Re: Carnet du téléphone Android et Linux

[Basile Starynkevitch]

On 4/10/19 6:30 AM, Basile Starynkevitch wrote:


On 4/10/19 3:23 AM, Samuel Cifuentes wrote:


Bonjour

Pour des clients, j'ai mis en place une solution à base de Nextcloud 
(sur un serveur ovh, mais tu peux en faire un à la maison)


plusieurs agendas, un seul carnet d'adresse centralisé sur l'instance 
Nextcloud


ils synchronisent leurs carnets d'adresse sur Thunderbird, et sur 
leur tel via un client Davdroid (ils ont à peu prés 4000 contacts 
avec des doublons, triplons, et j'en passe)
Que veut dire exactement "synchronisent leurs carnets d'adresse sur 
Thunderbird". Je comprends le concept, mais pas les actions concrètes à 
faire (sur le téléphone comme sur le poste Linux). Quelles actions 
précises sur le téléphone, quelles commandes sous Linux. Je suis un 
grand débutant avec mon téléphone, mais je comprends très bien la ligne 
de commande Linux (et serais capable de coder l'application idoine, si 
j'en comprennais les protocoles et les formats de données).


pour Thunderbird, CardBook est pas mal pour travailler sur les 
doublons, mais c'est quand même fastidieux s'il n'y a pas eu un 
minimum de "bonnes pratiques" lors de la collecte et de la saisie des 
contacts.




Mais concrètement, ça veut dire quoi. Quelle commandes Linux exactes 
sont à exécuter? Quelle action concrète sur le téléphone?



Mon problème concret, d'un point de vue d'informaticien, est en fait: 
quelles actions concrètes (comment faire sur le téléphone, quelles 
commandes taper sous Linux) faire pour transferer le carnet d'adresse 
depuis le téléphone puis vers le téléphone; et sous quels formats (JSON, 
XML, sqlite...) verrais-je les données du carnet d'adresse sur mon poste 
Linux.


Une fois que j'ai compris tout ça, modifier le carnet d'adresse doit 
être simple. Je me sens même capable, si c'était nécessaire, d'écrire 
l'outil (en Qt ou techo web) facilitant la modification des données sur 
le poste Linux.


Par exemple, la wikipage sur https://en.wikipedia.org/wiki/VCard m'est 
parfaitement compréhensible. Ce que j'ignore, c'est quoi faire pour 
avoir "un ensemble de vcard" sur mon poste Linux, et comment le 
transférer depuis et vers mon téléphone.


Merci beaucoup.

--
Basile STARYNKEVITCH   == http://starynkevitch.net/Basile
opinions are mine only - les opinions sont seulement miennes
Bourg La Reine, France

Re: Carnet du téléphone Android et Linux

[Basile Starynkevitch]

On 4/10/19 3:23 AM, Samuel Cifuentes wrote:


Bonjour

Pour des clients, j'ai mis en place une solution à base de Nextcloud 
(sur un serveur ovh, mais tu peux en faire un à la maison)


plusieurs agendas, un seul carnet d'adresse centralisé sur l'instance 
Nextcloud


ils synchronisent leurs carnets d'adresse sur Thunderbird, et sur leur 
tel via un client Davdroid (ils ont à peu prés 4000 contacts avec des 
doublons, triplons, et j'en passe)


pour Thunderbird, CardBook est pas mal pour travailler sur les 
doublons, mais c'est quand même fastidieux s'il n'y a pas eu un 
minimum de "bonnes pratiques" lors de la collecte et de la saisie des 
contacts.




Mais concrètement, ça veut dire quoi. Quelle commandes Linux exactes 
sont à exécuter? Quelle action concrète sur le téléphone?


Merci beaucoup

--
Basile STARYNKEVITCH   == http://starynkevitch.net/Basile
opinions are mine only - les opinions sont seulement miennes
Bourg La Reine, France

Re: Carnet du téléphone Android et Linux

[Samuel Cifuentes]

Bonjour

Pour des clients, j'ai mis en place une solution à base de Nextcloud 
(sur un serveur ovh, mais tu peux en faire un à la maison)


plusieurs agendas, un seul carnet d'adresse centralisé sur l'instance 
Nextcloud


ils synchronisent leurs carnets d'adresse sur Thunderbird, et sur leur 
tel via un client Davdroid (ils ont à peu prés 4000 contacts avec des 
doublons, triplons, et j'en passe)


pour Thunderbird, CardBook est pas mal pour travailler sur les doublons, 
mais c'est quand même fastidieux s'il n'y a pas eu un minimum de "bonnes 
pratiques" lors de la collecte et de la saisie des contacts.


bon courage


Le 09/04/2019 à 21:38, Basile Starynkevsitch a écrit :


Bonsoir,

On est trois (mon fils majeur, mon épouse, et moi) dans mon foyer à 
avoir tous des téléphones mobiles Android. Plutôt des modèles bas de 
gamme (un Sony f311 sous Android 7.0).


On est aussi tous les trois à avoir Linux (x86-64): Debian/Sid pour 
moi Mint pour mon fils, Ubuntu/Bionic pour mon épouse, sur des 
ordinateurs fixes avec des écrans assez large.


Comment gérer correctement nos contacts téléphoniques? Il y a plein de 
doublons ou contacts obsoletes chez chacun. On voudrait gérer le 
carnet du téléphone (le réorganiser, le nettoyer) sous notre 
ordinateur Linux. Principalement pour des raisons de confort: j'y vois 
mal (je me fais bientôt opérer de la cataracte, et j'attends ça avec 
impatience) et gérer toutes les addresses sur le téléphone m'est 
pénible (sur mon ordinateur Linux, j'ai deux grands écrans - et un 
AMD2970WX avec 64Go de RAM et 2To de SSD).


Je connais bien Debian (et j'utilise Linux depuis 1993), mais mon 
téléphone Android, pas trop. Je voudrais utiliser le cable USB pour 
transférer le carnet d'adresse sur le PC fixe, éditer ce carnet 
agréablement avec mon clavier sur le PC, et le retransférer vers le 
téléphone après nettoyage et réorganisation.



Je connais la ligne de commande, je suis chercheur en informatique 
(pour la petite histoire, je développe Bismon 
 sous GPLv3+) et je suis à l'aise 
avec sqlite ou XML ou JSON (j'imagine que le carnet du téléphone 
utilise ces technos là).


Par contre, j'investis assez peu mon téléphone mobile, car j'y vois mal.

Merci de vos lumières

Librement

--
Basile STARYNKEVITCH   ==http://starynkevitch.net/Basile
opinions are mine only - les opinions sont seulement miennes
Bourg La Reine, France

[Off-topic] Conta do Antigo Google Apps

[Henrique Fagundes]

Prezado Colegas,

Saudações "pinguianas" e boa noite!

Eu não se isso é permitido aqui, mas se não for, desde já, eu peço desculpas.

Eu gostaria de saber se alguém aqui tem alguma conta do antigo Google Apps 
(atual G-Suite) que não esteja usando...

E se pode ceder ou vender bem baratinho. É apenas para FINS DE ESTUDOS.

Atenciosamente, 

Henrique Fagundes 
Analista de Suporte Linux 
supo...@aprendendolinux.com 
Skype: magnata-br-rj 
Linux User: 475399 

https://www.aprendendolinux.com 
https://www.facebook.com/AprendendoLinux 
https://youtube.com/AprendendoLinux 
https://twitter.com/AprendendoLinux 
https://t.me/AprendendoLinux 
https://t.me/GrupoAprendendoLinux 
__ 
Participe do Grupo Aprendendo Linux 
https://listas.aprendendolinux.com/listinfo/aprendendolinux 

Ou envie um e-mail para: 
aprendendolinux-subscr...@listas.aprendendolinux.com 

BRASIL acima de tudo, DEUS acima de todos!

Re: build vs install

[David Wright]

On Tue 09 Apr 2019 at 20:25:20 (+0100), Brad Rogers wrote:
> On Tue, 9 Apr 2019 14:58:22 -0400 Lee  wrote:
> 
> >OK - good to know.  Somehow I'd got the impression that systemd was
> >moving to a windows registry type thing for config data instead of
> 
> I've no idea about that.  I don't think systemd is really relevant to
> the topic under discussion here, though.
> 
> >keeping it in plain text files.
> 
> For some years now, Grub (well, Grub2) has used a binary format for
> configs.

Really, which ones? I can't find any.

> Lilo has for done so for even longer, of course.  Mozilla
> stores much of its application and config data in SQlite databases, as do
> other organisations.  Ever more things are making use of XML in a
> similar way.  Obviously, that's still human readable(1).
> 
> (1)  Well, that is to say, humans can view it in a text editor rather
> than a binary file viewer/editor.  Whether the content makes any sense to
> the reader is another matter entirely, of course.   :-)

Cheers,
David.

Re: build vs install

[Lee]

On 4/9/19, Michael Howard wrote:
> On 09/04/2019 19:03, Lee wrote:
>> On 4/9/19, Michael Howard wrote:
>>> On 09/04/2019 16:35, Lee wrote:
 What are the downsides to getting the source code and doing the
 build/install myself vs. using a pre-built package other than I'm
 responsible for noticing the software needs to be updated?

 The latest example is ttcp
 http://nuttcp.net/nuttcp/latest/  has 8.1.4
 https://tracker.debian.org/pkg/nuttcp  has 6.1.2-4

 If I ever decide to go with the debian package I just uninstall the
 software I built and .. anything else that needs to be done before
 installing an official package?

 TIA
 Lee

>>> You could build a debian package from the source and let apt take care
>>> of install/updates/uninstall. Admittedly, it could be a bit of work
>>> initially with versions so far apart as above but it is an option.
>> That's an interesting idea.  Get the debian stuff from
>>https://packages.debian.org/stretch/nuttcp
>> and see which patches still apply to the current version of the software.
>>
>> But I'm missing what letting apt take care of
>> install/updates/uninstall gets me since I'd be the one creating &
>> updating the home-built package.
>>
> It's simply a case that apt then knows that ttcp is installed and at
> what version in case of conflicts or security aspects etc. If you simply
> do a ./configure && make && make install or what not, apt has no idea
> what you've done.

hrmmm..  I'm still at the using synaptics stage of installing/updating
packages but I can see how that might be the better way to go.  I'll
take a look at how to make & install a home-built package & see if it
looks doable in a few hours.  Otherwise it goes on my ever-expanding
would-be-nice-to-do-someday list

Thanks
Lee

Re: build vs install

[Lee]

On 4/9/19, Brad Rogers wrote:
> On Tue, 9 Apr 2019 14:58:22 -0400
> Lee wrote:
>
> Hello Lee,
>
>>OK - good to know.  Somehow I'd got the impression that systemd was
>>moving to a windows registry type thing for config data instead of
>
> I've no idea about that.  I don't think systemd is really relevant to
> the topic under discussion here, though.

I thought systemd was the new way to start services.  Having ttcp
automatically run as a service means I don't have to ssh into the
other machine & start the receiver manually.

But for things that don't run as a service, you're right - systemd
doesn't enter into the picture.  Maybe I can still use /etc/init.d or
/etc/services to get things started & keep systemd out of the picture?
 It's been a long time since I've had to do unix admin type stuff & it
seems like most everything I learned is out of date.

Regards,
Lee

Re: putting config files under revision control

[hdv@gmail]

On 09/04/2019 23.09, Lee wrote:
> On 4/9/19, hdv@gmail wrote:
>> On 09/04/2019 21.23, Lee wrote:
>>> On 4/9/19, Dan Ritter wrote:
 Lee wrote:
> What are people doing for putting config files in [under?] git?
>
> I'd like to have at least some system config files maintained in git
> so I can get a history of changes.
>   (and yes, I know, I really should be using a backup system for that,
> but I'm still at the 'rsync to usb drive' stage)

 apt install etckeeper. Choose the git backend. (I think it's the
 default these days.)
>>>
>>> Thank you!
>>>
 Chef or Puppet when you want to do this at scale.
>>>
>>> Maybe someday.  They'd be nice to learn, but they seem to be massive
>>> overkill for home use.  ..or at least for my home use.
>>
>> If you want to keep things simple, maybe this perl-script I wrote years ago
>> might be what you're looking for. Think of it as a visudo-style tool for
>> config
>> files (really just any file you can edit with vim). Just use the --manual
>> option
>> to read its man page. Basically it copies the file before editing to a
>> location
>> of your choosing, keeping its attributes if you want it to.
> 
> Thanks for the script.  It looks easy enough to understand, which is
> always nice :)

You're welcome. If you have any questions, feel free to ask.

P.S. I wrote it because I didn't want to have another daemon running only for
this. I have nothing against git or some such, but for this I thought it to be
overkill. Especially on servers.

Grx HdV

Re: build vs install

[Michael Howard]

On 09/04/2019 19:03, Lee wrote:

On 4/9/19, Michael Howard wrote:

On 09/04/2019 16:35, Lee wrote:

What are the downsides to getting the source code and doing the
build/install myself vs. using a pre-built package other than I'm
responsible for noticing the software needs to be updated?

The latest example is ttcp
http://nuttcp.net/nuttcp/latest/  has 8.1.4
https://tracker.debian.org/pkg/nuttcp  has 6.1.2-4

If I ever decide to go with the debian package I just uninstall the
software I built and .. anything else that needs to be done before
installing an official package?

TIA
Lee


You could build a debian package from the source and let apt take care
of install/updates/uninstall. Admittedly, it could be a bit of work
initially with versions so far apart as above but it is an option.

That's an interesting idea.  Get the debian stuff from
   https://packages.debian.org/stretch/nuttcp
and see which patches still apply to the current version of the software.

But I'm missing what letting apt take care of
install/updates/uninstall gets me since I'd be the one creating &
updating the home-built package.

It's simply a case that apt then knows that ttcp is installed and at 
what version in case of conflicts or security aspects etc. If you simply 
do a ./configure && make && make install or what not, apt has no idea 
what you've done.


--
Mike Howard

Re: putting config files under revision control

[Lee]

On 4/9/19, hdv@gmail wrote:
> On 09/04/2019 21.23, Lee wrote:
>> On 4/9/19, Dan Ritter wrote:
>>> Lee wrote:
 What are people doing for putting config files in [under?] git?

 I'd like to have at least some system config files maintained in git
 so I can get a history of changes.
   (and yes, I know, I really should be using a backup system for that,
 but I'm still at the 'rsync to usb drive' stage)
>>>
>>> apt install etckeeper. Choose the git backend. (I think it's the
>>> default these days.)
>>
>> Thank you!
>>
>>> Chef or Puppet when you want to do this at scale.
>>
>> Maybe someday.  They'd be nice to learn, but they seem to be massive
>> overkill for home use.  ..or at least for my home use.
>
> If you want to keep things simple, maybe this perl-script I wrote years ago
> might be what you're looking for. Think of it as a visudo-style tool for
> config
> files (really just any file you can edit with vim). Just use the --manual
> option
> to read its man page. Basically it copies the file before editing to a
> location
> of your choosing, keeping its attributes if you want it to.

