date:20210918

On Sat 18 Sep 2021 at 08:43:50 (-0400), Greg Wooledge wrote:
> On Sat, Sep 18, 2021 at 12:54:36PM +0200, Roger Price wrote:
> > In site.local I found
> > 
> >  # The following is a space-separated list of where additional user home
> >  # directories are stored, each must have a trailing '/'. Directories added
> >  # here are appended to @{HOMEDIRS}.  See tunables/home for details. Eg:
> >  #@{HOMEDIRS}+=/srv/nfs/home/ /mnt/home/
> > 
> > where curiously, the apparmor installation seems to have detected my
> > non-common /home and made the necessary addition, but appended to a
> > commented out example.
> 
> It wasn't "detected".  That's the generic site.local file that everyone
> has.  The commented-out line is provided as an example.
> 
> What you're supposed to do is either:
> 
> (a) Uncomment that last line, and edit it.
> 
> (b) Copy that last line, uncomment the copy, and edit the copy.
> 
> I prefer (b) myself.

Yes, it's pretty obvious what's going on if you actually do
"See tunables/home for details", because that has the =
definition that the += is appending to.

But — the last line of that comment above is actually inconsistent
with how comments are written in /etc/apparmor* files. It should
have a space after the #. #include lines are the only ones that don't.

The other files that look wrong are /etc/apparmor.d/tunables/*.d/*
and /etc/apparmor.d/tunables{kernelvars,sys}.

Cheers,
David.

Re: Make wider selectable edge in Mate windows

On Sun 19 Sep 2021 at 10:22:24 (+0800), Jeremy Ardley wrote:
> On 19/9/21 9:55 am, Dedeco Balaco wrote:
> > Em 18/09/2021 22:42, Jeremy Ardley escreveu:
> > > I'm running Mate on Debian 10.
> > > 
> > > The problem is that the width of the window edge that you pull to change
> > > the size of a window seems to be only 1 pixel wide.
> > > 
> > > I've heard of this problem in gnome along with some fixes, but I've not
> > > found any solution for Mate.
> > > 
> > > Suggestions?
> > > 
> > > 
> > I also use Mate. Until a few days ago, I was in Debian 9. Now, i updated
> > to Debian 11. Most things in Mate are basically the same, for me.
> > 
> > The issue with 1 pixel borders is not for every window. Mate terminal
> > has wider lines to resize it; Thunderbird 52 too; Pluma too; gcalculator
> > too; vim too; Telegram too; snes9x too; vivaldi too.
> > 
> > In my opinion, there are *bad programs* that do not build their window
> > with best practices. I have seen a program that has 1 pixel borders, but
> > has bigger corners to resize the window. Have you tried the corners? I
> > do not remember which program it was. All i tested, right now, are good
> > with their borders.
> > 
> 
> Testing a few just now, Mate Terminal, Thunderbird, Google Chrome all
> have the 1 pixel border, even in the corner.
> 
> My Mate version is 1.20.4 but this problem has been around for
> versions as long as I remember.

Perhaps it's in the themes? I've never used Mate, but
I downloaded bullseye mate-themes, and see files like
/usr/share/themes/Green-Submarine/metacity-1/metacity-theme-1.xml:

$ head -n 24 metacity-theme-1.xml 




  Green-Submarine
  Wolfgang Ulbrich
   2013
  Apil, 2013
  Dark theme based on Gilouche legacy.
  

 

  
  
  
  
  
  
  
  
  


$ 

whereas /usr/share/themes/TraditionalGreen/metacity-1/metacity-theme-1.xml:

$ head -n 24 metacity-theme-1.xml 




TraditionalGreen
Wolfgang Ulbrich chat-to...@raveit.de
 2013 Wolfgang Ulbrich
March, 2013
The Traditional Green Metacity Theme















$ 

"4 is greater than 1".

So perhaps confirm whether this is the answer by changing themes,
and then see whether you can modify your own theme, either the
"official" way or by hacking it. Of course, hacking it might
require a self-consistent set of changes further down the file. (Or
files: I have no idea of the significance of metacity-theme-[1-3].xml)

Cheers,
David.

Re: Colemak layout at boot time

On Sat 18 Sep 2021 at 21:56:34 (+0100), piorunz wrote:
> On 18/09/2021 20:00, David Wright wrote:
> 
> > A lot of hits from googling   grub colemak   including
> > https://forums.debian.net//viewtopic.php?f=16=76833
> > which uses dvorak as an example.
> 
> Thanks for your reply.
> Yes I seen this page.
> 
> ckbcomp dvorak command outputs the layout details and everything.
> 
> However:
> $ ckbcomp colemak
> /usr/bin/ckbcomp: Can not find file "symbols/colemak" in any known directory
> 
> > 
> > I don't recall the definition of "boot time password". Does this
> > denote something that Grub asks, or is it when dmcrypt is running
> > from the initrd?
> 
> By that I meant GRUB editor and Debian's standard whole disk encryption
> in Debian. I don't have Colemak there. I need to enter password in
> Colemak. Yes, I think that's called dmcrypt.
> 
> > Which is a reminder: is your keyboard definition
> > in /etc/default/keyboard getting incorporated into the initrd or not?
> 
> I don't know that. I only have Colemak in KDE. Everything else,
> including virtual terminals (Ctrl+Alt+F keys) are Qwerty.

How do you normally login, at a VC or in a Display Manager?

Where did you run "sudo dpkg-reconfigure keyboard-configuration" from?

Have you seen this line:
  update-initramfs: Generating /boot/initrd.img-4.19.0-17-amd64
since you changed /etc/default/keyboard and ran
  sudo dpkg-reconfigure keyboard-configuration?

Checking your current initrd is a little tedious: you run
unmkinitramfs to unpack the initrd, and you zcat your
/etc/console-setup/cached_UTF-8_del.kmap.gz to, say, /tmp.
Then run, eg:

$ diff -u …/main/etc/console-setup/cached_UTF-8_del.kmap 
/tmp/cached_UTF-8_del.kmap
$ 

where they are the unpacked and decompressed files respectively.

Cheers,
David.

Re: Mutt/Neomutt and mailcap

On Thu 16 Sep 2021 at 05:37:43 (-0500), Nate Bargmann wrote:
> * On 2021 15 Sep 21:36 -0500, David Wright wrote:
> > I think what your system is doing is as follows:
> > 
> > You probably have in your (default) /etc/Muttrc:
> 
> In my case, /etc/neomuttrc, and the following does exist.
> 
> My assumptions have been that the system RC is ignored when a user's RC
> exist.  I am finding this is not entirely true in all cases.

I can't speak for neomutt, but mutt has started by reading /etc/Muttrc
(unless you tell it not to, with -n) ever since it entered Debian in
hamm (1998). Being read after Muttrc, your own .muttrc can overwrite
mutt's defaults.

There are programs that behave in the way you suggested;
for example, xpdf comes to mind.

> Thanks for the helpful pointers through the maze!
> 
> The problem is that I visit something like this once about every several
> years and it's all new again.  Sigh...

Yes, same here, which is partly why I tried tracking it through.
I actually have a ~/.mutt/mailcap-mutt defined that matches only
a very select set of file types.

Cheers,
David.

Re: A bug in Vim, in Mate Terminal or in Debian 11?

2021-09-18 Thread Greg Wooledge

On Sat, Sep 18, 2021 at 10:14:39PM -0300, Dedeco Balaco wrote:
> My window manager is Mate Desktop. The terminal i most use is its own.
> And i use vim a lot,

> After the upgrade, when i quit vim, the terminal title becomes empty,
> instead of returning to the default title "Terminal".

The real question is how it worked in stretch.  There are several
moving pieces here, and it's not at all clear how they all fit together.

The terminal's title bar is set whenever some program writes the
appropriate bytes to the terminal.  Once the title is set to something,
it stays that way, until something changes it again.

So: under stretch, what program was writing what bytes to the terminal,
at what time?

We know that vim can be configured to set the title when you start it.
This is controlled by the "title" setting.  If you have "set title"
in your .vimrc file, then at startup, vim will change the terminal's
title to the name of the file you're editing.

But what happens when vim exits?  Vim doesn't know what the *previous*
title was, so how can it change the title back?  Therefore, when vim
exits (with "set title" in your .vimrc file), it changes the title
to "Thanks for flying Vim".

Vim has another setting called "titleold".  If this is set to a non-empty
string, vim will use the value of this variable as the new title.  So,
for example, if you would like the title to be changed to "Terminal"
every time you exit vim, you can put these lines in your .vimrc file:

set title
set titleold=Terminal

If titleold is set to the empty string, then the title isn't changed at
all on exit -- it just stays as the filename.

If there's a *third* option (to make vim somehow read the previous title
from the terminal, if that's even possible), then I don't know how to
achieve it.  It would probably be specific to the terminal as well.

If you want more answers than this, I suggest asking on a vim mailing
list.  They may have more expertise than we do.

A *fourth* option that some people use is to have their shell change the
terminal's title on a regular or semi-regular basis.  For example, some
people unilaterally override the title from the shell prompt every time
the prompt is drawn.  Other people override the title when the shell
changes directory.  These might be viable solutions for you as well, if
you choose.

Re: Make wider selectable edge in Mate windows

2021-09-18 Thread Jeremy Ardley



On 19/9/21 9:55 am, Dedeco Balaco wrote:


Em 18/09/2021 22:42, Jeremy Ardley escreveu:

I'm running Mate on Debian 10.

The problem is that the width of the window edge that you pull to change
the size of a window seems to be only 1 pixel wide.

I've heard of this problem in gnome along with some fixes, but I've not
found any solution for Mate.

Suggestions?



I also use Mate. Until a few days ago, I was in Debian 9. Now, i updated
to Debian 11. Most things in Mate are basically the same, for me.

The issue with 1 pixel borders is not for every window. Mate terminal
has wider lines to resize it; Thunderbird 52 too; Pluma too; gcalculator
too; vim too; Telegram too; snes9x too; vivaldi too.

In my opinion, there are *bad programs* that do not build their window
with best practices. I have seen a program that has 1 pixel borders, but
has bigger corners to resize the window. Have you tried the corners? I
do not remember which program it was. All i tested, right now, are good
with their borders.



Testing a few just now, Mate Terminal, Thunderbird, Google Chrome all 
have the 1 pixel border, even in the corner.


My Mate version is 1.20.4 but this problem has been around for versions 
as long as I remember.



--
Jeremy



OpenPGP_signature
Description: OpenPGP digital signature

Re: A bug in Vim, in Mate Terminal or in Debian 11?

2021-09-18 Thread Dedeco Balaco



Em 18/09/2021 22:49, Jeremy Hendricks escreveu:
> I apologize for assuming you upgraded from 9 to 11 directly (skipping
> 10). It’s a common misconception that is acceptable.
> 

No problem. Actually, my first try to update was directly from a not
completely updated Debian 9 to the current Debian 11. But this did not
work at all, and i ended up discovering, after a complicated fight, that
i needed to be completely updated with the 9, and then i should upgrade
to 10, and then check if it is updated too (which naturally happened,
since i was just upgrading to it, in that moment). After that, finally,
i upgraded to 11. There are strange things, now, though. A friend of
mine said they are possible, even if we are doing everything as it
should - something that is bad and sad, for me.

Re: A bug in Vim, in Mate Terminal or in Debian 11?

2021-09-18 Thread Jeremy Hendricks

I apologize for assuming you upgraded from 9 to 11 directly (skipping 10).
It’s a common misconception that is acceptable.

