You are not logged in.
If you install yt-dlp as Ron suggests, the update option is already built-in. As root, simply:
yt-dlp -U
If you installed it from the repos, that option is disabled. An error will show and it can only list the installed and available versions.
I run dnscrypt-proxy and choose resolvers list carefully (except when a trusted VPN is running). Rightly or wrongly, I view a corporate provider's DNS service as potentially unreliable *and* a security risk (surveillance, logging, passing data to who knows who)..
I use Ceres and update most days but never seen that problem before. Except yesterday..
I maintain a friend's machines which run chimaera (and use provider DNS):
Temporary failure resolving 'deb.devuan.org'
Immediately after editing resolv.conf with a known good alternative DNS server, all was well.
Later, at my own machine, again no problem. Why is that?
Thanks GNUser for this thread. Useful for me as I have only wireless here (in some areas only) from a neighbour's router.
It seems create_ap is no longer maintained, so let's not bother with it.
True but its been forked, and maintained (in the last 2 months) elsewhere:
https://github.com/lakinduakash/linux-wifi-hotspot
There is a deb package with python-based GUI (but usable from cli only). Or build your own from the source, as did I. Working here, tested only on ceres so far, at least as wireless repeater. I'm using network-manager, that seems to be detected and not permanantly interfered with. I now have optional connection in a previously dead zone via another Devuan box.
Using 2x cheap usb wireless adapters, couldnt get either to work alone.
You might want to disable automatic start of dnsmasq (sysv-rc-conf) unless you got reason not to.
I will try your more minimal script at some point.
No bug here, we just needed to watch the warnings and wait. Network-manager depends on libnm0 of the same version .. libnm0 was to be upgraded to 1.44.2-5 (at 23-11-22) but it seems network-manager 1.44.2-5 wasn't available till shortly afterwards. From my log:
The following packages will be REMOVED:
network-manager network-manager-gnome
#SNIP
Do you want to continue? [Y/n]
Seems something else is using 127.0.2.1:53 (maybe another instance of dnscrypt-proxy?) You could investigate using lsof and ps ..
Or just change 127.0.2.1:53 to 127.0.0.1:53, that's the default anyway. I only use 127.0.2.1 from an older release custom config, which simply works here. Remember to update network-manager configs also, if you use that. Should all be good after a reboot.
Working (for me) /etc/dnscrypt-proxy/dnscrypt-proxy.toml :
# Empty listen_addresses to use systemd socket activation
listen_addresses = ['127.0.2.1:53']
server_names = ['cloudflare']
[query_log]
file = '/var/log/dnscrypt-proxy/query.log'
[nx_log]
file = '/var/log/dnscrypt-proxy/nx.log'
[sources]
[sources.'public-resolvers']
url = 'https://download.dnscrypt.info/resolvers-list/v2/public-resolvers.md'
cache_file = '/var/cache/dnscrypt-proxy/public-resolvers.md'
minisign_key = 'RWQf6LRCGA9i53mlYecO4IzT51TGPpvWucNSCh1CBM0QTaLn73Y7GFO3'
refresh_delay = 72
prefix = ''
** I use network-manager. In ipv4 settings, "Method" is "Automatic (DHCP) addresses only" and "DNS servers" is "127.0.2.1" **
I have used https://ipleak.net/ to test but can't verify their reliability. Shows cloudfare dns.
root@ceres:~# cat /etc/resolv.conf
# Dynamic resolv.conf(5) file for glibc resolver(3) generated by resolvconf(8)
# DO NOT EDIT THIS FILE BY HAND -- YOUR CHANGES WILL BE OVERWRITTEN
# 127.0.0.53 is the systemd-resolved stub resolver.
# run "resolvectl status" to see details about the actual nameservers.
nameserver 127.0.2.1
You may also need to create ~/.asoundrc wih something like this (change "Port" and device "mac address" to suit.:
pcm.Port {
type bluealsa
device "77:20:1E:FE:4E:E4"
profile "a2dp"
hint { show on description "Port" }
}
I used bluez-alsa for a while with reasonable success but now finding pipewire more reliable. You can have both installed but if you try pipewire bluealsa service should stopped and disabled in sysv-rc-conf..
blueman-applet seems the easiest way to search available devices, get info, set up. Or bluetoothctl in a terminal. Setup sequence is trust>pair>connect .. good luck..
