In Jammy sound level reset after reboot

[corrado venturini]

After reboot sound level is reset ignoring the level set in previous session.

I manually set the sound output level
at next boot the sound is set at a different level (about 70%)

Expected: the sound level is persistent across power off/on

What happens : sound level is set at a different level (about 70%)

Note: this problem does not occur on different partitions of same PC running Ubuntu 20.04 or 21.10

corrado@corrado-p3-jj-1130:~$ apt policy gnome-control-center
gnome-control-center:
  Installed: 1:41.1-1ubuntu1
  Candidate: 1:41.1-1ubuntu1
  Version table:
 *** 1:41.1-1ubuntu1 500
        500 http://archive.ubuntu.com/ubuntu jammy/main amd64 Packages
        100 /var/lib/dpkg/status
corrado@corrado-p3-jj-1130:~$ inxi -Fx
System: Host: corrado-p3-jj-1130 Kernel: 5.13.0-19-generic x86_64 bits: 64 compiler: gcc v: 11.2.0
           Desktop: GNOME 40.5 Distro: Ubuntu 22.04 (Jammy Jellyfish)
Machine: Type: Laptop System: Dell product: Inspiron 3793 v: N/A serial: <superuser required>
           Mobo: Dell model: 0C1PF2 v: A00 serial: <superuser required> UEFI: Dell v: 1.5.0 date: 12/17/2019
Battery: ID-1: BAT0 charge: 13.9 Wh (42.1%) condition: 33.0/42.0 Wh (78.7%) volts: 11.3 min: 11.4
           model: SWD-ATL3.618 DELL WJPC403 status: Discharging
CPU: Info: Quad Core model: Intel Core i5-1035G1 bits: 64 type: MT MCP arch: Ice Lake rev: 5 cache:
           L2: 6 MiB
           flags: avx avx2 lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx bogomips: 19046
           Speed: 968 MHz min/max: 400/1000 MHz Core speeds (MHz): 1: 968 2: 710 3: 706 4: 714 5: 966 6: 712
           7: 724 8: 821
Graphics: Device-1: Intel Iris Plus Graphics G1 vendor: Dell driver: i915 v: kernel bus-ID: 00:02.0
           Device-2: Realtek Integrated_Webcam_HD type: USB driver: uvcvideo bus-ID: 1-6:3
           Display: wayland server: X.Org 1.21.1.2 compositor: gnome-shell driver: loaded: i915
           note: n/a (using device driver) resolution: 1920x1080~60Hz
           OpenGL: renderer: Mesa Intel UHD Graphics (ICL GT1) v: 4.6 Mesa 21.2.2 direct render: Yes
Audio: Device-1: Intel Ice Lake-LP Smart Sound Audio vendor: Dell driver: snd_hda_intel v: kernel
           bus-ID: 00:1f.3
           Sound Server-1: ALSA v: k5.13.0-19-generic running: yes
           Sound Server-2: PulseAudio v: 15.0 running: yes
           Sound Server-3: PipeWire v: 0.3.40 running: yes
Network: Device-1: Realtek RTL810xE PCI Express Fast Ethernet vendor: Dell driver: r8169 v: kernel port: 3000
           bus-ID: 01:00.0
           IF: enp1s0 state: down mac: 98:e7:43:59:19:19
           Device-2: Qualcomm Atheros QCA9377 802.11ac Wireless Network Adapter vendor: Dell driver: ath10k_pci
           v: kernel bus-ID: 02:00.0
           IF: wlp2s0 state: up mac: d8:12:65:b8:21:b9
Bluetooth: Device-1: Qualcomm Atheros type: USB driver: btusb v: 0.8 bus-ID: 1-10:4
           Report: hciconfig ID: hci0 rfk-id: 0 state: down bt-service: enabled,running rfk-block: hardware: no
           software: yes address: D8:12:65:B8:21:BA
Drives: Local Storage: total: 942.7 GiB used: 11.77 GiB (1.2%)
           ID-1: /dev/nvme0n1 vendor: Toshiba model: KBG40ZNS512G NVMe KIOXIA 512GB size: 476.94 GiB temp: 24.9 C
           ID-2: /dev/sda vendor: Crucial model: CT500MX500SSD1 size: 465.76 GiB
Partition: ID-1: / size: 46.95 GiB used: 11.73 GiB (25.0%) fs: ext4 dev: /dev/nvme0n1p3
           ID-2: /boot/efi size: 246 MiB used: 46.4 MiB (18.9%) fs: vfat dev: /dev/nvme0n1p1
Swap: Alert: No swap data was found.
Sensors: System Temperatures: cpu: 27.8 C mobo: N/A
           Fan Speeds (RPM): cpu: 0
Info: Processes: 263 Uptime: 31m Memory: 15.4 GiB used: 2.45 GiB (15.9%) Init: systemd runlevel: 5 Compilers:
           gcc: N/A Packages: 1785 Shell: Bash v: 5.1.12 inxi: 3.3.07
corrado@corrado-p3-jj-1130:~$
---
ProblemType: Bug
ApportVersion: 2.20.11-0ubuntu74
Architecture: amd64
CasperMD5CheckResult: pass
CurrentDesktop: ubuntu:GNOME
DistroRelease: Ubuntu 22.04
InstallationDate: Installed on 2021-11-30 (2 days ago)
InstallationMedia: Ubuntu 22.04 LTS "Jammy Jellyfish" - Alpha amd64 (20211130)
Package: gnome-control-center 1:41.1-1ubuntu1
PackageArchitecture: amd64
ProcVersionSignature: Ubuntu 5.13.0-19.19-generic 5.13.14
Tags: wayland-session jammy
Uname: Linux 5.13.0-19-generic x86_64
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin lxd plugdev sambashare sudo
_MarkForUpload: True