On Sat, Sep 18, 2021 at 9:46 PM Dedeco Balaco 
wrote:

> I did not leap any release. I did everything correctly, step by step.
>
> The issue is not being reprodutible. I do not find a clear condition to
> make it happen. Searching about your suggestion, I found that vim has a
> few very important arguments, where 2 of them interest me a lot, to
> isolate this issue (I guess):
>
> $ vim -u NORC # does not load any RC file, global or user's
>
> $ vim -u NONE # does not load any RC file, and also no plugin
>
> More information can be found in vim man page, and inside vim:
>
> :h initialization
>
>
> Em 18/09/2021 22:27, Jeremy Hendricks escreveu:
> > Generally it’s not recommended to leap frog over releases and you should
> > upgrade in order (Ex. From 9 to 10). I suspect you might need to delete
> > some of the config files in your home directory (rename them it .bak) as
> > the version of vim, etc might be considerable different from 9 to 11 and
> > might have odd results. I’d recommend first renaming .vimrc to
> > .vimrc.bak and see if that fixes it.
> >
> > The next step might be to rename the config files for the mate-terminal
> > and test to see if it’s fixed.
> >
> > On Sat, Sep 18, 2021 at 9:15 PM Dedeco Balaco  > > wrote:
> >
> > Hello,
> >
> > I recently upgraded my Debian 9 to Debian 11. But there are strange
> > or wrong things happening right now, that did not exist before.
> >
> > My window manager is Mate Desktop. The terminal i most use is its
> > own. And i use vim a lot, and i basically never use gvim, i prefer
> > it through a terminal window.
> >
> > Each mate-terminal window can have several tabs in it. I use this
> > feature a lot too. Each tab has its own title, which works pretty
> > fine. There are several programs that change the title shown while
> > they run.
> >
> > Vim is one of these programs. While it is running, the terminal
> > title shows the name of the file currently being edited, and the
> > number of files that was opened with it, when it was launched.
> >
> > After the upgrade, when i quit vim, the terminal title becomes
> > empty, instead of returning to the default title "Terminal".
> >
> > Why is this happening? A bug in Vim? In mate-terminal? Or in Debian
> 11?
> >
>
>

Re: A bug in Vim, in Mate Terminal or in Debian 11?

2021-09-18 Thread Dedeco Balaco

I did not leap any release. I did everything correctly, step by step.

The issue is not being reprodutible. I do not find a clear condition to
make it happen. Searching about your suggestion, I found that vim has a
few very important arguments, where 2 of them interest me a lot, to
isolate this issue (I guess):

$ vim -u NORC # does not load any RC file, global or user's

$ vim -u NONE # does not load any RC file, and also no plugin

More information can be found in vim man page, and inside vim:

:h initialization


Em 18/09/2021 22:27, Jeremy Hendricks escreveu:
> Generally it’s not recommended to leap frog over releases and you should
> upgrade in order (Ex. From 9 to 10). I suspect you might need to delete
> some of the config files in your home directory (rename them it .bak) as
> the version of vim, etc might be considerable different from 9 to 11 and
> might have odd results. I’d recommend first renaming .vimrc to
> .vimrc.bak and see if that fixes it.
> 
> The next step might be to rename the config files for the mate-terminal
> and test to see if it’s fixed.
> 
> On Sat, Sep 18, 2021 at 9:15 PM Dedeco Balaco  > wrote:
> 
> Hello,
> 
> I recently upgraded my Debian 9 to Debian 11. But there are strange
> or wrong things happening right now, that did not exist before.
> 
> My window manager is Mate Desktop. The terminal i most use is its
> own. And i use vim a lot, and i basically never use gvim, i prefer
> it through a terminal window.
> 
> Each mate-terminal window can have several tabs in it. I use this
> feature a lot too. Each tab has its own title, which works pretty
> fine. There are several programs that change the title shown while
> they run.
> 
> Vim is one of these programs. While it is running, the terminal
> title shows the name of the file currently being edited, and the
> number of files that was opened with it, when it was launched.
> 
> After the upgrade, when i quit vim, the terminal title becomes
> empty, instead of returning to the default title "Terminal".
> 
> Why is this happening? A bug in Vim? In mate-terminal? Or in Debian 11?
>

Make wider selectable edge in Mate windows

2021-09-18 Thread Jeremy Ardley


I'm running Mate on Debian 10.

The problem is that the width of the window edge that you pull to change 
the size of a window seems to be only 1 pixel wide.


I've heard of this problem in gnome along with some fixes, but I've not 
found any solution for Mate.


Suggestions?


--
Jeremy



OpenPGP_signature
Description: OpenPGP digital signature

Re: A bug in Vim, in Mate Terminal or in Debian 11?

2021-09-18 Thread Jeremy Hendricks

Generally it’s not recommended to leap frog over releases and you should
upgrade in order (Ex. From 9 to 10). I suspect you might need to delete
some of the config files in your home directory (rename them it .bak) as
the version of vim, etc might be considerable different from 9 to 11 and
might have odd results. I’d recommend first renaming .vimrc to .vimrc.bak
and see if that fixes it.

The next step might be to rename the config files for the mate-terminal and
test to see if it’s fixed.

On Sat, Sep 18, 2021 at 9:15 PM Dedeco Balaco 
wrote:

> Hello,
>
> I recently upgraded my Debian 9 to Debian 11. But there are strange or
> wrong things happening right now, that did not exist before.
>
> My window manager is Mate Desktop. The terminal i most use is its own. And
> i use vim a lot, and i basically never use gvim, i prefer it through a
> terminal window.
>
> Each mate-terminal window can have several tabs in it. I use this feature
> a lot too. Each tab has its own title, which works pretty fine. There are
> several programs that change the title shown while they run.
>
> Vim is one of these programs. While it is running, the terminal title
> shows the name of the file currently being edited, and the number of files
> that was opened with it, when it was launched.
>
> After the upgrade, when i quit vim, the terminal title becomes empty,
> instead of returning to the default title "Terminal".
>
> Why is this happening? A bug in Vim? In mate-terminal? Or in Debian 11?
>

A bug in Vim, in Mate Terminal or in Debian 11?

2021-09-18 Thread Dedeco Balaco

Hello,

I recently upgraded my Debian 9 to Debian 11. But there are strange or
wrong things happening right now, that did not exist before.

My window manager is Mate Desktop. The terminal i most use is its own.
And i use vim a lot, and i basically never use gvim, i prefer it through
a terminal window.

Each mate-terminal window can have several tabs in it. I use this
feature a lot too. Each tab has its own title, which works pretty fine.
There are several programs that change the title shown while they run.

Vim is one of these programs. While it is running, the terminal title
shows the name of the file currently being edited, and the number of
files that was opened with it, when it was launched.

After the upgrade, when i quit vim, the terminal title becomes empty,
instead of returning to the default title "Terminal".

Why is this happening? A bug in Vim? In mate-terminal? Or in Debian 11?

Re: Configuracao PT-BR em teclado US

2021-09-18 Thread Alex Porto dos Santos

Ola
Se for gnome
Em configurações > Fontes de entrada
No botão de mais > outro  e pesquisa a linguagem
Uso o atalho de teclado: bandeira(windows) + espaço para alternar entre as
linguagens
ajuda um monte
[image: image.png]

Em sáb., 18 de set. de 2021 às 17:14, Gerson 
escreveu:

> Prezados,
>
> Utilizo debian testing em um notebook com teclado US e layout BR.
> Na ultima atualizacao perdi esta configuracao e nao encontrei como
> retornar.
>
> Poderiam me ajudar, por favor?
>
> Grato,
> Gerson Luiz Haus
>

Re: Colemak layout at boot time

2021-09-18 Thread piorunz


On 18/09/2021 20:00, David Wright wrote:


A lot of hits from googling   grub colemak   including
https://forums.debian.net//viewtopic.php?f=16=76833
which uses dvorak as an example.


Thanks for your reply.
Yes I seen this page.

ckbcomp dvorak command outputs the layout details and everything.

However:
$ ckbcomp colemak
/usr/bin/ckbcomp: Can not find file "symbols/colemak" in any known directory



I don't recall the definition of "boot time password". Does this
denote something that Grub asks, or is it when dmcrypt is running
from the initrd?


By that I meant GRUB editor and Debian's standard whole disk encryption
in Debian. I don't have Colemak there. I need to enter password in
Colemak. Yes, I think that's called dmcrypt.


Which is a reminder: is your keyboard definition
in /etc/default/keyboard getting incorporated into the initrd or not?


I don't know that. I only have Colemak in KDE. Everything else,
including virtual terminals (Ctrl+Alt+F keys) are Qwerty.

--
With kindest regards, Piotr.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

Configuracao PT-BR em teclado US

2021-09-18 Thread Gerson

Prezados,
Utilizo debian testing em um notebook com teclado US e layout BR.
Na ultima atualizacao perdi esta configuracao e nao encontrei como retornar.
Poderiam me ajudar, por favor?
Grato,Gerson Luiz Haus

Re: How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

2021-09-18 Thread David Christensen


On 9/18/21 4:35 AM, Myron wrote:

Never done this one with Linux before.  I know that there is less than 16Gb
of data written to the Class 10 32Gb MicroSD card which is used as the
primary system storage on a single board system-on-a-chip computer.  What
I'm after is getting a 16 Gb Class 10 A1 MicroSD card and clone the entire
system from the 32Gb card to the 16Gb card.

What I'm after is when I start the SOC computer from the replacement 16Gv
Class 10 A1 MicroSD card, it will just start like there have been no
changes, well, apart from there being 16Gb storage and not slower 32Gb
storage.

This is relatively easy to do on Windows.  No clue how to do this with
Linux.




Backup your data.  I would take a raw binary image of the entire 32 GB 
MicroSD card as well.



My guess is that you should resize the contents of the 32 GB MicroSD 
card to fit onto the 16 GB MicroSD card, and then clone.



But, the devil is in the details and it would be helpful if we had more 
information...



What is the make and model of your "single board system-on-a-chip 
computer" (SBC)?  What CPU, memory, or other options does it have?  What 
is the technical support URL?



Is the SBC connected to a keyboard, mouse, and monitor, to a serial 
console, or to some other console?  Can you SSH into it?



How did you create a working Debian (?) GNU/Linux instance on the 32 GB 
MicroSD card?  If you followed some instructions, what is the URL?



On the 32 GB MicroSD card Debian instance, please login as root, run the 
following commands, and reply with the complete console session -- 
prompts, commands entered, output obtained:


# /bin/bash -l

# export PS1='\n\D{%Y-%m-%d %H:%M:%S} \u@\h \w\n\$ '

# cat /etc/debian_version ; uname -a

# egrep 'vendor_id|model name' /proc/cpuinfo | head -n 2

# grep MemTotal /proc/meminfo

# fdisk -l


David

Re: Lock tab on SD card?

2021-09-18 Thread Dan Ritter

John Conover wrote: 
> 
> Mounting an SD card with the lock tab enabled mounts as read only.
> 
> But can Linux still write to the SD card, under any circumstances?

Under *any* circumstances? Yes.

Those circumstances include:

- a broken lock position sensor on the SD card reader

- the SD card is actually a microSD card in an adaptor; moving
  it to a microSD reader or a different adaptor will work

- the reader has been altered away from spec

- the lock tab is moved

Note that Linux needs to remount a read-only filesystem to make
it read-write.

Now, what is the problem you are actually trying to solve?

-dsr-

Re: Pas de son

2021-09-18 Thread didier gaumet




Le samedi 18 septembre 2021 à 16:06 +0200, Georges a écrit :

[...]
> BFM a écrit :
[...]
> > load-module module-switch-on-port-available
> Et je n'ai pas celle la non plus  ??? 
> 
[...]

Bonjour,

Je pense que François-Marie t'a indiqué la bonne solution, si j'en
crois la doc Freedesktop:

"Depuis la version 2.0. Bascule automatiquement le profil de carte
et/ou le port matériel quand un matériel modifie son statut de
disponibilité. En pratique cela arrive quand on connecte/déconnecte une
fiche 3,5mm (pas sur une carte USB puisqu'elles ne supportent pas la
détection jack) ou un connecteur HDMI.
Ce module est recommandé dans la plupart des cas et fait partie de la
configuration par défaut. En particulier c'est très utile sur les
laptops pour basculer automatiquement entre le casque et les haut-
parleurs intégrés lorsque me casque est branché ou débranché."

https://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/User/Modules/#module-switch-on-port-available

Re: Colemak layout at boot time

On Sat 18 Sep 2021 at 12:02:49 (+0100), piorunz wrote:

> I use Colemak layout instead of Qwerty, because it's so much better and
> efficient.
> Today I repositioned keys on my laptop too, so I have two computers with
> Colemak.
> However, my laptop uses full disk encryption (Debian 11).
> At boot time, and in GRUB, I don't have Colemak, but default layout.
> How to change to Colemak? I read few very old tutorials, not one solves
> this problem in 2021. Not sure how I should proceed.
> 
> I already selected Colemak in KDE settings, and via sudo
> dpkg-reconfigure keyboard-configuration. But this does not affect boot
> time password question and GRUB editor. Also virtual consoles still use
> default Qwerty.

