Re: perte de la touche morte du clavier pour saisir les accents circonflexes depuis une mise à jour système

[Patrick ZAJDA]

Le 08/05/2024 à 23:33, Gaëtan Perrier a écrit :

Le mercredi 08 mai 2024 à 17:10 +0200, Patrick ZAJDA a écrit :

  Pour sid ça devrait arriver dans quelques heures, cinq jours dans testing
normalement. à moins qu'il y ait une procédure accélérée que je ne connaisse
pas pour ce genre de chose dans testing.
  C'est moins prévisible pour stable, il faut que l'équipe en charge de
bookworm-security accepte une nouvelle version qui corrige cette régression.
  En attendant, de mon côté j'ai créé un fichier
/etc/apt/preferences.d/glib.pref avec le contenu suivant :
  package: libglib2.0-dev-bin libglib2.0-bin libglib2.0-dev libglib2.0-data
libglib2.0-0
  Pin: version 2.74.6-2+deb12u1
  Pin-Priority: -1
  
  Ce qui empêche la mise à jour vers la version cassé tout en permettant à la

future mise à jour corrective de s'installer si j'ai bien compris ce qui m'a
été expliqué au moment de la mise en place de cette config :-)
  
  

C'est arrivé pour sid.



C'est aussi arrivé pour stable (version 2.74.6-2+deb12u2) et old-stable 
(version 2.66.8-1+deb11u3).


--
Patrick ZAJDA

Re: perte de la touche morte du clavier pour saisir les accents circonflexes depuis une mise à jour système

[Patrick ZAJDA]

Le 08/05/2024 à 17:00, Gaëtan Perrier a écrit :

Le mercredi 08 mai 2024 à 15:11 +0200, Gaëtan Perrier a écrit :

Bonjour,

J'ai aussi ce problème sous sid/gnome.
Ca semble être ce bug :

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1070736

A+

Gaëtan


Le bug est marqué résolu. Y a plus qu'a attendre que les paquets arrivent dans
les depots ...


Pour sid ça devrait arriver dans quelques heures, cinq jours dans 
testing normalement. à moins qu'il y ait une procédure accélérée que je 
ne connaisse pas pour ce genre de chose dans testing.
C'est moins prévisible pour stable, il faut que l'équipe en charge de 
bookworm-security accepte une nouvelle version qui corrige cette régression.
En attendant, de mon côté j'ai créé un fichier 
/etc/apt/preferences.d/glib.pref avec le contenu suivant :
package: libglib2.0-dev-bin libglib2.0-bin libglib2.0-dev 
libglib2.0-data libglib2.0-0

Pin: version 2.74.6-2+deb12u1
Pin-Priority: -1

Ce qui empêche la mise à jour vers la version cassé tout en permettant à 
la future mise à jour corrective de s'installer si j'ai bien compris ce 
qui m'a été expliqué au moment de la mise en place de cette config :-)


--
Patrick ZAJDA

Re: perte de la touche morte du clavier pour saisir les accents circonflexes depuis une mise à jour système

[Patrick ZAJDA]

Je m'aperçois que la formulation de mon message peut prêter à confusion.
Je reformule donc : Il faut non seulement remettre à la version 
précédente libglib2.0-0 mais aussi les autres paquets installés de 
libglib, c'est pour ça que ça en supprimait autant dans le retour de ta 
tentative.


Le 08/05/2024 à 15:30, Patrick ZAJDA a écrit :

Bonjour,

Pour ma part, sous Bookworm donc j'ai fait :
sudo apt install libglib2.0-dev-bin/bookworm libglib2.0-bin/bookworm 
libglib2.0-dev/bookworm libglib2.0-data/bookworm libglib2.0-0/bookworm


Il faut non seulement remettre à la version précédente libglib2.0-0 
mais aussi les autres paquets installés de libglib, c'est pour ça que 
ça en supprime autant.


En espérant que ça aide.

Patrick


Le 08/05/2024 à 15:21, Frédéric Baldit a écrit :

Bonjour,

déjà je vois que ne suis pas seul...ç'est rassurant.

Le changement de layout suggéré vers azerty-AFNOR (avec dpkg-reconfigure
keyboard-configuration) n'a rien fait.

Je tenterais bien le retour de libglib2.0-0 vers sa version antérieure,
sauf que apt me réponds qu'il va alors supprimer beaucoup de paquets,
dont certains plutt cruciaux, comme le montre la copie suivante de mon
terminal:

###  début copie 
fred@ThinkPadT15g:~$ sudo apt install libglib2.0-0=2.74.6-2
Lecture des listes de paquets... Fait
Construction de l'arbre des dépendances... Fait
Lecture des informations d'état... Fait
Les paquets suivants ont été installés automatiquement et ne sont plus
nécessaires :
   appstream apt-config-icons bolt fwupd fwupd-amd64-signed
gir1.2-accountsservice-1.0
   gir1.2-clutter-1.0 gir1.2-cogl-1.0 gir1.2-coglpango-1.0
gir1.2-evince-3.0 gir1.2-gck-1
   gir1.2-gcr-3 gir1.2-gdm-1.0 gir1.2-geoclue-2.0 gir1.2-geocodeglib-2.0
gir1.2-gmenu-3.0
   gir1.2-gnomebluetooth-3.0 gir1.2-gnomedesktop-4.0 gir1.2-goa-1.0
gir1.2-grilo-0.3
   gir1.2-gtkclutter-1.0 gir1.2-gweather-4.0 gir1.2-ibus-1.0
gir1.2-javascriptcoregtk-4.1
   gir1.2-javascriptcoregtk-6.0 gir1.2-json-1.0 gir1.2-mediaart-2.0
gir1.2-mutter-11 gir1.2-nm-1.0
   gir1.2-nma-1.0 gir1.2-poppler-0.18 gir1.2-rest-1.0 gir1.2-rsvg-2.0
gir1.2-shumate-1.0
   gir1.2-soup-3.0 gir1.2-tracker-3.0 gir1.2-upowerglib-1.0
gir1.2-webkit-6.0 gir1.2-webkit2-4.1
   gnome-session-bin gnome-session-common gnome-software-common
gobject-introspection
   gstreamer1.0-pipewire ibus ibus-data ibus-gtk ibus-gtk3 ibus-gtk4
im-config jq libblkid-dev
   libbrotli-dev libbz2-dev libcue2 libdjvulibre-dev libegl1-mesa-dev
libepoxy-dev libexif-dev
   libffi-dev libflashrom1 libfontconfig-dev libfreetype-dev
libfribidi-dev libftdi1-2 libfwupd2
   libgcab-1.0-0 libgnome-autoar-0-0 libgnome-menu-3-0 libgraphite2-dev
libharfbuzz-gobject0
   libimath-dev libiptcdata0 libjaylink0 libjcat1 libjq1 liblqr-1-0-dev
libmagick++-6-headers
   libmagickcore-6-arch-config libmagickcore-6-headers
libmagickwand-6-headers libmount-dev
   libnghttp2-dev libonig5 libopenexr-dev libopenjp2-7-dev
libosinfo-1.0-0 libosinfo-l10n
   libpixman-1-dev libportal-gtk4-1 libpsl-dev libshumate-1.0-1
libshumate-common libsmbios-c2
   libsqlite3-dev libsysprof-4 libsysprof-ui-5 libvulkan-dev
libwayland-bin libwayland-dev
   libwmf-dev libxcb-render0-dev libxcb-shm0-dev libxcomposite-dev
libxcursor-dev libxdamage-dev
   libxfixes-dev libxft-dev libxi-dev libxinerama-dev libxkbcommon-dev
libxrandr-dev libxrender-dev
   libxtst-dev nautilus-data osinfo-db pango1.0-tools python3-blinker
python3-dateutil
   python3-distro-info python3-ibus-1.0 python3-jwt
python3-lazr.restfulclient python3-lazr.uri
   python3-markdown python3-oauthlib python3-software-properties
python3-wadllib python3-yaml
   switcheroo-control uuid-dev wayland-protocols xwayland
Veuillez utiliser « sudo apt autoremove » pour les supprimer.
Les paquets supplémentaires suivants seront installés :
   policykit-1-gnome
Les paquets suivants seront ENLEVÉS :
   chrome-gnome-shell gdm3 gnome-browser-connector gnome-characters
gnome-core gnome-maps
   gnome-music gnome-session gnome-shell gnome-shell-extension-prefs
gnome-shell-extensions
   gnome-shell-mailnag gnome-software gnome-sushi gnome-weather
gstreamer1.0-packagekit
   libadwaita-1-dev libatk-bridge2.0-dev libatk1.0-dev libatspi2.0-dev
libcairo2-dev
   libgdk-pixbuf-2.0-dev libgirepository1.0-dev libglib2.0-bin
libglib2.0-dev libglib2.0-dev-bin
   libgraphene-1.0-dev libgtk-3-dev libgtk-4-dev libharfbuzz-dev
libjavascriptcoregtk-4.0-dev
   libjavascriptcoregtk-4.1-dev libjavascriptcoregtk-6.0-dev
libjson-glib-dev libmagick++-6.q16-dev
   libmagick++-dev libmagickcore-6.q16-dev libmagickcore-dev
libmagickwand-6.q16-dev libpango1.0-dev
   libpolkit-gobject-1-dev libpoppler-glib-dev librsvg2-dev
libsoup-3.0-dev libsoup2.4-dev
   libsysprof-4-dev libwebkit2gtk-4.0-dev libwebkit2gtk-4.1-dev
libwebkitgtk-6.0-dev nautilus
   packagekit packagekit-tools rygel-tracker software-properties-common
software-properties-gtk
   task-gnome-desktop tracker tracker-extract tracker-miner-fs
Les NOUVEAUX

Re: perte de la touche morte du clavier pour saisir les accents circonflexes depuis une mise à jour système

[Patrick ZAJDA]

Bonjour,

Pour ma part, sous Bookworm donc j'ai fait :
sudo apt install libglib2.0-dev-bin/bookworm libglib2.0-bin/bookworm 
libglib2.0-dev/bookworm libglib2.0-data/bookworm libglib2.0-0/bookworm


Il faut non seulement remettre à la version précédente libglib2.0-0 mais 
aussi les autres paquets installés de libglib, c'est pour ça que ça en 
supprime autant.


En espérant que ça aide.

Patrick


Le 08/05/2024 à 15:21, Frédéric Baldit a écrit :

Bonjour,

déjà je vois que ne suis pas seul...ç'est rassurant.

Le changement de layout suggéré vers azerty-AFNOR (avec dpkg-reconfigure
keyboard-configuration) n'a rien fait.

Je tenterais bien le retour de libglib2.0-0 vers sa version antérieure,
sauf que apt me réponds qu'il va alors supprimer beaucoup de paquets,
dont certains plutt cruciaux, comme le montre la copie suivante de mon
terminal:

###  début copie 
fred@ThinkPadT15g:~$ sudo apt install libglib2.0-0=2.74.6-2
Lecture des listes de paquets... Fait
Construction de l'arbre des dépendances... Fait
Lecture des informations d'état... Fait
Les paquets suivants ont été installés automatiquement et ne sont plus
nécessaires :
   appstream apt-config-icons bolt fwupd fwupd-amd64-signed
gir1.2-accountsservice-1.0
   gir1.2-clutter-1.0 gir1.2-cogl-1.0 gir1.2-coglpango-1.0
gir1.2-evince-3.0 gir1.2-gck-1
   gir1.2-gcr-3 gir1.2-gdm-1.0 gir1.2-geoclue-2.0 gir1.2-geocodeglib-2.0
gir1.2-gmenu-3.0
   gir1.2-gnomebluetooth-3.0 gir1.2-gnomedesktop-4.0 gir1.2-goa-1.0
gir1.2-grilo-0.3
   gir1.2-gtkclutter-1.0 gir1.2-gweather-4.0 gir1.2-ibus-1.0
gir1.2-javascriptcoregtk-4.1
   gir1.2-javascriptcoregtk-6.0 gir1.2-json-1.0 gir1.2-mediaart-2.0
gir1.2-mutter-11 gir1.2-nm-1.0
   gir1.2-nma-1.0 gir1.2-poppler-0.18 gir1.2-rest-1.0 gir1.2-rsvg-2.0
gir1.2-shumate-1.0
   gir1.2-soup-3.0 gir1.2-tracker-3.0 gir1.2-upowerglib-1.0
gir1.2-webkit-6.0 gir1.2-webkit2-4.1
   gnome-session-bin gnome-session-common gnome-software-common
gobject-introspection
   gstreamer1.0-pipewire ibus ibus-data ibus-gtk ibus-gtk3 ibus-gtk4
im-config jq libblkid-dev
   libbrotli-dev libbz2-dev libcue2 libdjvulibre-dev libegl1-mesa-dev
libepoxy-dev libexif-dev
   libffi-dev libflashrom1 libfontconfig-dev libfreetype-dev
libfribidi-dev libftdi1-2 libfwupd2
   libgcab-1.0-0 libgnome-autoar-0-0 libgnome-menu-3-0 libgraphite2-dev
libharfbuzz-gobject0
   libimath-dev libiptcdata0 libjaylink0 libjcat1 libjq1 liblqr-1-0-dev
libmagick++-6-headers
   libmagickcore-6-arch-config libmagickcore-6-headers
libmagickwand-6-headers libmount-dev
   libnghttp2-dev libonig5 libopenexr-dev libopenjp2-7-dev
libosinfo-1.0-0 libosinfo-l10n
   libpixman-1-dev libportal-gtk4-1 libpsl-dev libshumate-1.0-1
libshumate-common libsmbios-c2
   libsqlite3-dev libsysprof-4 libsysprof-ui-5 libvulkan-dev
libwayland-bin libwayland-dev
   libwmf-dev libxcb-render0-dev libxcb-shm0-dev libxcomposite-dev
libxcursor-dev libxdamage-dev
   libxfixes-dev libxft-dev libxi-dev libxinerama-dev libxkbcommon-dev
libxrandr-dev libxrender-dev
   libxtst-dev nautilus-data osinfo-db pango1.0-tools python3-blinker
python3-dateutil
   python3-distro-info python3-ibus-1.0 python3-jwt
python3-lazr.restfulclient python3-lazr.uri
   python3-markdown python3-oauthlib python3-software-properties
python3-wadllib python3-yaml
   switcheroo-control uuid-dev wayland-protocols xwayland
Veuillez utiliser « sudo apt autoremove » pour les supprimer.
Les paquets supplémentaires suivants seront installés :
   policykit-1-gnome
Les paquets suivants seront ENLEVÉS :
   chrome-gnome-shell gdm3 gnome-browser-connector gnome-characters
gnome-core gnome-maps
   gnome-music gnome-session gnome-shell gnome-shell-extension-prefs
gnome-shell-extensions
   gnome-shell-mailnag gnome-software gnome-sushi gnome-weather
gstreamer1.0-packagekit
   libadwaita-1-dev libatk-bridge2.0-dev libatk1.0-dev libatspi2.0-dev
libcairo2-dev
   libgdk-pixbuf-2.0-dev libgirepository1.0-dev libglib2.0-bin
libglib2.0-dev libglib2.0-dev-bin
   libgraphene-1.0-dev libgtk-3-dev libgtk-4-dev libharfbuzz-dev
libjavascriptcoregtk-4.0-dev
   libjavascriptcoregtk-4.1-dev libjavascriptcoregtk-6.0-dev
libjson-glib-dev libmagick++-6.q16-dev
   libmagick++-dev libmagickcore-6.q16-dev libmagickcore-dev
libmagickwand-6.q16-dev libpango1.0-dev
   libpolkit-gobject-1-dev libpoppler-glib-dev librsvg2-dev
libsoup-3.0-dev libsoup2.4-dev
   libsysprof-4-dev libwebkit2gtk-4.0-dev libwebkit2gtk-4.1-dev
libwebkitgtk-6.0-dev nautilus
   packagekit packagekit-tools rygel-tracker software-properties-common
software-properties-gtk
   task-gnome-desktop tracker tracker-extract tracker-miner-fs
Les NOUVEAUX paquets suivants seront installés :
   policykit-1-gnome
Les paquets suivants seront mis à une VERSION INFÉRIEURE :
   libglib2.0-0
0 mis à jour, 1 nouvellement installés, 1 remis à une version
inférieure, 59 à enlever et 0 non mis à jour.
Il est nécessaire de prendre 1 463 ko dans les archives.
Après cette opération, 130 Mo d'espace disque

Re: perte de la touche morte du clavier pour saisir les accents circonflexes depuis une mise à jour système

[Patrick ZAJDA]

Bonjour,

JE suis sous Bookworm et il semble que ça soit la version 
2.74.6-2+deb12u1 de libglib2.0-0 (et autres paquets libglib) disponibles 
dans bookworm-security qui pose problème.
Après downgrade de tous ces paquets vers la version 2.74.6-2 de 
bookworm, je peux à nouveau utiliser la touche accent circonflexe.
Il semble donc que cette mise à jour de sécurité entraîne une régression 
pour nos claviers.

Pour information, j'utilise Mate comme environnement de bureau.

Patrick


Le 08/05/2024 à 10:12, Txo a écrit :

Le 08/05/2024 à 05:42, Frédéric Baldit a écrit :


Bonjour,


Bonjour,



depuis la dernière mise à jour de mon système (Bookworm sous Gnome), je
n'ai plus accès avec mon clavier à la saisie des accents circonflexes
avec la touche morte de mon clavier (qui utilise le layout «Français,
obsolete, variante»).

Le phénomène est visible par exemple dans un terminal, dans emacs ou
depuis claws-mail. J'ai tenté un



Pareil ici avec Sid.


...sans succès. Curieusement, dans la zone de saisie d'un navigateur
cette touche fonctionne, ainsi que dans le terminal zutty. Si je me
connecte Aussi, si je me connecte sur tty3 (après Ctrl-Alt-F3), j'ai
bien l'usage de cette touche. Je sèche...



Par contre je n'ai mème pas cela, c'est partout qu'il me manque les 
accents. Merci au correcteur... Tout aussi sec.





--
Patrick ZAJDA

Re: Installer le noyau 6.6 sur Bookworm

[Patrick ZAJDA]

Bonjour,


Peut-on penser qu'il dépendra un jour d'un paquet
linux-image-6.6.X-Y.deb12.Z-amd64 ou bien cette dépendance est figée
pour tout le cycle d'utilisation de Bookworm ?


ça dépend si un backport est fait, pour ce qui est du kernel ça suit 
généralement la version de testing avec un décalage le temps que ça soit 
éprouvé dans testing et en suite après l'upload c'est assez long le 
temps du build et ça passe généralement par une validation des ftp masters.


En résumé la réponse est oui, c'est fort probable. Il faut patienter...


Le 08/03/2024 à 09:28, Olivier a écrit :

1. À Michel:
Je ne vois moi aussi que les paquets:
linux-image-6.6.13+bpo-amd64-dbg
linux-image-6.6.13+bpo-amd64-unsigned


2. Il y a quelques semaines, j'ai installé le noyau 6.5 avec la commande:
apt -y install linux-image-amd64 -t bookworm-backports

Ce paquet linux-image-amd64 du repo backports dépend du seul paquet
linux-image-6.5.0-0.deb12.4-amd64
Peut-on penser qu'il dépendra un jour d'un paquet
linux-image-6.6.X-Y.deb12.Z-amd64 ou bien cette dépendance est figée
pour tout le cycle d'utilisation de Bookworm ?

3. Au quotidien, comment met-on à jour un paquet linux-image-amd64 ou
linux-image-6.6.X-Y.deb12.Z-amd64 ?
Une commande apt upgrade ou apt dist-upgrade suffit-elle ?