[corrado venturini (corradoventu)]

apport information

[corrado venturini (corradoventu)]

apport information

[corrado venturini (corradoventu)]

apport information

[Sebastien Bacher (seb128)]

Saving the state and restoring it is alsa's job, gnome-control-center isn't a service and not doing anything when not started

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Ice-Tea (ice-tea)]

@corrado venturini

Thanks for reporting this bug as i was about to do the same.

[Fredrik Jacobsson (smask)]

I would like to add that my computer resets the volume to 70% after I resume from suspending.

DM: gdm
DE: gnome flashback

[Sebastien Bacher (seb128)]

Did the issue started after upgrading to the new serie or was it already there in impish?

[corrado venturini (corradoventu)]

My Jammy was not upgraded from impish but fresh install from ISO.
and i see the problem on 2 different install on same PC.
I don't have the problem on impish installed on different partition of same disk same PC.

[corrado venturini (corradoventu)]

Volume is reset to 70% also when i insert or remove headset.

[[e-mail] (sgage)]

This bug also affects me. I am using Ubuntu MATE (Jammy).

[Sebastien Bacher (seb128)]

It's a bit weird, trying in a VM in a simple wmaker session, changing the level in pavucontrol and restarting the machine the sound is correctly restored on the same position after the restart so the systemd unit seems to be working as expected.

Could someone having the issue change the level, restart, check that the level is indeed wrong and if it is get a log from the previous session with 'journalctl -b 1 > log' and add the log file to the report?

[corrado venturini (corradoventu)]

not 'journalctl -b 1 > log' but 'journalctl -b-1 > log'
.. log attached

[corrado venturini (corradoventu)]

I did some tests by changing the volume and inserting and removing the headphones but I don't see anything in the journalctl.

[corrado venturini (corradoventu)]

Problem disappeared after today updates(?) on my 2 desktop 3 different install of Jammy.
but NOT disappeared on my laptop also updated today.
Let me know what useful information I can provide.

[corrado venturini (corradoventu)]

Problem reappeared

[Igor V. Kovalenko (i-garrison)]

Do you have both pulseaudio and pipewire audio enabled simultaneously? At least system log suggests pipewire with session manager is also being activated, this could cause issues if both attempt to restore volumes for some reason.

[corrado venturini (corradoventu)]

how can i tell if i have pulseaudio and pipewire activated?

[Sebastien Bacher (seb128)]

what's the output of
$ dpkg -l | grep pipewire
?

[Nobuto Murata (nobuto)]

I can confirm that after stopping pipewire temporarily by:
$ systemctl --user stop pipewire.socket pipewire.service
The volume level is properly recovered across plugging in and out a headset for example, which is good.

