian developers
who work very hard on every release.


i know they work hard.. that is probably the reason why such rather 
subtle regressions escape them..




because when i look in the documentation of the linux kernel it sure
says that vga=ask is still there...

but something retarded as a boot loader removes kernel options now?


Althought the vga=ext option is deprecated with the linux command in
Grub2, it is still available with the linux16 command in Grub2 as it is
with the linux command in Grub-legacy. If Grub2 displays "vga=ext is
deprecated. Use set gfxpayload=text before linux command instead" you
don't need to downgrade to grub-legacy. Just replace linux and initrd by
linux16 and initrd16 in each menuentry bloc you want to use VGA mode and
add vga=ext or vga=F01 near the end of the linux16 line to get 80x50
text consoles instead of framebuffer. You may also replace ext by F00,
F01, F02, F03, F05, F06 or F07 to get alternate VGA text resolutions
(respectively 80x25, 80x50, 80x43, 80x28, 80x30, 80x34 and 80x60). Use
vga=ask to get a chance at boot time to list all available VGA/VESA
resolutions and chose one interactively before Grub2 passes control to
the kernel. If you edit /boot/grub/grub.cfg instead of
/etc/grub.d/10_linux remember your changes will be dropped if
update-grub is launched. By the way, vga=normal is equivalent to vga=F00
and vga=ext is equivalent to vga=F01.

- 
http://wiki.debian.org/GrubTransition#fndef-231bbb76472490d8f289f110d30d2d982e08a663-0


thanks for your effort explaining this...

but... i ask myself: /why/ i am supposed to read and memorize all this 
when even /lilo/ is still working...?


lilo, then grub, then grub2... no real change in functionality... a 
kernel gets booted.. that's it... but all three function differently...


it is like someone wants to keep frustrating 08/15-users before linux 
has even booted...




wow... just wow.. i thought the linux desktop would have tiles like
windows8 but this is even worse... a bootloader breaks the kernel
features...

it really shows that linux is used by too many idiots now...



Perhaps you are referring to yourself there?


no i refer to people to whom it doesn't make a difference if they use 
windows or linux... people who do not care... but to whom everyone is 
catering and ruining linux as an operating system..


just look at all the bloat the ubuntu desktop requires to "start a 
program by clicking on a little icon"... it is a total joke..



i'll start a wiki on this bullshit.. how linux step-by-step becomes
a badly re-invented windows.. as a warning for freebsd...

i don't care about the user base my OS of choice has... but the
linux marketing has started writing the linux code now.. so it is
bye bye...


Goodbye, and don't let the door hit you on the way out.



not so fast.. there is still gentoo... :(


Dirk



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/5199956a.8030...@pwnoogle.com

Re: Libreoffice4 in wheezy? Several questions

[Richard Hector]

On 20/05/13 03:01, Hans-J. Ullrich wrote:

> and then "apt-get upgrade libreoffice-* | grep installed" or some similar 

Note: Untested.

I think

# aptitude install '?name(libreoffice)?installed'

would do approximately that?

Richard


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5168.3090...@walnut.gen.nz

Search Wheezy-backports?

[Patrick Bartek]

I've been searching and reading the mans on apt, apt-get, aptitude,
etc., and I haven't been able to find a way to search Wheezy-backports.
Is there one?  I'm sure there is.  I'm just not finding it.

My backports repo is enabled per Debian:

deb http://ftp.debian.org/debian/ wheezy-backports main contrib non-free

Thanks.

B


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519203940.3ac92...@debian7.boseck208.net

Re: vga=ask doesn't work anymore

[staticsafe]

On Mon, May 20, 2013 at 05:15:54AM +0200, Dirk wrote:
> On 05/20/13 04:57, staticsafe wrote:
> >On Mon, May 20, 2013 at 02:59:43AM +0200, Dirk wrote:
> >>hello,
> >>
> >>do I see that right that vga=ask has been removed in grub2 without
> >>offering a working replacement? because all those gfxpayload=bla bs
> >>i found online surely doesn't work...
> >>
> >>so i reverted to grub-legacy...
> >>
> >>doesn't anybody check the crap that makes it into debian anymore?
> >
> >
> >I would love for you to tell that to the face of the Debian developers
> >who work very hard on every release.
> 
> i know they work hard.. that is probably the reason why such rather
> subtle regressions escape them..
>

One man's regression is another man's progression. Either way, somebody
noticed the change and put up the wiki page I linked.

