Re: Why or why not back up "/lost+found"

[tomas]

On Thu, Aug 10, 2023 at 03:42:01PM -0400, Default User wrote:
> Hi!
> 
> When backing up my system I have been using this exclusions list: 
> 
> /dev/*
> /proc/*
> /sys/*
> /tmp/*
> /run/*
> /mnt/*
> /media/*
> /lost+found
> 
> There are many sources online that suggest that "/lost+found" should be
> excluded from backups, but I can't seem to find a good explanation for
> why.

Squinting the other way (which doesn't contradict what others have
said in this thread): if you see anything in lost+found, the idea
is that you do something about it (rescue it, throw it away). Thus,
the content of lost+found is (or should be) as temporary as, say,
/tmp. And you don't back up that.

So I'd tend to exclude the special "lost+found" dirs (perhaps not
any dir called like that). But then, it's not that important.

Cheers
-- 
t


signature.asc
Description: PGP signature

Re: UNUBSCRIBE

[Charles Curley]

On Thu, 10 Aug 2023 19:20:30 -0700
glenn green  wrote:

> 

You would probably do better to send that to a special email address
intended for managing subscriptions. On most lists, you can find it by
searching the headers (CTL-H on many mail readers) for "unsubscribe".

-- 
Does anybody read signatures any more?

https://charlescurley.com
https://charlescurley.com/blog/

Re: apt policy / listing packages from repo

[Gareth Evans]

On Thu 10 Aug 2023, at 18:54, Sven Joachim  wrote:
> On 2023-08-10 09:30 -0500, David Wright wrote:
>
>>> I was looking for a way to list packages installed from a particular
>>> repo and/or sub-repo or whatever it's called (eg. main, non-free).
>>> 
>>> Does anyone know of a way to do this, with apt policy or otherwise?
>>
>> What I do in this situation is to type "apt" and press TAB twice.
>> Look at the resulting list of commands and check the man page for
>> the most likely looking, in this case apt-cache.
>>
>> An alternative method of course is to type   apt policy   into any
>> search engine. This will typically tell you not only how to invoke
>> the command, but also more about what it produces.
>>
>> As for your listing, I've done this in the past with a script that
>> runs apt-cache dump, grepping the Package/Version/File lines,
>> concatenating and sorting them, then filtering that list against
>> the output of dpkg-query -W -f to include only installed packages.
>> This yields a list like:
>>
>>   Package: acl Version: 2.2.53-10 File: 
>> /var/lib/apt/lists/deb.debian.org_debian_dists_bullseye_main_binary-amd64_Packages
>>   Package: adduser Version: 3.118 File: 
>> /var/lib/apt/lists/deb.debian.org_debian_dists_bullseye_main_binary-amd64_Packages
>>   [ … … ]
>>   Package: xtoolwait Version: 1.3-6.2 File: /var/lib/dpkg/status
>>   [ … … ]
>>   Package: yt-dlp Version: 2023.03.04-1~bpo11+1 File: /var/lib/dpkg/status
>>
>> which you can grep for particular subsets, though I'm usually more
>> interested in grep -v for packages originating from elsewhere,
>> like xtoolwait (squeeze) and yt-dlp (backports) there.
>>
>> There may well be better ways.
>
> I would probably use "apt list" with a search pattern described in
> apt-patterns(7), e.g. the following command lists all installed packages
> from non-free:
>
> $ apt list '~i ~snon-free/'
>
> Lots of interesting possibilities one can toy with. :-)

I will look into both approaches - my thanks to you and David.

G

Re: Weird messages on logs.

[Maureen L Thomas]

Sorry, it was late when I posted this.  I am using Bookworm and checked 
the log not figuring on finding anything and found this. My system has 
been upgrade to the last fixes for Bookworm.  I am using a Lanova all in 
one machine.  2TB hard drive with 8 for ram.


Moe

On 8/8/23 7:57 AM, Henning Follmann wrote:

On Tue, Aug 08, 2023 at 12:01:08AM -0400, Maureen L Thomas wrote:

 3:24:29 PM systemd: Failed to start tracker-extract-3.service - Tracker
metadata extractor.
 3:24:29 PM systemd: Failed to start tracker-extract-3.service - Tracker
metadata extractor.
 3:23:59 PM systemd: Failed to start tracker-extract-3.service - Tracker
metadata extractor.
 3:23:29 PM systemd: Failed to start
app-gnome-gnome\x2dkeyring\x2dsecrets-1958.scope - Application launched by
gnome-session-binary.
 3:23:29 PM systemd: Failed to start
app-gnome-gnome\x2dkeyring\x2dpkcs11-1959.scope - Application launched by
gnome-session-binary.

  These are the message I have in my logs.  I have no idea what it is about.
I would appreciate any help from you guys.

Thanks

Moe

Hello,
what release is this on?
On Buster I had also some issues with tracker mainly due to apparmor
missing some rules. I have not seen this so far on Bullseye.
-H

UNUBSCRIBE

[glenn green]

Re: logging no longer standard?

[Max Nikulin]

On 10/08/2023 16:53, gene heskett wrote:

On 8/9/23 21:15, Max Nikulin wrote:

On 08/08/2023 09:57, gene heskett wrote:
dbus-update-activation-environment: error: unable to connect to 
D-Bus: Failed to connect to socket /run/user/1000/bus: Connection 
refused

...
Try to figure out at which moment such messages appear. Try "busctl 
--user" and "busctl --user status" as a sanity check.