Le jeu. 7 mars 2024 à 17:57, Michel Verdier  a écrit :

Le 7 mars 2024 Olivier a écrit :


J'ai une machine UN305 (processeur Intel Alder lake) sur laquelle j'ai
installé Bookworm puis le noyau 6.5.
J'ai besoin d'y installer le noyau 6.6 (pour corriger un problème graphique).

En backport tu as la 6.6.13 disponible. Bizarrement il n'y a que la
unsigned ou alors j'ai mal regardé :
linux-image-6.6.13+bpo-amd64-unsigned
Les sources sont aussi dispo si tu compile ton kernel.
Et les headers si tu as des drivers à compiler :
linux-headers-6.6.13+bpo-amd64



--
Patrick ZAJDA

Re: Erreur nvidia suite upgrade noyau 6.1.0-18

[Patrick ZAJDA]

Le 19/02/2024 à 20:42, ajh-valmer a écrit :

On Monday 19 February 2024 19:17:22 Patrick ZAJDA wrote:
et epsilon

Merci :
  

La mise à jour est déjà disponibles sur les dépôts officiels, si
bookworm-updates est bien dans les listes de sources alors la version
525.147.05-7~deb12u1 devrait être disponible à l'installation

Il semblerait que la mise à jour du kernel 6.1.0-18-amd64,
ne concerne que la version Debian SID.
J'ai retenté un apt update  => apt upgrade en vain,
c'est un échec, ma Debian-12 refuse de dépasser le noyau 6.1.0-17-amd64 :
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1062932

Intéressant que tu dises que ça n'est que pour Sid alors qu'à la fin du 
bug que tu indiques, c'est clairement dit que c'est dispo pour stable...
Et je suis sous Bookworm, pas de Sid du tout chez moi ; mon kernel est 
bien 6.1.0-18. Peut-être que tu devrais revérifier tes listes sources 
(sources.list et autres dans sources.list.d) et si tu ne ferais pas du 
pining qui bloquerait quelque chose...
Utilises-tu apt-listbugs ? J'ai eu quelques surprises par le passé 
niveau pining.



Pour la partie drivers Nvidia :
https://lists.debian.org/debian-stable-announce/2024/02/msg2.html
The following packages have been updated to correct the problem:
Source package Fixed version
nvidia-graphics-drivers525.147.05-7~deb12u1
nvidia-graphics-drivers-tesla  525.147.05-7~deb12u1
nvidia-graphics-drivers-tesla-470  470.223.02-4~deb12u1
nvidia-settings525.147.05-1~deb12u1

Je ne peux rien tester tant que je ne peux upgrader.


Même chose que ce que j'ai dit ci-dessus.

Je crois qu'une (re ?)vérification s'impose de ton côté niveau config de 
apt.


Patrick


Bonne soirée.


$ apt policy nvidia-driver
nvidia-driver:
Installé : 525.147.05-7~deb12u1
Candidat : 525.147.05-7~deb12u1
Table de version :
*** 525.147.05-7~deb12u1 500
500https://deb.debian.org/debian  bookworm-updates/non-free
amd64 Packages
100 /var/lib/dpkg/status
525.147.05-4~deb12u1 500
  500https://deb.debian.org/debian  bookworm/non-free amd64 Packages


--
Patrick ZAJDA

Re: Erreur nvidia suite upgrade noyau 6.1.0-18

[Patrick ZAJDA]

Le 19/02/2024 à 15:13, ajh-valmer a écrit :

On Saturday 17 February 2024 10:49:04 Patrick ZAJDA wrote:
Quoi qu'il en soit, une mise à jour a été publié dans bookworm-updates

(SUA 252-1) pour corriger le problème de compilation lors de la mise à
jour vers le kernel 6.1.0-18.
apt update puis apt upgrade devrait donc pouvoir se passer sans
problème, à la limite un apt -f install si ça ne passe vraiment pas.

Je n'ai pas vu qu'une mise à jour a été publiée dans bookworm-updates,
(SUA 252-1) correctif de compilation, mise à jour vers le kernel 6.1.0-18.
Ou s'agit-il d'une opération de mise à jour en cours ?




https://lists.debian.org/debian-stable-announce/2024/02/msg2.html

La mise à jour est déjà disponibles sur les dépôts officiels, si 
bookworm-updates est bien dans les listes de sources alors la version 
525.147.05-7~deb12u1 devrait être disponible à l'installation



$ apt policy nvidia-driver
nvidia-driver:
  Installé : 525.147.05-7~deb12u1
  Candidat : 525.147.05-7~deb12u1
 Table de version :
 *** 525.147.05-7~deb12u1 500
    500 https://deb.debian.org/debian bookworm-updates/non-free 
amd64 Packages

    100 /var/lib/dpkg/status
 525.147.05-4~deb12u1 500
    500 https://deb.debian.org/debian bookworm/non-free amd64 Packages



--
Patrick ZAJDA

Re: Erreur nvidia suite upgrade noyau 6.1.0-18

[Patrick ZAJDA]

Bonjour,


Le 16/02/2024 à 20:36, ajh-valmer a écrit :

On Friday 16 February 2024 08:16:54 Michel Verdier wrote:

Le 14 février 2024 zithro a écrit :

- Michel a compris qu'André disait "je laisse tomber, ça montre bien
qu'il y a que des noobs/gens inutiles ici".

Effectivement ça me semblait une attaque de la liste, d'où la forme
abrupte de mon mail pour laquelle je m'excuse :

Et oui, il faut bien lire les contenus des mails.

Repassons au positif :
Ma carte Nvidia supportait le mode 1280x1024,
Avec le pilote Nouveau, le mode maxi est de 1024x768.
J'ai tout essayé, xrandr et compagnie... que des messages d'erreur.
Un problème succède à un autre, je ne peux plus upgrader
mon système du noyau 6.1.0-17-amd64 à 6.1.0-18-amd64.
J'ai effacé dans le /boot/ par :
# rm *6.1.0-18-amd64*
purgé par :
# apt purge *6.1.0-18-amd64*
"tout est à jour",
comme s'il y avait un résidu 6.1.0-18-amd64 quelque part.
Bonne soirée.



Le paquet linux-image-6.1.0-18 ne serait-il pas toujours installé ?
Quoi qu'il en soit, une mise à jour a été publié dans bookworm-updates 
(SUA 252-1) pour corriger le problème de compilation lors de la mise à 
jour vers le kernel 6.1.0-18.
apt update puis apt upgrade devrait donc pouvoir se passer sans 
problème, à la limite un apt -f install si ça ne passe vraiment pas.


--
Patrick ZAJDA

Debian Mobile cont'd from about 10 yeaes ago furey2310....

[Patrick Furey]

Can it be downloaded to usb, SD  and installed  at boot options via ADB?
I did successfully install the code via termux on 2 phones but that termux 
wasnt rooted i believe and my charging was subsequently ruined on my android 
o/s
The code is there on termux (terminal)
Regards
Patrick Shaun Furey.

Sent from Yahoo Mail on Android

Re: Délai de 25 secondes

[Patrick ZAJDA]

Hello,

Question que je ne pense pas avoir lu jusque là, quelle carte graphique 
utilises-tu ?


J'ai une carte NVIDIA et j'ai eu ce genre de souci quand j'avais voulu 
tester ce que ça donnerait avec le pilote nouveau.
Et installer le paquet du pilote propriétaire correspondant à ma carte 
avait résolu le souci.
C'est probablement pas la solution vu que j'y vais un peu au hasard, 
mais sait-on jamais si ça peut être utile.


Patrick


Le 10/11/2023 à 08:43, Seb a écrit :


Salut,


Il y a 6 mois (Debian 11), je n'avais aucun problème de ce type.

Il y a 3 mois (Debian 11), j'ai remarqué ce souci avec un Firefox que 
je venais de mettre à jour.


Aujourd'hui (Debian 12), j'ai le problème avec Firefox, Pavucontrol et 
Xdaliclock.


Le problème est croissant et menace de devenir handicapant en Debian 13.
Je ne pense pas qu'il soit lié spécifiquement à Pavucontrol.

Quelque chose qui rend mon installation peu courante, c'est que 
pendant l'install je décoche toutes les cases liées à des 
gestionnaires de fenêtres, et une fois l'install terminée j'installe 
le minimum (fvwm et ses dépendances). D'habitude, quand on décoche 
toutes ces cases, c'est pour un serveur (qui ne lancera jamais 
xdaliclock, donc on ne verra pas apparaître ce délai de 25 secondes).


De la sorte, il est probable que je n'installe pas un package qui est 
standard chez les autres utilisateurs et qui s'avère utile pour 
certains logiciels graphiques (quoique pas indispensable puisque ces 
logiciels finissent par se lancer).


Sinon, un rapport de bogue sur 
https://gitlab.freedesktop.org/pulseaudio/pavucontrol/-/issues 
pourrait être utile.


C'est vrai qu'un rapport de bug pourrait servir.
Peut-être plus avec reportbug toutefois, si c'est bien un package qui 
me manque ; il n'y aurait alors qu'une simple dépendance à ajouter.


Pour ma part, je cherche des contributeurs pour le moteur 
d'inférences http://refpersys.org/


Ça a l'air intéressant ; bonne chance !


Si tu as la flemme de débugger, j'imagine que tu as actuellement
pulseaudio d'installé. Tu pourrais le retirer au profit de
pipewire-pulse et voir si ça fonctionne mieux.


À vrai dire, j'utilise très peu pavucontrol ; c'est la multiplication 
des endroits où le délai se manifeste qui m'inquiète.


En tout état de cause, vu que tu as le fd et 25 secondes pour agir, 
un ls -al /proc/XXX/fd/YYY est sans doute utile pour savoir quelle 
socket/autre il poll.


Ben ça donne ça:
~> ls -l /proc/$(pidof pavucontrol)/fd/11
lrwx-- 1 seb seb 64 Nov 10 10:42 /proc/135045/fd/11 -> 
'anon_inode:[eventfd]'



Seb.


--
Patrick ZAJDA

Re: can you parse and "tail" at once? (and if you can't why not?)

[Patrick Wiseman]

On Sun, Oct 22, 2023, 10:39 PM Albretch Mueller  wrote:

>  OK, Greg's suggestion once again "made my day".
>  I know at some point I will have to code everything in some
> programming language, but for now I will just get things done as
> quickly as possible.
>  Also, Greg, please, I would like for you to understand that it is not
> my intention to upset you about what you find visually upsetting. We
> have talked about that before.
>  I am sure you could certainly understand that there are reasons why
> some people build certain habits and that not everyone gets upset
> about the same kinds of "secondary accidents" (as Aristotle would put
> it). Some people may have a hard time even noticing what upsets you.
>  lbrtchx
>
> Context? Totally lost. No need to reply, I haven't been following anyway.

Re: LTSP: commande introuvable

[Patrick Magnaud]

Le 16/10/2023 à 18:48, Alex PADOLY a écrit :

Bonsoir,


Bonjour,

J'ai fait une nouvelle installation du système et à l'installation, j'ai 
choisi en plus serveur ssh.


Une fois le système installé, j'ai installé le paquet ltsp.

En mode super-utilisateur, la commande n'est pas active et le système 
m'a répondu ce que vous avez


écrit.


Pour passer root, il faut utiliser la commande Switch User, su, avec 
l'option "-". Sinon, on reste dans l’environnement de l'utilisateur 
initial et PATH n'est pas modifié. Il faut donc taper :


su -

Pour que la commande ltsp soit reconnue, il faut être vraiment sous le 
compte root.


Je n'ai pas terminé la configuration du serveur ltsp, j'apprends 
beaucoup de choses.


Comment fait-on pour modifier la variable PATH?

Depuis que j'utilise Debian (Potato), je n'ai eu besoin de le faire.

Merci pour vos réponses.


De rien,
Bien cordialement,
Patrick

Re: Qt6 Location for Debian 12

[Patrick Franz]

Hej,

Am Donnerstag, 12. Oktober 2023, 01:17:44 CEST schrieb Petric Frank:
[...]
> I got a hint on a git project where Qt6Location was compiled for older
> Qt Versions (< 6.5):
>   https://github.com/ntadej/qtlocation
> 
> Maybe the 6.4.2 release of it could be a candidate for being
> repackaged as *.deb ?

We cannot add packages to Debian 12 anymore. I'd also like to stick to 
the official Qt releases and avoid 3rd-party releases.


-- 
Med vänliga hälsningar

Patrick Franz

Re: Qt6 Location for Debian 12

[Patrick Franz]

Hej,

Am Mittwoch, 11. Oktober 2023, 16:52:24 CEST schrieb Petric Frank:
> Hello,
> 
> is there a *.deb for Qt6 Location (-dev) module ?
> 
> I have seen there is one for Qt6 Positioning in the stable repository
> but i miss the above one.
> 
> Or do i have to compile the Qt6 system myself ?

There is no Location module for Qt 6.4. It never existed. However, it 
got reintroduced as a tech preview for 6.5.


-- 
Med vänliga hälsningar

Patrick Franz

Re: Creality don't sound great (Was: Re: Is it possible to downgrade openssl?)

[Patrick Wiseman]

On Sun, Jul 30, 2023 at 11:33 AM Dan Ritter  wrote:

> Andy Smith wrote:
> >
> > Sounds like an absolutely shit-tier company. I hope there are
> > alternatives in the 3d-printing world, a world that I know almost
> > nothing about.
>
> Many, but Creality makes really cheap 3D printers, so
> lots of people buy them.
>

I think it would be fairer to say that they make inexpensive 3D printers.
The hardware is generally of good quality. And they just sent me a working
version of the AppImage which had been segfaulting, so not entirely
unresponsive (if a bit slow to respond).

Patrick

Re: Is it possible to downgrade openssl?

[Patrick Wiseman]

On Sat, Jul 29, 2023 at 10:42 PM Max Nikulin  wrote:

>
> On 30/07/2023 05:16, Patrick Wiseman wrote:
> > I'd already reached the conclusion that messing with openssl was a bad
> > idea. Unfortunately, Creality is unresponsive to pleas to fix their
> > software.
>
> I have no experience with 3d printers at all, but I am curious
> concerning any progress with a suggestion from another thread:
>
> Patrick Wiseman. Re: qt.network.ssl problems (OT?) Wed, 21 Jun 2023
> 10:35:57 -0400.
>
> https://lists.debian.org/msgid-search/CAJVvKsO_MubTQidcQZ+vSNkYmO=sqghvvno0nlzu8ohxch7...@mail.gmail.com
>
> > Turns out that the K1 printer is Klipper under the hood and there's a
> > way to hack it to gain full access (so I don't need the broken app).
>

This is veering way off topic ... but, since you ask :)

It's true that the K1 is Klipper (one of several flavors of 3D printer
software) under the hood, and I have used the hack to get to it. But
Creality apparently disapproves of the hack and so has disabled it in the
latest firmware. For now, I'm living with the older firmware and the very
useful hack, but I might not need to do that if their software worked on my
laptop. Hence my attempts to get it working. (Creality has historically
been good about open source hardware and software, so I'm not sure what
they're up to with this particular printer.)

Cheers
Patrick

Re: Is it possible to downgrade openssl?

[Patrick Wiseman]

On Sat, Jul 29, 2023, 6:03 PM Andy Smith  wrote:

> Hello,
>
> On Fri, Jul 28, 2023 at 04:14:30PM -0400, Patrick Wiseman wrote:
> > On Fri, Jul 28, 2023, 4:10 PM Brian  wrote:
> > > On Fri 28 Jul 2023 at 16:04:10 -0400, Patrick Wiseman wrote:
> > > > any way to revert to an earlier version of OpenSSL? I'm on an
> up-to-date
> > > > bookworm system.
> > >
> > > Install from https://snapshot.debian.org/.
> >
> > Thanks for the quick reply. I'll try that as soon as I get back to the
> box.
>
> You've been handed a very effective footgun with no further
> instructions. Please do not replace your system's openssl package
> with an older one unless you know exactly what you are doing. In all
> likelihood you will completely break your whole system. Tons of
> things link to openssl.
>
> AppImages are supposed to include the libraries they depend upon,
> though I don't use them so am not sure about this, but if that is
> true then I think it's unlikely that your AppImage is using the
> system openssl anyway.
>
> You really need to get support from the supplier of the package.
>

I'd already reached the conclusion that messing with openssl was a bad
idea. Unfortunately, Creality is unresponsive to pleas to fix their
software. A Googled source claimed the problem was fixed by installation of
an earlier version of openssl, but I'm glad I asked here before trying that!

I'm mostly a lurker here but always impressed with how patiently helpful
y'all are.

Thanks for being here!

Cheers
Patrick

Re: Is it possible to downgrade openssl?

[Patrick Wiseman]

On Sat, Jul 29, 2023 at 10:42 AM  wrote:

> Eduardo M KALINOWSKI  wrote:
> > On 28/07/2023 17:04, Patrick Wiseman wrote:
> > > I have an AppImage from Creality which segfaults with a QT ssl
> > > error. Googling tells me that the latest version of OpenSSL (3.x)
> > > omits some X509 functionality, which can be found in OpenSSL-1.1.
> > > (And someone reports that installing it solves the problem.) But I
> > > can't find that package. Is there any way to revert to an earlier
> > > version of OpenSSL? I'm on an up-to-date bookworm system.
> >
> > AppImages bundle all the libraries used by the application, so
> > changing the "system" version of openssl probably won't work.
>
> It sounds like you need to contact Creality to update the AppImage.
> Although
>
> https://forum.manjaro.org/t/creality-slicer-appimage-not-loading-qt-network-ssl-errors/143726
> suggests that the AppImage does not contain the OpenSSL library.
>
> But the only 'omission' of X.509 functionality that I can see on
> https://www.openssl.org/news/openssl-3.0-notes.html is
> "X509 certificates signed using SHA1 are no longer allowed at security
> level 1 or higher. The default security level for TLS is 1, so
> certificates signed using SHA1 are by default no longer trusted to
> authenticate servers or clients."
>
> I'm not sure I'd want to be deliberately trying to undo a security
> upgrade.
>
> I'm sure you're right. Contacting Creality is a fool's game; they've been
alerted to the problem since they released the software, so I'll just have
to be patient.

Thanks to all for your suggestions.

Patrick

Re: Is it possible to downgrade openssl?

[Patrick Wiseman]

On Fri, Jul 28, 2023, 4:10 PM Brian  wrote:

> On Fri 28 Jul 2023 at 16:04:10 -0400, Patrick Wiseman wrote:
>
> > I have an AppImage from Creality which segfaults with a QT ssl error.
> > Googling tells me that the latest version of OpenSSL (3.x) omits some
> X509
> > functionality, which can be found in OpenSSL-1.1. (And someone reports
> that
> > installing it solves the problem.) But I can't find that package. Is
> there
> > any way to revert to an earlier version of OpenSSL? I'm on an up-to-date
> > bookworm system.
>
> Install from https://snapshot.debian.org/.
>

Thanks for the quick reply. I'll try that as soon as I get back to the box.

Patrick

>

Is it possible to downgrade openssl?

[Patrick Wiseman]

I have an AppImage from Creality which segfaults with a QT ssl error.
Googling tells me that the latest version of OpenSSL (3.x) omits some X509
functionality, which can be found in OpenSSL-1.1. (And someone reports that
installing it solves the problem.) But I can't find that package. Is there
any way to revert to an earlier version of OpenSSL? I'm on an up-to-date
bookworm system.