You don't need pulseaudio and should uninstall if you try this.
Re bluez-alsa: it's only in unstable and not so new. And no syvinit script.
https://dev1galaxy.org/viewtopic.php?id=4612
I made updated packages for chimaera with sysvinit script and posted here (please install libasound2-plugin-bluez first):
https://exegnulinux.net/apt/pool/main/b … _amd64.deb
https://exegnulinux.net/apt/pool/main/b … _amd64.deb
A better way might be with pipewire:
https://dev1galaxy.org/viewtopic.php?id=5385
BT sound working here both ways.
Thanks all here for the info on setting up pipewire for chimaera. Tried and failed till now. Although bluez-alsa does work for my BT speaker. I don't use pulse-audio.
I first installed the packages as described by Devarch. Note these packages actually get installed as deps: pipewire-bin libpipewire-0.3-modules libwireplumber-0.4-0 libpipewire-0.3-0 . These backports are almost up to current unstable versions.
With the latest packages, everything subsequently (at least after reboot) worked "out of the box" on my Toshiba laptop. No file copying no other configurations. The Debian wiki is outdated. The BT speaker is up and running, including firefox streams which failed with bluez-alsa.
Thanks HoaS for the start script. I adapted it quite a bit because I want to start/stop pipewire manually. Also, sometimes a restart is needed (here) to fix an occasional bt connection drop out. And I don't want processes not in use autostarting (I disabled also /etc/xdg/autostart/blueman.desktop).
Placed in ~/bin/ and called from a desktop launcher:
#!/bin/bash
# Script name: bluestart
# Place in ~/bin and make a desktop launcher
# Execute "bluestart stop" to just stop pipewire and related processes without restarting.
# (in case you used "stop" parameter)
MODE="$1"
# You probably need to restart your system mixer to register/deregister the pipewire device.
# Edit (or comment) the next line to suit whatever your DE uses:
MIXER=kmix
# Don't run this as root!
if [ $EUID -eq 0 ]; then exit 1; fi
restart_mixer () {
if [ -n "$MIXER" ]; then
pkill -u "$USER" "$MIXER" >/dev/null 2>&1
$MIXER &
fi
}
stop_pipewire () {
pkill -u "$USER" pipewire-pulse >/dev/null 2>&1
pkill -u "$USER" wireplumber >/dev/null 2>&1
pkill -u "$USER" pipewire >/dev/null 2>&1
killall blueman-applet blueman-tray blueman-mechanism 2>/dev/null
# delete blueman stat files in ~/.config
for i in $(ls ~/.config|grep blueman); do rm -f $i; done
}
stop_pipewire
if [ "$MODE" = "stop" ]; then
sleep 3
restart_mixer
exit 0
fi
pipewire &
until pgrep -f pipewire >/dev/null 2>&1 ; do
sleep 1
done
wireplumber &
pipewire-pulse &
blueman-applet &
sleep 2
restart_mixer
Hope that helps someone..
Sorted! Please mark as "solved".. Note it's actually /etc/init.d/rc2.d/ that's relevant (unless you manually changed default runlevel from 2). Disabled you will see K01cups, or enabled, S01cups..
root@exegnu:~# ls /etc/rc2.d|grep cups
K01cups
I did not touch the eudev service at all.
Please don't, for sure will break things! And I hope you're not running konqueror as root..
@OP: Please note Devuan default runlevel is 2 (you only change that if you got a reason)
I think the en_US.UTF-8 locale should be generated regardless of the selected system locale
It is! (and you can generate any others needed and choose your system default)
root@exegnu:~# dpkg-reconfigure locales
Generating locales (this might take a while)...
de_DE.UTF-8... done
en_GB.UTF-8... done
en_US.UTF-8... done
es_ES.UTF-8... done
fr_FR.UTF-8... done
it_IT.UTF-8... done
pl_PL.UTF-8... done
pt_PT.UTF-8... done
Generation complete.