Both of those get somewhat copious outputs, what should I be looking 
for?  Should it name the app?


You can compare usual output of the commands and their output just after 
the error message appears. You may get some impression of features 
exposed through D-Bus.



Looking in /etc/dbus-1, I see two dirs, one of which is empty:


What do you expect to find there? Packages put files into 
/usr/share/dbus-1. Systemd units may tell that they may be activated 
through D-Bus. Applications may express in .desktop files that they 
should handle arguments through D-Bus and Exec=... should be ignored.


dbus has always been a puzzle to me. It has no man pages to explain its 
functions.


There are a lot of docs for particular D-Bus APIs, e.g.
https://www.freedesktop.org/software/systemd/man/org.freedesktop.hostname1.html

If you are looking for some overview and introduction then starting 
points may be

https://en.wikipedia.org/wiki/D-Bus
and docs linked from
https://www.freedesktop.org/wiki/Software/dbus/
e.g.
https://www.freedesktop.org/wiki/IntroductionToDBus/

The point is that if some application can not connect to D-Bus then 
arbitrary features may be broken. However your error is either extremely 
severe since session D-Bus is not running at all or it is almost 
harmless since it appears due to some race on login or logout.

Re: Why or why not back up "/lost+found"

[Greg Wooledge]

On Thu, Aug 10, 2023 at 09:15:31PM -0400, Default User wrote:
> Fortunately, since my lost+found  directories are all empty, I have no
> "raw material" to practice extraction on.

There's no "extraction".  If a file is there, it will be a file.  It
will have a meaningful owner, group and permissions.  It will have a
meaningless name.  You will have to guess what it is based on the
owner, group, permissions, timestamps, and contents.

> Now at this point, it seems that I could:
> 
> 1)  Continue to back up /home/lost+found and /var/lost+found, but NOT 
> back up /lost+found.
> 
> 2)  Start backing up /home/lost+found AND /var/lost+found AND
> /lost+found. 
> 
> 3)  Start NOT backing up /home/lost+found OR /var/lost+found OR
> /lost+found.
> 
> What is everyone else doing about backing up, or not backing up, the
> various lost+found directories on their systems? 

You are asking the wrong question.

If you find a file in lost+found, SOMETHING BAD HAS HAPPENED.  You
will want to figure out how bad the damage is, and decide what to do
about it.

If the damage is extremely small, then you might be able to figure
out where the files go, and move them.

If the damage is large, you may need to replace the disk and restore
your last working backup.  Which will not have anything in lost+found.

Backing up lost+found is pretty meaningless, because if you're backing
up a lost+found with files in it, it means you're backing up a damaged
file system.

What you should be asking instead is whether your backup should ALERT
you if it finds stuff in lost+found.  Having it quietly back up the
files is not as helpful as you might be imagining.

Re: Why or why not back up "/lost+found"

[Default User]

On Thu, 2023-08-10 at 22:03 +, Minecraftchest1 wrote:
> This post on the U Stack Exchange sitr summs it up fairly well.
> https://unix.stackexchange.com/a/18157.
> 
> In short, `lost+found` is a place for fscheck to link filesystem
> entries that don't have an entry anywhere on the file system, during
> a filesystem check. Chances are, the files there were bring deleted,
> but weren't fully removed from the file system due to an open handle
> on 
> the file. If you aren't specifically looking for a file that got lost
> due to a system crash, the files there likely aren't relavent
> and can be safely ignored.
> 
> 
> On August 10, 2023 9:46:27 PM UTC, Default User
>  wrote:
> > On Thu, 2023-08-10 at 21:45 +0200, Nicolas George wrote:
> > > Default User (12023-08-10):
> > > > And, if /lost+found should be excluded, then shouldn't
> > > > "lost+found"
> > > > in
> > > > any other directories be excluded from backups as well? Why/why
> > > > not? 
> > > > 
> > > 
> > > Before anybody answers all your question, there is something that
> > > needs
> > > checking:
> > > 
> > > Do you know what lost+found is for?
> > > 
> > > If not, please read some documentation about it. The explanation
> > > of
> > > its
> > > function contains the answer to the question I left quoted above.
> > > 
> > > Regards,
> > > 
> > > 
> > 
> > 
> > Hi, Nicholas.
> > 
> > Well, I did read some documentation, and I have at least a
> > rudimentary
> > understanding of fsck and lost+found. 
> > 
> > Unfortunately, I regret to say that I did not find that the answer
> > to
> > the question(s) about lost+found in the original post were
> > contained in
> > the explanation(s) of its function - at least in what I have read
> > so
> > far. 
> > 
> > Or, maybe I'm just stupid. 
> > 
> > I'm sorry if I have bothered you. 
> > 


Three of my mount points, each of which is on a separate partition,
have lost+found directories:
/lost+found
/home/lost+found
/var/lost+found

All of these appear to be empty. 

Of these, I have rsnapshot configured to exclude:
/lost+found

but the other two:
/home/lost+found
/var/lost+found

ARE currently being backed up.

The backups are to an external usb hard drive, mounted as:
/media/[user]/MSD1
which has its own lost+found:
/media/[user]/MSD1/lost+found

which also appears to be empty. 