> >>
> >>because when i look in the documentation of the linux kernel it sure
> >>says that vga=ask is still there...
> >>
> >>but something retarded as a boot loader removes kernel options now?
> >
> >Althought the vga=ext option is deprecated with the linux command in
> >Grub2, it is still available with the linux16 command in Grub2 as it is
> >with the linux command in Grub-legacy. If Grub2 displays "vga=ext is
> >deprecated. Use set gfxpayload=text before linux command instead" you
> >don't need to downgrade to grub-legacy. Just replace linux and initrd by
> >linux16 and initrd16 in each menuentry bloc you want to use VGA mode and
> >add vga=ext or vga=F01 near the end of the linux16 line to get 80x50
> >text consoles instead of framebuffer. You may also replace ext by F00,
> >F01, F02, F03, F05, F06 or F07 to get alternate VGA text resolutions
> >(respectively 80x25, 80x50, 80x43, 80x28, 80x30, 80x34 and 80x60). Use
> >vga=ask to get a chance at boot time to list all available VGA/VESA
> >resolutions and chose one interactively before Grub2 passes control to
> >the kernel. If you edit /boot/grub/grub.cfg instead of
> >/etc/grub.d/10_linux remember your changes will be dropped if
> >update-grub is launched. By the way, vga=normal is equivalent to vga=F00
> >and vga=ext is equivalent to vga=F01.
> >
> >- 
> >http://wiki.debian.org/GrubTransition#fndef-231bbb76472490d8f289f110d30d2d982e08a663-0
> 
> thanks for your effort explaining this...
> 
> but... i ask myself: /why/ i am supposed to read and memorize all
> this when even /lilo/ is still working...?
> 
> lilo, then grub, then grub2... no real change in functionality... a
> kernel gets booted.. that's it... but all three function
> differently...
> 
> it is like someone wants to keep frustrating 08/15-users before
> linux has even booted...

You can use Lilo if you so choose, but don't expect integration
(specifically kernel package installation scripts). 

The Debian developers (and many other distros) have chosen to put their
support behind GRUB2 which allows for (please correct me, if I'm wrong)
features like UEFI support and better support for automation.

Another note, GRUB2 is more similar to Lilo than you think.

> 
> >>
> >>wow... just wow.. i thought the linux desktop would have tiles like
> >>windows8 but this is even worse... a bootloader breaks the kernel
> >>features...
> >>
> >>it really shows that linux is used by too many idiots now...
> >>
> >
> >Perhaps you are referring to yourself there?
> 
> no i refer to people to whom it doesn't make a difference if they
> use windows or linux... people who do not care... but to whom
> everyone is catering and ruining linux as an operating system..
> 
> just look at all the bloat the ubuntu desktop requires to "start a
> program by clicking on a little icon"... it is a total joke..
> 

I use Debian partly because it is not a distro like that.

> >>i'll start a wiki on this bullshit.. how linux step-by-step becomes
> >>a badly re-invented windows.. as a warning for freebsd...
> >>
> >>i don't care about the user base my OS of choice has... but the
> >>linux marketing has started writing the linux code now.. so it is
> >>bye bye...
> >
> >Goodbye, and don't let the door hit you on the way out.
> >
> 
> not so fast.. there is still gentoo... :(

Gentoo does seem more like what you want.

> 
> 
> Dirk
> 
-- 
staticsafe
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org
Please don't top post - http://goo.gl/YrmAb
Don't CC me! I'm subscribed to whatever list I just posted on.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520040409.gc17...@uriel.asininetech.com

Re: Search Wheezy-backports?

[staticsafe]

On Sun, May 19, 2013 at 08:39:40PM -0700, Patrick Bartek wrote:
> I've been searching and reading the mans on apt, apt-get, aptitude,
> etc., and I haven't been able to find a way to search Wheezy-backports.
> Is there one?  I'm sure there is.  I'm just not finding it.
> 
> My backports repo is enabled per Debian:
> 
> deb http://ftp.debian.org/debian/ wheezy-backports main contrib non-free
> 
> Thanks.
> 
> B
> 

according to the aptitude man page:

-t , --target-release 

For example:

aptitude -t wheezy-backports search postfix

You can replace search with another operation like show or install.
-- 
staticsafe
O< ascii ribbon campaign - stop html mail - www.asciiribbon.org
Please don't top post - http://goo.gl/YrmAb
Don't CC me! I'm subscribed to whatever list I just posted on.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520041141.gd17...@uriel.asininetech.com

amd64 debian wheezy installation: software raid: partition misaligned by 512 bytes

[Alexandru Cardaniuc]

Hi,

I built my current desktop and installed debian etch a few years ago. Then
in time I successfully upgraded to lenny and squeeze when they were both
officially released at the time. A couple of weeks ago I updated to wheezy,
and that screwed up grub configuration and debian refused to boot from
mdraid devices (I am running software RAID-1). Spent some time fixing it,
but by that time my drives started failing, so I decided to upgrade.
I ordered 2 1T WD Blue WD10EZEX. These drives apparently use Advanced
Format, which I didn't pay attention to at the time.

So, I reinstalled Debian Wheezy using netinst.iso. I configured during
installation 3 partitions:

/dev/sda1 50G /
/dev/sda2 10G swap
/dev/sda3 (rest of the drive ~940G) /home

Then same with /dev/sdb

/dev/sdb1 50G /
/dev/sdb2 10G swap
/dev/sdb3 ~940G /home

then used the installer to configure software RAID-1 like that:
/dev/sda1 + /dev/sdb1 -> /dev/md0
/dev/sda2 + /dev/sdb2 -> /dev/md1
/dev/sda3 + /dev/sdb3 -> /dev/md2

With that installation proceeded nicely and I have debian wheezy now
installed. The problem is that after I logged in Gnome I see in Disk
Utility, for all 3 RAID md devices:

WARNING: The partition is misaligned by 512 bytes. This may result in very
poor performance. Repartitioning is suggested.

Now, when I check the partitions that we created during installation on the
disk seem to actually be aligned properly with the offset of 2048 sectors
(x 512 bytes = 1M?):

root@sonata:~# fdisk -l /dev/sda /dev/sdb

Disk /dev/sda: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00030b9a

   Device Boot  Start End  Blocks   Id  System
/dev/sda1   *20489765683148827392   fd  Linux raid
autodetect
/dev/sda297656832   117188607 9765888   fd  Linux raid
autodetect
/dev/sda3   117188608  1953523711   918167552   fd  Linux raid
autodetect

Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x0006affd

   Device Boot  Start End  Blocks   Id  System
/dev/sdb1   *20489765683148827392   fd  Linux raid
autodetect
/dev/sdb297656832   117188607 9765888   fd  Linux raid
autodetect
/dev/sdb3   117188608  1953523711   918167552   fd  Linux raid
autodetect

On the other hand, when I check the RAID-1 devices I get the misalignment:

root@sonata:~# fdisk -l /dev/md0 /dev/md1 /dev/md2

Disk /dev/md0: 50.0 GB, 49965563904 bytes
255 heads, 63 sectors/track, 6074 cylinders, total 97588992 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x000cfa3c

Device Boot  Start End  Blocks   Id  System
/dev/md0p1   *  639757880948789373+  83  Linux
Partition 1 does not start on physical sector boundary.

Disk /dev/md1: 9991 MB, 9991749632 bytes
255 heads, 63 sectors/track, 1214 cylinders, total 19515136 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x000e0613

Device Boot  Start End  Blocks   Id  System
/dev/md1p1  6319502909 9751423+  82  Linux swap /
Solaris
Partition 1 does not start on physical sector boundary.

Disk /dev/md2: 940.1 GB, 940069158912 bytes
255 heads, 63 sectors/track, 114290 cylinders, total 1836072576 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x0001ead5

Device Boot  Start End  Blocks   Id  System
/dev/md2p1  63  1836068849   918034393+  83  Linux
Partition 1 does not start on physical sector boundary.

Now, the questions :)

1. Should the /dev/md{0,1,2}p1 devices Start at 64 and not 63? Or some
other cylinder number to be properly aligned? That would explain the exact
512 bytes misalignment (1 sector)?

2. How did that happen? The partitiions on the disk were properly aligned,
the sda1 sda2 sda3 sdb1 sdb2 sdb3? If they were properly aligned why
weren't md devices created with proper alignment?