A lot of hits from googling   grub colemak   including
https://forums.debian.net//viewtopic.php?f=16=76833
which uses dvorak as an example.

I don't recall the definition of "boot time password". Does this
denote something that Grub asks, or is it when dmcrypt is running
from the initrd? — Which is a reminder: is your keyboard definition
in /etc/default/keyboard getting incorporated into the initrd or not?

Cheers,
David.

Re: Lock tab on SD card?

2021-09-18 Thread David Christensen


On 9/18/21 10:28 AM, John Conover wrote:


Mounting an SD card with the lock tab enabled mounts as read only.

But can Linux still write to the SD card, under any circumstances?



My assumption is that the lock tab on SD cards is read by the controller 
inside the SD card, which then provides read-only or read-write mode.



I would test  the SD card in question to confirm.


David

Re: Bestanden behouden in live images (was: UEFI + LEGACY op USB)

2021-09-18 Thread Paul van der Vlis

Op 18-09-2021 om 18:48 schreef Roland Clobus:

Hallo Paul, en anderen,

On 18/09/2021 15:05, Paul van der Vlis wrote:

Op 18-09-2021 om 09:44 schreef Roland Clobus:

On 17/09/2021 13:43, Paul van der Vlis wrote:

Ik gebruik een USB-naar-SATA adapter en daaraan hang ik dan een SSD
waarop ik Debian installeer. En van daar wil ik booten.

Ik weet niet precies wat het doel van deze externe SSD is.

Vooral computers testen. Ik stop de SSD er in, en start hem op een
willekeurige computer. Dan wil ik een heleboel dingen testen, en als het
niet werkt de problemen oplossen. Als de computer het niet goed doet,
dan raad ik mensen af er Linux op te zetten. Soms gaat hij dan terug
naar de leverancier.

Wat ik er ook wel mee doe is bewerkingen op een computer uitvoeren,
zoals bijvoorbeeld secure_erase van een SSD (jammer dat dat niet in de
Debian installer zit, moet ik misschien eens als bug melden).

Soms test ik er ook software mee die een systeem nogal in de war gooien.

Van alles dus.

Als je van plan bent om daarmee zonder netwerkverbinding andere
computers van Debian te voorzien, denk ik dat een (zelf gebouwde) live
image een goede optie voor je is. De live images kunnen alle benodigde
.deb bestanden (en .udeb waar nodig) bevatten voor een installatie van
Debian.

Voor een installatie gebruik ik normaal de installer-DVD-image.

Het nadeel van live-images vind ik, dat ik er niets blijvend aan kan
veranderen. Na een reboot zijn mijn wijzigingen weer weg. Daarom kan er
veel niet.

Zie bijvoorbeeld sectie '10.3' [1], je kunt de kernel-boot-optie
'persistence' meegeven, waardoor je toch bestanden kunt behouden als
toevoeging op de live sessie.

Ik ken dat van bijvoorbeeld Tails, maar ik denk toch dat het zijn
beperkingen heeft. Tails moet je bijvoorbeeld toch vaak opnieuw installeren.

En ik zie ook niet echt het voordeel van een niet schrijfbaar systeem,
behalve dan dat het wat minder kwetsbaar is. Maar als de boel niet goed
werkt, dan maak ik in een half uur een nieuwe.

Als je zelf een live image bouwt, kun je de kernel-boot-optie ook per
default aanzetten (is standaard uitgeschakeld i.v.m. potentiële
veiligheidsrisico's).
Voor een standaard live image heb je een partitie met een speciaal label
nodig.

Ik heb weleens gekeken naar het zelf bouwen, maar vond het nog niet
simpel. Toen kwam er zo'n online service, dat vond ik wel fijn. Maar die
is er niet meer jammer genoeg.

Maar eigenlijk gebruik ik dus maar weinig live systemen, ik vind
schrijfbare systemen meestal fijner.

Groet,
Paul

Met vriendelijke groeten,
Roland Clobus

[1]
https://live-team.pages.debian.net/live-manual/html/live-manual/customizing-run-time-behaviours.en.html#547

--
Paul van der Vlis Linux systeembeheer Groningen
https://www.vandervlis.nl/

Lock tab on SD card?

2021-09-18 Thread John Conover



Mounting an SD card with the lock tab enabled mounts as read only.

But can Linux still write to the SD card, under any circumstances?

Thanks,

John

-- 

John Conover, cono...@rahul.net, http://www.johncon.com/

Re: Bestanden behouden in live images (was: UEFI + LEGACY op USB)

2021-09-18 Thread Roland Clobus

Hallo Paul, en anderen,

On 18/09/2021 15:05, Paul van der Vlis wrote:
> Op 18-09-2021 om 09:44 schreef Roland Clobus:
>> On 17/09/2021 13:43, Paul van der Vlis wrote:
>>> Ik gebruik een USB-naar-SATA adapter en daaraan hang ik dan een SSD
>>> waarop ik Debian installeer. En van daar wil ik booten.

Ik weet niet precies wat het doel van deze externe SSD is.
Als je van plan bent om daarmee zonder netwerkverbinding andere
computers van Debian te voorzien, denk ik dat een (zelf gebouwde) live
image een goede optie voor je is. De live images kunnen alle benodigde
.deb bestanden (en .udeb waar nodig) bevatten voor een installatie van
Debian.

> Het nadeel van live-images vind ik, dat ik er niets blijvend aan kan
> veranderen. Na een reboot zijn mijn wijzigingen weer weg. Daarom kan er
> veel niet.

Zie bijvoorbeeld sectie '10.3' [1], je kunt de kernel-boot-optie
'persistence' meegeven, waardoor je toch bestanden kunt behouden als
toevoeging op de live sessie.
Als je zelf een live image bouwt, kun je de kernel-boot-optie ook per
default aanzetten (is standaard uitgeschakeld i.v.m. potentiële
veiligheidsrisico's).
Voor een standaard live image heb je een partitie met een speciaal label
nodig.

Met vriendelijke groeten,
Roland Clobus

[1]
https://live-team.pages.debian.net/live-manual/html/live-manual/customizing-run-time-behaviours.en.html#547

OpenPGP_signature
Description: OpenPGP digital signature

Re: krijg Wireguard niet naar verwachting aan de gang met IPv6

2021-09-18 Thread Gĳs Hillenius




OW! Dommoor - ik maakte een fout in de wg0.conf op de server, vergat er
de AllowedIPs reeks aan te passen. Als ik dat alsnog doe:



AllowedIPs = 10.93.15.2/32, 2a01:4f8:200:546b:4653::2/128

(waarom /128, overigens?)

blijft de client klagen:

wg-quick up wg0
[#] ip link add wg0 type wireguard
[#] wg setconf wg0 /dev/fd/63
[#] ip -4 address add 10.93.15.2/24 dev wg0
[#] ip -6 address add 2a01:4f8:200:546b:4653::2/64 dev wg0
RTNETLINK answers: Network is unreachable

nog meer data:

ifconfig als wg0 aan is op de server


wg0: flags=209  mtu 1420
inet 10.93.15.1  netmask 255.255.255.0  destination 10.93.15.1
inet6 2a01:4f8:200:546b:4653::1  prefixlen 64  scopeid 0x0
unspec 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00  txqueuelen 1000 
 (UNSPEC)
RX packets 0  bytes 0 (0.0 B)
RX errors 0  dropped 0  overruns 0  frame 0
TX packets 0  bytes 0 (0.0 B)
TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0




On 18 September 2021 17:19 Gĳs Hillenius, wrote:

> On 18 September 2021 10:48 Geert Stappers, wrote:
>
>> On Sat, Sep 18, 2021 at 09:42:13AM +0200, Geert Stappers wrote:
>>> On Sat, Sep 18, 2021 at 08:51:32AM +0200, Gĳs Hillenius wrote:
>>   [...] knip
>>> > client
>>> > ,
>>> > | [Interface]
>>> > | Address = 10.93.15.2/24, fdab:9205:cf78:f608::2/64
>>> > | PrivateKey = 
>>> > | 
>>> > 
>>   
>>> > 
>>> > en dan testen met ping
>>> > ping6 ping.xs4all.nl
>>> > PING ping.xs4all.nl(ping.xs4all.nl (2001:888:0:5::1)) 56 data bytes
>>> > ^C
>>> > --- ping.xs4all.nl ping statistics ---
>>> > 5 packets transmitted, 0 received, 100% packet loss, time 4077ms
>>> > 
>>
>> Netwerkpakketten vanaf fdab:9205:cf78:f608::2 bereiken
>> misschien wel 2001:888:0:5::1, maar pakketen van 2001:888:0:5::1
>> vinden niet hun weg terug naar fdab:9205:cf78:f608::2.
>>
>> [1]
>>
>> Iets om te proberen:
>> Onder 2a01:4f8:200:546b/64 bijvoorbeeld 2a01:4f8:200:546b:4653/80
>> hangen. Aan wireguard server geef je 2a01:4f8:200:546b:4653::1
>> Aan wireguard client geef je 2a01:4f8:200:546b:4653::2  [2]
>
>
> Met andere woorden, je wilt dat ik dit probeer. (spoiler: het wil niet,
> zie onder)
>
>
> wg0 op de client
>
> ,
> | [Interface]
> | Address= 10.93.15.2/24, 2a01:4f8:200:546b:4653::2/64
> | PrivateKey = 
> | 
> | [Peer]
> | PublicKey = P3GrgaFCxj6gc6CnOUPo8vxBtKaOcKa7wa8LoL1oUl0=
> | Endpoint = [2a01:4f8:200:546b::9e15:1]:51820
> | AllowedIPs = 0.0.0.0/0, ::/0
> | 
> | PersistentKeepalive = 25
> `
>
> wg0 op de server
>
> ,
> | [Interface]
> | Address = 10.93.15.1/24, 2a01:4f8:200:546b:4653::1/64
> | PrivateKey = 
> | ListenPort = 51820
> | 
> | [Peer]
> | PublicKey = nRwfI98C+AFDaLZuaF1i7YWrj7yQDHrQO07XvivGn2U=
> | AllowedIPs = 10.93.15.2/32, fdab:9205:cf78:f608::2/128
> `
>
> Beiden aangezet, maar de client moppert meteen:
>
> server
>
> wg-quick up wg0
> [#] ip link add wg0 type wireguard
> [#] wg setconf wg0 /dev/fd/63
> [#] ip -4 address add 10.93.15.1/24 dev wg0
> [#] ip -6 address add 2a01:4f8:200:546b:4653::1/64 dev wg0
> [#] ip link set mtu 1420 up dev wg0
> [#] ip -6 route add fdab:9205:cf78:f608::2/128 dev wg0
>
>
> dan de client:
>
> wg-quick up wg0
> [#] ip link add wg0 type wireguard
> [#] wg setconf wg0 /dev/fd/63
> [#] ip -4 address add 10.93.15.2/24 dev wg0
> [#] ip -6 address add 2a01:4f8:200:546b:4653::2/64 dev wg0
> RTNETLINK answers: Network is unreachable
> [#] ip link set mtu 1420 up dev wg0
> [#] wg set wg0 fwmark 51820
> [#] ip -6 route add ::/0 dev wg0 table 51820
> [#] ip -6 rule add not fwmark 51820 table 51820
> [#] ip -6 rule add table main suppress_prefixlength 0
> [#] ip6tables-restore -n
> [#] ip -4 route add 0.0.0.0/0 dev wg0 table 51820
> [#] ip -4 rule add not fwmark 51820 table 51820
> [#] ip -4 rule add table main suppress_prefixlength 0
> [#] sysctl -q net.ipv4.conf.all.src_valid_mark=1
> [#] iptables-restore -n
>
> Network unreachable, maar eigenwijs toch ping6 geprobeerd:
>
> ping6 ping.xs4all.nl
> PING ping.xs4all.nl(ping.xs4all.nl (2001:888:0:5::1)) 56 data bytes
> ^C
> --- ping.xs4all.nl ping statistics ---
> 13 packets transmitted, 0 received, 100% packet loss, time 12294ms
>
>

-- 
I've no idea when Linus is going to release 2.0.24, but if he takes
too long Im going to release a 2.0.24unoff and he can sound off all
he likes.
-- Alan Cox

Re: Your Thoughts on Printer Replacement

2021-09-18 Thread James H. H. Lampert


On 9/18/21 2:19 AM, Jeremy Ardley wrote:
My experience is that toner does degrade over a period of years. To get 
full life you need to use your advertised pages within a year or so.