Both pulseaudio and pipewire are installed out of the box and running if I'm not mistaken.

$ apt show pulseaudio |& grep Task:
Task: ubuntu-desktop-minimal, ubuntu-desktop, ubuntu-desktop-raspi, kubuntu-desktop, xubuntu-core, xubuntu-desktop, lubuntu-desktop, ubuntustudio-desktop-core, ubuntustudio-desktop, ubuntukylin-desktop, ubuntu-mate-core, ubuntu-mate-desktop, ubuntu-budgie-desktop, ubuntu-budgie-desktop-raspi

$ apt show pipewire |& grep Task:
Task: ubuntu-desktop-minimal, ubuntu-desktop, ubuntu-desktop-raspi, kubuntu-desktop, xubuntu-core, xubuntu-desktop, lubuntu-desktop, ubuntustudio-desktop-core, ubuntustudio-desktop, ubuntukylin-desktop, ubuntu-mate-core, ubuntu-mate-desktop, ubuntu-budgie-desktop, ubuntu-budgie-desktop-raspi

[Igor V. Kovalenko (i-garrison)]

Sebastien, if distribution is intended to provide pipewire function for video only, there is no need to install audio parts. At very least 'with-pulse' 'with-jack' and other files directly or transitively enabling 'with-audio' pipewire parts should not be installed so pipewire media session daemon does not get hold of hardware controls.

Alternatively, if distribution is intended to provide pulseaudio interface via pipewire, that should be properly switching pulseaudio activation socket to pipewire-pulse, and/or disabling or uninstalling pulseaudio daemon and maybe disabling pulseaudio autospawn in client.conf

[corrado venturini (corradoventu)]

On my desktop after a lot of new updates the sound level is NOT reset at reboot.

I see pipewire and pulseaudio are both installed but i don't know if they are active.

corrado@corrado-p5-jj-0203:~$ dpkg -l | grep pipewire
ii gstreamer1.0-pipewire:amd64 0.3.45-1 amd64 GStreamer 1.0 plugin for the PipeWire multimedia server
ii libpipewire-0.3-0:amd64 0.3.45-1 amd64 libraries for the PipeWire multimedia server
ii libpipewire-0.3-common 0.3.45-1 all libraries for the PipeWire multimedia server - common files
ii libpipewire-0.3-modules:amd64 0.3.45-1 amd64 libraries for the PipeWire multimedia server - modules
ii pipewire:amd64 0.3.45-1 amd64 audio and video processing engine multimedia server
ii pipewire-bin 0.3.45-1 amd64 PipeWire multimedia server - programs
ii pipewire-media-session 0.4.1-2 amd64 example session manager for PipeWire
corrado@corrado-p5-jj-0203:~$ apt policy pipewire pulseaudio
pipewire:
  Installed: 0.3.45-1
  Candidate: 0.3.45-1
  Version table:
 *** 0.3.45-1 500
        500 http://archive.ubuntu.com/ubuntu jammy/main amd64 Packages
        100 /var/lib/dpkg/status
pulseaudio:
  Installed: 1:15.0+dfsg1-1ubuntu6
  Candidate: 1:15.0+dfsg1-1ubuntu6
  Version table:
 *** 1:15.0+dfsg1-1ubuntu6 500
        500 http://archive.ubuntu.com/ubuntu jammy/main amd64 Packages
        100 /var/lib/dpkg/status
corrado@corrado-p5-jj-0203:~$

and from system monitor i see both are active. see attachment

[corrado venturini (corradoventu)]

On a different Jammy install on a different desktop the problem dos still occur.

[Igor V. Kovalenko (i-garrison)]

Please have a look which apps are using audio devices: 'lsof /dev/snd/*

[Paul White (paulw2u)]

Igor, I'm responding as someone also affected by this issue. Using the command you gave in comment #25 I saw that both pulseaudio and pipewire were using the same audio device(s).

I used the command 'sudo systemctl disable --global pipewire' to disable pipewire and deleted the ~/.config/pulse/ directory. After I rebooted several times I can see that the volume is no longer set to maximum after rebooting or logging out and back in.

