The same one, open it in the editor.

I opened '/etc/init.d/dbus' in Mousepad. It reported "1 occurrence" when I searched for 'IDTYPE'.

$ uname -a
Linux ng3 6.1.0-22-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.94-1 (2024-06-21) x86_64 GNU/Linux

Where is the 'dbus' that you opened in KWrite?

Ah! No need to respond! It took a while, then I spotted that I mis-read: it is in /etc/default/dbus.

Many thanks. I've now changed that setting.

$ cat /etc/init.d/dbus | fgrep -i IDTYPE
  if [ "${IDTYPE}" = "RANDOM" ] && \

Here is the only bit of /etc/init.d/dbus that applies:

create_machineid() {
  # Create or recreate machine-id file
  
  UPTIME=$(cat /proc/uptime | cut -d " " -f 1 | sed 's/\..*//g')
  # re-create machine-id if it already exists and this is the first 
  # time the initscript is called (either PIDDIR does not exist or
  # uptime is less than GRACETIME)
  if [ "${IDTYPE}" = "RANDOM" ] && \
     [ -f "${MACHINEID}" ] && \
     [ ! -d "$PIDDIR" -o ${UPTIME} -lt ${GRACETIME} ]; then 
        rm -f "${MACHINEID}"
  fi 
  if [ -x $UUIDGEN ]; then
    $UUIDGEN $UUIDGEN_OPTS
  fi
}

Presumably something sets $IDTYPE to "RANDOM", but it is not in that script-file.

the IDTYPE=RANDOM was the problem for me. comment-out this line and machine_id
remains stable across boots.

this change keeps pulseaudio stable with no build-up of junk files in ~/.config/pulse.
it also keeps the ~/.dbus/session-bus/ directory clean as well, which is welcome.

i had scripts to handle the build-up of junk files in these two dirs, which have now been deleted.

FYI, after making the above change i came across this blog post, which adds a little more info:

https://kelar.org/~bandali/blog/pacify.html

You can edit /etc/default/dbus to comment the last line as shown below to disable the machine-id change.

# This is a configuration file for /etc/init.d/dbus; it allows you to
# perform common modifications to the behavior of the dbus daemon
# startup without editing the init script (and thus getting prompted
# by dpkg on upgrades).  We all love dpkg prompts.

# Parameters to pass to dbus.
PARAMS=""

# IDTYPE: how to deal with /var/lib/dbus/machine-id:
# 
# if IDTYPE="RANDOM": regenerate /var/lib/dbus/machine-id at each boot
# else keep it fixed across reboots
#IDTYPE="RANDOM"

It's also possible to rename those files so they use the current machine-id so they work.

defaults.ctl.card 0
defaults.pcm.card 0
defaults.timer.card 0

use your researched values...

~/.asoundrc has precedence over /etc/asound.conf

I use this to Not use hdmi (desktop-tower, not laptop)... I've sure it works both ways.

This being said, to each their own.

If you can get it working of course.

Some distros can, hard to know if Devuan is one of them, never tried before for whatever reason. Perhaps I didn't know if the service's name. Idk...

People who need pipewire and pulseaudio aren't dumb, they just like unneeded bloatware.

Still foolish? In my opinon, yes... but what can you do.

:shrug:

btw, nice necro. This page was dead 2 years ago.

If (like me,) you can not, for example, get sound out of your hdmi monitor speakers, all alsa configuration methods will fail, I tell you no lie, I am not the only one, we are many.

apt-get install alsa-topology-conf alsa-ucm-conf alsa-utils

/etc/init.d/alsa-utils reset

apt-get install pulseaudio pavucontrol

In GUI, run pavucontrol and setup configuration, e.g., I want sound out of line out on my box and my HDMI monitor speakers as well.

Purge pulseaudio and pavucontrol, I purge the flipping lot with debfoster (& rm /var/lib/debfoster/keepers.)

touch /etc/asound.conf

vi (or whatever,) /etc/asound.conf
defaults.pcm.card 0
defaults.pcm.device 3 # my flipping HDMI monitor speakers
defaults.ctl.card 0

pavucontrol does something that makes alsa asound.conf 'work.'

Do `/etc/init.d/alsa-utils reset` again, and it will trash it.

A very clever person could tell us what pavucontrol 'modifies,' so we wouldn't have to do the pavucontrol thing, and we are many.
If somebody tells us what pavucontrol does, without actually doing it themselves, that's not very clever.
AT ALL

P.S., try pavucontrol before jerking around with modules, and firmware-intel-sound firmware-misc-nonfree firmware-sof-signed

then at the end of /etc/pulse/default.pa I put:

# set-card-profile name-of-card-without-angle-brackets name-of-profile-without-trailing-colon
set-card-profile alsa_card.pci-0000_03_06.0 output:analog-stereo+input:analog-mono

After adding another PCI-Express card to the pc, I needed to comment out the line referencing the sound card in /etc/pulse/default.pa and re-run:

pacmd list-cards

which gave the new name of the card as:

alsa_card.pci-0000_04_06.0

and updated the line in /etc/pulse/default.pa to:

set-card-profile alsa_card.pci-0000_04_06.0 output:analog-stereo+input:analog-mono

Thanks to this thread I solved the issue for myself and filed a bug report:

https://bugs.debian.org/cgi-bin/bugrepo … bug=986211

My solution was to do:

pacmd list-cards

(program pacmd is in package pulseaudio-utils)

to find the actual output that I wanted to use (in my case, analogue stereo output on a Soundblaster Audigy 2 ZS)

then at the end of /etc/pulse/default.pa I put:

# set-card-profile name-of-card-without-angle-brackets name-of-profile-without-trailing-colon
set-card-profile alsa_card.pci-0000_03_06.0 output:analog-stereo+input:analog-mono

I also recommend installing the package pavucontrol and using the command pavucontrol if something gets accidentally muted.

This seems to be a problem of lack of documentation on setting the default audio profile.

How about neither...

Since they both are redhat garbage software.

I tire of the corporate ideal to eff things up that work well just to make people depend on software that is bloated and sux...

Smh...  greed does wonders... or horrors if you prefer.