Agreed. I've seen toner cartridges go bad. Of course, they had been 
sitting on a shelf for *many* years.


--
JHHL

Re: Your Thoughts on Printer Replacement

2021-09-18 Thread James H. H. Lampert


On 9/18/21 2:00 AM, Jonathan Dowland wrote:

The direction of travel for printing is entirely driverless, so this is
less important than it used to be.


Really? If true, that is exceptionally good news. The last time I looked 
at new printers, the "direction of travel" was entirely 
driver-dependent, RIPping the PostScript, PCL, or straight ASCII in the 
driver, rather than in the printer's own processor and firmware, and 
anything that could RIP a PostScript data stream directly would have 
cost a fortune.


--
JHHL

Re: krijg Wireguard niet naar verwachting aan de gang met IPv6

2021-09-18 Thread Gĳs Hillenius

On 18 September 2021 10:48 Geert Stappers, wrote:

> On Sat, Sep 18, 2021 at 09:42:13AM +0200, Geert Stappers wrote:
>> On Sat, Sep 18, 2021 at 08:51:32AM +0200, Gĳs Hillenius wrote:
>   [...] knip
>> > client
>> > ,
>> > | [Interface]
>> > | Address = 10.93.15.2/24, fdab:9205:cf78:f608::2/64
>> > | PrivateKey = 
>> > | 
>> > 
>   
>> > 
>> > en dan testen met ping
>> > ping6 ping.xs4all.nl
>> > PING ping.xs4all.nl(ping.xs4all.nl (2001:888:0:5::1)) 56 data bytes
>> > ^C
>> > --- ping.xs4all.nl ping statistics ---
>> > 5 packets transmitted, 0 received, 100% packet loss, time 4077ms
>> > 
>
> Netwerkpakketten vanaf fdab:9205:cf78:f608::2 bereiken
> misschien wel 2001:888:0:5::1, maar pakketen van 2001:888:0:5::1
> vinden niet hun weg terug naar fdab:9205:cf78:f608::2.
>
> [1]
>
> Iets om te proberen:
> Onder 2a01:4f8:200:546b/64 bijvoorbeeld 2a01:4f8:200:546b:4653/80
> hangen. Aan wireguard server geef je 2a01:4f8:200:546b:4653::1
> Aan wireguard client geef je 2a01:4f8:200:546b:4653::2  [2]


Met andere woorden, je wilt dat ik dit probeer. (spoiler: het wil niet,
zie onder)


wg0 op de client

,
| [Interface]
| Address= 10.93.15.2/24, 2a01:4f8:200:546b:4653::2/64
| PrivateKey = 
| 
| [Peer]
| PublicKey = P3GrgaFCxj6gc6CnOUPo8vxBtKaOcKa7wa8LoL1oUl0=
| Endpoint = [2a01:4f8:200:546b::9e15:1]:51820
| AllowedIPs = 0.0.0.0/0, ::/0
| 
| PersistentKeepalive = 25
`

wg0 op de server

,
| [Interface]
| Address = 10.93.15.1/24, 2a01:4f8:200:546b:4653::1/64
| PrivateKey = 
| ListenPort = 51820
| 
| [Peer]
| PublicKey = nRwfI98C+AFDaLZuaF1i7YWrj7yQDHrQO07XvivGn2U=
| AllowedIPs = 10.93.15.2/32, fdab:9205:cf78:f608::2/128
`

Beiden aangezet, maar de client moppert meteen:

server

wg-quick up wg0
[#] ip link add wg0 type wireguard
[#] wg setconf wg0 /dev/fd/63
[#] ip -4 address add 10.93.15.1/24 dev wg0
[#] ip -6 address add 2a01:4f8:200:546b:4653::1/64 dev wg0
[#] ip link set mtu 1420 up dev wg0
[#] ip -6 route add fdab:9205:cf78:f608::2/128 dev wg0


dan de client:

wg-quick up wg0
[#] ip link add wg0 type wireguard
[#] wg setconf wg0 /dev/fd/63
[#] ip -4 address add 10.93.15.2/24 dev wg0
[#] ip -6 address add 2a01:4f8:200:546b:4653::2/64 dev wg0
RTNETLINK answers: Network is unreachable
[#] ip link set mtu 1420 up dev wg0
[#] wg set wg0 fwmark 51820
[#] ip -6 route add ::/0 dev wg0 table 51820
[#] ip -6 rule add not fwmark 51820 table 51820
[#] ip -6 rule add table main suppress_prefixlength 0
[#] ip6tables-restore -n
[#] ip -4 route add 0.0.0.0/0 dev wg0 table 51820
[#] ip -4 rule add not fwmark 51820 table 51820
[#] ip -4 rule add table main suppress_prefixlength 0
[#] sysctl -q net.ipv4.conf.all.src_valid_mark=1
[#] iptables-restore -n

Network unreachable, maar eigenwijs toch ping6 geprobeerd:

ping6 ping.xs4all.nl
PING ping.xs4all.nl(ping.xs4all.nl (2001:888:0:5::1)) 56 data bytes
^C
--- ping.xs4all.nl ping statistics ---
13 packets transmitted, 0 received, 100% packet loss, time 12294ms

Re: Your Thoughts on Printer Replacement

2021-09-18 Thread Fred


On 9/18/21 2:00 AM, Jonathan Dowland wrote:

On Sat, Sep 18, 2021 at 10:38:03AM +0200, Klaus Singvogel wrote:

Charles Curley wrote:

I print rarely. Inkjet or laser? Or other? The carts on the L7700 tend
to go bad before they empty,


Rarely printing? I would suggest a laser


I agree. Toner doesn't spoil sat there doing nothing for long periods of
time, whereas the ink for inkjets can do.
The next question then becomes: mono or colour?


Brand? The situation current is this, that only HP released their printer
drivers as Open Source (at least 99% of them).


The direction of travel for printing is entirely driverless, so this is
less important than it used to be. And indeed, driverless printing is a
lot less of a headache to administrate, IMHO, than drivered - even with
open source drivers.

If I were looking for a new laser printer today I'd look at Lexmark (on
a recommendation from a friend), the firmware for which is apparently
based on Yocto Linux.



Hello,
I have a Lexmark E260dn which has Postscript, double side, Ethernet and 
USB, parallel, monochrome.  It usually works fine but apparently the 
toner stirrer doesn't work well.  The toner light comes on occasionally 
requiring the toner cartridge to be removed and shaken.  The printer is 
all plastic and not very expensive but the toner cartridges are a bit 
expensive.  The cartridge has a sensor built in that keeps up with the 
number of copies so you can't refill the cartridge as there is no way to 
reset the sensor.  You have to use Lexmark's exchange program.  I don't 
know if it needs drivers to work with CUPS.  I just ftp files to it and 
it always prints them.


I read on this list a few years ago that the best printers are used HP 
office models.


Best regards,
Fred

Re: Pas de son

 Bonjour et merci de la réponse,

Le Sat, 18 Sep 2021 14:36:27 +0200,

BFM a écrit :

> chez moi je n'ai pas cette option
> 
> > load-module module-switch-on-connect
> par contre dans /etc/pulse/default.pa
idem pour moi
> j'ai ceci
> 
> ### Should be after module-*-restore but before module-*-detect
> load-module module-switch-on-port-available
Et je n'ai pas celle la non plus  ??? 
> et je suis sous bullseye.
idem et pareil dans bookworm  ??? 
> François-Marie
> 
> 
> Le 18/09/2021 à 11:17, Haricophile a écrit :
> > Le Sat, 18 Sep 2021 09:52:38 +0200,
> > Billard François-Marie  a
> > écrit :
> >
> >> Bonjour
> >>
> >> par hasard dans l'outil tel que le contrôleur de volume de
> >> pulseaudio, pour moi sous xfce, avez-vous un possibilité de
> >> sélection entre haut-parleur ou sortie-audio pour casque ?
> >>
> >> François-Marie
> > En plus il me semble que sous Bullseye XFCE il faut régler
> > manuellement la bascule automatique de la sortie son au branchement
> > d'un périphérique.
> >
> > dans /etc/pulse/default.pa ou ~/.config/pulse/default.pa
> >
> > load-module module-switch-on-connect
> >
>

Re: Pas de son



 Bonjour et merci de la réponse

Le Sat, 18 Sep 2021 11:17:33 +0200,

H a écrit :

> Le Sat, 18 Sep 2021 09:52:38 +0200,
> Billard François-Marie  a
> écrit :
> 
> > Bonjour
> > 
> > par hasard dans l'outil tel que le contrôleur de volume de
> > pulseaudio, pour moi sous xfce, avez-vous un possibilité de
> > sélection entre haut-parleur ou sortie-audio pour casque ?
> > 
> > François-Marie
> 
> En plus il me semble que sous Bullseye XFCE il faut régler
> manuellement la bascule automatique de la sortie son au branchement
> d'un périphérique.
> 
> dans /etc/pulse/default.pa ou ~/.config/pulse/default.pa
> 
> load-module module-switch-on-connect
 
Sous Bullseye / bookworm et icewm je n'ai pas cette option
dans /etc/pulse/default.pa

Re: Pas de son

 Bonjour et merci de la réponse,

Le Sat, 18 Sep 2021 09:52:38 +0200,

BFM a écrit :

> Bonjour
> 
> par hasard dans l'outil tel que le contrôleur de volume de
> pulseaudio, pour moi sous xfce, avez-vous un possibilité de sélection
> entre haut-parleur ou sortie-audio pour casque ?
Oui et si je connecte le casque il en tient compte mais si je le
déconnecte il se met sur Speakers sans que j'ai le son des HP
???
> François-Marie
> 
> Le 18/09/2021 à 08:46, Georges a écrit :
> > Bonjour,
> >
> >   J'ai un portable récent avec Bullseye et Bookworm :
> >
> > Intel(R) Core(TM) i3-1005G1 CPU @ 1.20GHz
> >
> > Audio device: Intel Corporation Ice Lake-LP Smart Sound Technology
> > Audio Controller (rev 30)
> >
> > Je n'ai pas de son sur les haut-parleurs mais seulement sur la
> > sortie casque.
> >
> > La faute à qui ?
> >
> > Le bios EFI ? pulseaudio ? bluetooth ? blueman ? pipewire ?
> >
> > Merci de vos lumières
> >
> > --
> >
> > Georges
> >
>

Re: installing debian 11.0

On Sat, Sep 18, 2021 at 03:40:18PM +0700, Ngọc Dũng Tạ wrote:
> When installing debian 11.0, it asks to scan the additional media "debian
> gnu/linux 11.0 _ bullseye _ - official amd64 dvd binary-1", so where can I
> get it?
> Thanks.
> -- 

Hi,

How are you installing Debian 11?

What file are you using to install from?

if the install has completeed: can you show us the /etc/apt/sources.list file
please.

Sometimes, if you use just a CD / DVD image to install Debian, and do
not use the network, the only entry refers to the CD or DVD.

Any extra information you can give may be helpful.