Thanks for the script.  It looks easy enough to understand, which is
always nice :)

Lee

Re: date(1) in stretch and buster

[David Wright]

On Tue 09 Apr 2019 at 20:07:37 (-), Curt wrote:
> On 2019-04-09, Étienne Mollier  wrote:
> >
> > The output may differ depending on you operating system level,
> > given Reco's observations.  Feel free to have à look at
> > /usr/share/zoneinfo/, to have an idea of the available
> > locations.
> 
> I took a look.
> 
> I was confused to note the presence of the UCT zone. UCT? Shouldn't that
> be UTC?  But the latter zone proves to be a symlink to the former.

I imagine it gets called Universal Coordinated Time (whereas I always
pronounce it "you-tee-see"). My rationalisation would be that it's
Coordinated Time (so was GMT) that has been agreed as Universal.
If it was already Universal Time, it wouldn't need to be Coordinated.
So I prefer it to Wiki's Coordinated Universal Time. Anyway, cut time
is a musical concept. UTC is of course a French acronym.

But I don't have symlinks: they're separate files, thus preserving the
name of the zone for those who still use it. I think that officially
it has migrated to the fictitous continent of landfill called "Etc".

> Then there's the GMT0, GMT-0, and GMT+0 zones, all symlinks to plain old
> venerable GMT, which kind of makes sense, I guess, zero being what it is.

Yes, and a load of GMT+x stuff which suffers the same way as UTC+x:
you end up with the name GMT applied to times that aren't GMT.

> Anyway, it's getting late.

Not here: it's barely teatime.

Cheers,
David.

Re: build vs install

[Brad Rogers]

On Tue, 9 Apr 2019 14:58:22 -0400
Lee  wrote:

Hello Lee,

>OK - good to know.  Somehow I'd got the impression that systemd was
>moving to a windows registry type thing for config data instead of

I've no idea about that.  I don't think systemd is really relevant to
the topic under discussion here, though.

>keeping it in plain text files.

For some years now, Grub (well, Grub2) has used a binary format for
configs.  Lilo has for done so for even longer, of course.  Mozilla
stores much of its application and config data in SQlite databases, as do
other organisations.  Ever more things are making use of XML in a
similar way.  Obviously, that's still human readable(1).

(1)  Well, that is to say, humans can view it in a text editor rather
than a binary file viewer/editor.  Whether the content makes any sense to
the reader is another matter entirely, of course.   :-)

-- 
 Regards  _
 / )   "The blindingly obvious is
/ _)radnever immediately apparent"
Bet you thought you had it all worked out
Problem - Sex Pistols


pgpm2d92d3O83.pgp
Description: OpenPGP digital signature

Re: 'synaptic' removed from buster

[David Wright]

On Sat 06 Apr 2019 at 09:26:51 (-0400), Gene Heskett wrote:
> On Friday 05 April 2019 23:06:13 David Wright wrote:
> > On Fri 05 Apr 2019 at 14:21:21 (-0400), Gene Heskett wrote:
> > > On Friday 05 April 2019 11:05:00 Jonathan Dowland wrote:
> > >
> > > [...]
> > >
> > > > Indeed, and running the UI toolkit code as root was always
> > > > considered a bad design pattern, even whilst it works under X.
> > >
> > > But thats no longer possible with X, root cannot use the users
> > > display.
> > >
> > > And for me, the only user on this multiple machine network, that is
> > > a major PITA because of the extremely inconsistent workaround's.
> > >
> > > Give us a method to su or sudo root, and run the stuff needing root,
> > > at least a consistent procedure thats good for all releases.
> > > Changing it around makes applying the same user package update to
> > > every machine a different operation. LinuxCNC is updated in master
> > > at least 2x a week sometimes daily as a new feature gets added and
> > > needs debugged. So even if wheezy is dead, synaptic gets run quite
> > > frequently to keep that up to date. The jessie install on the pi for
> > > instance won't let me run synaptic from anyplace but its own
> > > keyboard. Anyplace else, and ssh'd in, I have to sudo apt etc.
> > >
> > > Consistency is the magic word, and we don't have it.
> >
> > I don't understand how you can expect consistency in an OS that
> > supports ~10 architectures and four releases, written over a period
> > of, say, six years or more. Particularly considering that it's
> > not possible to revise the earlier releases to take account of
> > changes forced by the evolution of software external to the project
> > and by increasing security exigencies.
> >
> If I'm willing to forgo web browsing, wheezy can still be running 20 
> years from  now, the only changes needed would be for ssh|l|tls stuff if 
> you are using it for email. I am behind a dd-wrt flashed router with no 
> local firewalls running anyplace on this local net.
> 
> No one has come past that that I didn't invite in, and I'm not doing a 
> thing to get that security that the rest of you can't do, probably even 
> better.
> 
> So I don't worry about security, I worry about interoperability on my 
> local network, but the way thats headed, there will not be the 
> possibility of my ssh'ing into one of my other machines and doing a 
> simple ls or pwd to see where I am.  That IMO is not real security, but 
> PARANOIA and should rightly be called as such.
> 
> This os was originally able to ignore whether the key/mouse stroke came 
> from its own keyboard, or one in Lisbon, Portugal. Or from this machine, 
> 100+ feet of cat5 away. This is NOT an M$ single user system, so quit 
> trying to restrict it to what a winders box can do.

I keep reading here that you have problems with ssh -X and -Y, but
find it difficult to replicate them here. I've been running a mix of
wheezy/jessie/stretch here with no problems (though I have retired
the wheezy versions recently and removed all references in
apt-cacher-ng now that the archives have moved).

> > You can hide any differences in how you obtain root by just wrapping
> > your command in a script, appropriate for each architecture/release.
> > Just one script to maintain if you use case or if/else.
> 
> Then Make it so...

I meant for you to do that, in the same way that I write scripts for
some user programs in order to cover up the differences between
versions of Debian, or the hardware (like sound cards) on different
hosts. It just means the differences are put into the scripts so you
can type the same thing without having to remember all the time.

$HOSTNAME is obvious, but I also have $Mycodename which is set from
the first line of sources.list. This gives a more consistent answer
the debian_version, os-release etc as it's under my control.

Cheers,
David.

Re: putting config files under revision control

[hdv@gmail]

On 09/04/2019 21.23, Lee wrote:
> On 4/9/19, Dan Ritter wrote:
>> Lee wrote:
>>> What are people doing for putting config files in [under?] git?
>>>
>>> I'd like to have at least some system config files maintained in git
>>> so I can get a history of changes.
>>>   (and yes, I know, I really should be using a backup system for that,
>>> but I'm still at the 'rsync to usb drive' stage)
>>
>> apt install etckeeper. Choose the git backend. (I think it's the
>> default these days.)
> 
> Thank you!
> 
>> Chef or Puppet when you want to do this at scale.
> 
> Maybe someday.  They'd be nice to learn, but they seem to be massive
> overkill for home use.  ..or at least for my home use.

If you want to keep things simple, maybe this perl-script I wrote years ago
might be what you're looking for. Think of it as a visudo-style tool for config
files (really just any file you can edit with vim). Just use the --manual option
to read its man page. Basically it copies the file before editing to a location
of your choosing, keeping its attributes if you want it to.

Grx HdV

#!/usr/bin/perl

#TODO : add support for settings stored in ~/.vicfrc or an explicitly given file

our $VERSION = '0.92';

use strict;
use warnings;
use Getopt::Long qw(:config bundling);
use Pod::Usage;
use Sys::Hostname;
use Cwd qw(realpath getcwd);
use POSIX qw(strftime sysconf _PC_CHOWN_RESTRICTED);
use File::Spec;
use File::Copy;

my $hostname = hostname();


#User-definable defaults


my $root = '';
SWITCH: {
  if ($hostname eq 'mjollnir') {
$root = '/home/hdv/backup/mjollnir/vicf_root';
last SWITCH;
  }
  if ($hostname eq 'odin') {
#$root = '/home/hdv/backup/odin/vicf_root';
$root = '/home/hdv/backup/odin/vicf_root';
last SWITCH;
  }
  if ($hostname eq 'sleipnir') {
$root = '/home/hdv/backup/sleipnir/vicf_root';
last SWITCH;
  }
}

my $datetime_format = '-%Y%m%d';   #d
my $sequencenr_format = '-%02d';   #n
my $append_sequencenr = 1; #a
my $keep_permissions = 1;  #p
my $keep_owner = 1;#o
my $keep_group = 1;#g
my $keep_times = 1;#t
my $backup_file = 0;   #b
my $x_editor = 0;  #x
my $editor_path = '/usr/bin/vim';
my $x_editor_path = '/usr/bin/gvim';
my $backup_option = '-c "set backup"';
my $no_backup_option = '-c "set nobackup"';


#Internal variables


#Defaults for commandline options
my $help = 0;
my $manual = 0;
my $show_version = 0;
my $debug = 0;
my $verbose = 0;


#Parse the commandline arguments


#Get all options
GetOptions(#Standard options
   'debug|D+', \$debug,
   'help!', \$help,
   'h|?', \$help,
   'manual!', \$manual,
   'version!', \$show_version,
   'V', \$show_version,
   'verbose|v+', \$verbose,
   #Options specific for this program
   'root|r=s', \$root,
   'datetime|d=s', \$datetime_format,
   'sequencenr|n=s', \$sequencenr_format,
   'append_sequencenr!', \$append_sequencenr,
   'a!', \$append_sequencenr,
   'permissions!', \$keep_permissions,
   'p!', \$keep_permissions,
   'owner!', \$keep_owner,
   'o!', \$keep_owner,
   'group!', \$keep_group,
   'g!', \$keep_group,
   'times!', \$keep_times,
   't!', \$keep_times,
   'backup!', \$backup_file,
   'b!', \$backup_file,
   'x_editor!', \$x_editor,
   'x!', \$x_editor
  ) or pod2usage(0);
pod2usage(verbose => 1, exitval => 0) if $help;
pod2usage(verbose => 2, exitval => 0) if $manual;
if ($show_version) {
  print "vicf version $VERSION (c) 2015 Jadev\n";
  exit 0;
}

#Assign the first non-option argument to a variable for easier use
die "No file to be edited was given.\n" unless $ARGV[0];
my $source = $ARGV[0];

#Sanitize paths for easier use
$root = expand_path($root);
die "Path pointing to the repository ($root) is invalid.\n" unless $root;
$source = expand_path($source);
die "Path pointing to the file to be edited ($source) is invalid.\n" unless 
$source;

#Check validity of given options
die "The given root directory does not exist.\n" unless -d $root;
die "The datetime formatstring may contain only alphanumeric or punctuation 
characters.\n" 
  unless $datetime_format =~ /^[[:print:]]*$/;#May be empty
die "The sequencenumber 

Carnet du téléphone Android et Linux

[Basile Starynkevitch]

Bonsoir,

On est trois (mon fils majeur, mon épouse, et moi) dans mon foyer à 
avoir tous des téléphones mobiles Android. Plutôt des modèles bas de 
gamme (un Sony f311 sous Android 7.0).


On est aussi tous les trois à avoir Linux (x86-64): Debian/Sid pour moi 
Mint pour mon fils, Ubuntu/Bionic pour mon épouse, sur des ordinateurs 
fixes avec des écrans assez large.


Comment gérer correctement nos contacts téléphoniques? Il y a plein de 
doublons ou contacts obsoletes chez chacun. On voudrait gérer le carnet 
du téléphone (le réorganiser, le nettoyer) sous notre ordinateur Linux. 
Principalement pour des raisons de confort: j'y vois mal (je me fais 
bientôt opérer de la cataracte, et j'attends ça avec impatience) et 
gérer toutes les addresses sur le téléphone m'est pénible (sur mon 
ordinateur Linux, j'ai deux grands écrans - et un AMD2970WX avec 64Go de 
RAM et 2To de SSD).


Je connais bien Debian (et j'utilise Linux depuis 1993), mais mon 
téléphone Android, pas trop. Je voudrais utiliser le cable USB pour 
transférer le carnet d'adresse sur le PC fixe, éditer ce carnet 
agréablement avec mon clavier sur le PC, et le retransférer vers le 
téléphone après nettoyage et réorganisation.



Je connais la ligne de commande, je suis chercheur en informatique (pour 
la petite histoire, je développe Bismon 
 sous GPLv3+) et je suis à l'aise 
avec sqlite ou XML ou JSON (j'imagine que le carnet du téléphone utilise 
ces technos là).


Par contre, j'investis assez peu mon téléphone mobile, car j'y vois mal.

Merci de vos lumières

Librement

--
Basile STARYNKEVITCH   == http://starynkevitch.net/Basile
opinions are mine only - les opinions sont seulement miennes
Bourg La Reine, France

Re: date(1) in stretch and buster

[Curt]

On 2019-04-09, Étienne Mollier  wrote:
>
> The output may differ depending on you operating system level,
> given Reco's observations.  Feel free to have à look at
> /usr/share/zoneinfo/, to have an idea of the available
> locations.

I took a look.

I was confused to note the presence of the UCT zone. UCT? Shouldn't that
be UTC?  But the latter zone proves to be a symlink to the former.

Then there's the GMT0, GMT-0, and GMT+0 zones, all symlinks to plain old
venerable GMT, which kind of makes sense, I guess, zero being what it is.

Anyway, it's getting late.

Out.

> Kind Regards,

Re: 'synaptic' removed from buster

[David Wright]

On Sat 06 Apr 2019 at 09:00:14 (-0400), Gene Heskett wrote:
> On Friday 05 April 2019 22:39:23 David Wright wrote:
> > On Fri 05 Apr 2019 at 17:01:33 (-0400), Lee wrote:
> > > On 4/5/19, Reco  wrote:
> > > > On Fri, Apr 05, 2019 at 04:01:50PM +0100, Jonathan Dowland wrote:
> > > >> On Thu, Apr 04, 2019 at 12:56:49PM -, Curt wrote:
> > > >> > My understanding is the problem lies in the Gnome/Wayland combo
> > > >> > (which is the default combo starting with Buster).
> > > >>
> > > >> The problem there, IMHO, is Wayland being the default desktop
> > > >> choice.
> > > >
> > > > ... on amd64/i386 only. Whenever Wayland is the problem or the
> > > > solution, punishing all other supported architectures seems
> > > > extreme.
> > >
> > > That's what I don't understand.  Why remove the package if it's only
> > > a problem with some desktops?
> > >
> > > https://wiki.debian.org/Wayland
> > > Unsupported Desktop environments:
> > > Cinnamon: discussed
> > > MATE: planned, source (2014)
> > > XFCE: planned
> > >
> > > Can't [whatever installs the software] notice that Gnome is
> > > installed/selected & not install synaptic?  Or patch synaptic to
> > > realize it's running under Gnome & spit out an error message and
> > > quit?
> >
> > I think a better solution would be what's suggested in message #60
> > of the bug report:
> >
> >"Why not make a [Conflicts:] with Wayland / Gnome? It's not
> > possible to make sure that synaptic installs on a [Conflicts:]
> > that would remove Wayland?"
> >
> > This is probably the easiest option to support as it should be
> > possible to implement just by adding a line to the two Packages
> > files for the architectures affected.
> >
> > > Everybody pays the price because it doesn't work with Gnome seems a
> > > bit much.
> >
> I 110% agree. 
> 
> > Given a straight toss-up though, I think synaptic has to give way
> > because there are plenty of alternatives.
> 
> Name a Good One! I don't believe you can, and be nameing one the rest of 
> us can agree is a good substitute for it.  One that works as well as or 
> better than synaptic.