Thanks
Patrick

Re: RFP: deadEarth RPG documentation - free and open RPG game

[Patrick Wiseman]

Not the list to make such requests

On Wed, Jun 28, 2023, 11:30 AM Joshua Allen  wrote:

> Package: wnpp
> Severity: wishlist
>
> * Package name: deadearth
>   Version : second edition
>   Upstream Contact: i...@thegamecrafter.com
> * URL : 
> https://web.archive.org/web/20021211104811/http://www.deadearth.com/index.pl/legal
> * License : License GNU Free Documentation License v1.1 or any later 
> version, no invariant sections or front and back text
>   Programming Lang: unknown
>   Description : tabletop rpg game
>
>
>  - deadEarth is a free and open TTRPG game for use in real life or online
>
> - I am aware of other TTRPGs but this looks to be the oldest one I could find 
> and it was released in 2002, since the developer no longer maintains the 
> site, I looked it up on the internet archive.
>
> - Debian doesnt have to be used but, it does qualify for inclusion in the 
> free documentation repositories.
>
>  - how do you plan to maintain it?
> I am just an end user and I am not able to maintain it per say, maybe clean 
> it up since the wording is difficult to see on the url listed above
>
> You would have to probably contact TGC LLC for official clarification, I did 
> months ago but Wikimedia commons needs confirmation from the source.
>
> Still it looks interesting, it probably already is in the repository.
>
>

Re: qt.network.ssl problems (OT?)

[Patrick Wiseman]

On Tue, Jun 20, 2023 at 5:30 PM gene heskett  wrote:

>
> I have no idea about the creality clone, but klipper runs perfectly fine
> on a bananapi-m5 w/4gigs of dram in front of the printer.  klipper
> itself is a 2 part thing, one part replacing the usually crippled marlin
> in the printers controller, by reflashing the controller card, the other
> part massages the gcode on the way to the printer enhancing the printers
> speed among other advantages.
>
> Turns out that the K1 printer is Klipper under the hood and there's a way
to hack it to gain full access (so I don't need the broken app).

Cheers
Patrick

Re: qt.network.ssl problems (OT?)

[Patrick Wiseman]

On Tue, Jun 20, 2023 at 2:18 PM Patrick Wiseman  wrote:

> On Tue, Jun 20, 2023, 2:02 PM gene heskett  wrote:
>
>> On 6/20/23 13:33, Patrick Wiseman wrote:
>> > Hello, all:
>> >
>> > This may be a little off-topic, in which case, apologies, but there's so
>> > much knowledge here, maybe someone can help.
>> >
>> > I recently acquired a Creality 3D printer, and Creality has supplied an
>> > AppImage to operate it. When I execute it, I get these errors and a
>> > segmentation fault:
>> >
>> > qt.network.ssl: QSslSocket: cannot resolve EVP_PKEY_base_id
>> > qt.network.ssl: QSslSocket: cannot resolve SSL_get_peer_certificate
>> > Segmentation fault
>> >
>> > Googling suggests this is a common problem for developers using Qt, but
>> I
>> > didn't find any obvious solution. Help would be much appreciated.
>> >
>> > Patrick
>>
>> That AppImage is probably a copyright violation copy of an old and
>> outdated klipper, Please install and use the real thing, many bugs have
>> been treated since it was forked and put under their proprietary
>> umbrella. I have an S1 coming tomorrow, but it will get run by the real
>> thing.
>>
>
> Thanks , but the AppImage is what it is and the firmware on the K1 printer
> is also what it is. Anything else I can do?
>
>>
>>
An acquaintance tells me it runs fine on Linux Mint 21.1 so what's
different about Debian 12 (bookworm)? And isn't an AppImage supposed to run
anywhere?

Patrick

Re: qt.network.ssl problems (OT?)

[Patrick Wiseman]

On Tue, Jun 20, 2023, 2:02 PM gene heskett  wrote:

> On 6/20/23 13:33, Patrick Wiseman wrote:
> > Hello, all:
> >
> > This may be a little off-topic, in which case, apologies, but there's so
> > much knowledge here, maybe someone can help.
> >
> > I recently acquired a Creality 3D printer, and Creality has supplied an
> > AppImage to operate it. When I execute it, I get these errors and a
> > segmentation fault:
> >
> > qt.network.ssl: QSslSocket: cannot resolve EVP_PKEY_base_id
> > qt.network.ssl: QSslSocket: cannot resolve SSL_get_peer_certificate
> > Segmentation fault
> >
> > Googling suggests this is a common problem for developers using Qt, but I
> > didn't find any obvious solution. Help would be much appreciated.
> >
> > Patrick
>
> That AppImage is probably a copyright violation copy of an old and
> outdated klipper, Please install and use the real thing, many bugs have
> been treated since it was forked and put under their proprietary
> umbrella. I have an S1 coming tomorrow, but it will get run by the real
> thing.
>

Thanks , but the AppImage is what it is and the firmware on the K1 printer
is also what it is. Anything else I can do?

Patrick


>

qt.network.ssl problems (OT?)

[Patrick Wiseman]

Hello, all:

This may be a little off-topic, in which case, apologies, but there's so
much knowledge here, maybe someone can help.

I recently acquired a Creality 3D printer, and Creality has supplied an
AppImage to operate it. When I execute it, I get these errors and a
segmentation fault:

qt.network.ssl: QSslSocket: cannot resolve EVP_PKEY_base_id
qt.network.ssl: QSslSocket: cannot resolve SSL_get_peer_certificate
Segmentation fault

Googling suggests this is a common problem for developers using Qt, but I
didn't find any obvious solution. Help would be much appreciated.

Patrick

Re: Evolution email (problem?)

[Patrick Wiseman]

Can't explain it, but it strikes me it's almost certainly a permissions
problem.

Patrick




On Wed, Apr 26, 2023 at 11:06 AM Default User 
wrote:

> Strange . . .
>
> I run Debian 11 (Bullseye) Stable, up to date, Gnome 3 desktop
> environment.
>
> I recently set up Evolution email.  Works okay.
>
> Two days ago, I realized that it absolutely refuses to delete email
> messages directly from the [Gmail]/All Mail folder of any email
> account.  To delete a message in the All Mail folder, that is not also
> in another folder, the message must be moved into another folder and
> then deleted from there.
>
> I sure think I remember being able to delete directly from the All Mail
> folder when I first set up Evolution.
>
> I re-started Evolution.  No change.
> I re-booted. No change.
> I did sudo aptitude reinstall evolution.  No Change.
> I moved ~/.local/share/evolution/mail/local/folders.db to Trash, then
> re-started Evolution to construct a new folders.db file. No change.
> I did sudo aptitude purge evolution, then sudo aptitude install
> Evolution.  No change.
> I moved ~/.local/share/evolution, ~/.cache/evolution, and
> ~/.config/evolution to Trash, then did sudo aptitude purge evolution,
> then sudo aptitude install Evolution.  No change.
> I did some research online, seeking a solution, or at least a "good"
> explanation.  No solution or "good" answer was found.
>
> But . . .   then I booted into a Debian 11.6 Live/Install usb thumb
> drive, and then installed Evolution into the Live session, upon which
> Evolution DID allow message deletion directly from the [Gmail]/All Mail
> folder!
>
> Back to my regular Debian 11 Stable install.  No change.
>
> It makes no sense to me that it would work from a Debian 11
> Live/Install session, but not (currently) from a regular Debian
> session.
>
> Even if it did not work from a Live Debian session, it makes no sense
> to me that Evolution could be designed this way.  I see no good reason
> that messages can be directly deleted from any other folder, but not
> [Gmail]/All Mail.
>
> Does anyone have a solution, so that messages must be moved to another
> folder (if they are not also there already), just to be deleted?  Or at
> least a "good" explanation as to why Evolution appears to be designed
> this way?
>
>
>
>

Re: aide au débogage : logged-gcc

[Ken-Patrick]

On 11/03/2023 13:44, Basile Starynkevitch wrote:


Non. Pour des tas de raisons (y compris que j'ai personnellement 
contribué à GCC mais pas à Clang, que j'en connais donc assez bien les 
internes, et aussi pour des raisons de licence -je préfère la GPL à BSD) 
je souhaite explicitement utiliser GCC (dans mon esprit, GCC 12 en début 
2023, sur Debian ou autre Linux)


  Cordialement



Hum, je n'ai peut-être pas été assez explicite.
La compilation database, c'est juste un json contenant les commandes 
utilisées pour compiler chaque fichier d'un projet.
bear peut te générer une compilation database quel que soit le 
compilateur ou le système de build utilisé (à noter que cmake sait le 
faire tout seul).
Le format des compilations database est décrit dans la doc de clang, 
parce que c'est utilisé dans la plupart de leur outils, mais c'est tout.


En plus, bear est en GPL ;-).

++
Ken-Patrick

Re: aide au débogage : logged-gcc

[Ken-Patrick]

Hello Basile,

On 10/03/2023 19:57, Basile Starynkevitch wrote:

Bonsoir,


[...]


Il est alors utile de pouvoir conserver la trace de toutes les 
compilations par GCC.



[...]

Est-ce que https://clang.llvm.org/docs/JSONCompilationDatabase.html et 
potentiellement https://github.com/rizsotto/Bear ne répondraient pas à 
ton besoin ? C'est à peu près standard il me semble.


++
Ken-Patrick

Re: Getting PC with Ubuntu; change to Debian?

[Patrick Wiseman]

Hello again:

I'm just getting around to firing up my new laptop (Lenovo X1 Carbon Gen
10), which came with Ubuntu installed. (By the way, I appreciate all of the
feedback I got.) Although Ubuntu is a Debian-derivative, I didn't much care
for the feel of it. This is entirely subjective, I realize, but it felt as
if it wanted to get in the way of my going "under the hood", something
which I've been doing with my computers for decades! So I resolved to
return to Debian.

After making a recovery USB stick (which I confirmed worked before doing
anything else), I tried to boot from a Debian 11 live USB. It dropped me
immediately to a grub prompt! As comfortable as I am with Debian, that
caused me some discomfort!

So I decided to try a Debian 12 net install, which worked flawlessly and
has given me a nice clean Xfce desktop. I haven't put it through any
serious paces yet, but I'm back on familiar territory.

Cheers
Patrick


On Sun, Dec 4, 2022, 9:51 PM Patrick Wiseman  wrote:

> Hello, fellow Debian users:
>
> I've had Debian on my computers for a very long time (can't remember
> exactly when but early 2000's for sure); and I've had Lenovo laptops for
> ages too. I finally need to replace my main laptop (an at least 10-year old
> ThinkPad), so I've bought an X1 from Lenovo, with Ubuntu pre-installed (to
> be delivered in January).
>
> So, a question. Should I side-grade to Debian (and, if so, how easy would
> it be to do that?)? Or will I be happy with Ubuntu (which is, after all, a
> Debian derivative)?
>
> I'm very familiar and comfortable with Debian (was happy with 'testing'
> for a long time, but have lately reverted to 'stable'). And, although I'm a
> rare participant on this list, I enjoy the lurk and would presumably need
> to go elsewhere if I had questions about my Ubuntu experience.
>
> Thoughts will be welcome.
>
> Patrick
>
>

Re: KDE Frameworks 5.103.0 and KDE Plasma 5.27

[Patrick Franz]

Hej,

Am Dienstag, 21. Februar 2023, 20:09:13 CET schrieb Timothy M 
Butterworth:
> Is Bookworm going to get Frameworks 5.103.0 and is there any estimated
> time when Bookworm will get Plasma 5.27?

Bookworm already has Plasma 5.27 beta (it's officially called 5.26.90), 
but both the Frameworks 5.103 and Plasma 5.27.0 have been uploaded to 
unstable and will be available in bookworm unless something weird 
happens.

The Frameworks 5.103 will migrate to bookworm towards the end of the 
week, while Plasma 5.27.0 will migrate in the middle of next week.

We hope we can get Plasma 5.27.2 in before the next freeze.


-- 
Med vänliga hälsningar

Patrick Franz

Re: [HS] confusion d'outils [Was] Re: Swap en mémoire RAM. Est-ce pertinent?

[Patrick ZAJDA]

Bonjour,


En fait, le principe n'est pas seulement de prendre de la RAM pour 
décharger la RAM pour la longévité de la carte SD car sinon, 
effectivement il y a comme un paradoxe.



Il faudrait que j'approfondisse ma recherche sur ce qui est fait sur le 
Raspberry Pi mais ce don je peux parler, c'est de Armbian.


Armbian utilise justement ZRAM.

Et oui on utilise toujours un espace en RAM comme swap mais celui-ci 
étant compressé, c'est toujours moins que si c'était un simple espace 
dédié genre tmpfs.


ça peut donc jouer sur les performances vu qu'il faut en suite 
décompressé mais il y a la possibilité de trouver un bon compromis entre 
performance de compression et performances à l'utilisation général.



En espérant que ça aide,


Patrick


Le 19/01/2023 à 09:50, didier gaumet a écrit :

Le jeudi 19 janvier 2023 à 08:16 +0100, Olivier backup my spare a
écrit :
[...]

Bêtement, j'ai voulu appliqué un usage lié au raspberry pi sur un
ordinateur portable.
Si on supprime ou met le swap d'un raspberry pi en RAM pour épargner
la carte SD, ça n'a aucun intérêt sur un PC.

[...]

C'est peut-être moi qui suis trop limité intellectuellement :-)

mais je ne vois pas plus l'intérêt sur un Raspberry (ou tout autre
ordinateur ayant une carte SD pour mémoire de masse): oui, ne pas
mettre le swap en RAM épargne la carte SD et évite ainsi une usure
précoce. Mais le but du swap n'est-il pas principalement (voire
exclusivement) d'éviter un plantage lors d'un manque de mémoire vive en
dédiant une partie de la mémoire de masse à un usage d'extension de la
mémoire vive? Donc à part un possible ralentissement, quel est le gain
attendu puisque la quantité de swap additionnée est dans ce cas
soustraite de la mémoire vive?

Du coup sur ce genre d'ordinateur, je suppose qu'il est possible de:
- soit privilégier la longévité de la carte SD en ne mettant pas le
swap dessus, mais dans ce cas mettre le swap en RAM n'apporte rien en
termes de quantité de mémoire globale disponible avant plantage. Dans
ce cas je verrais bien la suppression du swap?
- soit privilégier l'évitement de possibles plantages dus à une carence
mémoire disponible en créant un swap traditionnel sur la carte SD, en
intégrant que l'usure de la carte sera accélérée et à surveiller pour
remplacement périodique?



--
Patrick ZAJDA

Utiliser une version plus récente de Bluez sous Debian Bullseye

[Patrick ZAJDA]

Bonjour à tous,


Tout d'abord, j'en profite pour vous souhaiter une excellente année, 
dans la sérénité et qu'elle vous apporte ce que vous souhaitez tant dans 
vos projets informatiques que personnels.



J'en viens en suite à ma question :)

Vu que les mainteneurs des paquets bluez ne semblent pas avoir le projet 
de proposer un backport d'une version de Bluez supérieure à la 5.55 pour 
Bullseye, quelqu'un a-t-il déjà fait l'expérience, concluante de 
préférence, d'installer une version plus récente sous Debian Bullseye ?


Pour information, je leur en ai déjà fait la demande en écrivant à 
l'adresse indiquée en mainteneur du paquet bluetooth et sur la liste 
debian-backports sans avoir la moindre réponse de la part d'au moins un 
mainteneur.



Mes dernières tentatives de compilation m'ont donné des résultats peu 
concluants, il fallait que je force régulièrement le redémarrage du 
service bluetooth.


Je compilais directement depuis les sources téléchargées depuis 
bluez.org en suivant les instructions de la doc du même site.



J'utilise le bluetooth avec Home Assistant sur un Odroid N2+ sous 
Armbian Bullseye, qui propose un kernel 6.0. En somme, une architecture 
très proche de Debian de base.



Pensez-vous que télécharger les sources du paquet Debian proposé dans 
testing et construir les paquets puisse être plus concluant ?



Bonne journée,


Patrick

Getting PC with Ubuntu; change to Debian?

[Patrick Wiseman]

Hello, fellow Debian users:

I've had Debian on my computers for a very long time (can't remember
exactly when but early 2000's for sure); and I've had Lenovo laptops for
ages too. I finally need to replace my main laptop (an at least 10-year old
ThinkPad), so I've bought an X1 from Lenovo, with Ubuntu pre-installed (to
be delivered in January).

So, a question. Should I side-grade to Debian (and, if so, how easy would
it be to do that?)? Or will I be happy with Ubuntu (which is, after all, a
Debian derivative)?

I'm very familiar and comfortable with Debian (was happy with 'testing' for
a long time, but have lately reverted to 'stable'). And, although I'm a
rare participant on this list, I enjoy the lurk and would presumably need
to go elsewhere if I had questions about my Ubuntu experience.

Thoughts will be welcome.

Patrick

Re: nginx.conf woes

[Patrick Kirk]

On 02/10/2022 16:12, Greg Wooledge wrote:

On Sun, Oct 02, 2022 at 04:07:05PM +0300, Patrick Kirk wrote:

If I
try lynx http://cleardragon.com a similar redirect takes place and I get a
"Alert!: Unable to connect to remote host" error and lynx closes down.

I'm not an expert on nginx configuration, but I did notice one thing:


         server {
         listen                    443 ssl http2;
         listen                    [::]:443 ssl http2;
         server_name               server_name  cleardragon.com
*.cleardragon.com;

You've repeated "server_name" twice here.

Thanks.  Good spot but sadly that doesn't fix it.

nginx.conf woes

[Patrick Kirk]

Hi all,

I have 2 sites to run from one server.  Both are based on ASP.Net Core.  
Both have SSL certs from letsencrypt.  One works perfectly.  The other 
sort of works. If I go to http://localhost:5100 by redirecting to 
https://localhost:5101 and then it warns of an invalid certificate.  If 
I try lynx http://cleardragon.com a similar redirect takes place and I 
get a "Alert!: Unable to connect to remote host" error and lynx closes down.


When I do sudo tail -f /var/log/nginx/error.log I see: 2022/10/02 
12:44:22 [notice] 1624399#1624399: signal process started


I'm baffled as to how I can diagnose the problem.

Here's my nginx.conf just in case there is some error in it.

Thanks in advance,

Patrick

user www-data;
worker_processes auto;
pid /run/nginx.pid;
include /etc/nginx/modules-enabled/*.conf;
events {
        worker_connections 768;
# multi_accept on;
}
http {
    include        /etc/nginx/conf.d/proxy.conf;
    limit_req_zone $binary_remote_addr zone=one:10m rate=5r/s;
    server_tokens  off;
    charset utf-8;
    charset_types text/css text/xml text/plain text/vnd.wap.wml 
application/javascript application/rss+xml;

    sendfile on;
    keepalive_timeout   29;
    client_body_timeout 10; client_header_timeout 10; send_timeout 10;
    server {
    listen 80 default_server;
    listen [::]:80 default_server;
    server_name _;
    return 301 https://$host$request_uri;
    }
    upstream kirksnet{
        server localhost:5000;
    }
    server {
        listen                    443 ssl http2;
        listen                    [::]:443 ssl http2;
        server_name               kirks.net *.kirks.net;
        ssl_certificate           
/etc/letsencrypt/live/kirks.net/fullchain.pem;
        ssl_certificate_key       
/etc/letsencrypt/live/kirks.net/privkey.pem;

        add_header X-Frame-Options DENY;
        add_header X-Content-Type-Options nosniff;
        location / {
            proxy_pass http://kirksnet;
            limit_req zone=one burst=10 nodelay;
        }
    }
        upstream RazorHill{
        server localhost:5100;
    }
        server {
        listen                    443 ssl http2;
        listen                    [::]:443 ssl http2;
        server_name               server_name  cleardragon.com 
*.cleardragon.com;
        ssl_certificate           
/etc/letsencrypt/live/cleardragon.com/fullchain.pem;
        ssl_certificate_key       
/etc/letsencrypt/live/cleardragon.com/privkey.pem;

        add_header X-Frame-Options DENY;
        add_header X-Content-Type-Options nosniff;
        location / {
            proxy_pass http://RazorHill;
            limit_req zone=one burst=10 nodelay;
        }
    }
}

Re: .Net Core program crashes on Linux and not on Windows

[Patrick Kirk]

 Log
2:14:55 PM:- Starting process.
2:14:56 PM:- Auction run for Kazzak EU.
2:14:58 PM:- Connected realm id for Kazzak EU is 1305.
2:15:01 PM:- The realm data for Kazzak EU namespace is downloaded.
2:15:42 PM:- We have 0 to add and 95789 auctions to update and 1883 expired
or sold auctions in the database for Kazzak EU.
2:15:42 PM:- Saving changes for Kazzak EU.
2:16:13 PM:- Updated 95789 auction listings for Kazzak EU in 10 batches.
2:16:13 PM:- Marking expired auctions for Kazzak EU.
2:16:14 PM:- Getting Kazzak EU done took 1minutes 17 seconds.
2:16:14 PM:- Auction run for Illidan US.
2:16:14 PM:- Connected realm id for Illidan US is 57.
...snip...
2:18:18 PM:- Auctions scan 1 complete.