With every good wish

Andy Cater

> -
> *TA NGOC DUNG (Mr)*
> *Job:* Marketing Online
> *Mobile:* 0962 597 563

Re: How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

On Sat, Sep 18, 2021 at 12:35:13PM +0100, Myron wrote:
> Never done this one with Linux before.  I know that there is less than 16Gb
> of data written to the Class 10 32Gb MicroSD card which is used as the
> primary system storage on a single board system-on-a-chip computer.  What
> I'm after is getting a 16 Gb Class 10 A1 MicroSD card and clone the entire
> system from the 32Gb card to the 16Gb card.
> 
> What I'm after is when I start the SOC computer from the replacement 16Gv
> Class 10 A1 MicroSD card, it will just start like there have been no
> changes, well, apart from there being 16Gb storage and not slower 32Gb
> storage.
> 
> This is relatively easy to do on Windows.  No clue how to do this with
> Linux.

You might find it easier to do a reinstall and then copy any data over -
it will be at least as quick as shrinking the first data disk and using
dd to copy the data from one SD card to the other.

If you have any particular software that you must have - get a listing of 
what's on the first card and copy that list over as the list of packages you
 install

dpkg --get-selections > myfile

dpkg --set-selections < myfile

will probably do it followed by an apt update.

Then it's just a question fo copying over personal files and config files 
from the old SD card.

All the very best, as ever,

Andy Cater

Re: UEFI + LEGACY op USB

2021-09-18 Thread Paul van der Vlis


Hoi Roland, en anderen,

Op 18-09-2021 om 09:44 schreef Roland Clobus:

Hallo Paul,

On 17/09/2021 13:43, Paul van der Vlis wrote:

Ik gebruik een USB-naar-SATA adapter en daaraan hang ik dan een SSD
waarop ik Debian installeer. En van daar wil ik booten.

Die SSD zou ik graag zowel bootbaar hebben met UEFI, als via de oude
legacy methode. Dus dat het werkt onafhankelijk van de instelling van
het BIOS.

De Debian USB-sticks kunnen dit, weet iemand hier hoe je dat ook zelf
kunt realiseren?


Je zou kunnen kijken naar het pakket 'live-build'. Hiermee kun je zelf
live images maken (NB op een net iets andere manier dan de officiële
live images van Debian).


Het nadeel van live-images vind ik, dat ik er niets blijvend aan kan 
veranderen. Na een reboot zijn mijn wijzigingen weer weg. Daarom kan er 
veel niet.


Wat ik daarom doe is een echte installatie op een extern medium. Het 
lukt me echter tot nu toe niet om die zo te maken dat het zowel werkt 
met UEFI als met Legacy.


Bedankt voor je verhaal over live-images, ik moet er weer eens naar 
kijken want ze zijn ook leuk voor bepaalde toepassingen. Maar toch, ik 
gebruik veel vaker een echte installatie.


Groet,
Paul

--
Paul van der Vlis Linux systeembeheer Groningen
https://www.vandervlis.nl/

Re: How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

2021-09-18 Thread Reco

On Sat, Sep 18, 2021 at 08:39:41AM -0400, The Wanderer wrote:
> I'm used to seeing ISOLINUX for bootable CDs, and something (I've never
> been sure what) for bootable USB drives, but have/had never learned what
> was/is used for bootability on SD cards.

It's simple. First, you look at the processor's architecture.

GRUB is used on x86, modern POWER (think IBM p-Series) and ARM64 (but
you have to use a server hardware for that).
For x86 there's also syslinux, lilo, and direct UEFI kernel boot (aka EFIStub).
Older POWER (and PowerPCs) used kboot, but I have no personal experience
with them.
u-boot is ARM and ARM64, maybe RISC-V (no personal experience with that).
silo is for UltraSPARCs.

I even do not want to know what they're using on MIPS, HPPA or Itanium
for boot, but there's definitely something.

Next you look at the platform-specific detail.

For x86 the biggest (and only) choice you have is between BIOS and UEFI.
For ARM it's somewhat rough, but manageable if you happen to have an
appropriate datasheet. Theoretically, each SOC has its own, unique way of
booting. Practically acquiring right u-boot configuration and patches is
all it takes.

UltraSPARCs and modern POWERs are easy (you have no choice, that is),
but good luck on getting the hardware. It's costly, and barring that
Talos thing comes in the form of a typical rack-mount server.

And whatever you're using as a boot media does not matter at all, unless
you're doing something very exotic, like [1].

Reco

[1] https://github.com/raspberrypi/linux/issues/3178

Re: UEFI + LEGACY op USB

2021-09-18 Thread Paul van der Vlis


Hoi Geert en anderen,

Op 18-09-2021 om 09:13 schreef Geert Stappers:

On Fri, Sep 17, 2021 at 01:43:10PM +0200, Paul van der Vlis wrote:

Hallo,

Ik gebruik een USB-naar-SATA adapter en daaraan hang ik dan een SSD waarop
ik Debian installeer. En van daar wil ik booten.

Die SSD zou ik graag zowel bootbaar hebben met UEFI, als via de oude legacy
methode. Dus dat het werkt onafhankelijk van de instelling van het BIOS.

De Debian USB-sticks kunnen dit, weet iemand hier hoe je dat ook zelf kunt
realiseren?



Interresante uitdaging.

Wild idee:

De SSD voorzien met Debian-Installer plus een preseed-file.


Ik heb meerdere USB-poorten. Wat ik normaal zou doen is booten vanaf een 
USB-stick met de Debian installer, en als target de SSD gebruiken.


Omdat er maar 1 iets gemaakt wordt, lijkt me een preseed file overdreven.


In die preseed-file ook aanroep van aanvullende scripts.

Bij de eerste boot moet er dan dit gebeuren:

* Debian-Installer wordt gestart
* preseed file wordt gevonden
* inhoud van preseed file wordt gebruikt
* automatisch verloop van de installatie
   hierbij wordt de SSD overschreven
* Reboot


Hmm, kan dat niet handiger, directer?

Misschien mis ik iets.

Groeten,
Paul




--
Paul van der Vlis Linux systeembeheer Groningen
https://www.vandervlis.nl/

Re: Debian 11: evince and apparmor flood kernel log

2021-09-18 Thread Greg Wooledge

On Sat, Sep 18, 2021 at 12:54:36PM +0200, Roger Price wrote:
> In site.local I found
> 
>  # The following is a space-separated list of where additional user home
>  # directories are stored, each must have a trailing '/'. Directories added
>  # here are appended to @{HOMEDIRS}.  See tunables/home for details. Eg:
>  #@{HOMEDIRS}+=/srv/nfs/home/ /mnt/home/
> 
> where curiously, the apparmor installation seems to have detected my
> non-common /home and made the necessary addition, but appended to a
> commented out example.

It wasn't "detected".  That's the generic site.local file that everyone
has.  The commented-out line is provided as an example.

What you're supposed to do is either:

(a) Uncomment that last line, and edit it.

(b) Copy that last line, uncomment the copy, and edit the copy.

I prefer (b) myself.

Re: How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

2021-09-18 Thread The Wanderer

On 2021-09-18 at 08:34, Reco wrote:

> On Sat, Sep 18, 2021 at 08:01:34AM -0400, The Wanderer wrote:
>
>> On 2021-09-18 at 07:53, Reco wrote:

>>> 1) Plug-in source card, use dump(8) to backup the contents of its
>>> filesystem.
>>> 2) Plug-in target card, create appropriate partition(s) on it.
>>> 3) Make the needed amount of filesystems on a target SD card.
>>> For ext4 you'll want to use -U option of mkfs to clone filesystem UUIDs
>>> (i.e. UUID on the target card must be the same compared to the source
>>> one).
>>> 4) Use restore(8) to recreate filesystem(s) contents on a target card.
>>> 5) Unmount filesystems made on a target card.
>> 
>> Will this really be enough?

>> For hard-drive installs you're likely to have a GRUB installation, which
>> wouldn't be brought across by a measure like this.
> 
> I doubt that GRUB will come into play in this scenario. You probably
> meant u-boot.

No - I really don't know enough about this type of context to be able to
make that connection. I was mentioning GRUB as an example of the
non-filesystem-based boot configuration that's needed in the hard-drive
case; since this isn't the hard-drive case, if any such configuration is
needed, it would probably be something other than GRUB.

I'm used to seeing ISOLINUX for bootable CDs, and something (I've never
been sure what) for bootable USB drives, but have/had never learned what
was/is used for bootability on SD cards.

>> For a SD-card-based install I'm not sure, but I'd be a bit surprised
>> to learn that no such non-filesystem-based configuration is necessary.
> 
> A canonical example - Raspberry Pi. You just have to clone partition
> UUIDs, and fill first partition with Broadcom blobs. That's it, nothing
> more complex is required.
> RPi have their share of deficiencies and simply are broken by design in
> some regards, but the their boot process is as straightforward as
> possible.

So this isn't really a concern in practice, and I've learned something
today. Good to know!

-- 
   The Wanderer

The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man. -- George Bernard Shaw

signature.asc
Description: OpenPGP digital signature

Re: Pas de son

2021-09-18 Thread Billard François-Marie


chez moi je n'ai pas cette option


load-module module-switch-on-connect

par contre dans /etc/pulse/default.pa

j'ai ceci

### Should be after module-*-restore but before module-*-detect
load-module module-switch-on-port-available

et je suis sous bullseye.

François-Marie


Le 18/09/2021 à 11:17, Haricophile a écrit :

Le Sat, 18 Sep 2021 09:52:38 +0200,
Billard François-Marie  a écrit
:


Bonjour

par hasard dans l'outil tel que le contrôleur de volume de
pulseaudio, pour moi sous xfce, avez-vous un possibilité de sélection
entre haut-parleur ou sortie-audio pour casque ?

François-Marie

En plus il me semble que sous Bullseye XFCE il faut régler manuellement
la bascule automatique de la sortie son au branchement d'un
périphérique.

dans /etc/pulse/default.pa ou ~/.config/pulse/default.pa

load-module module-switch-on-connect

Re: How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

2021-09-18 Thread Reco

On Sat, Sep 18, 2021 at 08:01:34AM -0400, The Wanderer wrote:
> On 2021-09-18 at 07:53, Reco wrote:
> >> No clue how to do this with Linux.
> > 
> > 1) Plug-in source card, use dump(8) to backup the contents of its
> > filesystem.
> > 2) Plug-in target card, create appropriate partition(s) on it.
> > 3) Make the needed amount of filesystems on a target SD card.
> > For ext4 you'll want to use -U option of mkfs to clone filesystem UUIDs
> > (i.e. UUID on the target card must be the same compared to the source
> > one).
> > 4) Use restore(8) to recreate filesystem(s) contents on a target card.
> > 5) Unmount filesystems made on a target card.
> 
> Will this really be enough?

OP specified that it's a

>> single board system-on-a-chip computer

So it is usually enough.

> I'd expect that you'd also need to bring across the bootability
> configuration, which - depending on how it's set up on that particular
> device - might well require additional steps.

If it's a really broken SBC, like, for instance Ordroid N2 or Exynos
5422 (aka Odroid XU4) - then yes, you'll have to also write about 1Mb of
non-free blobs at the start of the card, and without overwriting any
partition contents.

> For hard-drive installs you're likely to have a GRUB installation, which
> wouldn't be brought across by a measure like this.

I doubt that GRUB will come into play in this scenario. You probably
meant u-boot.
GRUB does not have ARM port at all, unless it's an ARM64 server we're
talking about here. And even then it'll be GRUB-EFI, which merely
requires copying grub.aarch64 to EFI FAT fs.

> For a SD-card-based install I'm not sure, but I'd be a bit surprised
> to learn that no such non-filesystem-based configuration is necessary.

A canonical example - Raspberry Pi. You just have to clone partition
UUIDs, and fill first partition with Broadcom blobs. That's it, nothing
more complex is required.
RPi have their share of deficiencies and simply are broken by design in
some regards, but the their boot process is as straightforward as
possible.