They're all good in their own ways. I don't know who "the rest of us"
is, nor who should pass judgment on them.

> > I'd never heard of it until 
> > a few people started mentioning it here, and I'd never consider using
> > it myself on X except as an ordinary user.
> >
> That David, I think would be your mistake, for not experienceing a 
> package manager that for 99.99% of the job, Just Works. Lots of people 
> here will defend aptitude, but I've had aptitude totally destroy my 
> systems 3 times now, IMNSHO it should come with big warning labels, and 
> a chance to abort it when you hit the g, because once its started to do 
> what it thinks is best, and it doesn't always tell you what else its 
> going to rip out because of perceived interdependencies, it is totally 
> unresponsive to anything but the power switch. In short, aptitude can be 
> extremely dangerous to anyone who /thinks/ its a good package manager.

I thought we'd determined that something was messed up in your
terminal definitions back in June 2016, as you were also complaining
about massive screen corruption in aptitude at that time, which no one
else was reporting. In which case, it was never clear to what
questions were actually posed when you were answering g and q.

Cheers,
David.

Re: 'synaptic' removed from buster

[David Wright]

On Sat 06 Apr 2019 at 11:35:08 (+0100), Dominic Knight wrote:
> On Sat, 2019-04-06 at 19:56 +1100, David wrote:
> > On Sat, 6 Apr 2019 at 19:08, Curt  wrote:
> > > My impression from my general reading here is quite a few people
> > > rely on
> > > the synaptic package manager. I use apt-get; it's pie-like
> > > simplicity
> > > comforts me.
> >
> > Speaking in very broad terms to make a general and somewhat
> > obvious point, we could say that Gnome and synaptic are examples of
> > tools written by experts to assist lower-expertise users.
> >
> > It follows that most more-expert users (apart from the developers)
> > tend
> > not to use these kind of tools themselves. So support channels like
> > this one
> > and IRC tend to lack people who are able to answer questions based
> > on their own use of these tools, because they don't use, or even care
> > about, these kind of tools.
> >
> > I have seen this in IRC. People join there to ask questions
> > about Gnome for example, but no-one providing support in the
> > channel is actually using Gnome themselves, because they prefer more
> > sophisticatedenvironments, even though it's the default GUI for
> > Debian
> > that all the newbie questioners are using.
> >
> > Newbie asks "how do I do X in Gnome" ... and no-one there knows the
> > answer :)
> > This might be less of an issue in other distros than it is in Debian.
> >
> > > Thing is, beyond its innate and fundamental heresy (a gui app
> > > running as
> > > root!), synaptic is the only GUI package manager available in
> > > Debian
> > > AFAIK (I'm uncertain whether kpackage is defunct or not).
> >
> > If I understand correctly, Reco mentioned another one earlier in the
> > thread ...
> >
> > On Fri, 5 Apr 2019 at 00:02, Reco  wrote:
> > > The *unofficial* one is the existence of "gnome-packagekit". The
> > > thing
> > > needs users, and this is one of the ways of getting them.
> >
> > https://packages.debian.org/buster/gnome-packagekit
> > https://help.gnome.org/users/gnome-packagekit/stable/intro.html.en
> >
> > I know nothing about it, I never tried it :)
> > I prefer using shell tools for package management.
> >
> > I certainly do use some GUI tools. 'meld' for example, for side-by-
> > side
> > diffs. If that was dropped from buster then I would notice :)
> >
> Lets take a look at installing gnome-packagekit and dependencies in
> Buster;
> 
> Retrieving bug reports... Done
> Parsing Found/Fixed information... Done
> serious bugs of unattended-upgrades (→ 1.11) 
>  b2 - #905877 - regression in 1.4: upgrades random packages from
> testing to experimental (doesn't respect pinning?)
> Summary:
>  unattended-upgrades(1 bug)
> 
> 
> Then again perhaps not just yet

I'm not sure unattended-upgrades is wise for buster.
But I'm happy that there are people who are prepared
to test testing with such packages.

OTOH unattended downloads are a different matter. I have

0 */3 * * * apt-get -qq update && apt-get -qq -d upgrade && find 
/var/cache/apt/archives/ -name '*deb'

in root's crontab which also sends an email whenever packages are
sitting in the cache. (My .bash_profile also checks that.)

Cheers,
David.

Re: date(1) in stretch and buster

[David Wright]

On Tue 09 Apr 2019 at 14:55:37 (-0400), Greg Wooledge wrote:
> On Tue, Apr 09, 2019 at 01:28:40PM -0500, David Wright wrote:
> > My question about the OP's issue is whether I'm going to have to
> > change something to keep what I get in jessie and stretch, or is
> > this just a temporary bug in buster. Or has the US format been
> > wrong all along? (As an expat, I'm unfit to say.)
> 
> Looks to be intentional and permanent.
> 
> https://sourceware.org/bugzilla/show_bug.cgi?id=24046
> 
> So, you can live with it, or you can override LC_TIME, or you can create
> your own customized variant of en_US as I described earlier.  Other
> solutions are welcome, but that's all I can think of.

Perhaps I'll just set en_GB.UTF-8 as my default and US as the
additional, rather than the other way round as at present.
But thanks for the description earlier.

Cheers,
David.

Re: date(1) in stretch and buster

[David Wright]

On Tue 09 Apr 2019 at 20:46:53 (+0200), Étienne Mollier wrote:
> On 4/9/19 8:39 PM, David Wright wrote:
> > We avoided the problem when I went to sea by using the letter codes,
> > Z(ulu), A(lpha), B(ravo) etc. because there are no civil timezones,
> > daylight savings times or anything else. 
> 
> You mean, like this ?
> 
>   $ TZ=Zulu date
>   Tue Apr  9 18:46:20 UTC 2019

Yes, but only Zulu works here, not Z, nor all the rest.
But we were using it as the shortest unambiguous representation of
the timezone we were currently sailing in, suitable even for drunks
dozing off in the middle of the Middle Watch to get right. The last
thing you want is people trying to do timezone conversions in the
middle of the night. The confusion arises because the scientific
clocks displayed UTC whereas the scientists and crew worked in
shipboard/local time.

Cheers,
David.

Re: mais ou est passee la place manquante ?

[Étienne Mollier]

Pascal Hambourg, au 2019-04-09 :
> Au passage, la méthode pour écarter la ligne d'en-têtes n'est
> pas fiable car elle dépend de la langue d'affichage. Il aurait
> mieux valu utiliser tail pour extraire la dernière ligne.

Oui, une autre solution aurait pu aussi consister à s'en tenir à
la locale C :

#!/bin/bash
set -e
export LANG=C
homedevice="$(
df /home \
| grep -v '^Filesystem' \
| cut -f1 -d' '
)"
tune2fs -m 0 "$homedevice"

Amicalement,
-- 
Étienne Mollier 

Ce qui est bien, c'est qu'il y a plein de manières de faire.
Ce qui est terrible, c'est qu'il y a encore plus de manières de
se prendre les pieds dans le tapis.

Re: date(1) in stretch and buster

[Gene Heskett]

On Tuesday 09 April 2019 14:39:19 David Wright wrote:

> On Tue 09 Apr 2019 at 14:03:08 (-0400), Gene Heskett wrote:
> > On Tuesday 09 April 2019 09:38:43 Vincent Lefevre wrote:
> > > On 2019-04-08 18:26:23 +0300, Reco wrote:
> > > > stretch$ TZ=UTC date
> > > > Mon Apr  8 15:22:02 UTC 2019
> > > > buster$ TZ=UTC date
> > > > Mon 08 Apr 2019 03:22:04 PM UTC
> > >
> > > This is unrelated to your issue, but note that the correct TZ
> > > string for UTC is "UTC0", not "UTC". See
> > >
> > >
> > > http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08
> > >.htm l
> >
> > Yikes. Can that be actually put into English?
> >
> > I've read thru it, and the only thing I can come away is that 'if
> > UTC, then offset is required, and it can be anything from -23 to
> > +23, including your example 0 meaning no offset from UTC.
> >
> > As it now reads, theres a ton of ambiguity that needs further
> > clarification.
>
> I'd agree. I really can't see how you can use UTC6 as a timezone
> though it makes sense as an offset. But in that case, UTC is UTC,
> and the zero is just tautological.
>
> We avoided the problem when I went to sea by using the letter codes,
> Z(ulu), A(lpha), B(ravo) etc. because there are no civil timezones,
> daylight savings times or anything else.
>
> Cheers,
> David.

I've at times wondered about that, but I never did any military service, 
seems the were wanting cannon targets for Korea. I made a 98 on the AFQT 
in '53 or so, which automatically got me 4F'd. Obviously they didn't 
want anyone who could think. The next best score in just under 140 boys 
that took that test that day was 36.  He of course would take orders. 
They didn't have time to mess with somebody who would question orders. 
60+ years later I can say I've had a good ride thru several technical 
fields. I have a list of BTDT's that not many could match.

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)
Genes Web page 

Re: putting config files under revision control

[Lee]

On 4/9/19, Dan Ritter wrote:
> Lee wrote:
>> What are people doing for putting config files in [under?] git?
>>
>> I'd like to have at least some system config files maintained in git
>> so I can get a history of changes.
>>   (and yes, I know, I really should be using a backup system for that,
>> but I'm still at the 'rsync to usb drive' stage)
>
> apt install etckeeper. Choose the git backend. (I think it's the
> default these days.)

Thank you!

> Chef or Puppet when you want to do this at scale.

Maybe someday.  They'd be nice to learn, but they seem to be massive
overkill for home use.  ..or at least for my home use.

Thanks
Lee

Re: date(1) in stretch and buster

[Michael Stone]

On Tue, Apr 09, 2019 at 03:11:27PM -0400, Gene Heskett wrote:

The point I was trying to make Mike, is, if thats to be a std, its an
extremely obtuse way of writing that std. Consequently I suspect it will
be 100% ignored. And folks will continue to muddle along just fine. :)


It reads fine to me, but it's describing functionality that's been 
obsolete for decades so it doesn't really matter.


Mike Stone

Re: date(1) in stretch and buster

[Gene Heskett]

On Tuesday 09 April 2019 14:28:18 Michael Stone wrote:

> On Tue, Apr 09, 2019 at 02:03:08PM -0400, Gene Heskett wrote:
> >On Tuesday 09 April 2019 09:38:43 Vincent Lefevre wrote:
> >> On 2019-04-08 18:26:23 +0300, Reco wrote:
> >> > stretch$ TZ=UTC date
> >> > Mon Apr  8 15:22:02 UTC 2019
> >> > buster$ TZ=UTC date
> >> > Mon 08 Apr 2019 03:22:04 PM UTC
> >>
> >> This is unrelated to your issue, but note that the correct TZ
> >> string for UTC is "UTC0", not "UTC". See
> >>
> >>
> >> http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.
> >>htm l
> >
> >Yikes. Can that be actually put into English?
> >
> >I've read thru it, and the only thing I can come away is that 'if
> > UTC, then offset is required, and it can be anything from -23 to
> > +23, including your example 0 meaning no offset from UTC.
> >
> >As it now reads, theres a ton of ambiguity that needs further
> >clarification.
>
> There's a discrepency between the historic posix definition of the TZ
> environment variable, and how it works on a modern system. There's an
> official workaround that I've never actually seen used in the wild. It
> doesn't matter in practice.
>
> Mike Stone

The point I was trying to make Mike, is, if thats to be a std, its an 
extremely obtuse way of writing that std. Consequently I suspect it will 
be 100% ignored. And folks will continue to muddle along just fine. :) 

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)
Genes Web page 

Re: putting config files under revision control

[Dan Ritter]

Lee wrote: 
> What are people doing for putting config files in [under?] git?
> 
> I'd like to have at least some system config files maintained in git
> so I can get a history of changes.
>   (and yes, I know, I really should be using a backup system for that,
> but I'm still at the 'rsync to usb drive' stage)

apt install etckeeper. Choose the git backend. (I think it's the
default these days.)

Chef or Puppet when you want to do this at scale.

-dsr-

Re: build vs install

[Lee]

On 4/9/19, Brad Rogers  wrote:
> On Tue, 9 Apr 2019 13:30:45 -0400
> Lee  wrote:
>
> Hello Lee,
>
>>> package installer may write the configs somewhere other than where the
>>> creator does, for example.
>>But they're still all text files - right?  There's nothing like MS
>
> I'd hope so, yes.  IOW, I'd hope the Debian maintainers don't change
> config styles.  Of course, there are several organisations that seem to
> like SQLite for config files.  I'm not sure what, if any, advantage
> SQLite offers over plain text though.
>
>>Windows registry that you might have to clean after uninstalling
>>something or installing in a different location?
>
> Almost certainly not.  You should be awarethat distro maintained
> packages usually put binaries in /usr/bin/, libraries in /usr/lib/ and
> so on.  Self-compiled software most often puts stuff under
> /usr/local/bin/, /usr/local/libs/ etc.  Occasionally, upon removal, some
> stuff might get left behind, perhaps resulting in erratic behaviour.
> Such things are rare, thankfully.

OK - good to know.  Somehow I'd got the impression that systemd was
moving to a windows registry type thing for config data instead of
keeping it in plain text files.

Thanks
Lee

Re: date(1) in stretch and buster

[Greg Wooledge]

On Tue, Apr 09, 2019 at 01:28:40PM -0500, David Wright wrote:
> My question about the OP's issue is whether I'm going to have to
> change something to keep what I get in jessie and stretch, or is
> this just a temporary bug in buster. Or has the US format been
> wrong all along? (As an expat, I'm unfit to say.)

Looks to be intentional and permanent.

https://sourceware.org/bugzilla/show_bug.cgi?id=24046

So, you can live with it, or you can override LC_TIME, or you can create
your own customized variant of en_US as I described earlier.  Other
solutions are welcome, but that's all I can think of.

Re: date(1) in stretch and buster

[Étienne Mollier]

On 4/9/19 8:39 PM, David Wright wrote:
> We avoided the problem when I went to sea by using the letter codes,
> Z(ulu), A(lpha), B(ravo) etc. because there are no civil timezones,
> daylight savings times or anything else. 

You mean, like this ?

$ TZ=Zulu date
Tue Apr  9 18:46:20 UTC 2019

:)
-- 
Étienne Mollier 

putting config files under revision control

[Lee]

What are people doing for putting config files in [under?] git?