I took your advice and doubled it.  A program that had me baffled this
morning is now running perfectly.  Many thanks guys!

On Thu, 1 Sept 2022 at 16:34, Greg Wooledge  wrote:

> On Thu, Sep 01, 2022 at 03:47:52PM +0300, Patrick Kirk wrote:
> > /var/log/messages:Sep  1 12:41:34 debian-s-websites kernel:
> > [31104249.962672] .NET ThreadPool invoked oom-killer:
> > gfp_mask=0x6280ca(GFP_HIGHUSER_MOVABLE|__GFP_ZERO), nodemask=(null),
> > order=0, oom_score_adj=0
> > /var/log/messages:Sep  1 12:41:34 debian-s-websites kernel:
> > [31104249.973435]  oom_kill_process.cold.30+0xb/0x1cf
> >
> > Dan - you found my problem.  Thanks!
> >
> > When the program is running I have this:
> > Mem:  989Mi   579Mi64Mi   149Mi   346Mi
> > 120Mi
> >
> > It clearly thinks 64M is not enough.  I know my program can require up
> to a
> > Gig of memory for the bigger datasets.  What is the minimum GNU/Linux
> will
> > tolerate so I don't end up throwing money away on unused RAM?
>
> You said this is a VPS, so it's going to depend on which *type* of VPS,
> but mostly it's a matter of how much memory the VPS is allowed to use,
> by the host system.
>
> It looks like your VPS is limited to 1 GB (989 Mi, it says) of total
> memory.  If your program is coming close to that, then you may need to
> talk to your VPS provider and get more memory allocated to you (which
> may cost more money).
>
> Try doubling it.
>
>

Re: .Net Core program crashes on Linux and not on Windows

[Patrick Kirk]

Hi Dan,

Yes I've been coding for a living for 15 years in C#.


/var/log/messages:Sep  1 12:41:34 debian-s-websites kernel:
[31104249.962672] .NET ThreadPool invoked oom-killer:
gfp_mask=0x6280ca(GFP_HIGHUSER_MOVABLE|__GFP_ZERO), nodemask=(null),
order=0, oom_score_adj=0
/var/log/messages:Sep  1 12:41:34 debian-s-websites kernel:
[31104249.973435]  oom_kill_process.cold.30+0xb/0x1cf

Dan - you found my problem.  Thanks!

When the program is running I have this:
Mem:  989Mi   579Mi64Mi   149Mi   346Mi
120Mi

It clearly thinks 64M is not enough.  I know my program can require up to a
Gig of memory for the bigger datasets.  What is the minimum GNU/Linux will
tolerate so I don't end up throwing money away on unused RAM?

Patrick

On Thu, 1 Sept 2022 at 15:21, Dan Ritter  wrote:

> Patrick Kirk wrote:
> > I have a program that is meant to run on my Debian VPS which is hosted by
> > Digital Ocean.  My access to the server is via Putty. Unfortunately the
> > program simply stops, silently, after a few minutes.
> >
> > My code is littered with try catch statements. The console window shows
> no
> > errors. It just silently fails.
> >
> > One issue may be that the program is a resource hog working with datasets
> > of a million or so records at a time. Perhaps the system kills processes
> > that threaten to take over all resources?
>
> There is a kernel feature called the Out Of Memory Killer, which
> kills processes that try to use all the memory of the system.
>
> > Anyway, I would be grateful for any suggestion on how to track down this
> > problem. I'm happy to read the documentation but as you can probably
> tell I
> > don't know how to start framing the question.
>
> Did you write this program?
>
> Are you a fluent .Net Core programmer?
>
> Has it ever run on some other system?
>
> What happens if you try with a dataset of ten records?
>
> -dsr-
>

.Net Core program crashes on Linux and not on Windows

[Patrick Kirk]

Hi all,

I have a program that is meant to run on my Debian VPS which is hosted by
Digital Ocean.  My access to the server is via Putty. Unfortunately the
program simply stops, silently, after a few minutes.

My code is littered with try catch statements. The console window shows no
errors. It just silently fails.

One issue may be that the program is a resource hog working with datasets
of a million or so records at a time. Perhaps the system kills processes
that threaten to take over all resources?

Anyway, I would be grateful for any suggestion on how to track down this
problem. I'm happy to read the documentation but as you can probably tell I
don't know how to start framing the question.

Patrick

Re: chmod u+s

[Patrick ZAJDA]

Le 24/08/2022 à 18:21, Gabriel Moreau a écrit :


À ma connaissance, sous Linux, le chmod u+s sur un dossier ne sers à 
rien ! Ce qui sers sur un dossier est


 chmod g+s
 chmod o+t

Le premier affecte tout nouveau fichier au groupe du dossier (pratique 
dans un partage puisque les personnes ne savent plus faire newgroup de 
nos jours et que les interfaces graphiques ne savent pas gérer ça).


Le second interdit à une personne d'effacer un fichier qui ne lui 
appartient pas dans un dossier, même s'il a les droits d'écriture sur 
le dossier. Cela a été mis au point à l'origine principalement pour 
résoudre le problème du dossier /tmp partagé.


En effet, sous UNIX, on a le droit d'effacer un fichier même si on n'a 
pas le droit d'écriture sur le fichier... puisque en pratique, on 
écrit dans la table d'index et non dans le fichier.




Merci encore ;)

Du coup, je ne comprends pas pourquoi est-ce que les mainteneurs du 
paquet transmission-daemon ont mis :


drwsrwxr-x 51 debian-transmission debian-transmission 4096 23 août 13:30 
downloads


Qui semble du coup superflu :) alors que le comportement qui a été 
décrit, à savoir conserver le propriétaire, aurait été bien arrangeant.


Bonne soirée,

--
Patrick ZAJDA

Re: chmod u+s

[Patrick ZAJDA]

Le 24/08/2022 à 18:12, Gabriel Moreau a écrit :


Cela ne fonctionne que pour les exécutables binaires. Cela ne 
fonctionne pas pour les scripts. C'est une sécurité car un script est 
trop facilement modifiable.


C'est la raison pour laquelle il y avait il y a longtemps perlsuid... 
et ainsi de suite



Merci ! ;)

Du coup je comprends mieux pourquoi ma mise en application ne 
fonctionnait pas et j'ai pu le voir à l’œuvre avec un binaire.



--
Patrick ZAJDA

Re: chmod u+s

[Patrick ZAJDA]

Merci à tous pour vos réponses.


J'ai bien lu, en tout cas la page man de chmod.

Sauf que je ne comprends pas ce que ça implique sur un dossier, en tout 
cas je n'arrive pas à appliquer ça pour m'en rendre compte.


Pire encore, je n'arrive même pas à appliquer ça pour un fichier.


Du coup, en tentant d'appliquer ce que j'ai compris :

- Dans le cas d'un dossier, si j'y crée un fichier en tant qu'un autre 
utilisateur, le propriétaire n'a pas plus de droit malgré le changement +s.


- Dans le cas d'un fichier, peut-être qu'un script bash ne permet pas 
d'appliquer mais changer son propriétaire et le chmod u+s ne m'a pas 
permis d'avoir les droits de l'utilisateur propriétaire.



Du coup, je n'ai certainement pas compris ce que ça implique exactement, 
ou mon approche pour vérifier que j'ai bien compris n'est pas la bonne 
mais avant d'écrire ici, je me suis bel et bien documenté ;)





Le 24/08/2022 à 17:18, Jean-Pierre Giraud a écrit :

Bonjour,

Le 24/08/2022 à 17:07, Basile Starynkevitch a écrit :


On 8/24/22 15:10, Patrick ZAJDA wrote:


Hello,


Quelqu'un pourrait-il, en des mots simple, m'expliquer à quoi sert 
le mode s pour un dossier ?





J'aurais tendance à suggérer la lecture (en anglais) de :

https://man7.org/linux/man-pages/man7/inode.7.html

https://man7.org/linux/man-pages/man1/chmod.1.html




C'est en effet un bon réflexe de commencer par lire par lire la page 
du manuel...
J'en profite pour signaler que depuis 2020, l'équipe de traduction en 
français de Debian s'investit avec une forte intensité pour traduire 
ou mettre à jour les pages de manpages.debian.org, effort qui 
bénéficie aux d'autres distributions qui utilisent les mêmes sources 
pour construire leurs pages de manuel (Archlinux, Fédora, Mageia et 
Opensuse.


et je cherche toujours des partenaires intéressés par RefPerSys 
<http://refpersys.org/>.



Merci.

--
Basile Starynkevitch
(only mine opinions / les opinions sont miennes uniquement)
92340 Bourg-la-Reine, France
web page: starynkevitch.net/Basile/


Amicalement,
jipege


--
Patrick ZAJDA

chmod u+s

[Patrick ZAJDA]

Hello,


Quelqu'un pourrait-il, en des mots simple, m'expliquer à quoi sert le 
mode s pour un dossier ?



Je le remarque entre autre sur le dossier download de transmission-remote.

Après de nombreux tests, je ne saisi pas à quoi il sert, je pensais 
qu'on pouvait avec ça faire en sorte que les sous-répertoires de 
celui-ci appartiendraient au même propriétaire.



En faisant chmod g+s j'ai bien le comportement que je pense, à savoir 
que le groupe est conservé même si c'est un autre utilisateur qui crée 
un sous-répertoire et le mode g+s est également mis sur le répertoire créé.


Mais en faisant u+s, j'ai l'impression que ça ne change tout simplement 
rien.



Bonne journée,

--
Patrick ZAJDA

Re: google account say it will no longer deliver email

[Patrick Bartek]

On Wed, 1 Jun 2022 18:04:02 +0100
Brian  wrote:

> On Thu 12 May 2022 at 10:08:01 -, Virgo Pärna wrote:
> 
> > On Wed, 11 May 2022 20:09:14 +0200, Fero Dali 
> > wrote:  
> > > Sorry for misunderstanding: it seems that my account will
> > > continue to work but ability to download mail with POP3 without
> > > OAUTH2 will be unavailable. 
> > 
> > Actually, even without OAUTH2 it should be still possible.
> > With two factor authentication enabled it is possible to generate
> > app password for use with standard authentication.  
> 
> It's June 1st and my ability to collect mail via POP3 from gmail is
> unimpaired. No  OAUTH2 or 2FA at this site. Whatever Google intended
> the situation to be after May 30th, it appears the interpretation by
> some users of their mail was off the mark.
> 

Still works here, too. Claws-mail 3.17.3 IMAP.  No OAuth2 or 2FA.
Neither of which this version of Claws supports, IIRC. Of course,
notification email did say "may not" not won't.

FWIW: Yahoo mail ceased working with Claws several years ago due to
security changes.  Though still accessible via web browser with only a
password.

B

Re: Debian desktop environment

[Patrick Bartek]

On Tue, 24 May 2022 13:27:29 +0200
Antonino Saetta  wrote:

> Hi,
> 
> After my surrender to Jessie I've thought of moving on with Stretch.
> 
> Currently I've installed it through the net, no problems at all.
> 
> So I was wondering, why am I asked to choose (or not) a GNOME desktop
> environment, other than *Debian desktop environment*?

To give you a choice of which desktop or desktops to install, or no GUI
at all. With Linux you can install and run multiple desktops, if you
want to.

> I thought that Debian is GNOME by default...

It is, but you don't have to use/install it if you don't want to.

> Also, what's the lightest desktop? Default, XFCE or LXDE...?

XFCE, LXDE, LXQt are all considered "light" desktops. Running only a
window manager (I use Openbox. There are many others.) is even lighter.

B

Re: Rendre persistante la cmde echo 1 > /proc/sys/net/netfilter/nf_conntrack_acct

[Patrick ZAJDA]

Bonjour,


Pour être sûr que tout est appliqué, il peut être bon de rajouter --system :

sysctl -p --system

Le 04/05/2022 à 15:17, NoSpam a écrit :


Le 04/05/2022 à 15:04, Olivier a écrit :

Voici les infos demandées:

# uname -a
Linux gsapilot 5.10.0-13-amd64 #1 SMP Debian 5.10.106-1 (2022-03-17)
x86_64 GNU/Linux

# sysctl -a|grep nf_conntrack_acct
0

# sysctl -p /etc/sysctl.d/foo.conf
net.ipv4.ip_forward = 1
net.netfilter.nf_conntrack_acct = 1

# sysctl -a|grep nf_conntrack_acct
1

Si je comprends bien, la commande "sysctl -p " déclenche
la mise à jour du paramètre net.netfilter.nf_conntrack_acct
mais cette commande n'est pas exécutée au démarrage ou bien elle est
exécutée mais son exécution ne donne pas l'effet escompté.


C'est procps qui s'occupe de cela, systemctl status procps

Perso je fais sysctl -p (sans nom de fichier) après chaque 
modification(s) celle(s) ci persiste(nt) dans le temps.




Le mer. 4 mai 2022 à 11:28, NoSpam  a écrit :

Bonjour. Quelle version de kernel ? Que dit sysctl -p ? Puis sysctl -n
net.netfilter.nf_conntrack_acct ?

Le 04/05/2022 à 11:17, Olivier a écrit :

Bonjour,

Pour activer l'accounting de conntrack, j'utilise:
echo 1 > /proc/sys/net/netfilter/nf_conntrack_acct

Comment rendre cette commande persistante (ie qu'elle soit
automatiquement lancée au démarrage) ?

J'ai essayé avec un fichier /etc/sysctl.d/foo.conf et le contenu 
ci-après.

net.ipv4.ip_forward=1
net.netfilter.nf_conntrack_acct=1

J'observe que seule la première ligne produit l'effet escompté:
# sysctl net.ipv4.ip_forward
1
# sysctl net.netfilter.nf_conntrack_acct
0

Je ne vois aucun message d'erreur dans les logs.

Slts



--
Patrick ZAJDA
Certification NVDA 2019

Re: Issues running TigerVNC on Debian WSL-2

[Mark Wood-Patrick]

I need to run on Debian under WSL-2 and want to be able to use Microsoft
Wayland as display 0 and want to have VNC access to the system.

Can you elaborate on exactly what your concerns are and how you would
resolve them and on your comment about the sticky bit, any links on these
issues would be appreciated

Regards

Mark

On Sun, Apr 17, 2022 at 11:07 AM Greg Wooledge  wrote:

> On Sun, Apr 17, 2022 at 10:59:44AM -0700, mwoodpatr...@gmail.com wrote:
> > Many thanks for the response. Much appreciated
> >
> > Permissions look ok
> >
> > ls -ld /tmp /tmp/.X11-unix
> >  drwxrwxrwt 2 root root 4096 Apr 17 09:31 /tmp
> >  lrwxrwxrwx 1 root root   19 Apr 17 09:31 /tmp/.X11-unix ->
> > /mnt/wslg/.X11-unix
> >
> > ls -ld /mnt/wslg/.X11-unix
> > drwxrwxrwx 2 root root 60 Apr 17 09:31 /mnt/wslg/.X11-unix
> >
> > I have write access to both locations
>
> That looks incredibly not OK.  You're doing something unusual, and
> you are going to have to deal with the consequences of that.  This
> may include running afoul of various things like AppArmor that are
> restricting access to specific directory trees, which you are no longer
> in.
>
> I can't imagine what benefit you think you're deriving from this
> convoluted setup, but whatever it is, I hope it's worth the pain you're
> going to experience, trying to track down all of the things you've
> broken.
>
> (You've also forgotten the sticky bit on your mounted directory.)
>
>

-- 
Mark Wood-Patrick

Re: No dbg packages for Plasma?

[Patrick Franz]

Hi,

Am Mittwoch, 6. April 2022, 15:20:49 CEST schrieb Luiz Romário Santana 
Rios:
> I just noticed that there seem to be absolutely no debug packages for
> plasma-desktop or for plasma frameworks. The only thing I can find
> online is this plasma-desktop-dbgsym package[1], which is only
> available for Sid.
> 
> Why? Is this a bug?

The debug packages for Plasma are available where all other dbg packages 
for Debian are. Here you can find the ones for e.g. plasma-desktop:

http://deb.debian.org/debian-debug/pool/main/p/plasma-desktop/


-- 
Med vänliga hälsningar

Patrick Franz

Re: Can't create a password successfully.

[Patrick Wiseman]

Chrome does that.


On Mon, Apr 4, 2022 at 10:04 AM Curt  wrote:

> On 2022-04-03, Brian  wrote:
> >
> >> One of the bits of advice is to use long passwords made up of three
> >> random words and to use a different password per website / to use
> >> your web browser to generate an appropriate random password.
> >> Forcing passwords to change regularly may not be a good way to
> >> maintain security - it can mean that people use password01, password02
> >> and things like that.
> >
> > Changing passwords at frequent intervals? Total nonsensense as far as
> > advice goes.
> >
>
> What web browser generates 'random' passwords? Am I missing something?
>
>
>
>
>

Re: Installing minimal command line system with netinst.iso

[Patrick Bartek]

On Thu, 3 Mar 2022 14:24:29 -0600
Richard Owlett  wrote:

> I have limited internet connectivity.
> I have read over the years that installing a *MINIMAL* command line 
> system from netinst.iso [without internet is possible].
> The intent to add pieces later assumed.
> Is this process described somewhere?
> 
> I've got a system that appears to have a very minimal shell running.
> My potential internet connection is a USB connected hotspot from 
> T-Mobile. On a typical GUI installed Debian it runs fine.
> 
> My minimal system doesn't recognize it.
> What might be the missing software?
> 
> Any suggestions?
> My effective data cap at the moment is ~1 GB/month.
> Among my goals is installer for those with _*minimal*_ resources.

I'm assuming you're connecting your phone directly via its USB port to a
USB port on your computer.  You probably need to make an entry
in /etc/network/interfaces. Plus, enable "USB tethering" (not wifi) in
Settings/Network on your phone.  Here's my interfaces entry:

allow-hotplug usb0
iface usb0 inet dhcp

My phone gets recognized as usb0, your's my differ.  Make the
correction, if needed.

B



iface usb0 inet dhcp

Re: W10 sur kvm/qemu

[Patrick ZAJDA]

Hello,


J'ai une VM windows 10 installée en suivant ce tuto : 
https://dennisnotes.com/note/20180614-ubuntu-18.04-qemu-setup/


J'avais pris la dernière version.