Reco

Re: How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

2021-09-18 Thread The Wanderer

On 2021-09-18 at 07:53, Reco wrote:

>   Hi.
> 
> On Sat, Sep 18, 2021 at 12:35:13PM +0100, Myron wrote:
>> This is relatively easy to do on Windows.
> 
> This is true only if you're using that sad excuse for a filesystem
> called NTFS.
> 
>> No clue how to do this with Linux.
> 
> 1) Plug-in source card, use dump(8) to backup the contents of its
> filesystem.
> 2) Plug-in target card, create appropriate partition(s) on it.
> 3) Make the needed amount of filesystems on a target SD card.
> For ext4 you'll want to use -U option of mkfs to clone filesystem UUIDs
> (i.e. UUID on the target card must be the same compared to the source
> one).
> 4) Use restore(8) to recreate filesystem(s) contents on a target card.
> 5) Unmount filesystems made on a target card.

Will this really be enough?

I'd expect that you'd also need to bring across the bootability
configuration, which - depending on how it's set up on that particular
device - might well require additional steps.

For hard-drive installs you're likely to have a GRUB installation, which
wouldn't be brought across by a measure like this. For a SD-card-based
install I'm not sure, but I'd be a bit surprised to learn that no such
non-filesystem-based configuration is necessary.

-- 
   The Wanderer

The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man. -- George Bernard Shaw

signature.asc
Description: OpenPGP digital signature

Re: How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

2021-09-18 Thread Reco

Hi.

On Sat, Sep 18, 2021 at 12:35:13PM +0100, Myron wrote:
> This is relatively easy to do on Windows.

This is true only if you're using that sad excuse for a filesystem
called NTFS.

> No clue how to do this with Linux.

1) Plug-in source card, use dump(8) to backup the contents of its
filesystem.
2) Plug-in target card, create appropriate partition(s) on it.
3) Make the needed amount of filesystems on a target SD card.
For ext4 you'll want to use -U option of mkfs to clone filesystem UUIDs
(i.e. UUID on the target card must be the same compared to the source
one).
4) Use restore(8) to recreate filesystem(s) contents on a target card.
5) Unmount filesystems made on a target card.

Reco

How do I clone a Debian Distro from a 32Gb Class 10 MicroSD card to a 16Gb Class 10 A1 MicroSD card?

2021-09-18 Thread Myron

Never done this one with Linux before.  I know that there is less than 16Gb
of data written to the Class 10 32Gb MicroSD card which is used as the
primary system storage on a single board system-on-a-chip computer.  What
I'm after is getting a 16 Gb Class 10 A1 MicroSD card and clone the entire
system from the 32Gb card to the 16Gb card.

What I'm after is when I start the SOC computer from the replacement 16Gv
Class 10 A1 MicroSD card, it will just start like there have been no
changes, well, apart from there being 16Gb storage and not slower 32Gb
storage.

This is relatively easy to do on Windows.  No clue how to do this with
Linux.

Colemak layout at boot time

2021-09-18 Thread piorunz


Hi all,

I use Colemak layout instead of Qwerty, because it's so much better and
efficient.
Today I repositioned keys on my laptop too, so I have two computers with
Colemak.
However, my laptop uses full disk encryption (Debian 11).
At boot time, and in GRUB, I don't have Colemak, but default layout.
How to change to Colemak? I read few very old tutorials, not one solves
this problem in 2021. Not sure how I should proceed.

I already selected Colemak in KDE settings, and via sudo
dpkg-reconfigure keyboard-configuration. But this does not affect boot
time password question and GRUB editor. Also virtual consoles still use
default Qwerty.

My /etc/default/keyboard file:
# KEYBOARD CONFIGURATION FILE

# Consult the keyboard(5) manual page.

XKBMODEL="pc105"
XKBLAYOUT="gb"
XKBVARIANT="colemak"
XKBOPTIONS=""

BACKSPACE="guess"


Any tips appreciated!


--
With kindest regards, Piotr.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄

Re: Debian 11: evince and apparmor flood kernel log

2021-09-18 Thread Roger Price


On Sat, 18 Sep 2021, Klaus Singvogel wrote:


Roger Price wrote:

In Debian 11, evince has an appamor profile which floods the kernel log with
hundreds of messages of the style:


Not only at Debian 11, even Debian 10 has it.

[...]

 (evince:2869): GVFS-WARNING **: 22:18:18.510: can't init metadata tree 
/mnt/home/rprice/.local/share/gvfs-metadata/home: open: Permission denied

[...]

Is there some way of calming evince+appamor?


The location of your home is uncommon (as on my side).

Fix: edit /etc/apparmor.d/tunables/home.d/site.local


In site.local I found

 # The following is a space-separated list of where additional user home
 # directories are stored, each must have a trailing '/'. Directories added
 # here are appended to @{HOMEDIRS}.  See tunables/home for details. Eg:
 #@{HOMEDIRS}+=/srv/nfs/home/ /mnt/home/

where curiously, the apparmor installation seems to have detected my non-common 
/home and made the necessary addition, but appended to a commented out example.


I added line /mnt/home/ and tried to restart apparmor.service.  This failed with 
error messages such as


Sep 18 12:08:33 titan apparmor.systemd[5150]: AppArmor parser error for
 /etc/apparmor.d/lsb_release in /etc/apparmor.d/tunables/multiarch at line 13:
 syntax error
Sep 18 12:08:33 titan apparmor.systemd[5154]: AppArmor parser error for
 /etc/apparmor.d/nvidia_modprobe in /etc/apparmor.d/tunables/multiarch at line
 13: syntax error

So I tried replacing @{HOMEDIRS}=/home/ with @{HOMEDIRS}=/mnt/home/ in file 
/etc/apparmor.d/tunables/home


I restarted apparmor.service and some light testing shows that the problem is 
solved.


My error in site.local was probably to have added /mnt/home and not 
@{HOMEDIRS}+=/mnt/home


Thanks to all who responded!  Roger

Re: Your Thoughts on Printer Replacement

On Fri, Sep 17, 2021 at 04:38:27PM -0600, Charles Curley wrote:
> I have an HP Officejet Pro L7700, which is starting to show its age.
> Also HP has discontinued the standard size cartridges. I can get the
> large ones.
> 
> First question:
> 
> I print rarely. Inkjet or laser? Or other? The carts on the L7700 tend
> to go bad before they empty, 

Laser, assuredly. I bought a Samsung a few years ago - unfortunately
that line has now been bought by HP but it's still good and reasonable.

Colour laser - certainly but the consumables are hight cost when you
have to replace them. For you that might be once a decade.

Networked is very necessary: duplex with a high quality feeder is a thing 
I wish I'd been able to afford.

If you have an office supplies / print shop locally, there's a lot to be said
for printing long documents there and having them in a comb binder  even if you
have got a good printer. 

I have a separate scanner (or two): unless you
go up to the top quality office printers, the MFP quality is not great for
scanning.

HP is the most likely to be readily available. It's worth pricing up the
quality of desktop space/power/wiring taken up vs. actual usage: best 
printers do tend to be the larger models though I used a tiny mono printer
for a few years quite successfully.

Hope this helps,

Andy Cater

> 
> -- 
> Does anybody read signatures any more?
> 
> https://charlescurley.com
> https://charlescurley.com/blog/
>

Re: debian-installer RAID question

Many Thanks for the very helpful reply, Reco!
-- 
Felix Natter
debian/rules!

Re: debian-installer RAID question

Thanks Sven!
-- 
Felix Natter
debian/rules!

Re: debian-installer RAID question

Many Thanks for the very helpful reply Andy!
-- 
Felix Natter
debian/rules!

Re: debian-installer RAID question

Darac Marjal  writes:

> On 11/09/2021 17:55, Felix Natter wrote:
>> hello fellow Debian users,
>>
>> I have an SSD for the root filesystem, and two HDDs using RAID1 for
>> /storage running Debian10. Now I need a plan B in case the upgrade
>> fails.
>
> Just want to check that you've not missed something obvious here. You
> don't *need* to involve Debian-Installer in an upgrade. Debian fully
> supports in-place upgrades. The process is documented here:
> https://www.debian.org/releases/bullseye/amd64/release-notes/ch-upgrading.en.html

I wrote "Now I need a plan B in case the upgrade fails" ;-)

Thanks and Best Regards,

Felix
-- 
Felix Natter
debian/rules!

Re: debian-installer RAID question

hi Andrei,

Andrei POPESCU  writes:

thank you for your answer.

> On Sb, 11 sep 21, 18:55:56, Felix Natter wrote:
>> hello fellow Debian users,
>> 
>> I have an SSD for the root filesystem, and two HDDs using RAID1 for
>> /storage running Debian10. Now I need a plan B in case the upgrade
>> fails.
>> 
>> So I made an experiment with a VM and rougly the same setup (disk-wise),
>> and found out that when reinstalling Debian11, the d-i does recognize
>> the RAID1 (/storage) and can reuse it while keeping the data.
>
> Careful here! You can tell d-i to not format an existing partition, but 
> if you want to use it as the root partition the new system will probably 
> be installed in the same file structure as your other data.
>
> Is this what you're considering?

No, the RAID1 is for /storage, the root filesystem (which I intend to
format) is on a separate SSD.

Thanks and Best Regards,
Felix
-- 
Felix Natter
debian/rules!

Re: Pas de son

2021-09-18 Thread Haricophile

Le Sat, 18 Sep 2021 09:52:38 +0200,
Billard François-Marie  a écrit
:

> Bonjour
> 
> par hasard dans l'outil tel que le contrôleur de volume de
> pulseaudio, pour moi sous xfce, avez-vous un possibilité de sélection
> entre haut-parleur ou sortie-audio pour casque ?
> 
> François-Marie

En plus il me semble que sous Bullseye XFCE il faut régler manuellement
la bascule automatique de la sortie son au branchement d'un
périphérique.

dans /etc/pulse/default.pa ou ~/.config/pulse/default.pa

load-module module-switch-on-connect

Re: Your Thoughts on Printer Replacement

2021-09-18 Thread Jeremy Ardley



On 18/9/21 5:00 pm, Jonathan Dowland wrote:

On Sat, Sep 18, 2021 at 10:38:03AM +0200, Klaus Singvogel wrote:

Charles Curley wrote:

I print rarely. Inkjet or laser? Or other? The carts on the L7700 tend
to go bad before they empty,


Rarely printing? I would suggest a laser


I agree. Toner doesn't spoil sat there doing nothing for long periods of
time, whereas the ink for inkjets can do.
The next question then becomes: mono or colour?



My experience is that toner does degrade over a period of years. To get 
full life you need to use your advertised pages within a year or so.


Perhaps this is only for aftermarket cartridges? But overall, the cost 
per page is still 1-3 cents per page even with light use.


--
Jeremy



OpenPGP_signature
Description: OpenPGP digital signature

Re: Debian bullseye

On Sat, Sep 18, 2021 at 07:54:50AM +1200, Dawn Dorsett wrote:
> I have been using Debian for several years and have just upgraded to 
> Bullseye and I wish I hadn't.  It has some bug fixes in it (which were 
> more like minor inconveniences anyway), but Bullseye is SLOW.  Its getting 
> to be as bad as Windows and takes forever to start up.  Libreoffice is 
> particularly slow, and some days I see a delay between hitting the keyboard
>  and the results appearing on the screen.  None of this was a problem with 
> earlier releases. So right now I'm pretty pissed off with Debian.
> 
> Regards,
> _Dawn Dorsett_.
> 

Hi Dawn,

As others in the thread have said: what are the hardware/memory constraints
on your machine? What's the video card / graphic chipset? 

What earlier release were you moving from?

What window manager / desktop environment are you running now?

All of these may be factors: some software is becoming very memory intensive
- looking at you, here, Firefox and Chromium - and the difference is 
sometimes noticeable.

Any and all information you could give might be helpful. Often just writing 
it down is enough to remember "everything was so different before I changed
some one thing" :)

All the very best, as ever,

Andy Cater

Re: Your Thoughts on Printer Replacement

2021-09-18 Thread Jonathan Dowland


On Sat, Sep 18, 2021 at 10:38:03AM +0200, Klaus Singvogel wrote:

Charles Curley wrote:

I print rarely. Inkjet or laser? Or other? The carts on the L7700 tend
to go bad before they empty,


Rarely printing? I would suggest a laser


I agree. Toner doesn't spoil sat there doing nothing for long periods of
time, whereas the ink for inkjets can do. 