Fairly new Exe install here. I dont actually need cups always on so disabled it in all runlevels using sysv-rc-conf. After reboot:
root@exegnu:~# runlevel
N 2
root@exegnu:~# service cups status
cupsd is not running ... failed!
root@exegnu:~# ls /etc/rc5.d|grep cups
K01cups
That is normal, I can't reproduce aluma's problem!
GNU/Linux EXE installer doesn't seem to recognise virtio disks in QEMU
Exe GNU installer is quite basic. All Exe builds include also refractainstaller which may work in that case..
EDIT aluma, your sysv-rc-conf screenshot show cups only off in runlevel 5, that's not normal. What happens if you turn it off in all runlevels using arrow and spacebar keys?
? It boots here with or without "apparmor=0" ..
I always found "findiso" simpler to configure than "fromiso".. here's my grub.d/40_custom entry:
menuentry "devuan5 preview" {
insmod part_gpt
set isofile='(hd0,gpt7)/devuan_daedalus_5.0.preview_20221022_amd64_desktop-live.iso'
loopback loop $isofile
linux (loop)/live/vmlinuz boot=live tzdata=Europe/London locales=en_GB.UTF-8 username=devuan findiso=/devuan_daedalus_5.0.preview_20221022_amd64_desktop-live.iso
initrd (loop)/live/initrd.img
}
Running chimaera here on my vaio netbook but not testing/unstable at the moment.
I use bluetooth for android (lineageos) file transfer and for 3 different BT speaker sets. Eveything works.
$ dpkg -l|grep bluez
ii bluez 5.55-3.1 amd64 Bluetooth tools and daemons
ii bluez-alsa-utils 3.1.0~git20211121-1+exegnu1 amd64 Bluetooth Audio ALSA Backend (utils)
ii bluez-firmware 1.2-4 all Firmware for Bluetooth devices
ii bluez-obexd 5.55-3.1 amd64 bluez obex daemon
ii libasound2-plugin-bluez:amd64 3.1.0~git20211121-1+exegnu1 amd64 Bluetooth Audio ALSA Backend (plugins)
I don't know if bluez 5.55-3.1 will run on testing or if the problems mentioned are hardware-specific for all versions. Maybe someone wants to try package the latest source.
Re BT speakers/headsets: there was a thread here for those who don't want pulseaudio (bluez-alsa)
https://dev1galaxy.org/viewtopic.php?id=4612
I finally sorted debian sid packages' missing sysv script for bluez-alsa. The rebuilt debs do work in chimaera (can't test on Daedalus yet) :
The instructions linked above describe only Debian
That was the case before although the Debian packages were compatible with Devuan. TDE now have specific repos for all Devuan versions (and R14.0.11 is released today, in case anyone still thinks it's old, unmaintained code)
Thanks HoaS!
Looks like it's a dbus service so it probably won't work unless you can hook it into that somehow with sysvinit (systemd listens to the bus automatically). But this is just a guess.
It does here, maybe the dbus bits are built-in. That init script does work from a root shell after boot (although still will not detach, that was the issue)
Now using a simple manual /usr/local/bin start script with sudo permission so it only runs on demand, in preference.
Cable the speaker to the old vaio? The cable is always either lost or broken!
Worth a go here as my trusy old Vaio netbook has **** sound (and I don't want pulseaudio).
I got bluealsa working but it was not so simple. For a start packages bluez-alsa-utils libasound2-plugin-bluez (you need both) have only a systemd service file, no sysv script. I needed bluetooth daemon running first and a custom ~/.asoundrc
pcm.SoundCoreMini {
type bluealsa
device "XX:XX:XX:XX:XX:XX"
profile "a2dp"
hint { show on description "Sound Core Mini"}
}
Having already installed blueman I could use those tools to get the device ID. The device can then be manipulated using blueman-manager.