Vu que ça fait un moment, je ne me souviens pas suffisamment pour t'en 
dire plus que "j'ai utilisé ce tuto" ;)


Il explique pour installer le pilote dès l'installation de Windows.


Patrick


Le 24/02/2022 à 14:00, BERTRAND Joël a écrit :

Bonjour à tous,

Je fatigue...

J'essaie maintenant d'installer W10 sur une VM avec un disque virtio.
J'ai essayé plusieurs iso des pilotes et je me retrouve toujorus avec un
pilote virtio non installable... J'ai essayé :

virtio-win-0.1.172.iso  virtio-win-0.1.208.iso  virtio-win-0.1.215.iso

Une installation à la main en ligne de commande échoue aussi avec une
erreur cryptique.

Comment faire ?

Bien cordialement,

    JKB


--
Patrick ZAJDA

Re: Query

[Patrick Bartek]

On Mon, 7 Feb 2022 08:50:36 -0700
William Lee Valentine  wrote:

> I am wondering whether a current Debian distribution can be installed
> and run on an older Pentium III computer. (I have Debian 11.2 on a
> DVD.)

And have something usable? With the default GNOME desktop? Probably
not. With a lightweight desktop like XFCE? Maybe.  With a window
manager only? Better chance, but still iffy.  Main problem is lack of
RAM and slow CPU. You'll need the 32-bit version of Debian.

Here's a real example with equivalent hardware from around 2004 or 5:
1GHz Duron (PIII equivalent), 1 GB or so RAM, Fedora Core 6, GNOME
desktop.  My system at the time. After numerous upgrades from FC2,
system had become sluggish, particularly with menus -- a second or so
pause before appearing. I was able to get a usable system for another
year or so by a very custom install of Debian Wheezy. Started with a
terminal-only install, then added a minimal X with a window manager
(Openbox), a few utilities and lastly apps. However, I doubt if
Debian 11 would run well on it, even if you abandoned a desktop and
went with a window manager.

All you can do is try and see what happens. Good luck.

> The computer is
> 
>     Dell Dimension XPS T500: Intel Pentium III processor (Katnai)
>     memory: 756 megabytes, running at 500 megahertz
>     IDE disc drive: 60 gigabytes
>     Debian partition: currently 42 gigabytes
>     Debian 6.0: Squeeze
> 
> If I install Debian 11.2, will it run on this machine? Will it
> preserve the files and directories that I have on Squeeze?

Re: Package name to report bug with KDE file open dialog

[Patrick Dunford]

Ok

I will take this matter up with the application developers

thanks for your assistance to date.


On 29/01/22 06:10, Bijan Soleymani wrote:

On 2022-01-28 11:36, Patrick Dunford wrote:
It is not relevant whether the issue occurs in other applications 
because they all use the file open dialogs in ways that are specific 
to their application which the end user has no control over and 
therefore it cannot be tested for


If it doesn't occur in other applications we don't know if it is an 
app issue or a file chooser issue.


The application has been tested on various operating systems and this 
problem only occurs on the version of kde which is installed by the 
debian bookworm package manager.


Still could be an app issue.

Therefore just give me the name of the specific kde package that 
relates to that dialog, as I previously requested.


I believe a KDE app would use dolphin or possibly qt file chooser.

Bijan

Re: Package name to report bug with KDE file open dialog

[Patrick Dunford]

Hi

It is not relevant whether the issue occurs in other applications 
because they all use the file open dialogs in ways that are specific to 
their application which the end user has no control over and therefore 
it cannot be tested for


The application has been tested on various operating systems and this 
problem only occurs on the version of kde which is installed by the 
debian bookworm package manager.


Therefore just give me the name of the specific kde package that relates 
to that dialog, as I previously requested.


On 29/01/22 05:16, Bijan Soleymani wrote:

On 2022-01-28 11:10, Patrick Dunford wrote:
Which is the name of the package relating to the file open dialog to 
use in a bug report?


I think it depends on the application :)

Do you see it in other applications?

If so, you can figure out what GUI toolkit/environment it is gtk, qt, 
gnome, kde.


Otherwise probably file a bug on the app.

Bijan

Package name to report bug with KDE file open dialog

[Patrick Dunford]

Hello

I am using debian bookworm with kde and have noticed a bug in an 
application which is using a file open dialog. The problem only occurs 
in bookworm and is not present when using the same combination of kde 
and software on a system running bullseye.


Which is the name of the package relating to the file open dialog to use 
in a bug report?

Re: Bridger deux VLAN avec systemd network

[Patrick ZAJDA]

Bonjour,


Pour information, ça n'est pas moi qui ai décidé qu'il n'y aurait pas de 
subnet différent pour le réseau invité mais TP-Link.


On ne parle pas d'une solution offrant une souplesse si grande à 
l'utilisateur.


Je n'ai pas décidé de me compliquer la vie, je ne suis pas un pro du 
réseau mais j'aurais fait autrement pour le réseau invité :)


Autant dire que quand ce matériel me lâchera, je me délesterais d'un 
plus gros budget pour aller vers du Unify ou quelque chose de plus 
souple mais en attendant, je dois faire avec les décisions uni-latérales 
du constructeur qui me dira "Je vous ai compris" quand j'enverrais les 
suggestions d'une meilleure configuration réseau ;)



Bonne journée,


Patrick


Le 21/01/2022 à 20:20, Olivier Lange a écrit :
Euh, si t'es en wifi, non. Ton tag est sur tes équipements, et pas sur 
les équipements utilisateurs. Sauf si tu fournis un port réseau pour 
les visiteurs. Et encore, la tu configure en untag vid visiteur le 
port de ton switch.


Olivier

Le ven. 21 janv. 2022 à 14:03, Luc Novales <mailto:luc.nova...@enac.fr>> a écrit :


Bonsoir,

Je rajouterai qu'il vaut mieux taguer le VLAN normal. Taguer le
VLAN invité, c'est obliger les invit·és à utiliser une
configuration difficile à mettre en place, voire impossible sur
certains équipements.

Bonne soirée,

Luc.


Le 21/01/2022 à 16:26, Olivier Lange a écrit :

Bonjour

Je pense que tu as une ereur d'architecture. si tu veux avoir un
réseau invité dédié et sécurisé, il te faut un vlan dédié, et un
subnet (/24) dédié. Ce qui va te permettre de mettre en place des
règles spécifiques, autant au niveau 3/4/6 (via ton subnet et ton
routeur) que 2 (via ton vlan).

Tenter de bridger tout ca, au mieux ca devient une usine a gaz
ingérable avec plein de vulnérabilité, au pire ton réseau ne
fonctionnera pas.

C'est mon avis.
Olivier

Le ven. 21 janv. 2022 à 10:06, Patrick ZAJDA mailto:patr...@zajda.fr>> a écrit :

Bonjour,


Le contexte est le suivant :

J'utilise des TP-Link Deco M5 pour avoir un réseau wifi
maillé et souhaite activé le réseau wifi invité.

La seule chose qui sépare ces réseau c'est un VLAN tagué pour
le réseau invité.

Les adresses IP sont dans le même rang pour les deux réseaux,
les paramètres DHCP sont en tout point similaires pour les
réseau principal et invité. Il n'y a aucun paramètre DHCP
séparé pour le réseau invité.


Sur un Raspberry Pi, j'ai installé Adguard et souhaite donc
l'utiliser comme résolveur au niveau de tout le réseau.

Je pourais très bien spécifier l'adresse IP du Deco principal
comme résolveur et faire en sorte que celui-ci prenne Adguard
comme résolveur mais ça voudrait dire que toutes les requêtes
viendraient de la même adresse IP, ce qui ne convient pas.


Il faudrait donc que ce résolveur soit joignable depuis le
réseau principal comme invité et ce, en utilisant la même
adresse IP.


J'espère avoir pu expliquer clairement pourquoi il n'y a pas
d'autre solution qu'une seule IP qui serait joignable sur les
deux VLAN...



Le 21/01/2022 à 14:10, JUPIN Alain a écrit :

Bonjour,

Je comprends pas trop le concept.
Pour moi l’intérêt des VLAN c'est d'isoler les réseaux (en
utilisant les même équipements réseaux, switch par exemple).
En général on relie les VLAN entre eux par du routage, du
moins j'ai toujours pratiqué de cette manière !

Bridger des VLAN, si c'est pour du test/apprentissage
pourquoi pas, mais pour de la prod je me méfierait des
effets de bords au niveau de la sécurité.

Alain JUPIN

    Le 20/01/2022 à 14:51, Patrick ZAJDA a écrit :


Bonjour,


Est-il possible de bridger deux VLAN ?


Le but étant d'avoir une adresse IP commune aux deux VLAN
pour que les deux puissent utiliser le même résolveur entre
les réseaux principal et invité.

Si une autre méthode que le bridge est possible, je reste
bien entendu ouvert aux alternatives :)


Pour le moment, j'ai testé ceci (en anglais) sans succès :

https://www.variantweb.net/blog/creating-vlan-bridges-with-systemd-networkd/

<https://www.variantweb.net/blog/creating-vlan-bridges-with-systemd-networkd/>


Il utilise deux bridges mais le raspberry Pi sur lequel je
le fait ne passe jamais en ligne.

J'ai tenté de n'utiliser qu'un seul bridge, avec le même
résultat.


-- 
Patrick ZAJDA


-- 
Patrick ZAJDA



--
Patrick ZAJDA

Re: Bridger deux VLAN avec systemd network

[Patrick ZAJDA]

Bonjour,


Le contexte est le suivant :

J'utilise des TP-Link Deco M5 pour avoir un réseau wifi maillé et 
souhaite activé le réseau wifi invité.


La seule chose qui sépare ces réseau c'est un VLAN tagué pour le réseau 
invité.


Les adresses IP sont dans le même rang pour les deux réseaux, les 
paramètres DHCP sont en tout point similaires pour les réseau principal 
et invité. Il n'y a aucun paramètre DHCP séparé pour le réseau invité.



Sur un Raspberry Pi, j'ai installé Adguard et souhaite donc l'utiliser 
comme résolveur au niveau de tout le réseau.


Je pourais très bien spécifier l'adresse IP du Deco principal comme 
résolveur et faire en sorte que celui-ci prenne Adguard comme résolveur 
mais ça voudrait dire que toutes les requêtes viendraient de la même 
adresse IP, ce qui ne convient pas.



Il faudrait donc que ce résolveur soit joignable depuis le réseau 
principal comme invité et ce, en utilisant la même adresse IP.



J'espère avoir pu expliquer clairement pourquoi il n'y a pas d'autre 
solution qu'une seule IP qui serait joignable sur les deux VLAN...




Le 21/01/2022 à 14:10, JUPIN Alain a écrit :

Bonjour,

Je comprends pas trop le concept.
Pour moi l’intérêt des VLAN c'est d'isoler les réseaux (en utilisant 
les même équipements réseaux, switch par exemple).
En général on relie les VLAN entre eux par du routage, du moins j'ai 
toujours pratiqué de cette manière !


Bridger des VLAN, si c'est pour du test/apprentissage pourquoi pas, 
mais pour de la prod je me méfierait des effets de bords au niveau de 
la sécurité.


Alain JUPIN

Le 20/01/2022 à 14:51, Patrick ZAJDA a écrit :


Bonjour,


Est-il possible de bridger deux VLAN ?


Le but étant d'avoir une adresse IP commune aux deux VLAN pour que 
les deux puissent utiliser le même résolveur entre les réseaux 
principal et invité.


Si une autre méthode que le bridge est possible, je reste bien 
entendu ouvert aux alternatives :)



Pour le moment, j'ai testé ceci (en anglais) sans succès : 
https://www.variantweb.net/blog/creating-vlan-bridges-with-systemd-networkd/



Il utilise deux bridges mais le raspberry Pi sur lequel je le fait ne 
passe jamais en ligne.


J'ai tenté de n'utiliser qu'un seul bridge, avec le même résultat.


--
Patrick ZAJDA



--
Patrick ZAJDA

Re: Bridger deux VLAN avec systemd network

[Patrick ZAJDA]

Bonjour,



Le 20/01/2022 à 19:05, didier gaumet a écrit :


1) tu as redémarré le servive après les modifs?
# systemctl restart systemd-networkd


Oui, et juste après le Raspberry Pi passe hors ligne.



2) le wiki Debian avertit d'une particularité qui peut poser problème:
"Hint: in Debian Buster (probably also previous versions) systemd-
networkd may assign a different MAC-Address to the bridge than your
physical interface has. This may cause connection issues if your
service provider uses some kind of MAC-filtering when routing your
traffic. To circumvent such problems you may assign a MAC-address to
your bridge (probably the same as your physical device, replace the
'xx' with valid MAC): "
https://wiki.debian.org/SystemdNetworkd




Pas de filtrage particulier au niveau adresse MAC, tout au plus un bail DHCP 
statique assigné à la MAC de l'interface réseau.


J'essayerais d'assigner la MAC au bridge au cas où...


Merci beaucoup pour les pistes :)

--
Patrick ZAJDA

Re: TDE File Manager options

[Patrick Bartek]

On Thu, 20 Jan 2022 10:35:50 -0600
"c. marlow"  wrote:

> Hi,
> 
> Yes, I know that in previous emails I was using LXDE, but I thought
> that I would nuke and pave give TDE a try since I had never tried
> TDE  before.
> 
> And I am wondering what other file managers work with TDE 14 besides 
> Konqueror,  which ain't worth a dang! 

Take a look at XFE, a fairly full-featured, but lightweight, X-based
file manager. It seems compatible with all window-manager only or DTE
systems. I've been using it since I upgraded to Wheezy/Openbox (No
DTE just the window manager) from Fedore 12/Gnome years ago. XFE
continued to work fine with Stretch, Buster and now Devuan Beowulf.

B

Bridger deux VLAN avec systemd network

[Patrick ZAJDA]

Bonjour,


Est-il possible de bridger deux VLAN ?


Le but étant d'avoir une adresse IP commune aux deux VLAN pour que les 
deux puissent utiliser le même résolveur entre les réseaux principal et 
invité.


Si une autre méthode que le bridge est possible, je reste bien entendu 
ouvert aux alternatives :)



Pour le moment, j'ai testé ceci (en anglais) sans succès : 
https://www.variantweb.net/blog/creating-vlan-bridges-with-systemd-networkd/



Il utilise deux bridges mais le raspberry Pi sur lequel je le fait ne 
passe jamais en ligne.


J'ai tenté de n'utiliser qu'un seul bridge, avec le même résultat.


--
Patrick ZAJDA

Re: Nginx and ASP.Net Core

[Patrick Kirk]

On Mon, 13 Dec 2021 at 13:37, Dan Ritter  wrote:

> Patrick Kirk wrote:
>
> 6. If serving dynamic content, where to get it
>
>
> A program running on port 5000 is a fine place to serve dynamic
> content from, but you haven't told nginx about it.
>
> I'll bet you that this "kestrel" program comes with
> documentation about using a reverse proxy, which is what you
> want nginx to do.
>
> -dsr-
>
Hi Dan,

Nginx is working well for items 1 to 6 on your list.  I followed this
tutorial:
https://docs.microsoft.com/en-us/aspnet/core/host-and-deploy/linux-nginx?view=aspnetcore-6.0

As you can see from the logs, there is no problem serving static files.

But I can't find how to tell nginx where to get dynamic content.

Nginx and ASP.Net Core

[Patrick Kirk]

Hi all,

I am trying to run an ASP.Net Core site on my hosted Debian box and I seem
to have messed up the configuration.  All attempts to reach the page on
port 80 get 403 Forbidden messages.

My error log says: pk@debian-s-websites:~$ sudo tail -f
/var/log/nginx/error.log 2021/12/13 06:27:24 [error] 17999#17999: *675
directory index of "/var/www/cxxxn.org/" is forbidden, client: 213.7.43.53,
server: cxxxn.org, request: "GET / HTTP/1.1", host: "www.cxxxn.org"

If I try lynx http://localhost:5000 locally I get a perfect session. So
kestrel is working but nginx is not configured correctly. I think the
mistake is in my the config file for cxxxn.org has a line saying: "index
index.html index.htm index.nginx-debian.html" To test this, I put a Hello
World file in the root folder and it was served up correctly. If I
then try http://cxxxn.org/Privacy I get a 404 error.

Does anyone know if my guess that the "index" line in the configuration is
the problem is correct?

How can I configure Nginx to serve content from Asp.Net Core where there is
no static html file being generated?

Thanks in advance.

Re: lsusb

[Patrick Magnaud]

Le 26/10/2021 à 17:07, k6dedi...@free.fr a écrit :
> Bonjour,

Bonjour,

> suite à une réorganisation de mon PC, je réinstalle tout en GPT EFI.
> Après avoir installé Debian 11, je cherche à affiner mes paramètres.
> La  commande "lsusb" n'est pas présente, je ne peux l'atteindre ni en 
> utilisateur, ni en administrateur.
> Le message est :
> lsusb: commande introuvable dans les deux cas.
> 
> Quel paquet est sensé la contenir ?

Sur la page https://www.debian.org/distrib/packages, dans la partie
"Recherche dans le contenu des paquets" et en tapant lsusb, on trouve
que lsusb est dans le paquet usbutils.

> Merci de vos réponses.

de rien,

> Cassis

Patrick

Re: question from total newbie. a little help please

[Patrick Bartek]

On Sun, 17 Oct 2021 09:00:52 -0400
JAMES BOSWELL  wrote:

> if i divide my hard drive and install debian lynx on it. will i be
> able to effectively run debian on this laptop?

The best way to find that out is to get a Live version of Debian, and
see if boots and runs without problems.

> Device name LAPTOP-R4DB7V5U
> Processor Intel(R) Core(TM) i3-10110U CPU @ 2.10GHz   2.59 GHz
> Installed RAM 4.00 GB (3.81 GB usable)
> Device ID CAACC244-37B7-4294-84E4-E73B9C030FDF
> Product ID 00356-02325-39311-AAOEM
> System type 64-bit operating system, x64-based processor
> Pen and touch No pen or touch input is available for this display
> 
> Edition Windows 10 Home
> Version 21H1
> Installed on ‎4/‎2/‎2021
> OS build 19043.1288
> Experience Windows Feature Experience Pack 120.2212.3920.0
> 
> i know about enough to fill a thimble but i'm hopeful and any guidance
> would be greatly appreciated and i would follow it to the T's

Since your knowledge of Linux admittedly is severely lacking, I would
recommend thoroughly researching Linux, in general, and Debian, in
particularly, BEFORE attempting any install. And the first attempt be
on a system you don't mind trashing.

And always keep in mind: Linux is NOT Windows.  So never assume that
the way you did it on Windows will work on Linux.

Welcome to the neighborhood.

B

Re: First time WINE user looking for tutorial

[Patrick Bartek]

On Sun, 10 Oct 2021 05:14:36 -0500
Richard Owlett  wrote:

> On 10/09/2021 10:24 AM, Patrick Bartek wrote:
> > On Sat, 9 Oct 2021 09:40:21 -0500
> > Richard Owlett  wrote:
> >   
> >> [snip]
> 
> > 
> > Just be forewarned, WINE is not the catchall solution to running
> > Windows apps: The more involved codewise the program is like games
> > or Photoshop, the more problems you'll have.  
> 
> For context, I've been providing informal support to a local couple
> for decades. He is a retired pastor, now a missionary. They need a
> new computer and as part of my support, I'll be purchasing a
> replacement. As I've not used Windows since WinXP and they are pure
> Windows users I planned to dual boot Windows and Debian. Debian
> primarily for its maintenance tools. I hope WINE will run enough of
> their "must have" apps that I can use that as a selling point to move
> from Windows to Linux.