This workaround is documented at https://bugs.launchpad.net/ubuntu/+source/pipewire/+bug/1897965/comments/23

[Igor V. Kovalenko (i-garrison)]

OK that matches what we have observed elsewhere. In theory there is an audio device reservation protocol which is obeyed by both pulseaudio and pipewire, but in practice it looks like at least some versions of pipewire media server component do not obey the protocol in full (i.e. are still using control device) I cannot think of any other reason why both pulseaudio and pipewire are using the same control device.

[corrado venturini (corradoventu)]

After last updates problem disappeared again
corrado@corrado-n4-jj-0203:~$ lsof /dev/snd/*
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
pipewire 1611 corrado 43u CHR 116,1 0t0 621 /dev/snd/seq
pipewire 1611 corrado 44u CHR 116,1 0t0 621 /dev/snd/seq
pipewire- 1612 corrado 18u CHR 116,3 0t0 834 /dev/snd/controlC1
pipewire- 1612 corrado 24u CHR 116,14 0t0 1069 /dev/snd/controlC0
pulseaudi 1613 corrado 30u CHR 116,3 0t0 834 /dev/snd/controlC1
pulseaudi 1613 corrado 32u CHR 116,14 0t0 1069 /dev/snd/controlC0
pulseaudi 1613 corrado 39u CHR 116,14 0t0 1069 /dev/snd/controlC0
corrado@corrado-n4-jj-0203:~$

[corrado venturini (corradoventu)]

Problem seems solved for my 2 desktop and on my laptop:
corrado@corrado-n4-jammy-0202:~$ lsof /dev/snd/*
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
pipewire 1393 corrado 43u CHR 116,1 0t0 566 /dev/snd/seq
pipewire 1393 corrado 44u CHR 116,1 0t0 566 /dev/snd/seq
pipewire- 1394 corrado 18u CHR 116,13 0t0 1049 /dev/snd/controlC0
pulseaudi 1395 corrado 25u CHR 116,13 0t0 1049 /dev/snd/controlC0
pulseaudi 1395 corrado 32u CHR 116,13 0t0 1049 /dev/snd/controlC0
pulseaudi 1395 corrado 37u CHR 116,13 0t0 1049 /dev/snd/controlC0
corrado@corrado-n4-jammy-0202:~$

[Igor V. Kovalenko (i-garrison)]

Corrado, this is still not ideal - while it appears to work at the moment, having two audio servers with their own idea of volumes would probably cause trouble going on.

[Erich Eickmeyer (eeickmeyer)]

Ubuntu Studio lead here. We have been very closely watching the development of PipeWire as it is of interest to a large amount of our user base in the Audio Production world.

Please note that PipeWire, in its current iteration, is meant to be a drop-in replacement for both PulseAudio and Jack sound servers. Having PulseAudio and PipeWire (specifically the pipewire-pulse package) installed simultaneously is asking for trouble as the PulseAudio implementation of PipeWire is feature-complete. I'm actually surprised that the pipeiwire-pulse package doesn't have a "Provides:" or "Conflicts:" line for PulseAudio as it's literally a drop-in replacement.

For this reason, this is *strongly* looking like a support request more than an actual bug and I'm tempted to close it as invalid (support requests are not allowed on a bug tracker). Either that or a bug against the pipewire source package for not having the "Provides:" or "Conflicts:" lines mentioned above.

TL;DR: pipewire-pulse and PulseAudio should *not* be installed at the same time as they serve the same function and applications don't know the difference.

[Daniel van Vugt (vanvugt)]

This is definitely a bug and one we need to solve before the masses start using jammy. So just closing it wouldn't be helpful.

I'm reminded a little of bug 1897965. Maybe we need either:

  https://bugs.launchpad.net/ubuntu/+source/pipewire/+bug/1897965/comments/16

or to modify pipewire-media-session.service with something like:

  /usr/bin/pipewire-media-session -d bluez5,alsa-seq,alsa-pcm,alsa-acp

That way PulseAudio is always the only one trying to control the hardware.

Alternatively; don't install pipewire-media-session by default. So long as it's not needed for Wayland screencasting/recording like at least some of Pipewire is :S

[Erich Eickmeyer (eeickmeyer)]

Daniel, I disagree.