I repackaged it with a custom init script but havent got that quite right as yet. It won't detach from the shell and hangs on bootup. But I can start it manually after disabling the init daemon.
Running chimaera you *might* get away with just a few lib deps from ceres. The original packages will break beowulf. My crude "backport" rebuild from source can run on beowulf.
Here's the offending custom init script, if anyone wants to debug it. More later (if) I get time to revisit this. Or maybe it's better not to have this start automatically..
#!/bin/sh
### BEGIN INIT INFO
# Provides: bluez-alsa
# Default-Start: 2 3 4 5
# Default-Stop: 0 1 6
# Required-Start: $syslog $local_fs $remote_fs bluetooth
# Required-Stop: $syslog $local_fs $remote_fs
# Short-Description: Bluealsa daemon
### END INIT INFO
. /lib/lsb/init-functions
prog=bluez-alsa
if test -f -/etc/default/bluez-alsa; then
. -/etc/default/bluez-alsa
fi
PIDFILE=/var/run/$prog.pid
DESC="Bluealsa daemon"
start() {
log_daemon_msg "Starting $DESC" "$prog"
start_daemon -p $PIDFILE /usr/bin/bluealsa $OPTIONS
if [ $? -ne 0 ]; then
log_end_msg 1
exit 1
fi
if [ $? -eq 0 ]; then
log_end_msg 0
fi
exit 0
}
stop() {
log_daemon_msg "Stopping $DESC" "$prog"
killproc -p $PIDFILE /usr/bin/bluealsa
if [ $? -ne 0 ]; then
log_end_msg 1
exit 1
fi
if [ $? -eq 0 ]; then
log_end_msg 0
fi
}
force_reload() {
stop
start
}
case "$1" in
start)
start
;;
stop)
stop
;;
force-reload)
force_reload
;;
restart)
stop
start
;;
*)
echo "$Usage: $prog {start|stop|force-reload|restart}"
exit 2
esac
My installed bluetooth-related packages:
root@vaio:~# dpkg -l|grep blue
ii blueman 2.0.8-1+deb10u1 amd64 Graphical bluetooth manager
ii bluez 5.50-1.2~deb10u2 amd64 Bluetooth tools and daemons
ii bluez-alsa-utils 3.0.0-2 amd64 Bluetooth Audio ALSA Backend (utils)
ii bluez-firmware 1.2-4 all Firmware for Bluetooth devices
ii bluez-obexd 5.50-1.2~deb10u2 amd64 bluez obex daemon
ii libasound2-plugin-bluez:amd64 3.0.0-2 amd64 Bluetooth Audio ALSA Backend (plugins)
ii libbluetooth-dev:amd64 5.50-1.2~deb10u2 amd64 Development files for using the BlueZ Linux Bluetooth library
ii libbluetooth3:amd64 5.50-1.2~deb10u2 amd64 Library to use the BlueZ Linux Bluetooth stack
ii libpam-blue 0.9.0-3 amd64 PAM module for local authenticaction with bluetooth devices
EDIT: example use:
:~$ mpv --audio-device=alsa/SoundCoreMini 'Sweet Georgia Brown 1.m4a'
Or in VLC, pick it from the audio device GUI.
Cheers all (and thanks for chimaera), D
Cheers Nik..
Your exegnu remaster boots fine and fast here from usb. A short cmdline addition e.g. lang=en_GB sets both locale and keyboard for en_US/GB or ES on boot. I just discovered your rt kernel runs well on one old hp laptop here, better than the standard kernel which did not.
Although I don't need some of your special applications included, this is a good test ISO for some of the core exegnu functions, in particular internationalisation, installer and some live-config scripts (which maybe need some updating). It's also a good, successful test for refractasnapshot.
Well done and thanks!
D
I guess some have their use-case reasons to automount removables. I don't know why.
Mounting an mtp device should be as simple as:
jmtpfs /path/to/your/mountpoint/
fusermount -u /path/to/your/mountpoint/
Or you could script that with some checks and a fancy yad dialog.