If all they've ever used is Windows, leave it at that. Don't attempt
to switch them to Linux. It will be more trouble than it's worth. Get
them a laptop with Windows, clean the preinstalled and CPU
cycles eating background crap off of it, bring it up-to-date, and you're
done. 


> > In those cases, just run Windows in a virtual machine which is what
> > I do for ALL Windows apps I need. Less or virtually no gotchas!  
> 
> On my personal machines I would have no motivation to install a VM. 
> However, I'll investigate the pros/cons of having their machine run a
> VM in which I would run Debian as a demo.

Unless they really NEED Debian or Linux, don't bother.

If you want to demo Linux just get one of the many "live" versions to
boot the system.  Then they can play with it without installing.  And
it won't do anything to their existing Windows install.

> Are there good FOSS or low cost VMs for Windows machines?

Windows has it's own VM.  Can't remember its name, but I'd recommend
VirtualBox instead -- https://www.virtualbox.org/ .  It's free and has
versions that run on Linux or Windows, among others.

B

Re: First time WINE user looking for tutorial

[Patrick Bartek]

On Sun, 10 Oct 2021 03:28:42 +
"Russell L. Harris"  wrote:

> On Sat, Oct 09, 2021 at 08:24:38AM -0700, Patrick Bartek wrote:
> >On Sat, 9 Oct 2021 09:40:21 -0500
> >Richard Owlett  wrote:  
> 
> >Just be forewarned, WINE is not the catchall solution to running
> >Windows apps: The more involved codewise the program is like games or
> >Photoshop, the more problems you'll have. In those cases, just run
> >Windows in a virtual machine which is what I do for ALL Windows apps
> >I need. Less or virtually no gotchas!  
> 
> Richard, Can you recommend a virtual machine for Debian which can run
> Google Earth?
> 
> 

Richard didn't write that, I did -- Patrick.

I use VirtualBox for my VM needs, but why would you need to: Google
Earth has versions that run natively on Windows, OSX and Linux or you
can run it in most any web browsers -- https://earth.google.com/ --
regardless of OS.

B 

Re: First time WINE user looking for tutorial

[Patrick Bartek]

On Sat, 9 Oct 2021 09:40:21 -0500
Richard Owlett  wrote:

> I just installed WINE64 on a Bullseye system. I'm looking for a basic 
> tutorial. Got no promising hits from DuckDuckGo or Google. I did some 
> unproductive roaming of https://www.winehq.org/ .

Really? "No promising hits?" I did an "install set up wine" DuckDuckGo
search and got numerous useful hits.  Of course, if you're looking for
a Bullseye specific tutorial, I doubt if you' find one -- too new.

Maybe, these will help:

  https://itsfoss.com/use-windows-applications-linux/
  https://linuxhint.com/install-use-wine-linux/

Just be forewarned, WINE is not the catchall solution to running
Windows apps: The more involved codewise the program is like games or
Photoshop, the more problems you'll have. In those cases, just run
Windows in a virtual machine which is what I do for ALL Windows apps I
need. Less or virtually no gotchas!
 
B

Re: unhappy upgrade

[Patrick Bartek]

On Mon, 4 Oct 2021 14:49:10 -0400
"Roy J. Tellason, Sr."  wrote:

> On Sunday 03 October 2021 07:48:38 pm Patrick Bartek wrote:
> > On Sun, 3 Oct 2021 12:49:12 -0400
> > "Roy J. Tellason, Sr."  wrote:  
>
>[snip]
>  
> > > > Reboot after each version upgrade, then do an apt-get
> > > > update/upgrade, etc.
> > > 
> > > Are you suggesting that I go through that for all of the
> > > versions, one after another?  
> > 
> > Yes.  You want to make sure that each version is fully up to date
> > BEFORE dist-upgrading to the next.  
> 
> How specifically do I do that?

Make sure your repo list in /etc/apt/ is correctly configured for
the version you just dist-upgraded to, then as root apt update ... apt
upgrade (NOT dist-upgrade).  That should do it.
 
> > Also, check that the repo sources list is correct for each
> > version.  
> 
> I'm assuming that you refer to /etc/apt/sources.list?  At the

Yes. Also check /etc/apt/sources.list.d for anything just in case.

> [snip]
> 
> > You can use dpkg to install, but it won't automatically take care
> > of dependencies.  I use gdebi-core, a command line utility, that
> > will install the .deb file correctly and install any dependencies.
> > You must be root to install.  
> 
> Is that another package that I should install?

It isn't installed as part of the OS. So, yes. However, if you don't
want to, use dpkg to install and apt-get -f to resolve the dependencies.

> > > > > [snip]
> > > 
> > > I have done similar in the passt,  with Slackware.  Which isually
> > > involves stickinng another hard drive in the machine.  I don't
> > > have one handy at present,  though,  and probably won't be
> > > getting one in the near-term.  
> > 
> > You really only need enough free space on a single drive to install
> > the new version.  I share the swap partition to conserve space. So,
> > I only need / and /home partitions for the new install. Some share
> > the /home partition, too, but I never have. The installer should
> > take care of setting up dual or multibooting as required.  
> 
> Okay,  now it sounds like you're talking about setting up another
> partition and doing a fresh install to that.  Which isn't what I
> did,  and I don't particularly plan to on that machine.  I've already
> done the one upgrade here,  it's somewhat broken,  though partially
> functional,  and I'd much rather spend my energies to fix that,
> rather than going that route...

I am. That's what I do.  I avoid dist-upgrading as I usually use
the OS until Long Term Support ends ... about 5 years, which is about
2 release versions. Saves a lot of time and headaches. In your case, it
would be the easiest, less problematical way to go since your first
dist-upgrade is partially broken. Just install Bullseye as a new
install, install what apps you need, copy over your data ... Done.

B

Re: unhappy upgrade

[Patrick Bartek]

On Sun, 3 Oct 2021 12:49:12 -0400
"Roy J. Tellason, Sr."  wrote:

> On Sunday 03 October 2021 12:11:13 am Patrick Bartek wrote:
> > On Sat, 2 Oct 2021 11:57:51 -0400
> > "Roy J. Tellason, Sr."  wrote:
> >   
> > > In recent messaging here I touched on how I'd determined that my
> > > workstation was way behind being current.  Apparently I needed to
> > > go 8->9->10->11.  I tried the first of those steps,  and things
> > > did not go well in a number of ways...  
> > 
> > First, back up all your data before starting the dist-upgrade
> > series.  
> 
> I did back up a bunch of stuff,  and it's a good thing I did,  or I
> would have lost a lot...
> 
> > Also, read and follow the Release Notes of dist-upgrade for each
> > version.  
> 
> Where are these to be found?

https://www.debian.org/releases/jessie/releasenotes
https://www.debian.org/releases/stretch/releasenotes
https://www.debian.org/releases/buster/releasenotes
https://www.debian.org/releases/stable/releasenotes (Bullseye)

Here's the main document page:

https://www.debian.org/doc/

> > Reboot after each version upgrade, then do an apt-get
> > update/upgrade, etc.  
> 
> Are you suggesting that I go through that for all of the versions,
> one after another?

Yes.  You want to make sure that each version is fully up to date
BEFORE dist-upgrading to the next.

Also, check that the repo sources list is correct for each version.

> > > For some reason,  the software decided to remove virtualbox.
> > > Which is a real problem,  because it's inside a virtual machine
> > > that I do all of my mail,  so I couldn't get at my mail for a
> > > while.  In Synaptic Package Manager if I try to install it I get
> > > the following error message:
> > > 
> > > "Package virtualbox has no available version, but exists in the
> > > database.  This typically means that the package was mentioned in
> > > a dependency and never uploaded,  has been obsoleted or is not
> > > available with the contents of sources.list"
> > > 
> > > I got temporarily past this by installing 8 on this laptop I'm
> > > currently typimg on.  Some things are decidedly less convenient.  
> > 
> > Complete the dist-upgrade for all versions, then reinstall
> > VirtualBox. Be sure you saved all your VMs before the
> > dist-upgrade(s).  They may be deleted.   
> 
> Yes,  I did save them and will do so again before I proceed,  since
> I'm modifying this one as I type here... 
> > I don't use the VB version in the Debian repos, but download and
> > install the .deb directly from their web site.  During the install
> > the VB repo will be set up.  
> 
> I did do a download from their site.  But it's not clear to me where
> I need to put it and how to tell the package management software
> about it.

Did you download the .deb file of it?  You can use dpkg to install, but
it won't automatically take care of dependencies.  I use gdebi-core, a
command line utility, that will install the .deb file correctly and
install any dependencies. You must be root to install.
 
> > > Firefox is majorly different,  apparently having gone from 68.9.0
> > > --> 78.14.0esr.  It no longer uses the font that I'm telling it
> > > to,  and for some odd reason won't play any youtube videos any
> > > more.  Instead I see a message on the screen that says "if video
> > > doesn't start momentarily restart your device".  Huh?
> > > 
> > > Konqueror is also broken,  I get "Could not start proess Unable to
> > > create io-slave: klauncher said: Error loading 'kio_file'"...
> > > 
> > > Trying to open a file with Okular gets me a similar message.  
> > 
> > Complete the entire dist-upgrade before trying to fix problems.
> > The problems may fix themselves.  
> 
> For all of the versions,  straight through to 11?

More or less.  Check the repos are the correct for that version,
Update/upgrade, then reboot. The Release Notes gives full
instructions.
 
> > > I don't see any way to back out of these changes,  or any obvious
> > > place to look for where the problems are.  Suggestions welcomed.  
> > 
> > I avoid dist-upgrading, if I can because of such problems, and do a
> > clean install of the newest version on a free partition keeping the
> > old install as fallback in case something goes awry.  
> 
> I have done similar in the passt,  with Slackware.  Which isually
> involves stickinng another hard drive in the machine.  I don't have
> one handy at present,  though,  and probably won't be getting one in
> the near-term.

You really only need enough free space on a single drive to install
the new version.  I share the swap partition to conserve space. So, I
only need / and /home partitions for the new install. Some share
the /home partition, too, but I never have. The installer should take
care of setting up dual or multibooting as required.

B 

Re: unhappy upgrade

[Patrick Bartek]

On Sat, 2 Oct 2021 11:57:51 -0400
"Roy J. Tellason, Sr."  wrote:

> In recent messaging here I touched on how I'd determined that my
> workstation was way behind being current.  Apparently I needed to go
> 8->9->10->11.  I tried the first of those steps,  and things did not
> go well in a number of ways...

First, back up all your data before starting the dist-upgrade series.
Also, read and follow the Release Notes of dist-upgrade for each
version. Reboot after each version upgrade, then do an apt-get
update/upgrade, etc.

> For some reason,  the software decided to remove virtualbox.  Which
> is a real problem,  because it's inside a virtual machine that I do
> all of my mail,  so I couldn't get at my mail for a while.  In
> Synaptic Package Manager if I try to install it I get the following
> error message:
> 
> "Package virtualbox has no available version, but exists in the
> database.  This typically means that the package was mentioned in a
> dependency and never uploaded,  has been obsoleted or is not
> available with the contents of sources.list"
> 
> I got temporarily past this by installing 8 on this laptop I'm
> currently typimg on.  Some things are decidedly less convenient.

Complete the dist-upgrade for all versions, then reinstall VirtualBox.
Be sure you saved all your VMs before the dist-upgrade(s).  They
may be deleted. 

I don't use the VB version in the Debian repos, but download
and install the .deb directly from their web site.  During the
install the VB repo will be set up.

> Firefox is majorly different,  apparently having gone from 68.9.0 -->
> 78.14.0esr.  It no longer uses the font that I'm telling it to,  and
> for some odd reason won't play any youtube videos any more.  Instead
> I see a message on the screen that says "if video doesn't start
> momentarily restart your device".  Huh?
> 
> Konqueror is also broken,  I get "Could not start proess Unable to
> create io-slave: klauncher said: Error loading 'kio_file'"...
> 
> Trying to open a file with Okular gets me a similar message.

Complete the entire dist-upgrade before trying to fix problems.  The
problems may fix themselves.

> I don't see any way to back out of these changes,  or any obvious
> place to look for where the problems are.  Suggestions welcomed.

I avoid dist-upgrading, if I can because of such problems, and do a
clean install of the newest version on a free partition keeping the old
install as fallback in case something goes awry.

B

Bug : Evince - Impression PDF freezes

[patrick jollant]

Bonjour,

ne sachant où exposer le problème, j'ai ouvert un sujet sur ce forum

https://debian-facile.org/viewtopic.php?pid=366818#p366818

après différentes recherches, il apparaîtrait qu'il s'agisse réellement d'un bug

Le numériseur pose également problème.

Si vous désirez plus d'informations, n'hésitez pas à me contacter

Par avance, merci

Patrick JOLLANT

Re: Install Debian netinstall to HP Elitebook 840 G8 problem

[Patrick Bartek]

On Fri, 3 Sep 2021 09:46:24 +0200 (CEST)
Richard Forst  wrote:

> I purchased a new laptop HP Elitebook 840 G8, and am trying to
> install Debian to it. However I encounter a problem. 
> 
> I change the bios setting, but when booting from usb. What was shown
> on the screen is simply a grub env command line like

When in bios, did you disable Secureboot, Fastboot, and/or enable
Legacy option?

B

Re: Bulleye sécurité

[Patrick ZAJDA]

Bonjour,


C'est parce que le nom a changé, maintenant il faut utiliser 
stable-security à la place de stable/update



Sources :

https://www.debian.org/releases///bullseye/s390x/release-notes/ch-information.fr.html#security-archive

https://wiki.debian.org/fr/DebianBullseye#FAQ



Le 16/08/2021 à 12:50, MERLIN Philippe a écrit :


Bonjour,

Depuis deux jours j'ai ce message lorsque je lance  apt update :

Le dépôt http://security.debian.org stable/updates Release ne contient 
plus de fichier Release .


Ce message ne m'inquiéterai pas trop si je n'avais eu lors d'une mise 
à jour hier un message disant que le nom d'une release n'était pas 
Bullseye mais je ne m'en rappelle plus exactement un nom contenant 
...Worn il fallait répondre par O ou N N étant proposé par défaut j'ai 
répondu O.


Avez vous le même problème.

Merci d'avance.

Philippe Merlin



--
Patrick ZAJDA

Re: Debian svensk IRC-kanal

[Patrick Franz]

Hej,

Am Donnerstag, 27. Mai 2021, 15:04:48 CEST schrieb Andreas Rönnquist:
> Hej!
> 
> Som ni kanske har märkt har freenode lagt ner, och Debian har haft en
> IRC-kanal för svensk support där (som nu har flyttats till servern
> libera_chat), som länkas på debians webb. [1]
> 
> Ska vi fortsätta att länka #debian.se på både servern libera-chat och
> på OFTC från webbsidan? Det är så pass lite trafik på dessa
> sammanlagt, så jag anser väl att en skulle absolut räcka, men jag
> vill absolut inte ta ett beslut på egen hand och trampa på någons
> tår.

Jag tycker det räcker med att länka till en om det finns så pass lite 
trafik där. Då kän vi väl gärna ta den officiella OFTC-servern.

> Topic på OFTC-kanalen länkar också se.debian.nu, vilken även den
> verkar väldigt död, har någon här koll på det?

Ja, det är bara en tom sida. Man kunde länka till t.ex. 
https://www.debian.org/index.sv.html


-- 
Med vänliga hälsningar

Patrick Franz

Re: [HS] iptables et fork tail syslog

[Patrick CAO HUU THIEN]

Le 05 May 2021 a 12:05:45 +0200, Bernard Schoenacker a écrit :

Bonjour David,
> 
> > Bonjour,
> 
> > Je cherche le moyen de lancer un truc du genre à la fin de mon script
> > iptables, mais ça ne veux pas.
> 
> > gnome-terminal -e 'tail -f /var/log/syslog' &

Avec tmux je fais un truc dans le genre pour mettre a jour les aliases
postfix:

# split with 10 lines
test -n "$TMUX_PANE" && tmux split-window -l 10 ssh "root@${SERV}" tail -f 
/var/log/messages /var/log/maillog
# update postfix aliases
ssh "root@${SERV}" "postalias $FILE && echo DONE"

Si ca t'inspire :)

amicalement
patrick

Re: clamav installation question

[Patrick Bartek]

On Fri, 26 Feb 2021 10:19:56 +0300
Semih Ozlem  wrote:

> Sorry, stopping and restarting the service seems to have solved the
> problem.

Does it work after a cold system restart without restarting it manually?
If not, my guess is some clamav or clamav-daemon dependency is not
getting loaded.  I'm not running the daemon and clamav run manually
works fine. Clamav-freshclam works fine, too, without manually starting
it.  But I'm not running a standard desktop set up. Just a window
manager and a panel.

B



> Semih Ozlem , 26 Şub 2021 Cum, 10:18
> tarihinde şunu yazdı:
> 
> > user@debian:~$ sudo systemctl status clamav-daemon.service
> > ● clamav-daemon.service - Clam AntiVirus userspace daemon
> >Loaded: loaded (/lib/systemd/system/clamav-daemon.service;
> > enabled; vendor preset: enabled)
> >   Drop-In: /etc/systemd/system/clamav-daemon.service.d
> >└─extend.conf
> >Active: inactive (dead)
> > Condition: start condition failed at Fri 2021-02-26 10:15:57 +03;
> > 1min 30s ago
> >└─
> > ConditionPathExistsGlob=/var/lib/clamav/daily.{c[vl]d,inc} was not
> > met Docs: man:clamd(8)
> >man:clamd.conf(5)
> >https://www.clamav.net/documents/
> >
> > Feb 26 10:15:57 debian systemd[1]: Condition check resulted in Clam
> > AntiVirus userspace daemon being skipped.
> >
> > Charles Curley , 26 Şub 2021 Cum,
> > 09:54 tarihinde şunu yazdı:
> >  
> >> On Fri, 26 Feb 2021 08:58:06 +0300
> >> Semih Ozlem  wrote:
> >>  
> >> > I am able to install clamav and clamav-daemon. It appears in
> >> > dpkg -l results. The problem is it is not functioning properly.
> >> > So I am wondering if anyone else had similar issues or not.  
> >>
> >> Can you be a bit more explicit? What did you get, and what were you
> >> expecting?

Re: clamav installation question

[Patrick Bartek]

On Thu, 25 Feb 2021 23:17:52 +0300
Semih Ozlem  wrote:

> Hi everyone
> 
> When I try to install clamav I am getting errots
> When I try to follow installation instructions from debian's site
> instruction
> 
> (i) clamav-data package is not available in the repository
> (ii) freshclam gives the error
> "!checkdbdir: Can't ope directory /var/lib/clamav/"
> (iii) I can't be sure if it is running or not. systemctl status
> clamav-daemon.service states it to be inactive.
> 
> when I try to follow the instructions on clamav's site
> 
> check-devel package is a requirement but is unavailable in debian
> repository.

Check your apt sources.list is set up.  Also, you have to be root to
install packages.

B

Re: Monter le partage samba IPC$

[Patrick ZAJDA]

Le 18/02/2021 à 18:58, Th.A.C a écrit :

Bonjour,


ce n'est pas un partage de fichiers, ça s'apparente plus à un point 
d'entrée et un canal de communication


d'ailleurs tu ne trouveras pas sur ton serveur le dossier associé à 
IPC$, donc tu ne peux pas le monter comme un partage de fichiers.


http://cvez.free.fr/cours/formation%20linux%20samba%20&%20squid/samba.linuxbe.org/samba.linuxbe.org/fr/samba/learn/ipc.html 



Thierry



Merci, je comprends mieux. Je sous-estimais largement la fonction de IPC$...
Du coup, y a-t-il la possibilité de monter tous les partages au sein 
d'un point de montage ?
Ou est-ce qu'il n'y a pas d'autre possibilité que créer un point de 
montage par dossier partagé ?