I have read a number of things online about fsck and lost+found,
including the previously mentioned:
https://unix.stackexchange.com/a/18157. 

I think I understand the basic concept of how fsck puts stray files and
file parts in lost+found, and (in theory) how to manually extract whole
and partial files from lost+found. 

Fortunately, since my lost+found  directories are all empty, I have no
"raw material" to practice extraction on.

Now at this point, it seems that I could:

1)  Continue to back up /home/lost+found and /var/lost+found, but NOT 
back up /lost+found.

2)  Start backing up /home/lost+found AND /var/lost+found AND
/lost+found. 

3)  Start NOT backing up /home/lost+found OR /var/lost+found OR
/lost+found.

What is everyone else doing about backing up, or not backing up, the
various lost+found directories on their systems? 

Bonus points for why you do it that way!

:)

Re: Why or why not back up "/lost+found"

[David Wright]

On Thu 10 Aug 2023 at 15:52:02 (-0700), Bob McGowan wrote:
> On 8/10/23 03:03 PM, Nicolas George wrote:
> > Default User (12023-08-10):
> > > > > And, if /lost+found should be excluded, then shouldn't "lost+found"
> > > > > in any other directories be excluded from backups as well? Why/why
> > > > > not?
> > > Unfortunately, I regret to say that I did not find that the answer to
> > > the question(s) about lost+found in the original post were contained in
> > > the explanation(s) of its function - at least in what I have read so
> > > far.
> > The lost+found directory at the root of the filesystem is special, it is
> > created when the filesystem is created and set up to receive orphaned
> > files. A lost+found directory elsewhere… is just a directory with a
> > wacky name.
> > 
> Almost but not quite completely true.
> 
> If you have more than one filesystem on your host, they may also have
> lost+found directories which will show up in their mount points.
> 
> For example, with your home on a separate disk, mounted on /home,
> there may be a /home/lost+found.
> 
> The same thinking applies to them as to /lost+found.

But /home/lost+found is still called /lost+found on the filesystem
it belongs to; it's only now called /home/lost+found because you
mounted that filesystem on /home. That's why Nicolas wrote "at the
root of the filesystem"; that's any filesystem (that uses the concept).

The OP should also note that you musn't use mkdir to create such
directories: mklost+found should be used instead. Typically it
pre-allocates space so that fsck doesn't have to disturb the rest
of the filesystem when it runs.

Cheers,
David.

Re: Why or why not back up "/lost+found"

[Bob McGowan]

On 8/10/23 03:03 PM, Nicolas George wrote:

Default User (12023-08-10):

And, if /lost+found should be excluded, then shouldn't "lost+found"
in any other directories be excluded from backups as well? Why/why
not?

Unfortunately, I regret to say that I did not find that the answer to
the question(s) about lost+found in the original post were contained in
the explanation(s) of its function - at least in what I have read so
far.

The lost+found directory at the root of the filesystem is special, it is
created when the filesystem is created and set up to receive orphaned
files. A lost+found directory elsewhere… is just a directory with a
wacky name.

Regards,


Almost but not quite completely true.

If you have more than one filesystem on your host, they may also have 
lost+found directories which will show up in their mount points.


For example, with your home on a separate disk, mounted on /home, there 
may be a /home/lost+found.


The same thinking applies to them as to /lost+found.

Any other location, then it is a user creating a directory or file with 
a wacky name, as Nicolas suggests.


Bob

Re: Why or why not back up "/lost+found"

[Greg Wooledge]

On Fri, Aug 11, 2023 at 12:03:26AM +0200, Nicolas George wrote:
> Default User (12023-08-10):
> > > > And, if /lost+found should be excluded, then shouldn't "lost+found"
> > > > in any other directories be excluded from backups as well? Why/why
> > > > not? 
> 
> > Unfortunately, I regret to say that I did not find that the answer to
> > the question(s) about lost+found in the original post were contained in
> > the explanation(s) of its function - at least in what I have read so
> > far. 
> 
> The lost+found directory at the root of the filesystem is special, it is
> created when the filesystem is created and set up to receive orphaned
> files. A lost+found directory elsewhere… is just a directory with a
> wacky name.

Expanding on this: the only time there are ever going to be files in
the lost+found directory is if your file system has suffered some
damage/corruption, and fsck has found some files that are not
referenced by any directory.  Fsck puts those files in lost+found with
a procedurally generated name, since fsck doesn't have any idea what
the file's original names were.

If you ever see files in lost+found, then you know that this has
occurred.  Furthermore, it is on *you* to try to figure out what each
file is, where it came from, and what it should be called.  This will
be a 100% manual process.

Re: Why or why not back up "/lost+found"

[Nicolas George]

Default User (12023-08-10):
> > > And, if /lost+found should be excluded, then shouldn't "lost+found"
> > > in any other directories be excluded from backups as well? Why/why
> > > not? 

> Unfortunately, I regret to say that I did not find that the answer to
> the question(s) about lost+found in the original post were contained in
> the explanation(s) of its function - at least in what I have read so
> far. 

The lost+found directory at the root of the filesystem is special, it is
created when the filesystem is created and set up to receive orphaned
files. A lost+found directory elsewhere… is just a directory with a
wacky name.