PipeWire is meant to be a *drop-in replacement* for PulseAudio. Applications that recognize PulseAudio are supposed to use it instead if it's available in lieu of PulseAudio.

Both options need to be available, otherwise the entire goal of PipeWire's audio capabilities of unifying Linux audio under one audio server are lost.

While this might still be a bug, which is the entire reason I didn't follow through with closing it, I still believe this may be a packaging issue more than anything.

The points you bring up are probably upstream ones, so I'd take it up there. https://gitlab.freedesktop.org/pipewire/pipewire/

[Daniel van Vugt (vanvugt)]

I don't think the points I raised are upstream ones and I wouldn't take them upstream myself. I think it's good (in the long term) and appropriate that PipeWire upstream is a daemon, just not appropriate in Ubuntu 22.04 since we've decided to stick with PulseAudio. Hence the configuration changes I mention would be Ubuntu-specific.

But if we can just not install pipewire-media-session by default then that would be better than any config changes.

[Nobuto Murata (nobuto)]

> TL;DR: pipewire-pulse and PulseAudio should *not* be installed at the same time as they serve the same function and applications don't know the difference.

This is not the case at least for the default Ubuntu flavor (w/ GNOME).

$ curl -s https://cdimages.ubuntu.com/daily-live/current/jammy-desktop-amd64.manifest | egrep 'pipewire|pulseaudio'
gstreamer1.0-pipewire:amd64 0.3.44-1
gstreamer1.0-pulseaudio:amd64 1.18.5-1ubuntu3
libpipewire-0.3-0:amd64 0.3.44-1
libpipewire-0.3-common 0.3.44-1
libpipewire-0.3-modules:amd64 0.3.44-1
pipewire:amd64 0.3.44-1
pipewire-bin 0.3.44-1
pipewire-media-session 0.4.1-2
pulseaudio 1:15.0+dfsg1-1ubuntu6
pulseaudio-module-bluetooth 1:15.0+dfsg1-1ubuntu6
pulseaudio-utils 1:15.0+dfsg1-1ubuntu6

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Cam Cope (ccope)]

So I can confirm that copying /usr/share/pipewire/media-session.conf to /etc/pipewire/media-session.d/media-session.conf and deleting these lines fixes the issue:
$ diff /etc/pipewire/media-session.d/media-session.conf /usr/share/pipewire/media-session.d/media-session.conf
94a95,116
> with-audio = [
> metadata
> default-nodes
> default-profile
> default-routes
> alsa-seq
> alsa-monitor
> ]
> with-alsa = [
> with-audio
> ]
> with-jack = [
> with-audio
> ]
> with-pulseaudio = [
> with-audio
> bluez5
> bluez5-autoswitch
> logind
> restore-stream
> streams-follow-default
> ]

[Daniel van Vugt (vanvugt)]

Thanks Cam Cope, that's an interesting fix.

It's probably better if we can avoid starting pipewire-media-session in GNOME sessions without other config changes. I don't want to cripple pipewire-media-session to the point of it not being able to replace PulseAudio in other session types.

[Daniel van Vugt (vanvugt)]

The offending package seems to be coming via:

ubuntu-meta >
gnome-remote-desktop >
pipewire-media-session | wireplumber

So it's a dependency we probably can't drop. Maybe a config change is the only real option.

[Daniel van Vugt (vanvugt)]

Comparing the configs of impish (where it works) and jammy (where it fails) I can't find any differences of consequence. Most of the changes are either formatting, comments, or irrelevant.

I've also tried bisecting and refining the media session config to reduce it to the minimal change to fix the bug. It seems Cam Cope got it exactly right in comment #37. No smaller/simpler change resolves the issue in my testing. So unless anyone has a better idea I will propose a distro patch for that on Wednesday.

[Daniel van Vugt (vanvugt)]

It might be possible to refine further if not all of these need removing:

> with-audio = [
> metadata
> default-nodes
> default-profile
> default-routes
> alsa-seq
> alsa-monitor
> ]

Hard to tell because the docs either aren't built properly in places, don't exist at all for some things, or are just too hard to find.