--
Patrick ZAJDA

Monter le partage samba IPC$

[Patrick ZAJDA]

Bonjour,


Je souhaite monter le partage samba "IPC$" (donc la racine avec tous les 
dossiers de partage) d'un serveur, vu que j'utilise tous ses partages.


Cependant, mount -t cifs -o 
uid=1000,gid=1000,credentials=/path/to/credentials //ip_serveur/IPC$ 
/mnt me retourne "No such file or directory".


Si je remplace IPC$ par un des partages de ce serveur, ça fonctionne.

smbclient -L ip_serveur m'affiche bien IPC$ dans la liste.

Y a-t-il une subtilité pour pouvoir monter ce partage spécial ou 
n'est-ce tout simplement pas possible ?



--
Patrick ZAJDA

Audio broken on Lenovo ThinkPad T14 by recent upgrade on testing

[Patrick Elsen]

Hey all,

I'm not sure if this is the right way or place to report this, but I've
been running Debian bullseye (Testing) for some months now and it's
been a really great experience, everything works out of the box
(backlight, suspend, display brightness control, switching between
speakers and headphones when plugged in). 

However, in a recent update, my audio has been broken. Now, I have
three outputs that I didn't have before (labelled HDMI1, HDMI2, HDMI3),
it doesn't recognize my internal speakers or headphones anymore,
meaning that audio playback is completely broken.

I don't know what package specfically is responsible, but I have
attached the /var/log/apt/history.log entry that was the update that I
think bricked things. I can provide more logs and information if
needed.

My guess would be that alsa-topology-conf or alsa-ucm-conf are the
culprit, because those two are the only ones I see that are related to
sound and audio things?

Maybe someone here can help me or look into it.

Cheers,
-Patrick
Start-Date: 2020-12-27  23:06:21
Commandline: apt upgrade
Requested-By: patrick (1000)
Install: libdeflate0:amd64 (1.6-1, automatic), libadwaitaqtpriv1:amd64 
(1.2.0-2, automatic), llvm-11-runtime:amd64 (1:11.0.0-5+b1, automatic), 
libges-1.0-0:amd64 (1.18.2-1, automatic), libgit2-1.1:amd64 (1.1.0+dfsg.1-2, 
automatic), liblist-moreutils-xs-perl:amd64 (0.430-2, automatic), 
containerd:amd64 (1.4.3~ds1-1, automatic), libboost-locale1.74.0:amd64 
(1.74.0-3+b1, automatic), libadwaitaqt1:amd64 (1.2.0-2, automatic), 
llvm-11-dev:amd64 (1:11.0.0-5+b1, automatic), gedit-plugin-text-size:amd64 
(3.38.1-1, automatic), libplymouth5:amd64 (0.9.5-2, automatic), 
libxkbregistry0:amd64 (1.0.3-2, automatic), clang-format-11:amd64 
(1:11.0.0-5+b1, automatic), ruby-rubygems:amd64 (3.2.0~rc.2-5, automatic), 
llvm-11-tools:amd64 (1:11.0.0-5+b1, automatic), linux-image-5.9.0-5-amd64:amd64 
(5.9.15-1, automatic), llvm-11:amd64 (1:11.0.0-5+b1, automatic)
Upgrade: speech-dispatcher-espeak-ng:amd64 (0.10.2-1, 0.10.2-2), 
gir1.2-lokdocview-0.1:amd64 (1:7.0.3-4+b1, 1:7.0.4~rc2-1+b1), 
speech-dispatcher-audio-plugins:amd64 (0.10.2-1, 0.10.2-2), fdisk:amd64 
(2.36.1-2, 2.36.1-3), alsa-topology-conf:amd64 (1.2.3-1, 1.2.4-1), 
texlive-plain-generic:amd64 (2020.20200925-1, 2020.20201129-2), 
node-os-locale:amd64 (4.0.0-1, 5.0.0-1), libjson-glib-1.0-0:amd64 (1.6.0-1, 
1.6.0-2), libxencall1:amd64 (4.14.0+80-gd101b417b7-1, 4.14.0+88-g1d1d1f5391-2), 
manpages:amd64 (5.09-2, 5.10-1), perl-base:amd64 (5.32.0-5, 5.32.0-6), 
init:amd64 (1.59, 1.60), libreoffice-math:amd64 (1:7.0.3-4+b1, 
1:7.0.4~rc2-1+b1), init-system-helpers:amd64 (1.59, 1.60), libssh2-1:amd64 
(1.8.0-2.1, 1.9.0-2), node-resolve:amd64 (1.19.0+~cs4.20.2-1, 
1.19.0+~cs5.20.8-2), aspell-pt-pt:amd64 (20201103-2, 20201212-1), 
libgovirt2:amd64 (0.3.7-1, 0.3.7-2), python3-pyside2.qtwidgets:amd64 
(5.15.0-5+b1, 5.15.0-6), liblibreoffice-java:amd64 (1:7.0.3-4, 1:7.0.4~rc2-1), 
fonts-lyx:amd64 (2.3.5.2-1, 2.3.6-1), bsdextrautils:amd64 (2.36.1-2, 2.36.1-3), 
python3-cupshelpers:amd64 (1.5.12-1, 1.5.13-1), texlive-metapost-doc:amd64 
(2020.20200925-1, 2020.20201203-2), texlive-font-utils:amd64 (2020.20201129-1, 
2020.20201129-2), texlive-lang-spanish:amd64 (2020.20200925-2, 
2020.20201129-1), libseccomp2:amd64 (2.5.0-3+b1, 2.5.1-1), 
util-linux-locales:amd64 (2.36.1-2, 2.36.1-3), libencode-locale-perl:amd64 
(1.05-1, 1.05-1.1), texlive-publishers-doc:amd64 (2020.20201129-1, 
2020.20201129-2), uno-libs-private:amd64 (1:7.0.3-4, 1:7.0.4~rc2-1+b1), 
libao4:amd64 (1.2.2+20180113-1+b1, 1.2.2+20180113-1.1), libaudit-common:amd64 
(1:2.8.5-3.1, 1:3.0-1), uim-data:amd64 (1:1.8.8-6.1, 1:1.8.8-7), 
adwaita-qt:amd64 (1.1.4-1, 1.2.0-2), uim-fep:amd64 (1:1.8.8-6.1+b2, 1:1.8.8-7), 
texlive-latex-base:amd64 (2020.20200925-1, 2020.20201203-2), uuid-runtime:amd64 
(2.36.1-2, 2.36.1-3), libquvi-scripts-0.9:amd64 (0.9.20131130-1.1, 
0.9.20131130-1.2), libqt5multimediagsttools5:amd64 (5.15.1-2, 5.15.2-2), 
python3-zope.interface:amd64 (5.1.2-1, 5.1.2-1+b1), iportuguese:amd64 
(20201103-2, 20201212-1), libfdisk1:amd64 (2.36.1-1, 2.36.1-3), 
publicsuffix:amd64 (20200729.1725-1, 20201215.0006-1), javascript-common:amd64 
(11, 11+nmu1), libauthen-sasl-perl:amd64 (2.1600-1, 2.1600-1.1), 
libdrm-nouveau2:amd64 (2.4.103-1, 2.4.103-2), osinfo-db:amd64 (0.20201119-1, 
0.20201218-1), hunspell-fr:amd64 (1:6.4.1-1, 1:7.0-1), 
libreoffice-l10n-zh-cn:amd64 (1:7.0.3-4, 1:7.0.4~rc2-1), libridl-java:amd64 
(1:7.0.3-4, 1:7.0.4~rc2-1), node-jquery:amd64 (3.5.1+dfsg+~3.5.4-2, 
3.5.1+dfsg+~3.5.4-3), librygel-renderer-gst-2.6-2:amd64 (0.38.3-1, 0.40.0-1), 
libreoffice-l10n-zh-tw:amd64 (1:7.0.3-4, 1:7.0.4~rc2-1), linux-libc-dev:amd64 
(5.9.9-1, 5.9.15-1), liblist-moreutils-perl:amd64 (0.416-1+b6, 0.430-2), 
libxml-sax-base-perl:amd64 (1.09-1, 1.09-1.1), libpython3.8-dev:amd64 (3.8.6-1, 
3.8.7-1), llvm:amd64 (1:9.0-49.1, 1:11.0-51+nmu1), alsa-ucm-conf:amd64 
(1.2.3-1, 1.2.4-1), feynmf:amd64 (1.08-11, 1.08-12), 
libuno

Re: Open Source Flash Alternatives

[Patrick Bartek]

On Fri, 1 Jan 2021 18:17:32 -0500
Kenneth Parker  wrote:

> On Fri, Jan 1, 2021, 10:29 AM Kenneth Parker 
> wrote:
> 
> > Since Adobe Flash is going the way of the Dodo Bird, I thought I'd
> > check up on Open Source Alternatives, since I have some
> > Standalone .swf files (games, etc).
> >
> > Two came up in my searches, Lightspark (which seems to be, mainly a
> > Plug-in), and Ghu Gnash, which I like better, since it works as a
> > Standalone Program.  Unfortunately, it doesn't seem available in
> > Bullseye (and Search didn't find it for Buster either).  It DOES
> > show up in Stretch, but that's on a different machine than where
> > my .swf files are. 
> 
> Update:  I just installed Stretch in QEMU-KVM in my "Knockabout" Mint
> 20 system and was able to install Gnash on it.  I did a bit of
> research into Gnu Gnash, seeing that it may not be currently
> maintained.  I wonder if that decision was made before Flash was
> going Extinct?
> 
> I don't consider this closed, because of the large number of
> Standalone .swf files. But I have something to test and, may contact
> gnu.org about Gnash.

Take a look at Ruffle, a "Flash player emulator" or so they say.
Open source. Available for Windows, Mac and Linux.  Haven't tried it
myself. No old Flash files laying around to test.

  https://ruffle.rs/

B

Impossible de ddémarrer VLC avec Optirun

[Patrick ZAJDA]

Bonjour,


J'ai installé et configuré Bumblebee sous Debian Buster à fin de 
n'utiliser que macarte Nvidia quand nécessaire.


Je souhaite l'utiliser pour VLC car avec le GPU Intel, la machine freeze 
tout simplement, je n'ai plus qu'à redémarrer si j'essaye de lire une 
vidéo quelle qu'elle soit et je n'ai trouvé aucune autre solution.


En écrivant "ON" dans /proc/acpi/bbswitch et démarrant la vidéo dans 
VLC ça fonctionne.


Par contre, si je fais "optirun vlc", j'ai simplement l'interface de VLC 
en ligne de commande.


Voilà ce qui ressort dans le terminal :

$ optirun vlc
VLC media player 3.0.11 Vetinari (revision 3.0.11-0-gdc0c5ced72)
[55e80faab570] main libvlc: Lancement de vlc avec l'interface par 
défaut. Utiliser « cvlc » pour démarrer VLC sans interface.

../include/vlc_xlib.h:46:vlc_xlib_init: Xlib not initialized for threads.
This process is probably using LibVLC incorrectly.
Pass "--no-xlib" to libvlc_new() to fix this.
[55e80fb980c0] qt interface error: Xlib not initialized for threads
../include/vlc_xlib.h:46:vlc_xlib_init: Xlib not initialized for threads.
This process is probably using LibVLC incorrectly.
Pass "--no-xlib" to libvlc_new() to fix this.
[55e80fb980c0] skins2 interface error: Xlib not initialized for threads
[55e80fb980c0] skins2 interface error: initializing xlib for 
multi-threading failed

[55e80fb980c0] skins2 interface error: cannot initialize OSFactory
[55e80faaf4e0] main playlist: playlist is empty
[55e80fb980c0] [cli] lua interface: Listening on host "*console".
VLC media player 3.0.11 Vetinari
Command Line Interface initialized. Type `help' for help.
> quit
Shutting down.
[55e80fb980c0] [cli] lua interface: Requested shutdown.


à toutes fins utiles :

$ lspci | grep VGA
00:02.0 VGA compatible controller: Intel Corporation Device 9ba4 (rev 05)
01:00.0 VGA compatible controller: NVIDIA Corporation Device 1f99 (rev ff)


Autre constatation : si je fais le lspci ci-dessus après le optirun, la 
machine freeze.


En fait ça semble le faire si la carte NVIDIA est sur off mais pas 
toujours.



Quelques spécifications matériel :

- Processeur : Intel Core i5 10200H ;

- Carte graphique : NVIDIA GeForce GTX 1650


Dans /etc/bumblebee/bumblebee.conf j'ai spécifié Bridge=primus dans la 
section optirun mais ça n'a rien changé.


J'ai installé le driver propriétaire, version backports, pareil pour 
la version de bumblebee et primus.


J'ai utilisé les paquets bumblebee-nvidia et primus-nvidia pour 
m'assurer d'avoir tout vu que j'utilise le driver propriétaire.


Version du kernel : 5.9.0-0.bpo.2-amd64

J'ai pris le meta-paquet firmware-linux dans sa version backports.

Vu que la machine se figeait dès le démarrage, conformément à ce qui 
est indiqué sur le wiki fr de Debian j'ai appliquer la solution au bug 
suivant : https://github.com/Bumblebee-Project/Bumblebee/issues/1036


Pour être sûr, j'ai également créé et activé un service SystemD 
pour réactiver le GPU NVIDIA à l'arrêt de la machine.



Je pense n'avoir rien oublié de mes recherches et manipulations :)


En faisant une recherches spécifique sur le problème de VLC, je ne 
trouve que des résultats qui datent de versions antérieures et qui 
spécifient que le problème a été corrigé dans VLC...



Quelqu'un a-t-il rencontré ce genre de problème ou aurait une idée de 
piste à part faire echo ON|sudo tee /proc/acpi/bbswitch avant de 
démarrer VLC puis faire la même chose mais en écrivant OFF au lieu de 
ON en fermant VLC ?


à la limite, que puis-je faire pour ne pas avoir à activer la carte 
NVIDIA ? En somme pouvoir utiliser VLC avec la carte Intel...



Merci d'avoir eu le courage de lire ce pavé...


Bon dimanche,


Patrick

Re: Comment démarrer un démon sous Debian avec systemd?

[Patrick ZAJDA]

Le 21/11/2020 à 17:15, Gaëtan Perrier a écrit :

Le samedi 21 novembre 2020 à 15:39 +0100, Patrick ZAJDA a écrit :

Hello,



Histoire de commencer par la base et parce que je ne l'ai pas lu dans ton
message d'origine, as-tu bien fait :

systemctl enable sync-perioadically.service





Il ne faut pas faire un

systemctl  start sync-periodically

ensuite en plus ?

Gaëtan

Effectivement, pour démarrer immédiatement le service...
Vu que la question était de savoir comment faire pour que le service se 
lance automatiquement au démarrage, je n'avais pas pensé à apporter 
cette précision.

--
Patrick ZAJDA
Certification NVDA 2019

Re: Comment démarrer un démon sous Debian avec systemd?

[Patrick ZAJDA]

Hello,


Histoire de commencer par la base et parce que je ne l'ai pas lu dans 
ton message d'origine, as-tu bien fait :


systemctl enable sync-perioadically.service


à moins que c'était ce que tu voulais dire par installer...


Patrick


Le 21/11/2020 à 15:19, Basile Starynkevitch a écrit :


Bonjour la liste,

Sur un serveur domestique sous Debian/Buster (vielle machine de marque 
HP Proliant, qu'on m'a gentillement donné, processeur Intel(R) 
Xeon(R) CPU E5-2609, donc Debian amd64)


uname -a donne

Linux ravel 4.19.0-12-amd64 #1 SMP Debian 4.19.152-1 (2020-10-18) 
x86_64 GNU/Linux


Son /sbin/init est en fait un /lib/systemd/systemd (paquet 
systemd-sysv   241-7~deb10u4 ...). Le paquet /systemd/ est 
installé, en version 241. Et systemd --version sous root donne


systemd 241 (241)
+PAM +AUDIT +SELINUX +IMA +APPARMOR +SMACK +SYSVINIT +UTMP 
+LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS 
+KMOD -IDN2 +IDN -PCRE2 default-hierarchy=hybrid


J'ai codé un petit démon qui appelle sync(2) régulièrement et qui 
avec l'option -d utilise daemon(3). Le but étant de perdre peu de 
données en cas de coupure de courant Les disques sont rotatifs et 
anciens, mais RAID (il y a un onduleur, mais pas connecté en USB à 
cette machine)


Le code source en C (sous licence GPLv3) de ce démon est en 
https://github.com/bstarynk/misc-basile/blob/master/sync-periodically.c 
<https://github.com/bstarynk/misc-basile/blob/master/sync-periodically.c> 
et je l'ai compilé en /usr/local/bin/sync-periodically


Je voudrais qu'au démarrage soit lancé 
/usr/local/bin/sync-periodically -d -L 3600 -Y 3 -P 
/var/run/sync-periodically.pid -ou quelque chose de similiare- sous le 
user root (ou peut-être sync)



J'ai tenté de coder 
https://github.com/bstarynk/misc-basile/blob/master/sync-periodically.service 
que j'ai installé sous root  en 
/etc/systemd/system/sync-perioadically.service


Mais quand je reboote ce serveur ravel, sync-periodically ne redemarre 
pas.


Comment faire pour un redemarrage automatique?

Librement

--
Basile Starynkevitch
(only mine opinions / les opinions sont miennes uniquement)
92340 Bourg-la-Reine, France
web page: starynkevitch.net/Basile/


--
Patrick ZAJDA
Certification NVDA 2019

Re: MX-Linux 19.2

[Patrick Bartek]

On Tue, 17 Nov 2020 23:11:56 +0200
ellanios82  wrote:

> On 11/17/20 10:54 PM, Dan Ritter wrote:
> > ellanios82 wrote:  
> >> ??Dear List
> >>
> >>
> >> ??- am using MX-Linux 19.2 :
> >>
> >>  What steps to upgrade to 19.3 please ?
> >>  
> > 1. Find a mailing list or forum for MX-Linux.
> >
> > 2. Ask for instructions.
> >
> > You might consider reading www.mxlinux.org.
> >
> > -dsr-  
> 
> - Thank you
> 
> [ did look for mailing list for MX-Linux but could not locate]

I don't think MX-Linux has a mail list, but they do have a forum:

  https://forum.mxlinux.org/

Here's the Support page:

https://mxlinux.org/support/

B

Re: An old box running Debian 8

[Patrick Bartek]

On Wed, 11 Nov 2020 17:40:50 +0100
Miroslav Skoric  wrote:

> I have an old comp (CPU Pentium II Celeron 400 MHz, 224 MB RAM)
> running ham radio server in Debian 8. It works well in CLI, but very
> slow after starting GUI. I wonder whether it would be worth to try
> (if possible at all) to upgrade it to Debian 9. Any experience with
> such old boxes?

You are trying to do what we call in the US, a Fool's Errand, that
is, a fruitless undertaking.  If you could upgrade the RAM to 512MB or
even 1 GB, you might get usable performance with Debian 8 and a
lightweight GUI environment or, better yet, a window manager, but
certainly not with GNOME or KDE. Let me give you an example:

About 10 years ago, I installed Debian 7 on an Asus EeePC 900 with a
900MHz Celeron and 512MB RAM. I tried GNOME first, but even then it was
too much a resources behemoth to even work. LXDE was lighter;
however, even with only a browser running, system performance was
slow, but usable, if you were patient. Upgrading RAM to 1GB made all
the difference in the world turning a barely usable system into one
that while not screaming fast was adequate for simple web browsing,
video streaming, email, etc. which was what it was intended for.