Regards,

-- 
  Nicolas George


signature.asc
Description: PGP signature

Re: Why or why not back up "/lost+found"

[Default User]

On Thu, 2023-08-10 at 21:45 +0200, Nicolas George wrote:
> Default User (12023-08-10):
> > And, if /lost+found should be excluded, then shouldn't "lost+found"
> > in
> > any other directories be excluded from backups as well? Why/why
> > not? 
> 
> Before anybody answers all your question, there is something that
> needs
> checking:
> 
> Do you know what lost+found is for?
> 
> If not, please read some documentation about it. The explanation of
> its
> function contains the answer to the question I left quoted above.
> 
> Regards,
> 


Hi, Nicholas.

Well, I did read some documentation, and I have at least a rudimentary
understanding of fsck and lost+found.  

Unfortunately, I regret to say that I did not find that the answer to
the question(s) about lost+found in the original post were contained in
the explanation(s) of its function - at least in what I have read so
far. 

Or, maybe I'm just stupid. 

I'm sorry if I have bothered you. 

Re: apt-key is deprecated

[ajh-valmer]

On Thursday 10 August 2023 21:01:15 didier gaumet wrote:
> Rappel:
> - Quand tu vois un signe $ en début de ligne, c'est le prompt d'un 
> utilisateur ordinaire, ce sont des commandes à taper en utilisateur 
> ordinaire
> - quand tu vois un dièse en début de ligne, c'est le prompt de root, ce 
> sont des commandes à taper en tant que root :

Pour ce premier paragraphe ci-dessus, tu me prends pour un super débutant.
Ça fait 20 ans que je pratique GNU/Linux et suis admin-sys sur un serveur,
et on ne peut pas tout connaître.
Avec toi, faut pas être susceptible du tout :-)

(Que de lignes...) 
Voyons la suite, le 2ème paragraphe :
> => BON , ENCORE PLUS SIMPLE, TAPE ÇA EN TANT QUE ROOT:
> # dpkg -l trinity-keyring
> si il est installé, comme c'est peut-être une vieille version (apt-key 
> uniquement), tu l'enlèves:
> # dpkg -P trinity-keyring.deb
> ensuite, toujours en tant que root:
> # wget http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-keyring.deb
> # dpkg -i trinity-keyring.deb
> puis tu fais un apt update pour voir si tu as toujours le warning
> SI ÇA MARCHE, C'EST FINI
> SI ÇA MARCHE PAS TU FAIS ÇA EN TANT QUE ROOT:
> (pour bookworm, si tu as une autre version, par exemple buster, tu 
> remplaces bookworm par buster) :

Il aurait suffit d'écrire, tente les commandes de mon 1er mail en root 
ou .

La suite, le 3ème paragraphe :
> dans ton sources.list, tu remplaces les lignes :
> deb http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps
> deb-src http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps
> par les lignes:
> deb [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
> http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps
> deb-src [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
> http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps
> SI ÇA MARCHE, C'EST FINI
> SI ÇA MARCHE PAS TU FAIS ÇA EN TANT QUE ROOT:
> dans ton sources.list, tu remplaces maintenant les lignes :
> deb [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
> http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps
> deb-src [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
> http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps
> par les lignes:
> deb [signed-by=/usr/share/keyrings/trinity-archive-keyring.gpg] 
> http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps
> deb-src [signed-by=/usr/share/keyrings/trinity-archive-keyring.gpg] 
> http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
> bookworm main deps :
Il aurait suffit d'écrire :
Tente ces lignes dans sources.list :
deb [signed-by=/usr/share/keyrings/trinity-archive-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps
deb-src [signed-by=/usr/share/keyrings/trinity-archive-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
(en diézant avant celles présentes si échec).

Bon, allez, tu as rédigé une réponse très complète, qui prend du temps,
je t'en remercie, mais je ne suis pas optimiste sur la méthode...
Bonne soirée.

Re: Why or why not back up "/lost+found"

[Nicolas George]

Default User (12023-08-10):
> And, if /lost+found should be excluded, then shouldn't "lost+found" in
> any other directories be excluded from backups as well? Why/why not? 

Before anybody answers all your question, there is something that needs
checking:

Do you know what lost+found is for?

If not, please read some documentation about it. The explanation of its
function contains the answer to the question I left quoted above.

Regards,

-- 
  Nicolas George


signature.asc
Description: PGP signature

Why or why not back up "/lost+found"

[Default User]

Hi!

When backing up my system I have been using this exclusions list: 

/dev/*
/proc/*
/sys/*
/tmp/*
/run/*
/mnt/*
/media/*
/lost+found

There are many sources online that suggest that "/lost+found" should be
excluded from backups, but I can't seem to find a good explanation for
why.

I do not know if restoring /lost+found would mess up the file system,
or is it simply "not necessary" to back up/restore /lost+found. 

If backup/restore of /lost+found would not hurt anything, then why not
always back it up for the sake of thoroughness, "just in case".  After
all, you can't restore what you have not backed up. 

And, if /lost+found should be excluded, then shouldn't "lost+found" in
any other directories be excluded from backups as well? Why/why not? 

I think backups are important, so I would like to know now, before I
NEED to know later. 

BTW, I am currently using Timeshift for my main system backups (no home
directory), and am using rsync/rsnapshot to back up /, except for the
exclusions listed above. ("Belt and suspenders!"). 