I'd like to have at least some system config files maintained in git
so I can get a history of changes.
  (and yes, I know, I really should be using a backup system for that,
but I'm still at the 'rsync to usb drive' stage)

For stuff that multiple people are allowed to update, I'm thinking

mkdir /source
   # config files go in a /source/ dir
chgrp staff source
  # add whoever is allowed to update configs to the 'staff' group

And then for something like bind where the config files are kept in /etc/bind
  mkdir /source/bind
Set the appropriate permissions on /source/bind, copy the files from
/etc/bind & create a make file where  'make install' will do something
like
  sudo rsync --include-from=xxx (or maybe --files-from==xxx) /etc/bind
and then a 'sudo rndc reload' to tell bind to read the new files.

Which seems workable but not very general.  Any pointers on how to do it better?

Thanks
Lee

Re: date(1) in stretch and buster

[David Wright]

On Tue 09 Apr 2019 at 14:03:08 (-0400), Gene Heskett wrote:
> On Tuesday 09 April 2019 09:38:43 Vincent Lefevre wrote:
> > On 2019-04-08 18:26:23 +0300, Reco wrote:
> > > stretch$ TZ=UTC date
> > > Mon Apr  8 15:22:02 UTC 2019
> > > buster$ TZ=UTC date
> > > Mon 08 Apr 2019 03:22:04 PM UTC
> >
> > This is unrelated to your issue, but note that the correct TZ string
> > for UTC is "UTC0", not "UTC". See
> >
> >  
> > http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.htm
> >l
> 
> Yikes. Can that be actually put into English?
> 
> I've read thru it, and the only thing I can come away is that 'if UTC, 
> then offset is required, and it can be anything from -23 to +23, 
> including your example 0 meaning no offset from UTC.
> 
> As it now reads, theres a ton of ambiguity that needs further 
> clarification.

I'd agree. I really can't see how you can use UTC6 as a timezone
though it makes sense as an offset. But in that case, UTC is UTC,
and the zero is just tautological.

We avoided the problem when I went to sea by using the letter codes,
Z(ulu), A(lpha), B(ravo) etc. because there are no civil timezones,
daylight savings times or anything else. 

Cheers,
David.

Re: date(1) in stretch and buster

[Étienne Mollier]

Gene Heskett, on 2019-04-09 :
> On Tuesday 09 April 2019 09:38:43 Vincent Lefevre wrote:
> > This is unrelated to your issue, but note that the correct TZ string
> > for UTC is "UTC0", not "UTC". See
> >
> > http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.htm
> > l
>
> Yikes. Can that be actually put into English?

Good Day Gene,

For general use, you might prefer to stick to a geographical
location, instead of a time zone code.  This way you don't have
to wonder if you are either in Daylight savings or not, if your
offset is -4 or +4, etc.  Maintainers of `tzdata` handle that
for you (a great "Thank You!" to these people by the way).

Usually, people in my neighborhood would use the equivalent of
the following, for instance:

$ TZ=Europe/Paris LC_TIME=fr_FR.UTF-8 date
mardi 9 avril 2019, 20:14:07 (UTC+0200)

Perhaps you will prefer setting a location nearer to yourself:

$ TZ=America/New_York LC_TIME=en_US.UTF-8 date
Tue 09 Apr 2019 02:20:00 PM EDT

The output may differ depending on you operating system level,
given Reco's observations.  Feel free to have à look at
/usr/share/zoneinfo/, to have an idea of the available
locations.

Kind Regards,
-- 
Étienne Mollier 

Re: mais ou est passee la place manquante ?

[Pascal Hambourg]

Le 09/04/2019 à 17:04, Stephane Ascoet a écrit :


Hamster:

utiliser la sortie de df risque de faire agir sur / comme tu l'a si bien
dit.


Non, si /home n'est pas un point de montage, il n'a pas de raison 
d'apparaitre en sortie de df.


De l'aveu même de son auteur, le script fourni dans le message auquel 
Hamster répondait se contente d'extraire le périphérique contenant 
/home, sans vérifier le point de montage. Je le remets :


#!/bin/bash
set -e
homedevice="$(
df /home \
| grep -v '^Filesystem' \
| cut -f1 -d' '
)"
tune2fs -m 0 "$homedevice"

Au passage, la méthode pour écarter la ligne d'en-têtes n'est pas fiable 
car elle dépend de la langue d'affichage. Il aurait mieux valu utiliser 
tail pour extraire la dernière ligne.

Re: date(1) in stretch and buster

[David Wright]

On Tue 09 Apr 2019 at 15:38:43 (+0200), Vincent Lefevre wrote:
> On 2019-04-08 18:26:23 +0300, Reco wrote:
> > stretch$ TZ=UTC date
> > Mon Apr  8 15:22:02 UTC 2019
> > buster$ TZ=UTC date
> > Mon 08 Apr 2019 03:22:04 PM UTC
> 
> This is unrelated to your issue, but note that the correct TZ string
> for UTC is "UTC0", not "UTC". See
> 
>   http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html

I think it's unwise to write that here. AIUI GNU/linux date uses the
so-called 3rd format for timezones, which points into the time zone
database. There's an entry in that database for UTC, but not for UTC*.
And, of course, what would the civil name be for date to label that
timezone with? You can't use just "UTC".

BTW it prints the right answer because it only looks at the 0; the
UTC is ignored. (See BTS #646174.)

The database, of course, contains a lot more information than the
standard you have quoted (for the 1st and 2nd formats) which only
tells you what time zone you're in now, and when that zone last
switched and will switch to/from DST. For example, it should know that
at the time of the Unix Epoch, Britain was "enjoying" summer time,
which we called "British Standard Time" at the time, and it does.
So we celebrated the New Year at the "wrong time" on three occasions
in that period, 1969, 1970 & 1971.
$ TZ=Europe/London date -d '1970-01-01 00:00:00 +'
Thu Jan  1 01:00:00 BST 1970
$ TZ=Europe/London date -d '1972-01-01 00:00:00 +'
Sat Jan  1 00:00:00 GMT 1972
$ 

My question about the OP's issue is whether I'm going to have to
change something to keep what I get in jessie and stretch, or is
this just a temporary bug in buster. Or has the US format been
wrong all along? (As an expat, I'm unfit to say.)

$ TZ=UTC date
Tue Apr  9 16:53:28 UTC 2019
$ 
ie a 24-hour clock, and not a 12-hour clock trying to look like a
24-hour one.

Cheers,
David.

Re: date(1) in stretch and buster

[Michael Stone]

On Tue, Apr 09, 2019 at 02:03:08PM -0400, Gene Heskett wrote:

On Tuesday 09 April 2019 09:38:43 Vincent Lefevre wrote:


On 2019-04-08 18:26:23 +0300, Reco wrote:
> stretch$ TZ=UTC date
> Mon Apr  8 15:22:02 UTC 2019
> buster$ TZ=UTC date
> Mon 08 Apr 2019 03:22:04 PM UTC

This is unrelated to your issue, but note that the correct TZ string
for UTC is "UTC0", not "UTC". See


http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.htm
l


Yikes. Can that be actually put into English?

I've read thru it, and the only thing I can come away is that 'if UTC,
then offset is required, and it can be anything from -23 to +23,
including your example 0 meaning no offset from UTC.

As it now reads, theres a ton of ambiguity that needs further
clarification.


There's a discrepency between the historic posix definition of the TZ 
environment variable, and how it works on a modern system. There's an 
official workaround that I've never actually seen used in the wild. It 
doesn't matter in practice.


Mike Stone

Re: Mettre à jour la libc6

[Yahoo]

Bon c'est bon,

pour une raison qui m'??chappe Thunderbird m'a configur?? l'envoi de mail 
en encodage Occidental(ISO-8859-1), mais c'est quand m??me mieux en UTF-8.


donc re re voici le texte :

/Kubernets n'est pas un concurrent de docker, mais un orchestrer de 
conteneur docker, il permet de g??rer, automatiser les conteneurs docker. 
c'est une des solutions les plus aboutit pour la gestion de docker.

/

//

/Mais sans kubernets, il est possible d'utiliser docker dans des 
environnement de production, et m??me avec des gestions de redondance. De 
plus docker permet de red??ployer des services rapidement sur un syst??me, 
ce qui peut ??tre tr??s pratique pour des migrations et mise ?? jour, il 
suffit de cr??er sont conteneur avec toutes les configurations voulus de 
ses services, et le d??ploiement peut se faire sur n'importe quel OS 
acceptant docker (soit presque tous les OS)./


//

/Je ne sais pas si c'est la meilleur m??thode, et aucun cas je ne 
pourrais dire cela, chaque m??thode ?? son int??r??t./


//

/Pour ma part, n'utilisant pas de solution de virtualisation sur tout 
les serveurs, je ne peux pas me permettre de bloquer un serveur pour une 
seule version de php, ou d'un autre service. Et dans ce cas l??, docker 
est parfait./


//

/un exemple:/

//

/pour un client, j'ai due migrer un service d'agenda web utilisant du 
php5, et sur le m??me serveur j'ai due installer une application CRM 
fonctionnement en version php7.3./


//

/Mon syst??me ??tant une Debian Stretch avec une version de php7.1, 
compliqu?? de faire fonctionner tous le monde ensemble./


//

/J'ai donc cr??er un conteneur PHP5 et un conteneur PHP7.3, un service 
apache2 install?? sur le syst??me h??te permet de faire le proxy web./


//

/sch??matiquement :
/

//

/
/

//

/Debian Stretch -> apache2 mode proxy/

//

/-> docker-ce?? -> un conteneur avec php5/

//

/-> un conteneur avec php7.3/

//

/
/

//

/ensuite docker utilise une gestion de volume, qui permet de g??rer les 
r??pertoires web des conteneurs ?? partir du syst??me h??te./


//

/En esp??rant t'avoir donn?? envie d'essayer docker :)/

Le 09/04/2019 ?? 20:12, Yahoo a ??crit??:


effectivement, d??sol?? je n'ai pas fais attention.

Merci Bernard.

voici le texte, j???esp??re sans probl??me d'encodage:


Kubernets n'est pas un concurrent de docker, mais un orchestrer de 
conteneur docker, il permet de g??rer, automatiser les conteneurs 
docker. c'est une des solutions les plus aboutit pour la gestion de 
docker.


Mais sans kubernets, il est possible d'utiliser docker dans des 
environnement de production, et m??me avec des gestions de redondance. 
De plus docker permet de red??ployer des services rapidement sur un 
syst??me, ce qui peut ??tre tr??s pratique pour des migrations et mise 
?? jour, il suffit de cr??er sont conteneur avec toutes les 
configurations voulus de ses services, et le d??ploiement peut se 
faire sur n'importe quel OS acceptant docker (soit presque tous les OS).


Je ne sais pas si c'est la meilleur m??thode, et aucun cas je ne 
pourrais dire cela, chaque m??thode ?? son int??r??t.


Pour ma part, n'utilisant pas de solution de virtualisation sur tout 
les serveurs, je ne peux pas me permettre de bloquer un serveur pour 
une seule version de php, ou d'un autre service. Et dans ce cas l??, 
docker est parfait.


un exemple:

pour un client, j'ai due migrer un service d'agenda web utilisant du 
php5, et sur le m??me serveur j'ai due installer une application CRM 
fonctionnement en version php7.3.


Mon syst??me ??tant une Debian Stretch avec une version de php7.1, 
compliqu?? de faire fonctionner tous le monde ensemble.


J'ai donc cr??er un conteneur PHP5 et un conteneur PHP7.3, un service 
apache2 install?? sur le syst??me h??te permet de faire le proxy web.


sch??matiquement :


Debian Stretch -> apache2 mode proxy

?? -> docker-ce?? -> un 
conteneur avec php5


?? 
-> un conteneur avec php7.3



ensuite docker utilise une gestion de volume, qui permet de g??rer les 
r??pertoires web des conteneurs ?? partir du syst??me h??te.


En esp??rant t'avoir donn?? envie d'essayer docker :)


Le 09/04/2019 ?? 19:26, Bernard Schoenacker a ??crit??:

bonjour,

snip snip, d??sol?? de taper sur le pianiste
mais le texte est presque illisible ?? cause
de l'encodage du texte

merci
slt
bernard



- Mail original -


De: "Yahoo"
??:debian-user-french@lists.debian.org
Envoy??: Mardi 9 Avril 2019 19:00:43
Objet: Re: Mettre ?? jour la libc6
Kubernets n'est pas un concurrent de docker, mais un orchestrer de
conteneur docker, il permet de g??rer, automatiser les conteneurs
docker. c'est une des solutions les plus aboutit pour la gestion de
docker.
Mais sans kubernets, il est possible d'utiliser docker dans des
environnement de production, et m??me avec des gestions de
redondance. De plus docker permet de 

Re: build vs install

[Brad Rogers]

On Tue, 9 Apr 2019 13:30:45 -0400
Lee  wrote:

Hello Lee,

>> package installer may write the configs somewhere other than where the
>> creator does, for example.  
>But they're still all text files - right?  There's nothing like MS

I'd hope so, yes.  IOW, I'd hope the Debian maintainers don't change
config styles.  Of course, there are several organisations that seem to
like SQLite for config files.  I'm not sure what, if any, advantage
SQLite offers over plain text though.

>Windows registry that you might have to clean after uninstalling
>something or installing in a different location?

Almost certainly not.  You should be awarethat distro maintained
packages usually put binaries in /usr/bin/, libraries in /usr/lib/ and
so on.  Self-compiled software most often puts stuff under
/usr/local/bin/, /usr/local/libs/ etc.  Occasionally, upon removal, some
stuff might get left behind, perhaps resulting in erratic behaviour.
Such things are rare, thankfully.

-- 
 Regards  _
 / )   "The blindingly obvious is
/ _)radnever immediately apparent"
Gary don't need his eyes to see, Gary and his eyes have parted company
Gary Gilmore's Eyes - The Adverts


pgpu0fhAYsXaU.pgp
Description: OpenPGP digital signature

Re: Mettre à jour la libc6

[Yahoo]

Bon, j'ai un petit soucis d'encodage,

avant de ré envoyer le texte, je fais un test avec ce message contenant 
des accents é à è ,


car je n'ai pas le soucis avec des tests sur d'autre boite mail.

veillez m'excuser pour cette pollution de mail

Le 09/04/2019 à 20:12, Yahoo a écrit :


effectivement, d??sol?? je n'ai pas fais attention.

Merci Bernard.

voici le texte, j???esp??re sans probl??me d'encodage:


Kubernets n'est pas un concurrent de docker, mais un orchestrer de 
conteneur docker, il permet de g??rer, automatiser les conteneurs 
docker. c'est une des solutions les plus aboutit pour la gestion de 
docker.