So, first, before changing anything else, see if upgrading RAM does any
good with 8.

B 

Re: Replacement Email Client

[Patrick Bartek]

On Mon, 26 Oct 2020 21:23:59 -0400
Carl Fink  wrote:

> On 10/26/20 6:16 PM, Patrick Bartek wrote:
> > On Sun, 25 Oct 2020 20:45:50 -0400
> > Carl Fink  wrote:
> >  
> >> On 10/25/20 8:28 PM, Patrick Bartek wrote:  
> >>> I'm not referring to viewing HTML emails. I already can do that in
> >>> Claws-Mail using its Dillo plugin. I'm talking about filling in
> >>> forms, etc. that are part of the HTML email and sending just the
> >>> data without "replying" in the normal sense.  This is beyond
> >>> Claws' and Dillo's capabilities.  I have to use a real browser,
> >>> log into that particular web mail account (like gmail), click on
> >>> that particular email, etc. to do so.
> >>>
> >>> I'm getting the sense that I may not be able to find a client that
> >>> can do that.  
> >> I don't remember ever getting an emailed form that was anything but
> >> a link to a web page. Who is sending you emailed inline forms?  
> > The ones I respond to are known to me and are legit --
> > organizations, businesses, government agencies, etc. -- that I do
> > business with.  To respond, I must switch to a web browser, login to
> > my email account (like gmail), find that particular email, and
> > enter the requested data either by keyboard, drop-down menu,
> > buttons, etc., then SUBMIT it.  This happens all within the email.
> > I never get forwarded to another web site. I always stay on the web
> > mail page. As far as I can tell only the data is sent. The email
> > itself is not replied to.  That is, there's nothing in the "Sent"
> > folder.  
> 
> Meaning no offense, I doubt it. I have been using email since before
> Gopher, and I have literally never received an HTML  email with an
> embedded form (that I opened, at least). I find it hard to believe
> that you get many of them.

I don't know if there is a  or not.  Someone else suggested
that.  It may be javascript.  Never checked email's code all that
closely.  Next time I get one of those type emails, I'll look.

> I think you're getting normal email with a link to a form, but (as
> you say) Dillo doesn't let you click the link, so you never realize
> what it is. Those "rate us from 1-5" things are normally five
> different links to the same online poll, with a parameter telling the
> page what rating you selected. If You opened the links in an
> HTML-aware mailer like Mutt, you could just select one of those to
> open the form in a browser.

Yes, sometimes when I use a browser for these HTML emails and click on
something, a new tab with a new URL opens.  Sometime not. Next time,
I'll check the code.

Not that it matters all that much.  All I want is a lightweight email
client that works with HTML emails, too, so I don't have to switch back
and forth to a browser, login, etc. to get things done.  

B

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 21:57:07 -0400
Dan Ritter  wrote:

> Carl Fink wrote: 
> > On 10/25/20 9:17 PM, John Hasler wrote:  
> > > Carl writes:  
> > > > I don't remember ever getting an emailed form that was anything
> > > > but a link to a web page.  
> > > I think that may be what he means.  
> > 
> > Can't be. He refers to having to log into a mail account
> > in the browser. That is never required for these mailed
> > links to forms--the form is not in your mailbox and can't
> > require you to log into GMail or whatever.  
> 
> I think we are all confused and the original questioner needs to 
> provide an example of the very strange email that they want
> to work with.

Okay.  Here's a trivial one to keep it simple:

Recently I went to my bank in person. A few days later I get an HTML
email wanting to know my level of satisfaction for the service I
received: 1 to 10, worse to best. There was a gadget in the email to
enter the number, and a SUBMIT button. That's it. I could view all this
in Claws through the Dillo plugin, but of course nothing is clickable.
So, would have to do the browser thing to respond. Not that I responded.
Too much a bother.

FWIW, all emails I get now from businesses, government agencies,
Windows users, etc. are HTML-based.  Text-based has mostly gone the way
of the dodo.

B

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 20:45:50 -0400
Carl Fink  wrote:

> On 10/25/20 8:28 PM, Patrick Bartek wrote:
> > I'm not referring to viewing HTML emails. I already can do that in
> > Claws-Mail using its Dillo plugin. I'm talking about filling in
> > forms, etc. that are part of the HTML email and sending just the
> > data without "replying" in the normal sense.  This is beyond Claws'
> > and Dillo's capabilities.  I have to use a real browser, log into
> > that particular web mail account (like gmail), click on that
> > particular email, etc. to do so.
> >
> > I'm getting the sense that I may not be able to find a client that
> > can do that.  
> 
> I don't remember ever getting an emailed form that was anything but
> a link to a web page. Who is sending you emailed inline forms?

The ones I respond to are known to me and are legit --
organizations, businesses, government agencies, etc. -- that I do
business with.  To respond, I must switch to a web browser, login to
my email account (like gmail), find that particular email, and
enter the requested data either by keyboard, drop-down menu, buttons,
etc., then SUBMIT it.  This happens all within the email. I never get
forwarded to another web site. I always stay on the web mail page. As
far as I can tell only the data is sent. The email itself is not
replied to.  That is, there's nothing in the "Sent" folder.

This happens often enough now to make having to switch to the browser
instead of being able to respond with my usual email client an annoying
inconvenience. Hence, my search for a new email client.

B 

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 14:07:00 -0500
John Hasler  wrote:

> Patrick Bartek writes:
> > But I need to view the entire email: images, graphics, etc. and be
> > able to interact with all the links, etc. and not just view them.
> > Want to get away from having to login to the mail account with a
> > browser to do so.  So, EMACS won't work for me.  
> 
> Log in to what mail account?  Gnus calls the browser and passes the
> HTML attachment to it.  No logging in involved.

I'm not referring to viewing HTML emails.  I already can do that in
Claws-Mail using its Dillo plugin. I'm talking about filling in forms,
etc. that are part of the HTML email and sending just the data without
"replying" in the normal sense.  This is beyond Claws' and Dillo's
capabilities.  I have to use a real browser, log into that particular
web mail account (like gmail), click on that particular email, etc. to
do so.

I'm getting the sense that I may not be able to find a client that can
do that.

B

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 11:42:20 -0700 (PDT)
didier gaumet  wrote:

> Le dimanche 25 octobre 2020 à 19:00:08 UTC+1, Patrick Bartek a écrit :
> 
> > Already have Dillo set up, but I need more than a viewer. I want a 
> > client that handles HTML as well as plain text emails, so I don't
> > have to login into the mail account via a browser to interact with
> > the email.  
> 
> I just tested the dillo plugin: I do not know if it is sufficient for
> your usage but after setting the plugin up, I was able to click on
> the links into the emails and most (not all) images were displayed

I have Dillo configured NOT to show images initially. Do that for
security. Reloading page usually brings them up. It will follow some
links to other pages, but for viewing purposes only. One cannot
interact and transmit data as I need to.  I have to use a real browser
for that. And that means logging in through the browser to that
particular email account.

> [...]
> > I installed Balsa in Devuan Beowulf which I'm testing, and it only 
> > installed a few libraries. Don't know if the same is true with
> > Buster. I'm still evaluating it. No opinion as yet. Wasn't able to
> > get it to receive mails, but could send. Probably an erroneous
> > setting somewhere. It was late.   
> 
> Just tested Balsa here (Buster+Gnome): I could read my emails but
> HTML was not rendered

Probably a missed setting somewhere.  But the more I look at Balsa, the
more I think it's just, for all practical purposes, a clone of
Claws-Mail.  I could be wrong.  We'll see.  Installing it for the third
try after purging it twice.

> If Claws HTML  rendering with the Dillo plugin is not satisfying
> enough for you, I fear you will need a heavy client like Thunderbird,
> Evolution or Kmail...

Oh, I hope not!!

B 

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 13:10:04 +0100
Michael  wrote:

> On Sunday, October 25, 2020 1:04:00 AM CEST, Patrick Bartek wrote:
> 
> > Any suggestions?
> 
> since this is a debian user mailinglist, my answer will be slightly
> off topic.
> 
> but i use 'trojita' (https://trojita.flaska.net) as a pure and simple
> imap based email client. but it is not in the debian repositories.

I've heard of it.  Haven't researched it.

> it does only one thing: handling email of one single imap account,
> but it does it well enough for me, so i stick with it. i don't like
> the other bloatware out there (kmail, thunderbird, claws-mail, etc.).

I have multiple imap email accounts.  If it can only handle a single
account, will be unsuitable.

> maybe it's worth a shot. ymmv
> 
>

Thanks

B

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 01:51:43 -0700 (PDT)
didier gaumet  wrote:

> Le dimanche 25 octobre 2020 à 01:10:06 UTC+2, Patrick Bartek a écrit :
> > Hi! All, 
> > 
> > Looking for recommendations for a lightweight email client that
> > will handle HTML as well as plain text to replace Claws-Mail. Have
> > been using Claws-Mail for years and before it Sylpheed. Claws used
> > to have a basic HTML plugin renderer which was sufficient, but
> > latest version does not.   
> 
> It seems that in the past, Debian provided the
> claws-mail-fancy-plugin package (a GTK2 HTML viewer). That is no
> longer the case but Debian is now providing  the
> claws-mail-dillo-viewer (a Dillo HTML viewer), and in addition to

Already have Dillo set up, but I need more than a viewer.  I want a
client that handles HTML as well as plain text emails, so I don't have
to login into the mail account via a browser to interact with the email.


> this, from Bulleye on, will be providing the
> claws-mail-litehtml-viewer package (a HTML viewer based on the
> litehml library).
> https://packages.debian.org/search?suite=all=all=any=all=claws+html
> 
> So, apparently, you do not need to replace claws-mail

Probably, I do.  Just a viewer plugin is insufficient.  Been using
viewers for years.  Time to move on.  Or, maybe, an interactive HTML
interpreter plugin?

> [...]
> > Both Thunderbird and Balsa have been rejected as T'bird is a
> > behemoth and no longer in development; and with Balsa, I don't want
> > to have to deal with GNOME-systemd, etc. dependencies. (I don't run
> > GNOME anyway, only a window manager Openbox, and use sysvinit and
> > not systemd as init.)   
> [...]
> 
> - As others have already stated, Thunderbird is still in development
> - I somewhat doubt installing Balsa will draw on any real Gnome or
> Systemd stuff
> https://packages.debian.org/search?suite=all=all=any=names=balsa

I installed Balsa in Devuan Beowulf which I'm testing, and it only
installed a few libraries.  Don't know if the same is true with Buster.
I'm still evaluating it. No opinion as yet. Wasn't able to get it to
receive mails, but could send. Probably an erroneous setting somewhere.
It was late.

Thanks for your input.

B

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 07:19:28 +0200
Teemu Likonen  wrote:

> * 2020-10-24 16:04:00-07, Patrick Bartek wrote:
> 
> > Looking for recommendations for a lightweight email client that will
> > handle HTML as well as plain text to replace Claws-Mail.  Have been
> > using Claws-Mail for years and before it Sylpheed.  
> 
> GNU Emacs mail clients "Gnus" and "Notmuch Emacs" automatically render
> HTML mail nicely as plain text. User can can also open HTML and other
> MIME parts in external viewer like web browser.
> 

I can already view the text of HTML emails with Claws-Mail.  But I need
to view the entire email: images, graphics, etc. and be able to interact
with all the links, etc. and not just view them.  Want to get away from
having to login to the mail account with a browser to do so.  So, EMACS
won't work for me.

Thanks for your suggestion.

B

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 05:55:02 +0100
Oliver Schoede  wrote:

> On Sat, 24 Oct 2020 16:04:00 -0700
> Patrick Bartek  wrote:
> 
> >
> >Looking for recommendations for a lightweight email client that will
> >handle HTML as well as plain text to replace Claws-Mail.  Have been
> >using Claws-Mail for years and before it Sylpheed.  Claws used to
> >have a basic HTML plugin renderer which was sufficient, but latest
> >version does not.  
> 
> I didn't even know it was gone, but apparently 3.17.7 just
> (re)introduced a light viewer duly taking care of people who need it.
> LiteHTML. It's packaged as claws-mail-litehtml-viewer, can't say if
> it's any good, there are other options though. Even something
> based on Dillo! ;) Not bad for a project with a different set of
> priorities. Don't like a plugin? Feeling more like using behemoth
> Firefox ('hate accessing mail through a browser'):
> 
> Config -> Preferences -> Message View -> External Programs
> 
> [snip]

I need more than just an HTML viewer.  I have dillo set up for that. It
is the default with the install. I need to be able to interact with the
HTML email without having to login via a browser to do so which I've
grown tired of.

Thanks for your reply.

B

Re: Replacement Email Client

[Patrick Bartek]

On Sat, 24 Oct 2020 21:52:54 -0400
Gene Heskett  wrote:

> On Saturday 24 October 2020 19:04:00 Patrick Bartek wrote:
> 
> > Hi! All,
> >
> > Looking for recommendations for a lightweight email client that will
> > handle HTML as well as plain text to replace Claws-Mail.  Have been
> > using Claws-Mail for years and before it Sylpheed.  Claws used to
> > have a basic HTML plugin renderer which was sufficient, but latest
> > version does not. And I'm getting more and more important emails in
> > HTML where just the plain text is insufficient to fully read the
> > email.  That is, text in (or as) images contain some (or much) of
> > the content, etc. Also, hate accessing email through a browser. So
> > that option is out.
> >
> > Both Thunderbird and Balsa have been rejected as T'bird is a
> > behemoth and no longer in development; and with Balsa, I don't want
> > to have to deal with GNOME-systemd, etc. dependencies. (I don't run
> > GNOME anyway, only a window manager Openbox, and use sysvinit and
> > not systemd as init.)
> >
> > Any suggestions?
> >  
> Yes, kmail, as supplied NOT from KDE but from TDE. Its the older
> 

I'll look into it.

B

Re: Replacement Email Client

[Patrick Bartek]

On Sat, 24 Oct 2020 19:50:14 -0500
Leslie Rhorer  wrote:

> On 10/24/2020 6:04 PM, Patrick Bartek wrote:
> > 
> > Hi! All,
> > 
> > Looking for recommendations for a lightweight email client that will
> > handle HTML as well as plain text  
> 
>   I want to know that, myself.
> 
> > deal with GNOME-systemd, etc. dependencies. (I don't run GNOME
> > anyway, only a window manager Openbox, and use sysvinit and not
> > systemd as  
> 
>   I don't run Gnome, either, but I did take the plunge and
> learn enough about systemd to get along.  I still don't like it, and
> it remains a very obtuse system designed to do things that fail to
> impress me.  It seems it may be inevitable to become universal.
> 

As long as there are those who don't want systemd, someone out there
will fill that need. That's why Devuan exists (and some others).

I've been testing Devuan Beowulf (Buster) with the Openbox window
manager (my normal configuration) for several weeks now and it's stable,
has all the apps that Debian does. Works just fine without systemd. So,
it can be done. Time will tell.

B 

Re: Replacement Email Client

[Patrick Bartek]

On Sun, 25 Oct 2020 17:22:35 +1100
Keith Bainbridge  wrote:

> On 25/10/20 12:52 pm, Gene Heskett wrote:
> > Both Thunderbird and Balsa have been rejected as T'bird is a
> > behemoth and no longer in development;  
> 
> 
> Um.   Why then am I getting new versions every few weeks?
> Currently 83.0a1.  Sure it's bigger than claws, but many processes are
> easier in day to day operation.

Mozilla has ceased the development; however, it is still "supported" --
security, bug fixes? -- but privately.  Maybe, a group other than
Mozilla has taken over development.  Don't know.

B

Replacement Email Client

[Patrick Bartek]

Hi! All,

Looking for recommendations for a lightweight email client that will
handle HTML as well as plain text to replace Claws-Mail.  Have been
using Claws-Mail for years and before it Sylpheed.  Claws used to have
a basic HTML plugin renderer which was sufficient, but latest version
does not. And I'm getting more and more important emails in HTML where
just the plain text is insufficient to fully read the email.  That is,
text in (or as) images contain some (or much) of the content, etc.
Also, hate accessing email through a browser. So that option is out.

Both Thunderbird and Balsa have been rejected as T'bird is a behemoth
and no longer in development; and with Balsa, I don't want to have to
deal with GNOME-systemd, etc. dependencies. (I don't run GNOME anyway,
only a window manager Openbox, and use sysvinit and not systemd as
init.)

Any suggestions?  

Thanks

B

Re: can't boot to a graphical interface.

[Patrick Bartek]

On Sat, 3 Oct 2020 13:39:22 -0400
Frank McCormick  wrote:

> On 10/3/20 11:20 AM, Kent West wrote:
> > 
> > 
> > On Sat, Oct 3, 2020 at 8:01 AM songbird  > > wrote:
> > 
> > Frank McCormick wrote:  
> >  > While compiling an application today my Debian bullseye
> >  > system  
> > somehow  
> >  > got messed up. It will boot to a CLI but no X, apparently
> >  > because  
> > for  
> >  > some reason the system is unable to access some files in
> >  > /usr/share/dbus-1. It keeps saying access denied. The
> >  > directories  
> > and  
> >  > files are owned by root, and if I noot to a CLI I have no
> >  > trouble accessing them using sudo and midnight commander.
> >  >
> >  > I tried reinstalling systemd but it ended with the same
> >  > problem.  
> > 
> > 
> > On my bullseye/sid lappie, for comparison to yours:
> > 
> > kent@westk-9463:~$ ls -lah /usr/share/dbus-1/
> > total 84K
> > drwxr-xr-x   8 root root 4.0K Sep 17 19:55 .
> > drwxr-xr-x 610 root root  20K Sep  5 16:26 ..
> > drwxr-xr-x   2 root root 4.0K Aug  6 20:50 accessibility-services
> > drwxr-xr-x   2 root root  12K Aug  6 21:03 interfaces
> > drwxr-xr-x   2 root root  20K Aug  6 20:55 services
> > -rw-r--r--   1 root root 3.6K Jul  2 08:19 session.conf
> > drwxr-xr-x   2 root root 4.0K Sep 17 19:55 session.d
> > -rw-r--r--   1 root root 5.7K Jul  2 08:19 system.conf
> > drwxr-xr-x   2 root root 4.0K Sep 17 19:55 system.d
> > drwxr-xr-x   2 root root 4.0K Aug  6 21:08 system-services
> > 
> > kent@westk-9463:~$ ls -lahd /usr/share/dbus-1/
> > drwxr-xr-x 8 root root 4.0K Sep 17 19:55 /usr/share/dbus-1/
> >   
> 
> 
> My system now is back to **normal**. I re-installed
> about a dozen packages including systemd.
> I also changed one mode  to match yours.
> 
> frank@franklin:~$ ls -lah /usr/share/dbus-1/
> total 44K
> drwxr-xr-x   6 root root 4.0K Oct  3 10:08 .
> drwxr-xr-x 246 root root  12K Oct  2 16:56 ..
> drwxr-xr-x   2 root root 4.0K Oct  3 10:08 interfaces
> drwxr-xr-x   2 root root 4.0K Oct  3 09:20 services
> -rw-r--r--   1 root root 3.6K Jul  2 09:19 session.conf
> -rw-r--r--   1 root root 5.7K Jul  2 09:19 system.conf
> drwxr-xr-x   2 root root 4.0K Oct  3 10:08 system.d
> drwxr-xr-x   2 root root 4.0K Oct  3 10:08 system-services
> 
> Still don't know what the h*ll happened.

You are using "testing" and an Alpha 2 Release after all.  So, expect
the occassional "gotcha."  It's the nature of the beast.

B