Re: apt-key is deprecated

[didier gaumet]

Rappel:
- Quand tu vois un signe $ en début de ligne, c'est le prompt d'un 
utilisateur ordinaire, ce sont des commandes à taper en utilisateur 
ordinaire
- quand tu vois un dièse en début de ligne, c'est le prompt de root, ce 
sont des commandes à taper en tant que root


=> BON , ENCORE PLUS SIMPLE, TAPE ÇA EN TANT QUE ROOT:

# dpkg -l trinity-keyring

si il est installé, comme c'est peut-être une vieille version (apt-key 
uniquement), tu l'enlèves:

# dpkg -P trinity-keyring.deb

ensuite, toujours en tant que root:

# wget http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-keyring.deb
# dpkg -i trinity-keyring.deb

puis tu fais un apt update pour voir si tu as toujours le warning

SI ÇA MARCHE, C'EST FINI

SI ÇA MARCHE PAS TU FAIS ÇA EN TANT QUE ROOT:
(pour bookworm, si tu as une autre version, par exemple buster, tu 
remplaces bookworm par buster)


dans ton sources.list, tu remplaces les lignes :
deb http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps
deb-src http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps


par les lignes:
deb [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps
deb-src [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps


SI ÇA MARCHE, C'EST FINI

SI ÇA MARCHE PAS TU FAIS ÇA EN TANT QUE ROOT:

dans ton sources.list, tu remplaces maintenant les lignes :
deb [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps
deb-src [signed-by=/usr/share/keyrings/trinity-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps


par les lignes:
deb [signed-by=/usr/share/keyrings/trinity-archive-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps
deb-src [signed-by=/usr/share/keyrings/trinity-archive-keyring.gpg] 
http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-r14.1.x 
bookworm main deps

Re: apt policy / listing packages from repo

[Sven Joachim]

On 2023-08-10 09:30 -0500, David Wright wrote:

>> I was looking for a way to list packages installed from a particular
>> repo and/or sub-repo or whatever it's called (eg. main, non-free).
>> 
>> Does anyone know of a way to do this, with apt policy or otherwise?
>
> What I do in this situation is to type "apt" and press TAB twice.
> Look at the resulting list of commands and check the man page for
> the most likely looking, in this case apt-cache.
>
> An alternative method of course is to type   apt policy   into any
> search engine. This will typically tell you not only how to invoke
> the command, but also more about what it produces.
>
> As for your listing, I've done this in the past with a script that
> runs apt-cache dump, grepping the Package/Version/File lines,
> concatenating and sorting them, then filtering that list against
> the output of dpkg-query -W -f to include only installed packages.
> This yields a list like:
>
>   Package: acl Version: 2.2.53-10 File: 
> /var/lib/apt/lists/deb.debian.org_debian_dists_bullseye_main_binary-amd64_Packages
>   Package: adduser Version: 3.118 File: 
> /var/lib/apt/lists/deb.debian.org_debian_dists_bullseye_main_binary-amd64_Packages
>   [ … … ]
>   Package: xtoolwait Version: 1.3-6.2 File: /var/lib/dpkg/status
>   [ … … ]
>   Package: yt-dlp Version: 2023.03.04-1~bpo11+1 File: /var/lib/dpkg/status
>
> which you can grep for particular subsets, though I'm usually more
> interested in grep -v for packages originating from elsewhere,
> like xtoolwait (squeeze) and yt-dlp (backports) there.
>
> There may well be better ways.

I would probably use "apt list" with a search pattern described in
apt-patterns(7), e.g. the following command lists all installed packages
from non-free:

$ apt list '~i ~snon-free/'

Lots of interesting possibilities one can toy with. :-)

Cheers,
   Sven

svnadmin dump much slower in Bookworm

[kjohnson]

A set of svnadmin dump commands that run as part of a backup procedure seem to 
be _much_ slower in Bookworm than in Bullseye.  Prior to the upgrade to 
Bullseye, these commands took slightly less than one hour.  After the upgrade, 
similar commands (dumping a few more revisions) require more than six hours.  
The script that performs these commands has not been changed.  Has anyone else 
seen this problem?

Thanks,

Ken
 
 

Re: Gérer l'état d'un serveur avec update-alternatives

[Michel Verdier]

Le 10 août 2023 Olivier a écrit :

> Je n'ai aucune différence en terme de paquet installé entre les
> versions de prod ou de test, uniquement des fichiers de configuration
> différents.

Mais si tu testes sur des fichiers de conf qu'au final tu remplaces par
d'autres, à quoi te servent les conf de test ? Ou alors tu as des conf de
test sur certains services pour tester un autre service ?
Je posais la question des paquets pour savoir, mais je ne pense pas que
ce soit forcément adapté pour faciliter un changement d'état du serveur.
Mais si tu choisis d'utiliser des paquets, mets dedans directement le
/etc/foo/foo.conf qui va bien, les liens ne feraient que charger
inutilement le truc.

Screensaver in KDE (Plasma5)

[Hans]

Hi folks,

it would be nice if someone could send me the correct settings of libs and 
binaries, which are related for the screensaver in plasma5.

Also I would like to know, if the user hast to be member in a special group 
related to the screensaver.

The problem here is, that I can not deactivate a started screensaver of my own 
session, although I know the correct password (of course).

For me, it looks some wrong settings in the rights, so that I myself can not 
stop the screensaver process. The process can be stopped by root.

Thanks for any help.

Best regards

Hans

 

Re: keybase upgrade / install fails

[Christian Britz]

This package is not part of debian, so you should probably try their
support channels.