Mais sans kubernets, il est possible d'utiliser docker dans des 
environnement de production, et m??me avec des gestions de redondance. 
De plus docker permet de red??ployer des services rapidement sur un 
syst??me, ce qui peut ??tre tr??s pratique pour des migrations et mise 
?? jour, il suffit de cr??er sont conteneur avec toutes les 
configurations voulus de ses services, et le d??ploiement peut se 
faire sur n'importe quel OS acceptant docker (soit presque tous les OS).


Je ne sais pas si c'est la meilleur m??thode, et aucun cas je ne 
pourrais dire cela, chaque m??thode ?? son int??r??t.


Pour ma part, n'utilisant pas de solution de virtualisation sur tout 
les serveurs, je ne peux pas me permettre de bloquer un serveur pour 
une seule version de php, ou d'un autre service. Et dans ce cas l??, 
docker est parfait.


un exemple:

pour un client, j'ai due migrer un service d'agenda web utilisant du 
php5, et sur le m??me serveur j'ai due installer une application CRM 
fonctionnement en version php7.3.


Mon syst??me ??tant une Debian Stretch avec une version de php7.1, 
compliqu?? de faire fonctionner tous le monde ensemble.


J'ai donc cr??er un conteneur PHP5 et un conteneur PHP7.3, un service 
apache2 install?? sur le syst??me h??te permet de faire le proxy web.


sch??matiquement :


Debian Stretch -> apache2 mode proxy

?? -> docker-ce?? -> un 
conteneur avec php5


?? 
-> un conteneur avec php7.3



ensuite docker utilise une gestion de volume, qui permet de g??rer les 
r??pertoires web des conteneurs ?? partir du syst??me h??te.


En esp??rant t'avoir donn?? envie d'essayer docker :)


Le 09/04/2019 ?? 19:26, Bernard Schoenacker a ??crit??:

bonjour,

snip snip, d??sol?? de taper sur le pianiste
mais le texte est presque illisible ?? cause
de l'encodage du texte

merci
slt
bernard



- Mail original -


De: "Yahoo"
??:debian-user-french@lists.debian.org
Envoy??: Mardi 9 Avril 2019 19:00:43
Objet: Re: Mettre ?? jour la libc6
Kubernets n'est pas un concurrent de docker, mais un orchestrer de
conteneur docker, il permet de g??rer, automatiser les conteneurs
docker. c'est une des solutions les plus aboutit pour la gestion de
docker.
Mais sans kubernets, il est possible d'utiliser docker dans des
environnement de production, et m??me avec des gestions de
redondance. De plus docker permet de red??ployer des services
rapidement sur un syst??me, ce qui peut ??tre tr??s pratique pour
des migrations et mise ?? jour, il suffit de cr??er sont conteneur
avec toutes les configurations voulus de ses services, et le
d??ploiement peut se faire sur n'importe quel OS acceptant docker
(soit presque tous les OS).
Je ne sais pas si c'est la meilleur m??thode, et aucun cas je ne
pourrais dire cela, chaque m??thode ?? son int??r??t.
Pour ma part, n'utilisant pas de solution de virtualisation sur tout
les serveurs, je ne peux pas me permettre de bloquer un serveur pour
une seule version de php, ou d'un autre service. Et dans ce cas
l??, docker est parfait.
un exemple:
pour un client, j'ai due migrer un service d'agenda web utilisant du
php5, et sur le m??me serveur j'ai due installer une application
CRM fonctionnement en version php7.3.
Mon syst??me ??tant une Debian Stretch avec une version de php7.1,
compliqu?? de faire fonctionner tous le monde ensemble.
J'ai donc cr??er un conteneur PHP5 et un conteneur PHP7.3, un
service apache2 install?? sur le syst??me h??te permet de faire
le proxy web.
sch??matiquement :
Debian Stretch -> apache2 mode proxy
??
-> docker-ce?? -> un conteneur avec php5
??
-> un conteneur avec php7.3
ensuite docker utilise une gestion de volume, qui permet de g??rer
les r??pertoires web des conteneurs ?? partir du syst??me 

Re: Mettre à jour la libc6

[Yahoo]

effectivement, d??sol?? je n'ai pas fais attention.

Merci Bernard.

voici le texte, j???esp??re sans probl??me d'encodage:


Kubernets n'est pas un concurrent de docker, mais un orchestrer de 
conteneur docker, il permet de g??rer, automatiser les conteneurs docker. 
c'est une des solutions les plus aboutit pour la gestion de docker.


Mais sans kubernets, il est possible d'utiliser docker dans des 
environnement de production, et m??me avec des gestions de redondance. De 
plus docker permet de red??ployer des services rapidement sur un syst??me, 
ce qui peut ??tre tr??s pratique pour des migrations et mise ?? jour, il 
suffit de cr??er sont conteneur avec toutes les configurations voulus de 
ses services, et le d??ploiement peut se faire sur n'importe quel OS 
acceptant docker (soit presque tous les OS).


Je ne sais pas si c'est la meilleur m??thode, et aucun cas je ne pourrais 
dire cela, chaque m??thode ?? son int??r??t.


Pour ma part, n'utilisant pas de solution de virtualisation sur tout les 
serveurs, je ne peux pas me permettre de bloquer un serveur pour une 
seule version de php, ou d'un autre service. Et dans ce cas l??, docker 
est parfait.


un exemple:

pour un client, j'ai due migrer un service d'agenda web utilisant du 
php5, et sur le m??me serveur j'ai due installer une application CRM 
fonctionnement en version php7.3.


Mon syst??me ??tant une Debian Stretch avec une version de php7.1, 
compliqu?? de faire fonctionner tous le monde ensemble.


J'ai donc cr??er un conteneur PHP5 et un conteneur PHP7.3, un service 
apache2 install?? sur le syst??me h??te permet de faire le proxy web.


sch??matiquement :


Debian Stretch -> apache2 mode proxy

-> docker-ce?? -> un conteneur avec php5

-> un conteneur avec php7.3


ensuite docker utilise une gestion de volume, qui permet de g??rer les 
r??pertoires web des conteneurs ?? partir du syst??me h??te.


En esp??rant t'avoir donn?? envie d'essayer docker :)


Le 09/04/2019 ?? 19:26, Bernard Schoenacker a ??crit??:

bonjour,

snip snip, d??sol?? de taper sur le pianiste
mais le texte est presque illisible ?? cause
de l'encodage du texte

merci
slt
bernard



- Mail original -


De: "Yahoo" 
??: debian-user-french@lists.debian.org
Envoy??: Mardi 9 Avril 2019 19:00:43
Objet: Re: Mettre ?? jour la libc6
Kubernets n'est pas un concurrent de docker, mais un orchestrer de
conteneur docker, il permet de g??rer, automatiser les conteneurs
docker. c'est une des solutions les plus aboutit pour la gestion de
docker.
Mais sans kubernets, il est possible d'utiliser docker dans des
environnement de production, et m??me avec des gestions de
redondance. De plus docker permet de red??ployer des services
rapidement sur un syst??me, ce qui peut ??tre tr??s pratique pour
des migrations et mise ?? jour, il suffit de cr??er sont conteneur
avec toutes les configurations voulus de ses services, et le
d??ploiement peut se faire sur n'importe quel OS acceptant docker
(soit presque tous les OS).
Je ne sais pas si c'est la meilleur m??thode, et aucun cas je ne
pourrais dire cela, chaque m??thode ?? son int??r??t.
Pour ma part, n'utilisant pas de solution de virtualisation sur tout
les serveurs, je ne peux pas me permettre de bloquer un serveur pour
une seule version de php, ou d'un autre service. Et dans ce cas
l??, docker est parfait.
un exemple:
pour un client, j'ai due migrer un service d'agenda web utilisant du
php5, et sur le m??me serveur j'ai due installer une application
CRM fonctionnement en version php7.3.
Mon syst??me ??tant une Debian Stretch avec une version de php7.1,
compliqu?? de faire fonctionner tous le monde ensemble.
J'ai donc cr??er un conteneur PHP5 et un conteneur PHP7.3, un
service apache2 install?? sur le syst??me h??te permet de faire
le proxy web.
sch??matiquement :
Debian Stretch -> apache2 mode proxy
??
-> docker-ce?? -> un conteneur avec php5
??
-> un conteneur avec php7.3
ensuite docker utilise une gestion de volume, qui permet de g??rer
les r??pertoires web des conteneurs ?? partir du syst??me h??te.
En esp??rant t'avoir donn?? envie d'essayer docker :)
Le 09/04/2019 ?? 16:01, Nicolas Malgat a ??crit??:

Re: date(1) in stretch and buster

[Greg Wooledge]

On Tue, Apr 09, 2019 at 02:03:08PM -0400, Gene Heskett wrote:
> Yikes. Can that be actually put into English?

"Use date -u instead."

Re: build vs install

[Lee]

On 4/9/19, Michael Howard wrote:
> On 09/04/2019 16:35, Lee wrote:
>> What are the downsides to getting the source code and doing the
>> build/install myself vs. using a pre-built package other than I'm
>> responsible for noticing the software needs to be updated?
>>
>> The latest example is ttcp
>>http://nuttcp.net/nuttcp/latest/  has 8.1.4
>>https://tracker.debian.org/pkg/nuttcp  has 6.1.2-4
>>
>> If I ever decide to go with the debian package I just uninstall the
>> software I built and .. anything else that needs to be done before
>> installing an official package?
>>
>> TIA
>> Lee
>>
> You could build a debian package from the source and let apt take care
> of install/updates/uninstall. Admittedly, it could be a bit of work
> initially with versions so far apart as above but it is an option.

That's an interesting idea.  Get the debian stuff from
  https://packages.debian.org/stretch/nuttcp
and see which patches still apply to the current version of the software.

But I'm missing what letting apt take care of
install/updates/uninstall gets me since I'd be the one creating &
updating the home-built package.

Thanks,
Lee

Re: date(1) in stretch and buster

[Gene Heskett]

On Tuesday 09 April 2019 09:38:43 Vincent Lefevre wrote:

> On 2019-04-08 18:26:23 +0300, Reco wrote:
> > stretch$ TZ=UTC date
> > Mon Apr  8 15:22:02 UTC 2019
> > buster$ TZ=UTC date
> > Mon 08 Apr 2019 03:22:04 PM UTC
>
> This is unrelated to your issue, but note that the correct TZ string
> for UTC is "UTC0", not "UTC". See
>
>  
> http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.htm
>l

Yikes. Can that be actually put into English?

I've read thru it, and the only thing I can come away is that 'if UTC, 
then offset is required, and it can be anything from -23 to +23, 
including your example 0 meaning no offset from UTC.

As it now reads, theres a ton of ambiguity that needs further 
clarification.

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)
Genes Web page 

Re: retour expérience sur Buster ??

[Étienne Mollier]

Bruno, au 2018-04-09 :
> pour le menu je m'explique un peu mieux
>
> sur une appli tu as la  barre de titre puis les menu ( fichier ,
> édition ,.), puis la tool bar
>
> en haut ( ou en bas)  de ton écran tu as la barre 'tableau de
> bord ou tu as le launcher, et autres infos.
>
> avec buster et kde je n'ai trouvé que deux alternative :
>
> 1 - le menu s'affiche dans la barre 'tableau de bord' et il faut
> jongler entre la fenêtre appli et le tableau de bord

Ah d'accord, je crois que je commence à comprendre le coup du
menu s'affichant dans le tableau de bord au lieu de la fenêtre :
c'est un peu comme dans Ubuntu classique, ou historiquement à la
Apple Macintosh (en tout cas c'était présent dans System 7.5).

Je n'ai pas eu ce comportement là par défaut, mais j'ai pu
l'activer en suivant :

Clic droit sur le tableau de bord
-> Options de Tableau de bord
-> Ajouter un panneau
-> Barre de menu des applications

Les applications Qt relancées après coup voient leur barre de
menu insérée dans le nouveau panneau, à la place de la fenêtre.
Pour revenir à l'état initial, je fais :

Clic droit sur la nouvelle barre de menu externe
-> Configurer Tableau de bord
-> Clic sur le bouton en forme de clé de mécano "Plus de
   paramètres"
-> Supprimer le tableau de bord.

Ensuite je relance mes applications, et la barre de menu est de
retour à l'intérieur des fenêtres.  Est-ce que ça aide ?

Amicalement,
-- 
Étienne Mollier 

Re: putty go slow

[Celejar]

On Tue, 09 Apr 2019 08:46:07 +0200
Peter Wiersig  wrote:

...

> For reference, here's my output:
> 
> eth0: flags=4163  mtu 1500
> inet 217.172.177.159  netmask 255.255.255.0  broadcast 217.172.177.255
> ether 00:19:66:f1:43:9e  txqueuelen 1000  (Ethernet)
> RX packets 81994186  bytes 14753508445 (13.7 GiB)
> RX errors 14  dropped 0  overruns 14  frame 0
> TX packets 107524155  bytes 14836289080 (13.8 GiB)
> TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
> 
> 
> Take note of the 2nd line of RX and TX status lines, there should be no
> counters there, in my case 14 overruns in regard to 819 million packets
> is a very low error rate.  I suspect that if there's a network problem

I think you mean 81.9 million packets.

Celejar

Re: build vs install

[Lee]

Hi Brad,

On 4/9/19, Brad Rogers wrote:
> On Tue, 9 Apr 2019 11:35:05 -0400
> Lee wrote:
>
> Hello Lee,
>
>>What are the downsides to getting the source code and doing the
>>build/install myself vs. using a pre-built package other than I'm
>>responsible for noticing the software needs to be updated?
>
> Mostly, it's time.  Also, you're responsible for installing any required
> new dependencies, be they development packages or binaries.  Which,
> again, comes down to time.  Such package requirements aren't always
> immediately obvious, and may not even be mentioned in release notes.
>
> For this reason, there's only one piece of software I compile and install
> 'by hand' as it were.

Yes, building & installing yourself takes time.  Hopefully not much
after working the kinks out on the first build/install, but still..

>>The latest example is ttcp
>>  http://nuttcp.net/nuttcp/latest/  has 8.1.4
>>  https://tracker.debian.org/pkg/nuttcp  has 6.1.2-4
>
> I was going to suggest adding stable-backports to your sources.list.
> Assuming, of course, that you're running stable.  A quick check of
> Debian packages shows that wouldn't help in this case.  A fact, no doubt
> you're aware of, hence your query.

Yes.  gawk being the other package I'd like to have on the latest
version, but doesn't seem to be in backports :(

>>If I ever decide to go with the debian package I just uninstall the
>>software I built and .. anything else that needs to be done before
>
> Yes, made easier if the install mechanism creates a useful log and you
> can later 'make --uninstall'.
>
>>installing an official package?
>
> I'd back up any config file(s) just in case anything 'blows up'.

Which raises another question, but I'll start another thread for that.

>  That
> is to say, if there's any incompatibility between Debian's way of writing
> the config files, and the package creator's way of doing so.  The Debian
> package installer may write the configs somewhere other than where the
> creator does, for example.

But they're still all text files - right?  There's nothing like MS
Windows registry that you might have to clean after uninstalling
something or installing in a different location?

Thank you!
Lee

Re: retour expérience sur Buster ??