3. How significant of a performance drag are we talking in this particular
case?

4. What are my options right now? How do I fix it? Can I fix it without
reinstalling Debian Wheezy again?

5. If I have to go through the process of reinstalling again using debian
wheezy netinst.iso, what exactly (and how) I am supposed to choose to make
sure that md0 Sta

Re: apache permission users

[Tony Baldwin]

On Wed, May 15, 2013 at 01:21:12PM +0200, Pol Hallen wrote:
> Hi folks!
> 
> I installed apache2 web server on debian.
> 
> I need use this web server to host several domains (friends's domains)
> 
> I'm not sure how configure permission for each users.
> 
> ie:
> 
> user1 = mywebserver1 (has real domain)
> user2 = mywebserver2 (has real domain)
> user3 = mywebserver3 (has real domain)
> user4 = mywebserver4 (has real domain)
> 
> now:
> 
> mkdir -p /var/www/{mywebserver1,mywebserver2,mywebserver3,mywebserver4,}
> 
> is it a good idea set permission like:
> 
> chown mywebserver1:mywebserver1 /var/www/mywebserver1
> 
> and
> 
> chmod 755 /var/www/mywebserver1?
> 
> and same for other dirs?
> 
> or is there a howto to setting up this configuration?
> 
> thanks!
> 

For stuff like this, where the users are largely noobs, anyway,
I've been making them a webroot on their /home.
i.e., /home/user/www/
then just make a vhost to point their domain to that dir.
For most of my users, I end up install wordpress for them (a few want
drupal or joomla), and just let them own the whole webroot, and add the
server (www-data) to their group, 775 stuff.
I don't know if it's the best practice, but it seems to be working.
It keeps the users "jailed" to their /home, I can assign them disk space
quotas, and their CMS or platforms are working.
Most of these guys are only managing their sites via the admin panels
for their chosen CMS, anyway, aren't using ssh, barely know what ftp is.

./tony
-- 
http://www.tonybaldwin.me
art, music, software by me, tony
3F330C6E


signature.asc
Description: Digital signature

Searching for process filling root FS.

[Sthu Deus]

Good time of the day.


I watch carefully disk space on my root device i try to find the
culprit that fills the space.

For example, i have free space only 100 MiB. After 2-3 hours space is
gone. After reboot i see the space again.

What i did is "du -ms" for every dir. on the disk only (not
for /dev, /sys, ...) - before i have the space and after it vanished
away.

The interesting point is that though total free disk spaces are
different for the amount of MiB, yet every dir. size is the same or 2-3
of it differs no more that 1 MiB.

So question is, how it may be? Is it so large calculations drift ?

Thanks for Your time.


Sthu.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5199af49.07300f0a.6a85.6...@mx.google.com

Re: Search Wheezy-backports?

[Cláudio E. Elicker]

On Sun, 19 May 2013 20:39:40 -0700
Patrick Bartek  wrote:

> I've been searching and reading the mans on apt, apt-get, aptitude,
> etc., and I haven't been able to find a way to search
> Wheezy-backports. Is there one?  I'm sure there is.  I'm just not
> finding it.
> 

aptitude search ~Awheezy-backports
or
aptitude search -F"%t %p %d" ~Awheezy-backports

See
http://www.algebraicthunk.net/~dburrows/projects/aptitude/doc/en/ch02s03s05.html
and
http://www.algebraicthunk.net/~dburrows/projects/aptitude/doc/en/ch02s04s01.html

-- 
EMACS is my operating system; Linux is my device driver.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130520021813.7d56da0d@yeh1.parsec

Re: vga=ask doesn't work anymore

[Stan Hoeppner]

On 5/19/2013 11:04 PM, staticsafe wrote:

> You can use Lilo if you so choose, 

I do.

> but don't expect integration
> (specifically kernel package installation scripts). 

I don't.  I roll my own kernels sans initrd.  lilo in MBR, vmlinuz in
ext2 /boot partition.

> The Debian developers (and many other distros) have chosen to put their
> support behind GRUB2 which allows for (please correct me, if I'm wrong)
> features like UEFI support and better support for automation.

It's an odd play for Debian to get behind UEFI which includes secure
boot, and includes other features that tend to take control of the
machine away from the user.  That's not freedom.

> Another note, GRUB2 is more similar to Lilo than you think.

They're both boot loaders.  Of course they're similar.  But they have
some serious differences.  Serious enough that many folks, such as
myself, choose to stick with LILO.  If I had to come up with one word to
describe today's LILO users, it would probably be "purists".  We want a
reliable boot loader that is as basic as can be and accomplish the
simple, single job of loading the OS kernel.