Gary Dale wrote:
> I'm running Debian/Bookworm on an AMD64 system. For the last two or
> three weeks I've been getting messages like below when I use apt: The
> keybase package doesn't seem to configure properly so apt keeps trying,
> and failing, to finish the package installation. Removing it (not
> purging) then reinstalling doesn't help.
> 
> Setting up keybase (6.2.2-20230726175256.4464bfb32d) ...
> Autorestarting Keybase via systemd for 
> mkdir: cannot stat ‘/keybase’: Transport endpoint is not connected
> chown: cannot access '/keybase': Transport endpoint is not connected
> chmod: cannot access '/keybase': Transport endpoint is not connected
> dpkg:error processing package keybase (--configure):
> installed keybase package post-installation script subprocess returned
> error exit status 1
> Errors were encountered while processing:
> keybase
> E: Sub-process /usr/bin/dpkg returned an error code (1)
> 
> Any suggestions?
> 

-- 
https://www.cb-fraggle.de

keybase upgrade / install fails

[Gary Dale]

I'm running Debian/Bookworm on an AMD64 system. For the last two or 
three weeks I've been getting messages like below when I use apt: The 
keybase package doesn't seem to configure properly so apt keeps trying, 
and failing, to finish the package installation. Removing it (not 
purging) then reinstalling doesn't help.


Setting up keybase (6.2.2-20230726175256.4464bfb32d) ...
Autorestarting Keybase via systemd for 
mkdir: cannot stat ‘/keybase’: Transport endpoint is not connected
chown: cannot access '/keybase': Transport endpoint is not connected
chmod: cannot access '/keybase': Transport endpoint is not connected
dpkg:error processing package keybase (--configure):
installed keybase package post-installation script subprocess returned 
error exit status 1

Errors were encountered while processing:
keybase
E: Sub-process /usr/bin/dpkg returned an error code (1)

Any suggestions?

Re: apt-key is deprecated

[ajh-valmer]

On Thursday 10 August 2023 14:43:35 didier gaumet wrote:
> le site web Trinity conseille pour la gestion des clés:
> (ne te fie pas aux retours à la ligen automatique: tout ce qui suit un 
> signe $ (dollar) est sur la même ligne)
> commence par regarder si un paquet trinity-keyring.deb est installé chez 
> toi: 
> $ dpkg -l trinity-keyring
> si il est installé, comme c'est peut-être une vieille version (apt-key 
> uniquement), tu l'enlèves:
> $ sudo dpkg -P trinity-keyring.deb
> $ wget http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-keyring.deb
> $ sudo dpkg -i trinity-keyring.deb
> normalement, d'après le contenu du paquet, ça devrait satisfaire 
> l'ancien fonctionnement avec apt-key et le nouveau fonctionnement sans 
> apt-key
> sinon, il doit falloir modifier les lignes trinity dan,s le sources.list.
> tiens-nous au courant

Merci.
J'ai fait les manips, puis apt update, et les warning "apt-key" sont toujours 
présents.
Faut-il faire les manips en mode  ou en root ?

Re: apt policy / listing packages from repo

[David Wright]

On Thu 10 Aug 2023 at 10:46:27 (+0100), Gareth Evans wrote:
> There seems to be no mention of 
> 
> apt policy
> 
> in either man apt or apt --help.

As   man apt   says:

 "Much like apt itself, its manpage is intended as an end user
  interface and as such only mentions the most used commands and
  options partly to not duplicate information in multiple places
  and partly to avoid overwhelming readers with a cornucopia of
  options and details."

> I was looking for a way to list packages installed from a particular repo 
> and/or sub-repo or whatever it's called (eg. main, non-free).
> 
> Does anyone know of a way to do this, with apt policy or otherwise?

What I do in this situation is to type "apt" and press TAB twice.
Look at the resulting list of commands and check the man page for
the most likely looking, in this case apt-cache.

An alternative method of course is to type   apt policy   into any
search engine. This will typically tell you not only how to invoke
the command, but also more about what it produces.

As for your listing, I've done this in the past with a script that
runs apt-cache dump, grepping the Package/Version/File lines,
concatenating and sorting them, then filtering that list against
the output of dpkg-query -W -f to include only installed packages.
This yields a list like:

  Package: acl Version: 2.2.53-10 File: 
/var/lib/apt/lists/deb.debian.org_debian_dists_bullseye_main_binary-amd64_Packages
  Package: adduser Version: 3.118 File: 
/var/lib/apt/lists/deb.debian.org_debian_dists_bullseye_main_binary-amd64_Packages
  [ … … ]
  Package: xtoolwait Version: 1.3-6.2 File: /var/lib/dpkg/status
  [ … … ]
  Package: yt-dlp Version: 2023.03.04-1~bpo11+1 File: /var/lib/dpkg/status

which you can grep for particular subsets, though I'm usually more
interested in grep -v for packages originating from elsewhere,
like xtoolwait (squeeze) and yt-dlp (backports) there.

There may well be better ways.

Cheers,
David.

Re: Gérer l'état d'un serveur avec update-alternatives

[Olivier]

Je n'ai aucune différence en terme de paquet installé entre les
versions de prod ou de test, uniquement des fichiers de configuration
différents.