[Bruno Volpi]

salut,

pour service c'est ok , je n'avais pas suivi la mise à jour su / su -

pour le menu je m'explique un peu mieux

sur une appli tu as la  barre de titre puis les menu ( fichier , édition 
,.), puis la tool bar


en haut ( ou en bas)  de ton écran tu as la barre 'tableau de bord ou tu 
as le launcher, et autres infos.


avec buster et kde je n'ai trouvé que deux alternative :

1 - le menu s'affiche dans la barre 'tableau de bord' et il faut jongler 
entre la fenêtre appli et le tableau de bord


2 - barre de menu se comporte comme un menu déroulant et se trouve dans 
la barre de titre de l'application


et comme dans un cas comme dans l'autre c'est plus que 'pas très pratique' .

ceci ne survient que sur les applications développée en  QT

je vais voir avec les dev de KDE s'il y a un bypass possible, sinon 
retour sur gnome.


EN tous les cas merci pour ton aide.


Bruno...

Le 08/04/2019 à 23:48, Étienne Mollier a écrit :

Bruno Volpi, au 2019-04-08 :

1 - menu des applications  dans la bar du tableau de bord. il
semblerai que seulement celles en QT sont affectées ( firefox et
thunderbird fonctionne normalement)

possibilité de changer dans configuration du système ->
apparences des applications ->  Décoration de fenêtres (onglet)
boutons -> add menu

Re,

N'étant un utilisateur quotidien ni de KDE, ni de Gnome, j'ai
peut-être compris le problème de travers.  Mais si je résume,
KDE dispose d'un réglage pour cacher la barre de menu d'une
application (contenant typiquement "Fichier", "Édition",
"Affichage", etc) dans un menu situé dans la barre de titre de
la fenêtre.  Ce composant fonctionne très bien avec les
applications Qt mais pas avec les autres applications, notamment
celles en GTK.

De ce que je comprends, et vu nos petits essais, ce réglage a
l'air spécifique à KDE/Qt, et les applications ne bénéficiant
pas d'une intégration KDE n'en font pas partie ; par exemple le
paquet libreoffice-kde5 fournit cette intégration à LibreOffice,
mais sans ce paquet, ce menu, ainsi que plein d'autres
fonctionnalités, n’apparaît pas.  Il est à noter que GTK a un
onglet dédié dans la fenêtre de configuration de KDE, mais ce
genre réglage n'y apparaît pas (pas encore?)


si quelqu'un à un peu d'expérience je serai très heureux de
savoir s'il existe un bypass, sinon retour sur gnome.

En la matière, mon expérience est limitée.  Mais étant donné les
tours de passe-passe pour intégrer les applications à KDE,
notamment LibreOffice, je crains qu'il n'y ait pas de bypass
simple actuellement pour faire de même avec les programmes en
GTK.  Peut-être dans le futur, si une demande de nouvelle
fonctionnalité est faite auprès des développeurs de KDE ?

Amicalement,

--
logogite*Roseline & Bruno Volpi**
**Chaillac**
**19450 Chamboulive*

F: 05 55 21 27 57
M: 06 78 07 40 27

web : http://www.brunovolpi.com/gites

Re: mais ou est passee la place manquante ?

[ajh-valmer]

On Tuesday 09 April 2019 17:04:34 Stephane Ascoet wrote:
> Le 04/04/2019 à 20:30, ajh-valmer a écrit :
> > M$-Windows : Il faut tout réinstaller (maudit copyright).

> ce n'est pas qu'une question de copyright, mais aussi de 
> conception merdique. Meme un deplacement a l'identique du dossier d'un 
> utilisateur d'un ordinateur a l'autre ne fonctionne pas... alors que 
> c'est une operation presque de routine sous GNU/Linux...

Surtout pour raisons financières :-)
empêcher de réinstaller windows sur un  autre ordinateur.
On a le droit de changer d'ordinateur, de disque dur... et de garder 
windows dont on a une licence.

Merdique aussi, car même si c'était possible,
il faut tout réinstaller windows depuis zéro,
alors que sous Linux, il suffit de réinstaller avec rsync
et on retrouve son linux au complet.

Re: Mettre à jour la libc6

[Yahoo]

Kubernets n'est pas un concurrent de docker, mais un orchestrer de 
conteneur docker, il permet de g�rer, automatiser les conteneurs docker. 
c'est une des solutions les plus aboutit pour la gestion de docker.


Mais sans kubernets, il est possible d'utiliser docker dans des 
environnement de production, et m�me avec des gestions de redondance. De 
plus docker permet de red�ployer des services rapidement sur un syst�me, 
ce qui peut �tre tr�s pratique pour des migrations et mise � jour, il 
suffit de cr�er sont conteneur avec toutes les configurations voulus de 
ses services, et le d�ploiement peut se faire sur n'importe quel OS 
acceptant docker (soit presque tous les OS).



Je ne sais pas si c'est la meilleur m�thode, et aucun cas je ne pourrais 
dire cela, chaque m�thode � son int�r�t.


Pour ma part, n'utilisant pas de solution de virtualisation sur tout les 
serveurs, je ne peux pas me permettre de bloquer un serveur pour une 
seule version de php, ou d'un autre service. Et dans ce cas l�, docker 
est parfait.



un exemple:

pour un client, j'ai due migrer un service d'agenda web utilisant du 
php5, et sur le m�me serveur j'ai due installer une application CRM 
fonctionnement en version php7.3.


Mon syst�me �tant une Debian Stretch avec une version de php7.1, 
compliqu� de faire fonctionner tous le monde ensemble.


J'ai donc cr�er un conteneur PHP5 et un conteneur PHP7.3, un service 
apache2 install� sur le syst�me h�te permet de faire le proxy web.



sch�matiquement :


Debian Stretch -> apache2 mode proxy

����������������������� -> 
docker-ce� -> un conteneur avec php5

-> un conteneur avec php7.3


ensuite docker utilise une gestion de volume, qui permet de g�rer les 
r�pertoires web des conteneurs � partir du syst�me h�te.



En esp�rant t'avoir donn� envie d'essayer docker :)




Le 09/04/2019 � 16:01, Nicolas Malgat a �crit�:
Merci pour l'id�e de Docker, je ne sais pas l'utiliser mais semble 
tr�s int�ressant.

J'ai vu que l'un de ses concurrents est *kubernets*.
Quel serait le plus int�ressant professionnellement ?

J'ai suivi une vid�o de pr�sentation du logiciel Docker:
https://www.youtube.com/watch?v=XgKOC6X8W28
L'utilisation d'un conteneur est de d�velopper un projet d'application 
web par exemple.
Cela implique la cr�ation d'un dossier commun entre le conteneur et 
l'ordinateur physique comme pr�sent� dans la vid�o.

Est-ce la meilleure m�thode ? / Quelle m�thode utilisez-vous ?

Pour mon soucis avec la libc6, j'ai choisi de compiler php7.3 finalement.


*De :* Yahoo 
*Envoy� :* mardi 9 avril 2019 09:58
*� :* debian-user-french@lists.debian.org
*Objet :* Re: Mettre � jour la libc6

Bonjour,

a mon avis il est toujours un peu compliqu� de mettre � jour libc, 
car le noyau Linux, et tous les programmes l'utilisent. Cela implique 
de mettre alors � jour une grande partie de son syst�me, juste 
pour php7.3.



Une solution est de passer par Docker, pour faire tourner php7.3 dans 
un conteneur qui utilisera les bonnes libraires C.


Pour ma part je fonctionne comme cela pour faire fonctionner sur une 
m�me machine php5, php7.1 et php7.3.



En esp�rant aider.

Lo�c.




Le 08/04/2019 � 18:09, Nicolas Malgat a �crit�:

Bonjour,

Je voulais installer php7.3 seulement cela requiert l'installation de 
la libc6 2.29 tandis que j'utilise actuellement la libc6 2.24


spikespiegel@debian:~/Documents/projets/demo$ sudo apt upgrade libc6
Lecture des listes de paquets... Fait
Construction de l'arbre des d�pendances � � �
Lecture des informations d'�tat... Fait
libc6 is already the newest version (2.24-11+deb9u4).
Calcul de la mise � jour... Fait
0 mis � jour, 0 nouvellement install�s, 0 � enlever et 0 non 
mis � jour.



J'ai lu certains �changes sur les forums mais les explications sont 
un peu confuses pour moi.

Quelle serait la m�thode � suivre ?

j'ai �galement cru comprendre que la version de Debian avait une 
importance.

Je fourni donc ma version:

spikespiegel@debian:~/Documents/projets/demo$ �lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description: Debian GNU/Linux 9.8 (stretch)
Release: 9.8
Codename: stretch

Re: build vs install

[Michael Howard]

On 09/04/2019 16:35, Lee wrote:

What are the downsides to getting the source code and doing the
build/install myself vs. using a pre-built package other than I'm
responsible for noticing the software needs to be updated?

The latest example is ttcp
   http://nuttcp.net/nuttcp/latest/  has 8.1.4
   https://tracker.debian.org/pkg/nuttcp  has 6.1.2-4

If I ever decide to go with the debian package I just uninstall the
software I built and .. anything else that needs to be done before
installing an official package?

TIA
Lee

You could build a debian package from the source and let apt take care 
of install/updates/uninstall. Admittedly, it could be a bit of work 
initially with versions so far apart as above but it is an option.


--
Mike Howard

Re: build vs install

[Brad Rogers]

On Tue, 9 Apr 2019 11:35:05 -0400
Lee  wrote:

Hello Lee,

>What are the downsides to getting the source code and doing the
>build/install myself vs. using a pre-built package other than I'm
>responsible for noticing the software needs to be updated?

Mostly, it's time.  Also, you're responsible for installing any required
new dependencies, be they development packages or binaries.  Which,
again, comes down to time.  Such package requirements aren't always
immediately obvious, and may not even be mentioned in release notes.

For this reason, there's only one piece of software I compile and install
'by hand' as it were.

>The latest example is ttcp
>  http://nuttcp.net/nuttcp/latest/  has 8.1.4
>  https://tracker.debian.org/pkg/nuttcp  has 6.1.2-4

I was going to suggest adding stable-backports to your sources.list.
Assuming, of course, that you're running stable.  A quick check of
Debian packages shows that wouldn't help in this case.  A fact, no doubt
you're aware of, hence your query.

>If I ever decide to go with the debian package I just uninstall the
>software I built and .. anything else that needs to be done before

Yes, made easier if the install mechanism creates a useful log and you
can later 'make --uninstall'.

>installing an official package?

I'd back up any config file(s) just in case anything 'blows up'.  That
is to say, if there's any incompatibility between Debian's way of writing
the config files, and the package creator's way of doing so.  The Debian
package installer may write the configs somewhere other than where the
creator does, for example.

-- 
 Regards  _
 / )   "The blindingly obvious is
/ _)radnever immediately apparent"
I guess I shouldn't have strangled her to death
Ugly - The Stranglers


pgphBEaULnkry.pgp
Description: OpenPGP digital signature

Re: OpenSSH not closing idle sessions.

[Thomas Pircher]

Greg Wooledge wrote:
>
> I suggest reading what ClientAliveCountMax and ClientAliveInterval
> actually do in sshd_config(5).  Take particular note of the word
> "unresponsive".  It is not the same as "idle".

Yes, you are right, this setting won't disconnect idle sessions. So I
guess it's mostly useful to clean up server resources.

Thomas

Re: Tracking the next Stable release

[Francisco M Neto]

On Tue, 2019-04-09 at 00:02 +0200, Peter Wiersig wrote:
> It's done when it's done and https://lists.debian.org/debian-announce/
> get's notified.  You'd probably want to subscribe there.

I am subscribed there. Being notified of the release is not the point.

Being aware of the progress towards it is.

-- 
[]'s,

Francisco M Neto

GPG: 4096R/D692FBF0


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

Re: Tracking the next Stable release

[Francisco M Neto]

On Tue, 2019-04-09 at 23:47 +0900, Mark Fletcher wrote:
> https://twitter.com/debian_tracker
> > 
> Nice! What does the level of release-critical bugs need to fall to 
> before a release can happen -- it's not zero is it?

There is no defined number, I think. If I'm not mistaken it happens when
the Release Team decides the bugs in that list - if any - are not relevant
enough to extend the freeze any longer.

-- 
[]'s,

Francisco M Neto

GPG: 4096R/D692FBF0


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

build vs install

[Lee]

What are the downsides to getting the source code and doing the
build/install myself vs. using a pre-built package other than I'm
responsible for noticing the software needs to be updated?

The latest example is ttcp
  http://nuttcp.net/nuttcp/latest/  has 8.1.4
  https://tracker.debian.org/pkg/nuttcp  has 6.1.2-4

If I ever decide to go with the debian package I just uninstall the
software I built and .. anything else that needs to be done before
installing an official package?

TIA
Lee

Re: OpenSSH not closing idle sessions.

[mick crane]

On 2019-04-08 18:25, timothylegg wrote:



Ideas?
I've not really used screen but isn't it that you want to start where 
you left off ?


mick


--
Key ID4BFEBB31

Re: ntfs-3g - mounting as normal user?

[Reco]

Hi.

On Tue, Apr 09, 2019 at 04:13:15PM +0200, Hans wrote:
> Hi folks,
> 
> I am not sure, but when I want to mount a ntfs-filesystem using ntfs-3g, I 
> always need to input the password of root.
> 
> According to the manual, if ntfs-3g is setuid-root, it shall allow normal 
> users to mount ntfs-files.

It does for me:

$ dd if=/dev/zero of=1.raw bs=1024 count=0 seek=10240
$ /sbin/mkfs.ntfs -F 1.raw
...
Creating NTFS volume structures.
mkntfs completed successfully. Have a nice day.
$ ntfs-3g 1.raw .fuse
$ mount
...
...1.raw on .../.fuse type fuse (rw,nosuid,nodev,relatime...)


> So, I have rwsr-xr-x with owner root:root, which IMO should do the trick, but 
> still asking for the password.

Your user does have read-write access to the underlying block device,
its not it?


> However, I remember dark, there was something with "hardcoded in ntfs-3g", 
> which does this no more let work. This was years ago and I only remember weak.

That only applies if:

a) ntfs-3g is suid root-owned binary.
b) ntfs-3g uses 'external' libfuse.so (i.e. dynamic link to a library).

But they build ntfs-3g with '--with-fuse=internal' flag, so last one
should not apply.

Reco

Re: mais ou est passee la place manquante ?

[Stephane Ascoet]

Le 04/04/2019 à 20:30, ajh-valmer a écrit :

Il faut tout réinstaller (maudit copyright).


Bonjour, ce n'est pas qu'une question de copyright, mais aussi de 
conception merdique. Meme un deplacement a l'identique du dossier d'un 
utilisateur d'un ordinateur a l'autre ne fonctionne pas... alors que 
c'est une operation presque de routine sous GNU/Linux...