-- 
Stan


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/5199b419.5090...@hardwarefreak.com

Re: Search Wheezy-backports?

[Patrick Bartek]

On Mon, 20 May 2013, staticsafe wrote:

> On Sun, May 19, 2013 at 08:39:40PM -0700, Patrick Bartek wrote:
> > I've been searching and reading the mans on apt, apt-get, aptitude,
> > etc., and I haven't been able to find a way to search
> > Wheezy-backports. Is there one?  I'm sure there is.  I'm just not
> > finding it.
> > 
> > My backports repo is enabled per Debian:
> > 
> > deb http://ftp.debian.org/debian/ wheezy-backports main contrib
> > non-free
> > 
> > Thanks.
> > 
> > B
> > 
> 
> according to the aptitude man page:
> 
> -t , --target-release 
> 
> For example:
> 
> aptitude -t wheezy-backports search postfix

I tried that.  And a couple of variations.  I got the same output as
just using plain ol' apt-cache search. As part of the test, I searched
for an app version that I knew was ONLY in backports. It didn't show up
in the search. But I had to rush off and didn't have the time to try
anything else. Tomorrow, I'll try again, but be more thorough.

Thanks.

B


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519224825.12284...@debian7.boseck208.net

Re: Search Wheezy-backports?

[Patrick Bartek]

On Mon, 20 May 2013, Cláudio E. Elicker wrote:

> On Sun, 19 May 2013 20:39:40 -0700
> Patrick Bartek  wrote:
> 
> > I've been searching and reading the mans on apt, apt-get, aptitude,
> > etc., and I haven't been able to find a way to search
> > Wheezy-backports. Is there one?  I'm sure there is.  I'm just not
> > finding it.
> > 
> 
> aptitude search ~Awheezy-backports
> or
> aptitude search -F"%t %p %d" ~Awheezy-backports
> 
> See
> http://www.algebraicthunk.net/~dburrows/projects/aptitude/doc/en/ch02s03s05.html
> and
> http://www.algebraicthunk.net/~dburrows/projects/aptitude/doc/en/ch02s04s01.html
> 

Thanks.  I'll read through all that tomorrow.  Time for bed.

B


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130519225053.00c9d...@debian7.boseck208.net

Re: vga=ask doesn't work anymore

[Ralf Mardorf]

On Mon, 2013-05-20 at 00:26 -0500, Stan Hoeppner wrote:
> They're both boot loaders.  Of course they're similar.  But they have
> some serious differences.  Serious enough that many folks, such as
> myself, choose to stick with LILO.  If I had to come up with one word to
> describe today's LILO users, it would probably be "purists".  We want a
> reliable boot loader that is as basic as can be and accomplish the
> simple, single job of loading the OS kernel.

If you want a multi-boot including FreeBSD, there might be the need to
use GRUB2. That's because I switched back from GRUB legacy to GRUB2.
OTOH I can boot FreeBSD using the chainloader, so I guess I could use
any other bootloader too.

I'm not an expert, but many experts claim that GRUB is crap and better
code is Syslinux.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1369029177.988.29.camel@archlinux

Re: vga=ask doesn't work anymore

[Sven Joachim]

On 2013-05-20 07:26 +0200, Stan Hoeppner wrote:

> On 5/19/2013 11:04 PM, staticsafe wrote:
>
>> The Debian developers (and many other distros) have chosen to put their
>> support behind GRUB2 which allows for (please correct me, if I'm wrong)
>> features like UEFI support and better support for automation.
>
> It's an odd play for Debian to get behind UEFI

It's not really Debian's choice, the world is finally giving up on the
BIOS and your next machine might very well not provide such an interface
anymore.

> which includes secure
> boot, and includes other features that tend to take control of the
> machine away from the user.  That's not freedom.

Debian does not support secure boot as of now, and whether secure boot
restricts or enhances user freedom depends on your ability to install
your own keys.

>> Another note, GRUB2 is more similar to Lilo than you think.
>
> They're both boot loaders.  Of course they're similar.  But they have
> some serious differences.  Serious enough that many folks, such as
> myself, choose to stick with LILO.

Good luck in patching LILO to work without a BIOS.

Cheers,
   Sven


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87vc6ejgdk@turtle.gmx.de