Par contre, justement, la question me fait penser que justement,
j'aurai peut être intérêt à créer un paquet .deb (cf [1]) et à
utiliser un répertoire comme /etc/alternatives/maconfig pour y
regrouper  tous les liens devant changer en cas de changement d'état.
Au lieu d'avoir un lien direct entre /etc/foo/foo.conf et
/etc/foo/prod.foo.conf, j'ai deux liens successifs /etc/foo/foo.conf
vers /etc/alternatives/maconfig/foo.conf et un autre de
/etc/alternatives/maconfig/foo;conf vers /etc/foo/prod.foo.conf.
Le premier lien ne change jamais. Le second change quand le serveur
change d'état.

Pourquoi utiliser un paquet ? Pour y cacher les scripts d'installation
et de désinstallation et bénéficier de dpkg-reconfigure pour changer
d'état.
Qu'en pensez-vous ?

[1] https://wiki.debian.org/ConfigPackages

Le mer. 9 août 2023 à 09:44, Michel Verdier  a écrit :
>
> Le 9 août 2023 Olivier a écrit :
>
> > Que pensez-vous d'utiliser le programme update-alternatives (que je
> > connais très mal) et ses sbires pour cela ?
>
> Les alternatives sont là pour permettre de changer de paquet pour un
> fichier donné, par exemple le shell, ou pour une version donnée. Donc ça
> concerne les versions ou le choix des paquets installés.
> Ta notion préparation/production me semble s'écarter un peu de ça.
> Ou alors si tu as des paquets -prep et des paquets -prod ?
> Tu peux donner un exemple de ce que tu gères avec cette différence
> préparation/production ?
>

The Debian Administrator's Handbook

[Cecil Westerhof]

Het lijkt me handig om wat dieper te duiken in Debian. Ik heb daarom
The Debian Administrator's Handbook gedownload. Die is echter in PDF
niet later als 10 beschikbaar. (Ik zie epub en mobipocket niet
zitten.) Ik gebruik 11 en moet eigenlijk overstappen op 12. Is het
handboek voor 10 goed genoeg, of zijn er grote verschillen?

-- 
Cecil Westerhof
Senior Software Engineer
LinkedIn: http://www.linkedin.com/in/cecilwesterhof

Re: apt-key is deprecated

[didier gaumet]

le site web Trinity conseille pour la gestion des clés:
(ne te fie pas aux retours à la ligen automatique: tout ce qui suit un 
signe $ (dollar) est sur la même ligne)


commence par regarder si un paquet trinity-keyring.deb est installé chez 
toi:

$ dpkg -l trinity-keyring

si il est installé, comme c'est peut-être une vieille version (apt-key 
uniquement), tu l'enlèves:

$ sudo dpkg -P trinity-keyring.deb

ensuite

$ wget http://mirror.ppa.trinitydesktop.org/trinity/deb/trinity-keyring.deb
$ sudo dpkg -i trinity-keyring.deb

normalement, d'après le contenu du paquet, ça devrait satisfaire 
l'ancien fonctionnement avec apt-key et le nouveau fonctionnement sans 
apt-key


sinon, il doit falloir modifier les lignes trinity dan,s le sources.list.
tiens-nous au courant

Re: logging no longer standard?

[gene heskett]

On 8/9/23 21:24, Max Nikulin wrote:

On 08/08/2023 19:07, gene heskett wrote:
digikam for example, does report what I assume is the package name, 
just running it, reports a couple screens full of Exiv2 errors, but 
Exiv2 is installed.


I have an impression that properly built AppImage should come with all 
necessary libraries included and should not rely on packages installed 
in the system.


We also think about AppImages as complete, satisfying ALL their own 
dependencies. No disagreement there. I do see them as huge storage 
resource hogs, but at the same time papers over individual system 
differences IF done right. FWIW, the repo version of digikam also 
suffers the can't write to local storage problem, but its supposedly a 
known bug in 7.9 that has since been fixed ack the digikam folks, but 
their AppImage builds for version 8.2 still don't work /here/.
That leaves the fact that /home is a raid here, but surely I am not the 
only one on the planet doing that.


Your camera and EXIF parsing library may have different notion on EXIF 
tags format and allowed values. You may discuss whether the errors are 
critical with developers or packagers of the applications and to filter 
out these errors otherwise.


Throw in that we have no control over the camera maker, in all cases 
here they are well known, reputable makers of at least 3 cameras that 
have saved pix in that tree, plus probably a thou or more of pix saved 
from the net in there, I guess its up to us to configure this stuff so 
it works.  Is there a tut describing how someplace?


cuda is also named in the errors, and none of its dozen or so variations 
is installed.  Which one to install?


Thank you Max.


.


Cheers, Gene Heskett.
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 

Re: apt-key is deprecated

[ajh-valmer]

On Wednesday 09 August 2023 19:21:00 Lamourec Alain wrote:
> Tu fais apt-key list en root
> Tu recherches toutes celles qui sont dans /etc/apt/trusted.gpg
> Puis tu les mets dans le dossier trusted.gpg.d
> apt-key export 1378B444 | gpg --dearmour -o  
> /etc/apt/trusted.gpg.d/Libreoffice.gpg
> en prenant comme dans l'exemple 1378B444 pour les 8 chiffres de 
> l'empreinte et tu lui donnes le nom que tu veux.

Merci de ton aide.