[Sebastien Bacher (seb128)]

We do need pipewire-media-session since according to GNOME upstream without it gnome shell recording would just stay pause and not actually work, unsure if that's still true so worth testing

[Daniel van Vugt (vanvugt)]

I think this is a better fix: Just comment out the default-profile line in
/usr/share/pipewire/media-session.d/media-session.conf

    with-audio = [
        metadata
        default-nodes
        #default-profile
        default-routes
        alsa-seq
        alsa-monitor
    ]

Seems to be enough to fix this bug without having to cripple PipeWire in any significant way.

If the above fix can be adapted to only omit default-profile in the presence of another audio server like pulse then I think that would be even better... But I don't know enough about the config semantics yet.

[Daniel van Vugt (vanvugt)]

I'm still not confident about the syntax of the solution we want so have asked upstream: https://gitlab.freedesktop.org/pipewire/media-session/-/issues/6

[Sebastien Bacher (seb128)]

Debian bug about splitting the audio configuration to its own binary which pipewire-pulse could depends on, https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1006364

[Daniel van Vugt (vanvugt)]

It seems the developers didn't really intend for media-session.conf to be edited. Instead it's the existence of the files 'with-jack', 'with-pulseaudio' or 'with-alsa' that trigger the 'with-audio' settings.

So I now think the cleanest solution is for us to just not install these files:

  with-jack
  with-pulseaudio

Then the bug is fixed. Anyone who wishes to re-enable audio support need only create an empty file called 'with-pulseaudio', 'with-jack' or 'with-alsa'.

P.S. I half expect long term the fix might be to move 'default-profile' (and others) into 'with-alsa'. 'with-alsa' is disabled by default but appears to be the intended method of enabling native audio support between pipewire-media-session and the kernel.

[Daniel van Vugt (vanvugt)]

Oh right, Igor mentioned that already in comment #22. :)

[Daniel van Vugt (vanvugt)]

Debian fix proposed in: https://salsa.debian.org/utopia-team/pipewire-media-session/-/merge_requests/1

[Daniel van Vugt (vanvugt)]

If anyone wanted PipeWire audio working at the same time as PulseAudio with Pulse as the main server then I suspect the module you want is 'pulse-tunnel'. I "suspect" that but the documentation is not great:

https://docs.pipewire.org/page_module_pulse_tunnel.html

It probably doesn't matter. We're likely to stay all-PulseAudio for now, and later switch to all-PipeWire so pulse-tunnel should never be required (unless the fix in comment #48 breaks audio in screen recording/sharing but even then I'm not really sure that's a thing).

[Daniel van Vugt (vanvugt)]

Seems Debian Salsa has been down for days. If anyone is curious what my final proposed fix looks like, here it is.

[Ubuntu Foundations Team Bug Bot (crichton)]

The attachment "Don-t-load-audio-modules-by-default-LP-1953052.patch" seems to be a debdiff. The ubuntu-sponsors team has been subscribed to the bug report so that they can review and hopefully sponsor the debdiff. If the attachment isn't a patch, please remove the "patch" flag from the attachment, remove the "patch" tag, and if you are member of the ~ubuntu-sponsors, unsubscribe the team.

[This is an automated message performed by a Launchpad user owned by ~brian-murray, for any issue please contact him.]

[Daniel van Vugt (vanvugt)]

No please don't treat it as a debdiff because it should be reviewed in https://salsa.debian.org/utopia-team/pipewire-media-session/-/merge_requests/1 instead.

[Sebastien Bacher (seb128)]

Thanks Daniel, I uploaded to Ubuntu now to not block longer on upstream or Debian. I suspect that Debian is rather going to want to go the road of changing the default, which we will likely do after the LTS, the delta isn't going to have a cost at this point. I also slightly simplied to rules change on having the option directly on the dh_auto_configure line since we usually use a variable only when the option needs to be dynamically set which isn't the case here, let me know if I missed something though

[Launchpad Janitor (janitor)]

This bug was fixed in the package pipewire-media-session - 0.4.1-2ubuntu1

---------------
pipewire-media-session (0.4.1-2ubuntu1) jammy; urgency=medium

  * Configure with-module-sets="" to disable audio by default (LP: #1953052)
    You can reenable it by creating an empty file:
    /usr/share/pipewire/media-session.d/with-pulseaudio

 -- Daniel van Vugt <email address hidden> Sat, 05 Mar 2022 12:01:39 +0100