(OT) thanks xinomilo for mentioning "lineageOS" (android AOSP based without bloat/spyware), probably as clean as you will get on such devices. The only network connection here at the moment is 4g. I did quite a bit of research on android (lack of) security and run lineage (without gapps).
Re inxi and smxi:
I have always had great respect for HH and thank him for his contributions
+1.. A founding member of the Sidux project, well remembered here from the early days. https://smxi.org/site/smxi-story.htm is an interesting read, for those who didn't already know.
Recent Youtube changes broke gtk-youtube-viewer (maybe other clients). It was forked to "gtk straw viewer" and works again. Well here and for now at least.. https://github.com/trizen/straw-viewer
Using a youtube client avoids heavyweight browser loading, ads, tracking... This one is quite light on deps and system resources but has a very good descriptive gui. No qt deps. Probably needs latest youtube-dl to run properly https://github.com/ytdl-org/youtube-dl
I posted (probably the only) deb package: http://exegnulinux.net/apt/pool/main/s/straw-viewer/
Sources, dsc et al are at the same url if anyone wants to inspect, compile or modify yourself. If interested, please post back if you see anything wrong with the package.
EDIT: youtube-dl (and anything else that uses it) will puke unless you update it quite often.
The default setup results in the grub menu showing 'Debian' instead of 'Devuan'.
If you're using legacy bios boot, you can just change /etc/os-release to show ID=devuan and then run update-grub
Also package lsb-release must be present. If you installed from debootstrap or maybe did a dist-upgrade it may not be. It's otherwise unnecessary and brings some weighty python deps. I don't know if the official installer includes it. Snip from stock /etc/default/grub (which sets the menu title) :
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
Or you could just edit /etc/default/grub (my usual way)
Of course, /etc/os-release needs to be correct in any other devuan installs (for os-prober, see /usr/lib/os-probes/mounted/90linux-distro )
Maybe or not your problem is related but:
My ceres live builds, with *default* 5x kernels (all of them), hang every time in seemingly random stages of bootup. Often while still in initramfs. This older HP (Intel) laptop is all I have around at the moment. They boot fine with some combinations of "failsafe graphics" parameters on the cmdline (maybe breaks some acpi-related functions):
noapic noapm nodma nomce nolapic nosmp
Seems a major kernel issue. A search for, e.g. "linux 5x intel hangs", returns much info but little good news, at least not soon. No such issues here with Beowulf and Ascii systems.
:~$ uname -r && inxi -CG
5.4.0-4-amd64
CPU:
Topology: Single Core model: Intel Core2 T5500 bits: 64 type: MCP
L2 cache: 2048 KiB
Speed: 1662 MHz min/max: N/A Core speed (MHz): 1: 1662
Graphics:
Device-1: Intel Mobile 945GM/GMS 943/940GML Express Integrated Graphics
driver: i915 v: kernel
Display: x11 server: X.Org 1.20.7 driver: intel
unloaded: fbdev,modesetting,vesa resolution: 1024x768~60Hz
Message: Unable to show advanced data. Required tool glxinfo missing.
The prompt is something like "Enter passphrase for <your-persistence-media>" after syslinus/isolinux/grub but while still in initramfs. The console doesn't show what you type. If it's incorrect after (if I remember right) 3 attempts then yes, barebones live image only boots.
The excellent tool fsmithred linked uses syslinux, IMO a better way than dd or cat (the usual "mainstream" methods), to get a bootable USB for live images..
EDIT: The prompt (in my newest beowulf image) is actually "Please unlock disk /dev/whatever" (here it's loop0, I use a persistence image file). It prints an error if you get it wrong and invites to retry or not. The boot initrd must have dmcrypt support built in.
My usual method is an out-of-the-box live image (Refracta is the prime candidate but there are others) with LUKS encrypted full persistence. The live image is tailored to suit as many machines as possible. The persistence is a custom overlayfs, can be a (decent size) file or a partition. On boot it will prompt for the LUKS key, if you don't have that it will continue to boot to only the basic, default live image. The security of the rest is down to the strength of your encryption key.
It's quite surprising how quickly and efficiently such a system can run..