J'ai lancé apt-key export... et à la fin le système dit :
"apt-key is deprecated".
apt-key list
Il y a tellement de clés à exporter que j'abandonne...

Bonne journée

rtkit-daemon problemen

[Cecil Westerhof]

Ik zie met journalctl meldingen als:
rtkit-daemon[1749779]: Warning: Reached maximum concurrent process limit 
for user '1000', denying request.

Dit is voor realtime behaviour als ik het goed begrijp. Volgens mij
heb ik dit helemaal niet geïnstalleerd. Is het een probleem als ik
rtkit verwijder? Moet ik me zorgen maken over deze meldingen?

-- 
Cecil Westerhof
Senior Software Engineer
LinkedIn: http://www.linkedin.com/in/cecilwesterhof

Re: logging no longer standard?

[gene heskett]

On 8/9/23 21:15, Max Nikulin wrote:

On 08/08/2023 09:57, gene heskett wrote:

Xsession: X session started for gene at Tue 27 Jun 2023 02:58:23 PM EDT

   ^^^
dbus-update-activation-environment: error: unable to connect to D-Bus: 
Failed to connect to socket /run/user/1000/bus: Connection refused


What fixes that?


These messages might be old, but they might mean that you broke D-Bus. 
Applications, especially self-containing packages, may heavily rely on 
its availability.


Try to figure out at which moment such messages appear. Try "busctl 
--user" and "busctl --user status" as a sanity check.


Both of those get somewhat copious outputs, what should I be looking 
for?  Should it name the app?



Looking in /etc/dbus-1, I see two dirs, one of which is empty:

gene@coyote:/etc/dbus-1$ ls -R
.:
session.d  system.d

./session.d:

./system.d:
bluetooth.conf 
com.ubuntu.SoftwareProperties.conf  org.freedesktop.DisplayManager.conf 
org.freedesktop.ModemManager1.conf  org.opensuse.CupsPkHelper.Mechanism.conf
com.redhat.NewPrinterNotification.conf   dnsmasq.conf 
org.freedesktop.GeoClue2.Agent.conf 
org.freedesktop.PackageKit.conf sddm_org.freedesktop.DisplayManager.conf
com.redhat.PrinterDriversInstaller.conf  net.hadess.SensorProxy.conf 
org.freedesktop.GeoClue2.conforg.freedesktop.realmd.conf 
 wpa_supplicant.conf


dbus has always been a puzzle to me. It has no man pages to explain its 
functions. Do you want to see the outputs of busctl?  Under what conditions?


I see busctl does have a man page. Pretty complex. What should I check next?

Thanks Max.

Cheers, Gene Heskett.
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 

apt policy / listing packages from repo

[Gareth Evans]

There seems to be no mention of 

apt policy

in either man apt or apt --help.

I was looking for a way to list packages installed from a particular repo 
and/or sub-repo or whatever it's called (eg. main, non-free).

Does anyone know of a way to do this, with apt policy or otherwise?

Thanks,
Gareth

Problem using pmount

[Sridhar M A]

Currently am using debian testing.

So far, I was using pmount to mount my usb drive without any issues
(am also a member of the plugdev group).

Since a couple of days, pmount started failing. Before I explain
further, the same drive can be mounted via thunar.

This is the error I am seeing:

$ pmount /dev/sdb1 mas
NTFS signature is missing.
Failed to mount '/dev/sdb1': Invalid argument
The device '/dev/sdb1' doesn't seem to have a valid NTFS.
Maybe the wrong device is used? Or the whole disk instead of a
partition (e.g. /dev/sda, not /dev/sda1)? Or the other way around?
NTFS signature is missing.
Failed to mount '/dev/sdb1': Invalid argument
The device '/dev/sdb1' doesn't seem to have a valid NTFS.
Maybe the wrong device is used? Or the whole disk instead of a
partition (e.g. /dev/sda, not /dev/sda1)? Or the other way around?
Error: could not delete mount point: No such file or directory.

I even formatted my pen drive, but the same problem persists. In fact,
the drive is formatted for vfat, not NTFS.

$ lsblk -o name,fstype,size,mountpoint,uuid
NAMEFSTYPE   SIZE MOUNTPOINT   UUID
sdb  7.5G
└─sdb1  vfat 7.5G  3BE4-3417

Could not find any bug reports about such an issue?

Any pointers will be helpful.

-- 
Sridhar M. A.

Re: debian add mirror after installation

[Luna Jernberg]

https://wiki.debian.org/SourcesList

Den tors 10 aug. 2023 kl 09:01 skrev Marco :
>
> Am 10.08.2023 schrieb Weijun Lu :
>
> > 1. Is there a way not to select the network mirror site in the
> > installation wizard, and reconfigure the network mirror site through
> > dpkg-reconfig xxx and other methods after the system installation is
> > complete.
>
> Simply edit /etc/apt/sources.list after the installation.
>

Re: Sound loses my analog speakers after suspend, and power settings don't affect monitor poweroff

[Marco]

Am 09.08.2023 schrieb Carl Fink :

> I suspended my system from the System menu Monday night. When I woke
> it up Tuesday morning, sound was coming from the HDMI monitor. The
> Sound Settings didn't know about any other sound system.

This sounds like a driver problem with the soundcard. HDMI is realized
with the graphics card, analog with a onboard sound card, mostly
integrated into the chipset.

Run lspci -nnk
when it works and when it doesn't work.