Par contre, dans la technique "transfert de Debian", la partie Grub peut 
etre galere, surtout avec UEFI :-(


Hamster:

D'un autre coté
utiliser la sortie de df risque de faire agir sur / comme tu l'a si bien
dit.


Non, si /home n'est pas un point de montage, il n'a pas de raison 
d'apparaitre en sortie de df.



Je note le lien. J'ai déjà fait des recherches de tutos pour faire des
scripts shell mais comme souvent je me suis perdu dans la profusion.


Je trouve les documentations en ligne bien moins pratiques que les 
livres pour ma part :-( Et, par rapport a une autre remarque, j'ai du 
mal a me passer des bashismes. C'est deja tellement rustique, alors si 
en plus on doit se limiter a la syntaxe sh, ne pas utiliser d'outils GNU 
ajoutes apres 1992... au secours.



Mon doigt a glissé et comme le h est juste a coté du f c'est lui qui a pris.


Bepoiste? ;-)

Pascal:

Non, tu as bien compris. Par défaut un système de fichiers ext* réserve 5% de 
l'espace à son utilisateur créateur, c'est-à-dire root la plupart du temps. 
Concrètement, cela signifie que quand l'espace utilisé atteint 95%, les autres 
utilisateurs ne peuvent plus allouer d'espace.


Je n'utilise pas Ext, mais du coup je me demande, pour les adeptes de 
cette famille de FS, l'affichage de df change t-il selon qui est 
connecte? Ou alors on peut continuer a ecrire en root alors que la place 
disponible indiquee est nulle?



Yann Serre:

Sans volonté particulière de raviver ce sujet, en ce moment je lis et j'entends 
9 GAFA pour 1 GAFAM (sans doute moins encore).


Je confirme :-(
--
Cordialement, Stephane Ascoet

Re: OpenSSH not closing idle sessions.

[Greg Wooledge]

On Tue, Apr 09, 2019 at 04:01:20PM +0100, Thomas Pircher wrote:
> > > ClientAliveInterval 5
> 
> This is the setting that the STIG ID RHEL-07-040320 in [2] suggests to
> edit.
> 
> Thomas
> 
> [1] https://iase.disa.mil/stigs
> [2] 
> https://rhel7stig.readthedocs.io/en/latest/medium.html#v-72237-all-network-connections-associated-with-ssh-traffic-must-terminate-at-the-end-of-the-session-or-after-10-minutes-of-inactivity-except-to-fulfill-documented-and-validated-mission-requirements-rhel-07-040320

Pathetic.

I suggest reading what ClientAliveCountMax and ClientAliveInterval
actually do in sshd_config(5).  Take particular note of the word
"unresponsive".  It is not the same as "idle".

Re: OpenSSH not closing idle sessions.

[Thomas Pircher]

Greg Wooledge wrote:
> Most people want the exact opposite of that.

I don't really know the OP's rationale, but terminating an idle ssh
session is a step in the requirements/guidelines (STIG [1]) for
hardening systems for the US Department of Defense.

> Basically, what you're asking for is directly hostile to any kind of
> sane operation of a computer.

I'm not going to defend this requirement, merely showing one example
where one would want (or would have to) configure the ssh server this
way.

> > ClientAliveInterval 5

This is the setting that the STIG ID RHEL-07-040320 in [2] suggests to
edit.

Thomas

[1] https://iase.disa.mil/stigs
[2] 
https://rhel7stig.readthedocs.io/en/latest/medium.html#v-72237-all-network-connections-associated-with-ssh-traffic-must-terminate-at-the-end-of-the-session-or-after-10-minutes-of-inactivity-except-to-fulfill-documented-and-validated-mission-requirements-rhel-07-040320

Re: Tracking the next Stable release

[Mark Fletcher]

On Mon, Apr 08, 2019 at 05:42:46PM -0300, Francisco M Neto wrote:
> Greetings!
> 
> 
> https://twitter.com/debian_tracker
> 
Nice! What does the level of release-critical bugs need to fall to 
before a release can happen -- it's not zero is it?

Mark

Re: OpenSSH not closing idle sessions.

[David Wright]

On Mon 08 Apr 2019 at 13:39:36 (-0400), Greg Wooledge wrote:
> On Mon, Apr 08, 2019 at 12:25:28PM -0500, timothylegg wrote:
> > I need to have the session expire and the ssh client terminate after
> > an idle time.
> 
> Most people want the exact opposite of that.
> 
> Basically, what you're asking for is directly hostile to any kind of
> sane operation of a computer.

The first thought that crossed my mind when I read the OP was a
teletype clattering out the message "FIVE MINUTE WARNING".
Expiration was, of course, standard practice when interactive sessions
were a scarce resource that required reclaiming after, say, ten
minutes of inactivity. How times change.

Of course we all hated it. One of the advantages of the old KSR33s
was this audible wake-up call when debugging was taking a little
longer than you expected. After they were replaced by "glass"
teletypes, you'd turn your attention back to the screen only to
find ten minutes had passed and you were logged off. With something
like four times as many terminals as sessions, it could take a while
to get back on at busy times of the day.

Cheers,
David.

ntfs-3g - mounting as normal user?

[Hans]

Hi folks,

I am not sure, but when I want to mount a ntfs-filesystem using ntfs-3g, I 
always need to input the password of root.

According to the manual, if ntfs-3g is setuid-root, it shall allow normal 
users to mount ntfs-files.

So, I have rwsr-xr-x with owner root:root, which IMO should do the trick, but 
still asking for the password.

However, I remember dark, there was something with "hardcoded in ntfs-3g", 
which does this no more let work. This was years ago and I only remember weak.

Thus led me to the conclusion, that either I am doing something wrong, or the 
manual is wrong (and of course should be corrected then).

Does someone know better or should I file a bugreport?

Best regards

Hans

 

 



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

Re: Qual plataforma de inscrição de eventos vocês usam?

[Jeann Wilson]

Aqui em Brasília nós usamos o Doity para inscrição e controle dos
participantes.

https://doity.com.br/


site: http://www.melhordetudo.com
blog: http://jeann.melhordetudo.com

Linux User: #471295

" A imaginação é mais importante que o conhecimento " ( Albert Einstein )

___

Antes de imprimir pense em sua responsabilidade e compromisso com o Meio
Ambiente


On Tue, Apr 9, 2019 at 11:02 AM Paulo Henrique de Lima Santana <
p...@debian.org> wrote:

> Olá,
>
> On 4/9/19 10:57 AM, Adriano Henrique wrote:
> > Prezados,
> >
> > Estou colaborando com a organização do FLISoL na minha cidade e gostaria
> > de saber qual plataforma de inscrição de eventos vocês utilizam?
>
> Talvez seja melhor você perguntar na lista de discussão do FLISOL.
>
> > Sei que a Even3 não é vista com bons olhos.
>
> Não conheço, mas ela é software livre?
>
> Abs
>
> --
> Paulo Henrique de Lima Santana (phls)
> Curitiba - Brasil
> Debian Developer
> Diretor do Instituto para Conservação de Tecnologias Livres
> Membro da Comunidade Curitiba Livre
> Site: http://www.phls.com.br
> GNU/Linux user: 228719  GPG ID: 0443C450
>
> Organizador da DebConf19 - Conferência Mundial de Desenvolvedores(as)
> Debian
> Curitiba - 21 a 28 de julho de 2019
> http://debconf19.debconf.org
>
>

Re: Qual plataforma de inscrição de eventos vocês usam?

[Daniel Lenharo de Souza]

Aqui em Curitiba,

Estamos utilizando o LimeSurvey.

Fizemos um form simples mesmo.


Em 09-04-2019 10:57, Adriano Henrique escreveu:

Prezados,

Estou colaborando com a organização do FLISoL na minha cidade e
gostaria de saber qual plataforma de inscrição de eventos vocês
utilizam?
Sei que a Even3 não é vista com bons olhos.

Atenciosamente,
--

ADRIANO HENRIQUE M. FRANÇA



[]'s

--
Daniel Lenharo de Souza
DebConf19 - Curitiba-BR
http://www.lenharo.eti.br
GPG: 31D8 0509 460E FB31 DF4B
 9629 FB0E 132D DB0A A5B1

Re: Qual plataforma de inscrição de eventos vocês usam?

[Paulo Henrique de Lima Santana]

Olá,

On 4/9/19 10:57 AM, Adriano Henrique wrote:
> Prezados,
> 
> Estou colaborando com a organização do FLISoL na minha cidade e gostaria
> de saber qual plataforma de inscrição de eventos vocês utilizam? 

Talvez seja melhor você perguntar na lista de discussão do FLISOL.

> Sei que a Even3 não é vista com bons olhos. 

Não conheço, mas ela é software livre?

Abs

-- 
Paulo Henrique de Lima Santana (phls)
Curitiba - Brasil
Debian Developer
Diretor do Instituto para Conservação de Tecnologias Livres
Membro da Comunidade Curitiba Livre
Site: http://www.phls.com.br
GNU/Linux user: 228719  GPG ID: 0443C450

Organizador da DebConf19 - Conferência Mundial de Desenvolvedores(as) Debian
Curitiba - 21 a 28 de julho de 2019
http://debconf19.debconf.org



signature.asc
Description: OpenPGP digital signature

Qual plataforma de inscrição de eventos vocês usam?

[Adriano Henrique]

Prezados,

Estou colaborando com a organização do FLISoL na minha cidade e gostaria de
saber qual plataforma de inscrição de eventos vocês utilizam?
Sei que a Even3 não é vista com bons olhos.

Atenciosamente,
-- 
*Adriano Henrique M. França*

Re: date(1) in stretch and buster

[Vincent Lefevre]

On 2019-04-08 18:26:23 +0300, Reco wrote:
> stretch$ TZ=UTC date
> Mon Apr  8 15:22:02 UTC 2019
> buster$ TZ=UTC date
> Mon 08 Apr 2019 03:22:04 PM UTC

This is unrelated to your issue, but note that the correct TZ string
for UTC is "UTC0", not "UTC". See

  http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html

-- 
Vincent Lefèvre  - Web: 
100% accessible validated (X)HTML - Blog: 
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

Re: installation Debian 9 et Grub [Résolu]

[sTriX]

le lundi 08 avril 2019 à 20:14 (+0200), Pascal Hambourg a écrit:
> Ce n'est pas GRUB mais l'initramfs, donc Linux.
> La procédure ci-dessous est donc applicable.
> 
> > > Le menu de GRUB s'affiche-t-il ? Si oui, appuie sur "e", remplace sdb2 par
> > > sda2 (QWERTY : touche Q pour taper A), appuie sur F10 pour démarrer et une
> > > fois dans le système, exécute update-grub en root pour corriger
> > > /boot/grub/grub.cfg.
> 
> Précision : la mention root=/dev/sdb2 à modifier en root=/dev/sda2 se trouve
> dans une ligne qui commence par "linux".
> 
La procédure de modification fonctionne bien, merci.

Re: Borg [was Re: Sauvegarde "classique" vs synchronisation ?]

[didier gaumet]

Le 09/04/2019 à 09:58, David BERCOT a écrit :
[...]
> Maintenant, j'ai beaucoup d'erreurs :
> Local Exception
> OSError: [Errno 28] No space left on device
[...]

d'après
https://stackoverflow.com/questions/6998083/python-causing-ioerror-errno-28-no-space-left-on-device-results-32766-h
ça a l'air d'une erreur Python qui n'a rien de spécifique à Borgbackup
et pour laquelle quelques pistes de résolution sont évoquées (tu en
trouveras sûrement d'autres en lançant une recherche du message d'erreur
sur internet)

Re: USB Fax modem yardım

[Umit Bozkir]

Selamlar,


On 4/9/19 6:10 AM, Recai Uyan wrote:

Merhaba, Conexant Systems (Rockwell), Inc. SoftK56 Data Fax Voice CARP
usb fax modem i günlerdir ubuntu 14.04 ve raspbian jessie ye
tanıtamıyorum. Bu konuda daha önce çalışmış olan yardımcı olabilecek
bir arkadaşa ihtiyacım var. Lütfen biliyorsanız yardımınızı rica
ediyorum.



Burada soyle bir sayfa [1] var. Yardimci olabilir (Sayfada 14.04 icin 
bir aciklama goremedim).



Iyi calismalar


1) https://help.ubuntu.com/community/DialupModemHowto/Conexant

Re: putty go slow

[Dan Ritter]

mick crane wrote: 
> On 2019-04-09 07:46, Peter Wiersig wrote:
> > mick crane  writes:
> > > 
> > > the PCs are physically adjacent connected with the RJ45 ( isn't it )
> > > cables through what is supposed to be a switch I got in B several
> > > years ago.
> > 
> i
> how do I tell if it's a switch or a hub ?
> I thought a switch sends data over the required port but a hub sends over
> all the ports ?

That is correct.


> It's got "8 port switch" printed on it but if there is network activity all
> the lights seem to flash.
> 
> /sbin/ifconfig
> enp0s25: flags=4163  mtu 1500
> inet 10.0.0.3  netmask 255.255.255.0  broadcast 10.0.0.255
> inet6 fe80::219:d1ff:fe41:c769  prefixlen 64  scopeid 0x20
> ether 00:19:d1:41:c7:69  txqueuelen 1000  (Ethernet)
> RX packets 47732498  bytes 13998322190 (13.0 GiB)
> RX errors 0  dropped 5642  overruns 0  frame 0
> TX packets 73403188  bytes 102853469866 (95.7 GiB)
> TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
> device interrupt 21  memory 0xdffe-e000

You see the RX dropped 5642? That's bad.

There's a chance that it's a bad ethernet cable; otherwise the
hub/switch is faulty. Given that it's labelled "switch" but all
the ports flash on every packet, it might well be that.

Unmanaged gigabit switches are cheap; I recommend that you
replace it if swapping cables doesn't solve the issue.

-dsr-

Event invitation: How do we collaborate to build a human-centric next generation internet? - Saturday 11th of May - Time: 14.00-17.00 - Stockholm

[Mattias Axell]

Hi Debian Sweden!

You are invited to help shape the future of the Internet!

How do we collaborate to build a human-centric next generation Internet? 
You are invited to an alliance working towards this goal. Edgeryders are 
a member of the NGI consortium, a key project informing future EU policy 
and funding for internet technologies.


Location: House Blivande, Södra Hamnvägen 9, Stockholm
Date: Saturday 11th of May
Time: 14.00-17.00

To register for the event: Post a reply to the invitation on our 
platform. https://edgeryders.eu/t/next-generation-internet-on-the-edge/9705


Edgeryders is a company living in symbiosis with an online community of 
thousands of hackers, activists, radical thinkers and doers, and just 
normal people that want to make a difference. We believe that a smart 
community outperforms any of its members; this is the result of people 
working together, improving on each other’s work.


Best regards,
Mattias Axell and Hugi Ásgeirsson
Edgeryders.eu
https://edgeryders.eu/c/ngi

Re: putty go slow

[mick crane]

On 2019-04-09 07:46, Peter Wiersig wrote:

mick crane  writes:


the PCs are physically adjacent connected with the RJ45 ( isn't it )
cables through what is supposed to be a switch I got in B several
years ago.