The next question then becomes: mono or colour?


Brand? The situation current is this, that only HP released their printer
drivers as Open Source (at least 99% of them).


The direction of travel for printing is entirely driverless, so this is
less important than it used to be. And indeed, driverless printing is a
lot less of a headache to administrate, IMHO, than drivered - even with
open source drivers.

If I were looking for a new laser printer today I'd look at Lexmark (on
a recommendation from a friend), the firmware for which is apparently
based on Yocto Linux.


--
Please do not CC me for listmail.

  Jonathan Dowland
✎j...@debian.org
   https://jmtd.net

installing debian 11.0

2021-09-18 Thread Ngọc Dũng Tạ

When installing debian 11.0, it asks to scan the additional media "debian
gnu/linux 11.0 _ bullseye _ - official amd64 dvd binary-1", so where can I
get it?
Thanks.
-- 
-
*TA NGOC DUNG (Mr)*
*Job:* Marketing Online
*Mobile:* 0962 597 563

Re: krijg Wireguard niet naar verwachting aan de gang met IPv6

2021-09-18 Thread Geert Stappers

On Sat, Sep 18, 2021 at 09:42:13AM +0200, Geert Stappers wrote:
> On Sat, Sep 18, 2021 at 08:51:32AM +0200, Gĳs Hillenius wrote:
  [...] knip
> > client
> > ,
> > | [Interface]
> > | Address = 10.93.15.2/24, fdab:9205:cf78:f608::2/64
> > | PrivateKey = 
> > | 
> > 

> > 
> > en dan testen met ping
> > ping6 ping.xs4all.nl
> > PING ping.xs4all.nl(ping.xs4all.nl (2001:888:0:5::1)) 56 data bytes
> > ^C
> > --- ping.xs4all.nl ping statistics ---
> > 5 packets transmitted, 0 received, 100% packet loss, time 4077ms
> > 

Netwerkpakketten vanaf fdab:9205:cf78:f608::2 bereiken
misschien wel 2001:888:0:5::1, maar pakketen van 2001:888:0:5::1
vinden niet hun weg terug naar fdab:9205:cf78:f608::2.

[1]

Iets om te proberen:
Onder 2a01:4f8:200:546b/64 bijvoorbeeld 2a01:4f8:200:546b:4653/80
hangen. Aan wireguard server geef je 2a01:4f8:200:546b:4653::1
Aan wireguard client geef je 2a01:4f8:200:546b:4653::2  [2]

> > 
> > ping4 ping.xs4all.nl
> > PING  (194.109.6.8) 56(84) bytes of data.
> > 64 bytes from ping.xs4all.nl (194.109.6.8): icmp_seq=1 ttl=56 time=11.3 ms
> > 64 bytes from ping.xs4all.nl (194.109.6.8): icmp_seq=2 ttl=56 time=11.3 ms
> > 

Dat is betekent:
* Of de pakketen vertrekken NIET vanaf 10.93.15.2
* Of de pakketen vertrekken vanaf 10.93.15.2 en ergens gebuerd
  nog Network Address Translation ...

Groeten
Geert Stappers

[1] Tja, de verwachtingen van een "VPN"
[2] In het oorspronkelijke bericht stond o.a.
  Gebruik ik de verkeerde IPv6 reeks?
-- 
https://1optic.io

Re: Your Thoughts on Printer Replacement

2021-09-18 Thread Klaus Singvogel

Charles Curley wrote:
> I print rarely. Inkjet or laser? Or other? The carts on the L7700 tend
> to go bad before they empty, 

Rarely printing? I would suggest a laser, even if the price for printouts
is higher.

Brand? The situation current is this, that only HP released their printer
drivers as Open Source (at least 99% of them). This means that you will
get forever support for these printers with OpenSource - at least in
theory. :-) Think: Even after the CUPS has been replaced in Debian or CUPS
is changing it's internal data/communication/API, the source can be
adapted by someone, who can program, even if HP won't.

Also a printer, which is capable of PostScript oder PDF, should be an
invest in future. As PostScript (PS) is still the most supported printer
language at the moment - PDF is/was only a derived language from PS.

On the other side: I own two lasers from HP because of the OpenSource
situation. I'm not so happy with my MFP device (CM415 - comparable with
MFP 227), as the scanner (!) part is cheap and shows this after some time:
a limited number of pages can be scanned by the sheet-fed and then its no
longer working (typical: 1000 pages), the scans are a way too light by b/w
only pages.

Real PostScript laser printers by Kyocera are not cheap, even I had a very
good experience with them under Linux in a former company.

Conclusion:

It's not easy to recommend a good printer model.

I would never recommend to buy a printer, where you don't get OpenSource
drivers. I'm not speaking of Linux drivers, I speak of Open >Source<
drivers.

Best regards,
Klaus.
-- 
Klaus Singvogel
GnuPG-Key-ID: 1024R/5068792D  1994-06-27

Re: USB Install Fails: does not detect the installer USB drive

2021-09-18 Thread Teemu Likonen

I am replying to an old thread which described the same or similar
problem that I had. I have a solution which may be useful to store in
the mailing list archives. The thread started with this message:

https://lists.debian.org/debian-user/2018/05/msg00391.html

* 2018-05-11 11:11:42-0500, Kent West wrote:

> I have a Dell Latitude E7250 laptop. I'm trying to install Debian to
> it using a USB stick.

> The real problem is that after going through the first three or four
> screens, the install halts, complaining about not being able to read
> the CD-ROM.

I have Dell Latitude E6230 and the same problem: The installer boots and
I can choose country and language settings. Then the installer tries to
detect and mount the installer drive but fails to detect it. I confirmed
that in installer's console: command "ls /dev" lists only the main hard
disk device files /dev/sda and its partitions.

The solution that worked for me: I removed the installer USB stick and
inserted it again. Then the installer finds the device and proceeds with
the install. Technically there is now device file /dev/sdb and its
partitions.

-- 
/// Teemu Likonen - .-.. https://www.iki.fi/tlikonen/
// OpenPGP: 4E1055DC84E9DFF613D78557719D69D324539450

signature.asc
Description: PGP signature

Re: Pas de son

2021-09-18 Thread Billard François-Marie


Bonjour

par hasard dans l'outil tel que le contrôleur de volume de pulseaudio, 
pour moi sous xfce, avez-vous un possibilité de sélection entre 
haut-parleur ou sortie-audio pour casque ?


François-Marie

Le 18/09/2021 à 08:46, Georges a écrit :

Bonjour,

  J'ai un portable récent avec Bullseye et Bookworm :

Intel(R) Core(TM) i3-1005G1 CPU @ 1.20GHz

Audio device: Intel Corporation Ice Lake-LP Smart Sound Technology
Audio Controller (rev 30)

Je n'ai pas de son sur les haut-parleurs mais seulement sur la sortie
casque.

La faute à qui ?

Le bios EFI ? pulseaudio ? bluetooth ? blueman ? pipewire ?

Merci de vos lumières

--

Georges

Re: UEFI + LEGACY op USB

2021-09-18 Thread Roland Clobus

Hallo Paul,

On 17/09/2021 13:43, Paul van der Vlis wrote:
> Ik gebruik een USB-naar-SATA adapter en daaraan hang ik dan een SSD
> waarop ik Debian installeer. En van daar wil ik booten.
> 
> Die SSD zou ik graag zowel bootbaar hebben met UEFI, als via de oude
> legacy methode. Dus dat het werkt onafhankelijk van de instelling van
> het BIOS.
> 
> De Debian USB-sticks kunnen dit, weet iemand hier hoe je dat ook zelf
> kunt realiseren?

Je zou kunnen kijken naar het pakket 'live-build'. Hiermee kun je zelf
live images maken (NB op een net iets andere manier dan de officiële
live images van Debian).

Images gemaakt met live-build kun je zowel met UEFI als legacy starten.
De scripten van live-build installeren zowel een UEFI map als een
isolinux map in de gebouwde images.

De algemene handleiding voor live images is hier:
https://live-team.pages.debian.net/live-manual/html/live-manual/index.en.html

Live-build is hier:
https://packages.debian.org/bullseye/live-build

Met vriendelijke groeten,
Roland Clobus

PS: Als je live-images wilt gaan maken, kijk dan ook eens op
https://wiki.debian.org/ReproducibleInstalls/LiveImages. Daar heb ik een
beschrijving geschreven die het mogelijk maakt om met relatief weinig
netwerkverkeer te experimenteren bij het bouwen van zulke images.
PS2: Disclaimer: ik ben een contributor voor live-build

OpenPGP_signature
Description: OpenPGP digital signature

Re: krijg Wireguard niet aan de gang met IPv6

2021-09-18 Thread Geert Stappers



Op drie kwart van bericht pas een reactie.
En helemaal onderaan ook.

On Sat, Sep 18, 2021 at 08:51:32AM +0200, Gĳs Hillenius wrote:
 [...] knip
> > In configuration van de client staat "ListenPort = 21841".
> > Er is geen reden voor de client om te luisteren.
> 
> Ow! Da's alvast een ding! En ik maar braaf internethandleidingen volgen.
> 
> huidige confs:  (ik laat de regels met comments weg).
> 
> client
> ,
> | [Interface]
> | Address = 10.93.15.2/24, fdab:9205:cf78:f608::2/64
> | PrivateKey = 
> | 
> | [Peer]
> | PublicKey = P3GrgaFCxj6gc6CnOUPo8vxBtKaOcKa7wa8LoL1oUl0=
> | Endpoint = [2a01:4f8:200:546b::9e15:1]:51820
> | AllowedIPs = 0.0.0.0/0, ::/0
> | 
> | PersistentKeepalive = 25
> `
> 
> server
> ,
> | [Interface]
> | Address = 10.93.15.1/24, fdab:9205:cf78:f608::1/64
> | PrivateKey = 
> | ListenPort = 51820
> | 
> | [Peer]
> | PublicKey = nRwfI98C+AFDaLZuaF1i7YWrj7yQDHrQO07XvivGn2U=
> | AllowedIPs = 10.93.15.2/32, fdab:9205:cf78:f608::2/128
> `
> 
> > De client configuratie regel
> >   Endpoint = 2a01:4f8:200:546b::9e15:1:51820
> > vind ik erg vreemd.
> 
> Helemaal mee eens! Ik was (gewoon) onder de indruk van software die al
> die : uit elkaar wist te houden.
> 
> > Een websearch op "IPv6 port syntax"
> > ( https://duckduckgo.com/?q=IPv6+port+syntax )
> > leert dat
> >   Endpoint = [2a01:4f8:200:546b::9e15:1]:51820
> > een grotere kans van slagen heeft.
> 
> Eveneens aangepast. Maar ik ben er nog niet.
> 
> > Hoe er rekening mee dat na die twee veranderingen
> > het geheel nog niet compleet hoeft te zijn.
> > In dat geval a.u.b. de bijgewerkte versie van beide configuraties laten
> > zien, alleen de Private key verborgen, de Public keys wel laten zien. ook
> > de output van `sudo wg` van beide kanten. De `wg` output in zijn geheel
> > laten zien, want de private key wordt verborgen gehouden.
> > Plus wat tekst hoe het kapot gaat danwel wat er zou moeten werken
> > maar dat net niet doet.
> 
> Op de client deed ik net:
> 
> 
> wg-quick up wg0
> [#] ip link add wg0 type wireguard
> [#] wg setconf wg0 /dev/fd/63
> [#] ip -4 address add 10.93.15.2/24 dev wg0
> [#] ip -6 address add fdab:9205:cf78:f608::2/64 dev wg0
> [#] ip link set mtu 1420 up dev wg0
> [#] wg set wg0 fwmark 51820
> [#] ip -6 route add ::/0 dev wg0 table 51820
> [#] ip -6 rule add not fwmark 51820 table 51820
> [#] ip -6 rule add table main suppress_prefixlength 0
> [#] ip6tables-restore -n
> [#] ip -4 route add 0.0.0.0/0 dev wg0 table 51820
> [#] ip -4 rule add not fwmark 51820 table 51820
> [#] ip -4 rule add table main suppress_prefixlength 0
> [#] sysctl -q net.ipv4.conf.all.src_valid_mark=1
> [#] iptables-restore -n
> 
> 
> en dan testen met ping
> ping6 ping.xs4all.nl
> PING ping.xs4all.nl(ping.xs4all.nl (2001:888:0:5::1)) 56 data bytes
> ^C
> --- ping.xs4all.nl ping statistics ---
> 5 packets transmitted, 0 received, 100% packet loss, time 4077ms
> 
> 
> ping4 ping.xs4all.nl
> PING  (194.109.6.8) 56(84) bytes of data.
> 64 bytes from ping.xs4all.nl (194.109.6.8): icmp_seq=1 ttl=56 time=11.3 ms
> 64 bytes from ping.xs4all.nl (194.109.6.8): icmp_seq=2 ttl=56 time=11.3 ms
> 
> 
> Nog steeds op de client:
> 
> /etc/wireguard# wg
> interface: wg0
>   public key: nRwfI98C+AFDaLZuaF1i7YWrj7yQDHrQO07XvivGn2U=
>   private key: (hidden)
>   listening port: 48540
>   fwmark: 0xca6c
> 
> peer: 
>   endpoint: [2a01:4f8:200:546b::9e15:1]:51820
>   allowed ips: 0.0.0.0/0, ::/0
>   latest handshake: 4 seconds ago
>   transfer: 11.93 KiB received, 24.81 KiB sent
>   persistent keepalive: every 25 seconds
> 
> Op de server:
> 
> wg-quick up wg0
> [#] ip link add wg0 type wireguard
> [#] wg setconf wg0 /dev/fd/63
> [#] ip -4 address add 10.93.15.1/24 dev wg0
> [#] ip -6 address add fdab:9205:cf78:f608::1/64 dev wg0
> [#] ip link set mtu 1420 up dev wg0
> 
> 
> Gek of niet? Di's veel korter dan op de client, niets over routes en sysctl
 
Ja, dat viel me ook op.
Het kan zijn dat we dat later nog moeten najagen.

Eerste het oorspronkelijke probleem.

 
> als root, de output van wg
> 
> interface: wg0
>   public key: P3GrgaFCxj6gc6CnOUPo8vxBtKaOcKa7wa8LoL1oUl0=
>   private key: (hidden)
>   listening port: 51820
> 
> peer: 
>   endpoint: [2a02:a03f:6b2d:b400:5fb2:6d1d:1a9a:dc69]:48540
>   allowed ips: 10.93.15.2/32, fdab:9205:cf78:f608::2/128
>   latest handshake: 50 seconds ago
>   transfer: 106.50 KiB received, 249.61 KiB sent
> 
> ^^^ die received sent cijfers groeien als ik ping4 doe. Ze groeien OOK
> als ik ping6 doe.
> 

Configuratie gezien en goed bevonden   :-)
 

Groeten
Geert Stappers
-- 
https://1optic.io

Re: Your Thoughts on Printer Replacement

2021-09-18 Thread mick crane


On 2021-09-17 23:41, Charles Curley wrote:

I have an HP Officejet Pro L7700, which is starting to show its age.
Also HP has discontinued the standard size cartridges. I can get the
large ones. I suspect I could buy a printer for what four large
cartridges would cost me.

Requirements:

* I print rarely, and I do use color. The carts on the L7700 tend to
  go bad before they empty, Inkjet or laser? Or other?

* Buster and Bullseye should both support any recommendations. I don't
  use Windows.

I've had good results with HP over the years, and they support Linux
well.

Reccomendations?


I use elderly HP LaserJet with a PC doing Cups server which is easy 
enough once you find out the name of the print queue.
Did have a Xerox Phaser for colour which worked very well until it had 
an accident to do with gravity.

mick
--
Key ID4BFEBB31

Re: UEFI + LEGACY op USB

2021-09-18 Thread Geert Stappers

On Fri, Sep 17, 2021 at 01:43:10PM +0200, Paul van der Vlis wrote:
> Hallo,
> 
> Ik gebruik een USB-naar-SATA adapter en daaraan hang ik dan een SSD waarop
> ik Debian installeer. En van daar wil ik booten.
> 
> Die SSD zou ik graag zowel bootbaar hebben met UEFI, als via de oude legacy
> methode. Dus dat het werkt onafhankelijk van de instelling van het BIOS.
> 
> De Debian USB-sticks kunnen dit, weet iemand hier hoe je dat ook zelf kunt
> realiseren?


Interresante uitdaging.

Wild idee:

De SSD voorzien met Debian-Installer plus een preseed-file.
In die preseed-file ook aanroep van aanvullende scripts.

Bij de eerste boot moet er dan dit gebeuren:

* Debian-Installer wordt gestart
* preseed file wordt gevonden
* inhoud van preseed file wordt gebruikt
* automatisch verloop van de installatie
  hierbij wordt de SSD overschreven
* Reboot



Groeten
Geert Stappers
-- 
Silence is hard to parse

Re: buster => bullseye, upgrade vs reinstall

2021-09-18 Thread Jean Bernon

Bonjour,

Ayant souvent fait auparavant des ugrade, j'ai fait une réinstallation complète 
de buster à bullseye. J'ai créé 2 nouvelles partitions / et /home. Après 
l'installation de bullseye, j'ai monté l'ancienne partition et recopié ce qui 
m'est utile (dossier musique, documents, images ...) et réinstallé quelques 
applis. Au final c'est plus performant qu'un upgrade parce que l'installation 
va très vite, la recopie et la réinstallation d'applis prend un peu de temps, 
mais c'est un ménage salutaire, et surtout le système est plus cohérent et tout 
fonctionne plus vite. Pour être complet, au passage la partition / est passée 
du disque dur à un petit SSD qui était inutilisé sur ma machine de 2013, ce qui 
explique en partie les meilleures performances, mais le démarrage de la machine 
jusqu'à l'ouverture de Firefox prend un peu plus de 30 s contre 3 minutes 
auparavant.

Jean

Re: krijg Wireguard niet aan de gang met IPv6

2021-09-18 Thread Gĳs Hillenius



[...] knip

>> 
>> Op IRC werd ik gewezen op een bug:
>> 
>> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=993716
>
> Package: bridge-utils
> Titel: IPv6 network bridge fails after upgrading Buster to Bullseye
>  
>> ik ben er nog niet helemaal zeker van of dat is wat me hier overkomt.
>
> Mijn inschatting is van niet.
>  
>
> In configuration van de client staat "ListenPort = 21841".
> Er is geen reden voor de client om te luisteren.

Ow! Da's alvast een ding! En ik maar braaf internethandleidingen volgen.

> Althans niet in de set-up die ik succesvol ingebruik heb.

huidige confs:  (ik laat de regels met comments weg).

client

,
| [Interface]
| Address = 10.93.15.2/24, fdab:9205:cf78:f608::2/64
| PrivateKey = 
| 
| [Peer]
| PublicKey = P3GrgaFCxj6gc6CnOUPo8vxBtKaOcKa7wa8LoL1oUl0=
| Endpoint = [2a01:4f8:200:546b::9e15:1]:51820
| AllowedIPs = 0.0.0.0/0, ::/0
| 
| PersistentKeepalive = 25
`

server
,
| [Interface]
| Address = 10.93.15.1/24, fdab:9205:cf78:f608::1/64
| PrivateKey = 
| ListenPort = 51820
| 
| [Peer]
| PublicKey = nRwfI98C+AFDaLZuaF1i7YWrj7yQDHrQO07XvivGn2U=
| AllowedIPs = 10.93.15.2/32, fdab:9205:cf78:f608::2/128
`



>
> De client configuratie regel
>   Endpoint = 2a01:4f8:200:546b::9e15:1:51820
> vind ik erg vreemd.

Helemaal mee eens! Ik was (gewoon) onder de indruk van software die al
die : uit elkaar wist te houden.

> Mijn (te korte??) websearch leverde
> geen voorbeeld op dat het zo zou kunnen werken.
>
> Een websearch op "IPv6 port syntax"
> ( https://duckduckgo.com/?q=IPv6+port+syntax )
> leert dat
>   Endpoint = [2a01:4f8:200:546b::9e15:1]:51820
> een grotere kans van slagen heeft.

Eveneens aangepast. Maar ik ben er nog niet.


> Hoe er rekening mee dat na die twee veranderingen
> het geheel nog niet compleet hoeft te zijn.
> In dat geval a.u.b. de bijgewerkte versie van beide configuraties laten
> zien, alleen de Private key verborgen, de Public keys wel laten zien. ook
> de output van `sudo wg` van beide kanten. De `wg` output in zijn geheel
> laten zien, want de private key wordt verborgen gehouden.
> Plus wat tekst hoe het kapot gaat danwel wat er zou moeten werken
> maar dat net niet doet.

Op de client dee ik net:


wg-quick up wg0
[#] ip link add wg0 type wireguard
[#] wg setconf wg0 /dev/fd/63
[#] ip -4 address add 10.93.15.2/24 dev wg0
[#] ip -6 address add fdab:9205:cf78:f608::2/64 dev wg0
[#] ip link set mtu 1420 up dev wg0
[#] wg set wg0 fwmark 51820
[#] ip -6 route add ::/0 dev wg0 table 51820
[#] ip -6 rule add not fwmark 51820 table 51820
[#] ip -6 rule add table main suppress_prefixlength 0
[#] ip6tables-restore -n
[#] ip -4 route add 0.0.0.0/0 dev wg0 table 51820
[#] ip -4 rule add not fwmark 51820 table 51820
[#] ip -4 rule add table main suppress_prefixlength 0
[#] sysctl -q net.ipv4.conf.all.src_valid_mark=1
[#] iptables-restore -n


en dan testen met ping
ping6 ping.xs4all.nl
PING ping.xs4all.nl(ping.xs4all.nl (2001:888:0:5::1)) 56 data bytes
^C
--- ping.xs4all.nl ping statistics ---
5 packets transmitted, 0 received, 100% packet loss, time 4077ms


ping4 ping.xs4all.nl
PING  (194.109.6.8) 56(84) bytes of data.
64 bytes from ping.xs4all.nl (194.109.6.8): icmp_seq=1 ttl=56 time=11.3 ms
64 bytes from ping.xs4all.nl (194.109.6.8): icmp_seq=2 ttl=56 time=11.3 ms


Nog steeds op de client:

/etc/wireguard# wg
interface: wg0
  public key: nRwfI98C+AFDaLZuaF1i7YWrj7yQDHrQO07XvivGn2U=
  private key: (hidden)
  listening port: 48540
  fwmark: 0xca6c

peer: 
  endpoint: [2a01:4f8:200:546b::9e15:1]:51820
  allowed ips: 0.0.0.0/0, ::/0
  latest handshake: 4 seconds ago
  transfer: 11.93 KiB received, 24.81 KiB sent
  persistent keepalive: every 25 seconds

Op de server:

wg-quick up wg0
[#] ip link add wg0 type wireguard
[#] wg setconf wg0 /dev/fd/63
[#] ip -4 address add 10.93.15.1/24 dev wg0
[#] ip -6 address add fdab:9205:cf78:f608::1/64 dev wg0
[#] ip link set mtu 1420 up dev wg0


Gek of niet? Di's veel korter dan op de client, niets over routes en sysctl


als root, de output van wg

interface: wg0
  public key: P3GrgaFCxj6gc6CnOUPo8vxBtKaOcKa7wa8LoL1oUl0=
  private key: (hidden)
  listening port: 51820

peer: 
  endpoint: [2a02:a03f:6b2d:b400:5fb2:6d1d:1a9a:dc69]:48540
  allowed ips: 10.93.15.2/32, fdab:9205:cf78:f608::2/128
  latest handshake: 50 seconds ago
  transfer: 106.50 KiB received, 249.61 KiB sent

^^^ die received sent cijfers groeien als ik ping4 doe. Ze groeien OOK
als ik ping6 doe.

Pas de son