Almost, RJ-45 is the specification for the plug and jacks, what you're
having here is ethernet wiring in twisted pairs between client and
server connected by a hub.  Or switch, please clarify that as that
changes a lot.

Is anything else connected to this hub?  If your problems occur, is
anything else using the hub concurrently?  Can you reduce the
connections only to server and client and maybe a internet uplink?
Network printers can do unimaginably bad things in regard to hubs, and
even switches.

Can you change the hub to a real switch?


but I think it might be a hub.
Perhaps that is the culprit ?


Perhaps. Check the "ifconfig" output on the Linux side, maybe reset the
server, connect from windows and if you're having problems check the
dmesg output regarding the interface.


i
how do I tell if it's a switch or a hub ?
I thought a switch sends data over the required port but a hub sends 
over all the ports ?
It's got "8 port switch" printed on it but if there is network activity 
all the lights seem to flash.


/sbin/ifconfig
enp0s25: flags=4163  mtu 1500
inet 10.0.0.3  netmask 255.255.255.0  broadcast 10.0.0.255
inet6 fe80::219:d1ff:fe41:c769  prefixlen 64  scopeid 0x20
ether 00:19:d1:41:c7:69  txqueuelen 1000  (Ethernet)
RX packets 47732498  bytes 13998322190 (13.0 GiB)
RX errors 0  dropped 5642  overruns 0  frame 0
TX packets 73403188  bytes 102853469866 (95.7 GiB)
TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
device interrupt 21  memory 0xdffe-e000

I've got a Buster PC with apache2, cups, dovecot, roundcube as mail and 
print server.
a Buster PC I mess about on, a win 10 PC and a printer all connected to 
the "switch"
the gateway is a PC with pfsense on it which is connected to "switch" 
and its other network card connected to ISP router thing.


Maybe the sluggishness I sometimes observed is Windows updating itself ?

mick



For reference, here's my output:

eth0: flags=4163  mtu 1500
inet 217.172.177.159  netmask 255.255.255.0  broadcast 
217.172.177.255

ether 00:19:66:f1:43:9e  txqueuelen 1000  (Ethernet)
RX packets 81994186  bytes 14753508445 (13.7 GiB)
RX errors 14  dropped 0  overruns 14  frame 0
TX packets 107524155  bytes 14836289080 (13.8 GiB)
TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0


Take note of the 2nd line of RX and TX status lines, there should be no
counters there, in my case 14 overruns in regard to 819 million packets
is a very low error rate.  I suspect that if there's a network problem
it would manifest in some higher relative values on your side.

If in doubt verify that both sides are set to auto-negotiate and 
replace

both wires from the machines to the hub with new cables.

Peter


--
Key ID4BFEBB31

USB Fax modem yardım

[Recai Uyan]

Merhaba, Conexant Systems (Rockwell), Inc. SoftK56 Data Fax Voice CARP
usb fax modem i günlerdir ubuntu 14.04 ve raspbian jessie ye
tanıtamıyorum. Bu konuda daha önce çalışmış olan yardımcı olabilecek
bir arkadaşa ihtiyacım var. Lütfen biliyorsanız yardımınızı rica
ediyorum.


Recai Uyan

Re: Tracking the next Stable release

[Curt]

On 2019-04-08, Peter Wiersig  wrote:
> Francisco M Neto  writes:
>>  Sometimes people ask me when is Debian going to release its next Stable;
>> that is not an easy answer, since it is not time-based but rather based on 
>> the
>> number of Release-Critical bugs. 
>
> It's done when it's done and https://lists.debian.org/debian-announce/
> get's notified.  You'd probably want to subscribe there.

Left field, meet Pete.

> Peter
>
>

Re: OpenSSH not closing idle sessions.

[Dan Ritter]

Richard Hector wrote: 
> On 9/04/19 12:14 PM, timothylegg wrote:
> > I have two residences and one
> > has a port forwarding issue.  I want to make an SSH tunnel to the
> > other site.  If I am at one place for multiple weeks, it's asking too
> > much for the SSH tunnel to stay live that long (I've seen many
> > complaints of SSH connections dropping).  I want to put it on a
> > periodic cron job, but if I don't answer into the tunnel within 30
> > minutes, it becomes idle and drops off.  Every hour a new tunnel is
> > initiated.  The old session must either be closed or expire else I'll
> > have multiple SSH sessions live simultaneously.
> 
> Have you considered using a VPN instead of SSH tunnelling?
> 
> Eg I have OpenVPN set up to my parents' place, so that I can do
> (automatic) backups of their system.
> 
> I did run into issues with address space conflicts (I can't renumber
> either end), so in the end I did it all with IPv6, but you might not
> need that.

If you love SSH: autossh.

If you want the easiest VPN config: wireguard (currently available
only in unstable, but known to work in stable)

-dsr-

Nowoczesny FanPage na Facebook'u

[Krzysztof Kasica | Twój Facebook]

Dzień dobry,



Kontaktuję się z Państwem, ponieważ jestem świadomy, jak ważny jest obecnie 
profesjonalnie prowadzony *Fanpage* na *Facebook’u.
*


W związku z powyższym, będzie mi bardzo miło, jeśli wyrazicie Państwo zgodę na 
przesłanie niezobowiązującej propozycji na ten temat.



Aby wyrazić zgodę, proszę o przesłanie słowa *TAK *w odpowiedzi na tego e-maila.



Pozdrawiam Serdecznie,
Krzysztof Kasica
Dział Marketingu

Re: Tracking the next Stable release

[Jonathan Dowland]

On Mon, Apr 08, 2019 at 05:42:46PM -0300, Francisco M Neto wrote:

https://twitter.com/debian_tracker


Very nice, thanks for sharing!


--

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Jonathan Dowland
⢿⡄⠘⠷⠚⠋⠀ https://jmtd.net
⠈⠳⣄ Please do not CC me, I am subscribed to the list.

Re: Mettre à jour la libc6

[Yahoo]

Bonjour,

a mon avis il est toujours un peu compliqu� de mettre � jour libc, car 
le noyau Linux, et tous les programmes l'utilisent. Cela implique de 
mettre alors � jour une grande partie de son syst�me, juste pour php7.3.



Une solution est de passer par Docker, pour faire tourner php7.3 dans un 
conteneur qui utilisera les bonnes libraires C.


Pour ma part je fonctionne comme cela pour faire fonctionner sur une 
m�me machine php5, php7.1 et php7.3.



En esp�rant aider.

Lo�c.




Le 08/04/2019 � 18:09, Nicolas Malgat a �crit�:

Bonjour,

Je voulais installer php7.3 seulement cela requiert l'installation de 
la libc6 2.29 tandis que j'utilise actuellement la libc6 2.24


spikespiegel@debian:~/Documents/projets/demo$ sudo apt upgrade libc6
Lecture des listes de paquets... Fait
Construction de l'arbre des d�pendances
Lecture des informations d'�tat... Fait
libc6 is already the newest version (2.24-11+deb9u4).
Calcul de la mise � jour... Fait
0 mis � jour, 0 nouvellement install�s, 0 � enlever et 0 non mis � jour.


J'ai lu certains �changes sur les forums mais les explications sont un 
peu confuses pour moi.

Quelle serait la m�thode � suivre ?

j'ai �galement cru comprendre que la version de Debian avait une 
importance.

Je fourni donc ma version:

spikespiegel@debian:~/Documents/projets/demo$ �lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description: Debian GNU/Linux 9.8 (stretch)
Release: 9.8
Codename: stretch

Borg [was Re: Sauvegarde "classique" vs synchronisation ?]

[David BERCOT]

Bonjour Bernard,

J'ai regardé la doc et, en effet, l'outil semble intéressant et correspond bien 
à mon besoin (sauvegarde distante, via SSH, avec déduplication, etc.)

Maintenant, j'ai beaucoup d'erreurs :
Local Exception
OSError: [Errno 28] No space left on device

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/borg/archiver.py", line 4455, in main
exit_code = archiver.run(args)
  File "/usr/lib/python3/dist-packages/borg/archiver.py", line 4387, in run
return set_ec(func(args))
  File "/usr/lib/python3/dist-packages/borg/archiver.py", line 145, in wrapper
assert_secure(repository, kwargs['manifest'], self.lock_wait)
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 206, in 
assert_secure
sm.assert_secure(manifest, manifest.key, lock_wait=lock_wait)
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 171, in 
assert_secure
with cache_config:
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 238, in __enter__
self.open()
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 261, in open
kill_stale_locks=hostname_is_unique()).acquire()
  File "/usr/lib/python3/dist-packages/borg/locking.py", line 351, in acquire
self._roster.modify(EXCLUSIVE, ADD)
  File "/usr/lib/python3/dist-packages/borg/locking.py", line 286, in modify
self.save(roster)
  File "/usr/lib/python3/dist-packages/borg/locking.py", line 258, in save
json.dump(data, f)
OSError: [Errno 28] No space left on device

Platform: Linux DBE 4.19.0-4-amd64 #1 SMP Debian 4.19.28-2 (2019-03-15) x86_64
Linux: debian buster/sid 
Borg: 1.1.9  Python: CPython 3.7.3
PID: 30522  CWD: /home/david.bercot
sys.argv: ['/usr/bin/borg', 'create', '--list', '--one-file-system', 
'--compression', 'auto,zlib,5', '--stats', 
'serveur:/data/borgbackup::home_david_{now:%Y-%m-%d_%H:%M}', 
'/home/david.bercot/', '--exclude', '/home/david.bercot/rep']
SSH_ORIGINAL_COMMAND: None

Killed stale lock DBE@110352858462797.30522-0.
Local Exception
OSError: [Errno 28] No space left on device

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/borg/archiver.py", line 4455, in main
exit_code = archiver.run(args)
  File "/usr/lib/python3/dist-packages/borg/archiver.py", line 4387, in run
return set_ec(func(args))
  File "/usr/lib/python3/dist-packages/borg/archiver.py", line 145, in wrapper
assert_secure(repository, kwargs['manifest'], self.lock_wait)
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 206, in 
assert_secure
sm.assert_secure(manifest, manifest.key, lock_wait=lock_wait)
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 171, in 
assert_secure
with cache_config:
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 238, in __enter__
self.open()
  File "/usr/lib/python3/dist-packages/borg/cache.py", line 261, in open
kill_stale_locks=hostname_is_unique()).acquire()
  File "/usr/lib/python3/dist-packages/borg/locking.py", line 351, in acquire
self._roster.modify(EXCLUSIVE, ADD)
  File "/usr/lib/python3/dist-packages/borg/locking.py", line 286, in modify
self.save(roster)
  File "/usr/lib/python3/dist-packages/borg/locking.py", line 258, in save
json.dump(data, f)
OSError: [Errno 28] No space left on device

Platform: Linux DBE 4.19.0-4-amd64 #1 SMP Debian 4.19.28-2 (2019-03-15) x86_64
Linux: debian buster/sid 
Borg: 1.1.9  Python: CPython 3.7.3
PID: 30586  CWD: /home/david.bercot
sys.argv: ['/usr/bin/borg', 'prune', '--list', '--keep-within=14d', 
'--keep-weekly=4', '--keep-monthly=-1', 'serveur:/data/borgbackup']
SSH_ORIGINAL_COMMAND: None

Je ne sais pas si tu as déjà rencontré ça et si tu as une idée ?
Les messages "no space left" me semblent très bizarres (pas du tout de souci 
d'espace disque)...

Merci d'avance.

David.

P.S. : sinon, je tenterai backuppc...

Le 09/03/2019 à 18:51, Bernard Schoenacker a écrit :
> 
> 
> - Mail original -
>> De: "David BERCOT" 
>> À: debian-user-french@lists.debian.org
>> Envoyé: Samedi 9 Mars 2019 18:42:04
>> Objet: Sauvegarde "classique" vs synchronisation ?
>>
>> Bonjour,
>>
>> Afin d'avoir une copie à jour (ou du moins, le plus à jour possible)
>> de
>> mon ordinateur (ce qui m'intéresse est uniquement mon /home),
>> j'hésite
>> entre des sauvegardes classiques et une synchronisation type "cloud".
>>
>> De mon point de vue, la sauvegarde a l'avantage d'être "consistante"
>> et
>> d'intégrer potentiellement des versions différentes de mes documents.
>> En revanche, si le crash ou la perte ou le vol de la machine se
>> produit
>> "relativement longtemps" après la dernière sauvegarde, les données ne
>> sont pas vraiment fraîches.
>>
>> La synchronisation (partons sur un serveur personnel de type
>> NextCloud
>> pour illustrer) permet justement de répondre à ce besoin de fraîcheur
>> mais peut poser 

Re: How to change GDM3 login screen background

[Curt]

On 2019-04-08, Andrew Clark  wrote:
>
> Thanks curt, that works for me.
>
> So the commentary in the comments in the source file supplied by the gdm3
> package don't work, along with the instructions in the wiki page pointing
> to them, should I log a bug?
>

I don't know much about bug filing. I've heard of documentation bugs,
but never of an inherent source file commentary documentation bug?
Other clearer, more informed heads might chime in here.

The wiki, of course, being what it is (a wiki) is theoretically
modifiable by anyone (which means, if you have the time and courage,
you, I guess).

Re: putty go slow

[Peter Wiersig]

mick crane  writes:
>
> the PCs are physically adjacent connected with the RJ45 ( isn't it )
> cables through what is supposed to be a switch I got in B several 
> years ago.

Almost, RJ-45 is the specification for the plug and jacks, what you're
having here is ethernet wiring in twisted pairs between client and
server connected by a hub.  Or switch, please clarify that as that
changes a lot.

Is anything else connected to this hub?  If your problems occur, is
anything else using the hub concurrently?  Can you reduce the
connections only to server and client and maybe a internet uplink?
Network printers can do unimaginably bad things in regard to hubs, and
even switches.

Can you change the hub to a real switch?

> but I think it might be a hub.
> Perhaps that is the culprit ?

Perhaps. Check the "ifconfig" output on the Linux side, maybe reset the
server, connect from windows and if you're having problems check the
dmesg output regarding the interface.

For reference, here's my output:

eth0: flags=4163  mtu 1500
inet 217.172.177.159  netmask 255.255.255.0  broadcast 217.172.177.255
ether 00:19:66:f1:43:9e  txqueuelen 1000  (Ethernet)
RX packets 81994186  bytes 14753508445 (13.7 GiB)
RX errors 14  dropped 0  overruns 14  frame 0
TX packets 107524155  bytes 14836289080 (13.8 GiB)
TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0


Take note of the 2nd line of RX and TX status lines, there should be no
counters there, in my case 14 overruns in regard to 819 million packets
is a very low error rate.  I suspect that if there's a network problem
it would manifest in some higher relative values on your side.

If in doubt verify that both sides are set to auto-negotiate and replace
both wires from the machines to the hub with new cables.

Peter