You are not logged in.
not sure if submitting that to debian or straight up go and try to get it into devuan?
Looks like the Debian zram-tools maintainer doesn't care about non-systemd support:
https://bugs.debian.org/cgi-bin/bugrepo … bug=960524
anyone who wants to help adding proper support for openrc and runit?
Devuan's runit will use sysvinit scripts. Search the boards if you want a method for setting up proper supervision with runit, it has been covered here before.
For OpenRC see https://github.com/vaeth/zram-init/blob … /zram-init
in theory pulseaudio is probably one of the best development for am universal use of powerful computer in multimedia as it allow to mix or and separate divers sounds in and output objects
PipeWire is *much* better than crusty old PulseAudio.
See https://trac.ffmpeg.org/wiki/Capture/PulseAudio for the ffmpeg documentation on the subject. I think each browser tab should be exposed as a separate source, unlike with pure ALSA, but I can't test this because I use pure ALSA.
The same method can be used with PipeWire if the pipewire-pulse package is installed.
There is also a page for ALSA but that seems to just allow recording the entire output from the browser so your stream would be interrupted if you listened to anything else. I might be wrong though.
assuming Devuan actually ships a browser with proper ALSA support
The firefox-esr package is compiled with --enable-alsa and works just fine with pure ALSA.
Exec=eval "$(/usr/lib/policykit-1-gnome/polkit-gnome-authentication-agent-1)"
Why the eval? The agent is directly executable.
A simpler approach is to just add this line to ~/.xsessionrc:
/usr/lib/policykit-1-gnome/polkit-gnome-authentication-agent-1 &Reference: https://wiki.debian.org/Xsession
This will clear USB sticks quicker than dd (install the gdisk package first):
sudo sgdisk --zap-all /dev/sdX
sudo wipefs -a /dev/sdXAs with dd be very careful with the device letter because those commands can nuke your system if you get it wrong.
Trackpad desired behaviour is right click to select options etc for the context, left click to select the folder/file/text/url etc,
In what way does your current configuration differ from this? How is it lacking? I can see that tap-to-click is disabled, is that what you want?
ditto for mouse, it is also erratic in that regard.
What does "erratic" mean, exactly? It is almost impossible to help you if I can't understand what the problem is.
I would actually recommend switching from the crusty old synaptics & evdev drivers to the newer libinput version:
# apt install xserver-xorg-input-{libinput,synaptics-,evdev-}If you're a fan of multi-touch gestures then xserver-xorg-input-mtrack is another possibility.
Do I actually need the xinput package to get this functionality to work
No.
I really cannot work out yet what needs to be changed.
Well explain exactly how you want the touchpad to behave (in as much detail as possible please) and post the full output of
synclientPlease use code tags when posting terminal output. Use the "quote" button on this post to see how I have made the above code block. No need to include the quote in your reply though. Thanks.
What "refuses to show its files"? Is it the file manager?
Can you mount the stick manually:
sudo mount /dev/sdXY /mnt
ls /mnt
sudo umount /mnt # wait until this command returns the prompt before removing the stickReplace X & Y with the drive letter and partition number for the USB stick (eg, /dev/sdb1). GParted should show this, I think. If not then check dmesg just after plugging the stick in.
when I type xinput it tells me the command is not found
The xinput command is supplied by the xinput package ![]()
See also https://wiki.debian.org/WhereIsIt#A.22I … tall.3F.22
synclient gives me a sensible list of many values, but I need to get the other parts corrected
What parts need to be corrected? In what way is the current configuration unsatisfactory?
sometimes triggering as if pressed when you merely scroll
Have you tried disabling the TapButtons? See the last post in the linked thread but set each of those options to zero (0).
If you still want taps to be recognised as clicks then perhaps try changing the MaxTapTime option.
@MLEvD: I only have a very vague idea of how sysvinit scripts work or I would have shared an attempt.
I could share an OpenRC script that would work but that's not very useful for our (now absent) OP.
EDIT: and even then all that /etc/rc?.d/ crap confuses me so it's probably best I don't ![]()
.targets ftw!
my thinking is just copy over to say /dev/sda3/debian11/debian11.iso then use that menu config like so below ?
That would only work if GRUB is installed on /dev/sda3 (ie, if grub.cfg is being read from that partition).
Otherwise the partition it must be identified on the loopback line, like this:
menuentry '[loopback]debian11' {
search --set=isopart --fs-uuid $uuid
set isofile='/debian11/debian11.iso'
loopback loop ($isopart)$isofile
linux (loop)/live/vmlinuz boot=live config fromiso=/dev/sda3/$isofile
initrd (loop)/live/initrd.gz
}Replace $uuid with the actual UUID of the partition.
Manual version here:
Okay, I've just seen those symbol errors in an Arch user's journal and they aren't complaining about any boot delays so the errors might be irrelevant.
Check /etc/initramfs-tools/conf.d/resume and confirm that the UUID printed there matches the UUID of the swap partition. Also check all entries in /etc/fstab for incorrect UUIDs, as per https://lists.debian.org/debian-user/20 … 00866.html
Use this command to list all UUIDs for your drives:
lsblk -o name,uuidI want to thank you
You're welcome ![]()
Is it bad to use an unsigned kernel?
The signed kernels can start with SecureBoot enabled but the unsigned versions cannot unless you sign them with your own key and enrol that into the UEFI firmware. The signature in question has been approved by Microsoft.
I don't understand why it isn't working cause aren't these the kernels from the unstable version 5? Yet I didn't have this problem running that version.
So you don't see the errors with other distributions that use the 5.15 kernel?
That kernel version is considered "stable" by the developers: https://www.kernel.org/. De{vu,bi}an use an LTS kernel version to suit their "stable" releases for which the API version should not change (which is what "stable" means in that context — it is not a synonym for "reliability" even though that tends to be a happy side-effect).
I guess the only thing now to do is try Liquorix.
You can try it but I'm pretty sure the ACPI interface will be the same as for the De{vu,bi}an kernels of the same API version (ie, 5.15).
I assume I need to do the following:
Actually I've just checked the script and it won't work for Devuan because it runs lsb_release -cs to check the distribution version.
This will modify the script to make it work:
curl 'https://liquorix.net/add-liquorix-repo.sh' | sed 's/^codename=.*/codename=bullseye/' | sudo bashThe apt-transport-https package can be removed afterwards though because the source uses http and that package is no longer needed for https sources in chimaera. I should probably tell the Liquorix folks about that...
I notice an apostrophe before https. I never noticed anything like that before. Is that correct or a typo?
It's correct, the quote is so curl parses it correctly. The command still works without the apostrophes though.
I would advise checking the script for any malicious code before piping it through to a privileged shell. I've vetted it myself but you shouldn't trust random people on the interweb ![]()
If the new kernels don't help at all go back to the original with
# apt install linux-image-amd64=5.10.84-1 linux-{image,headers}-liquorix-amd64-Then remove the added sources and the file at /etc/apt/trusted.gpg.d/liquorix-keyring.gpg and run apt update.
The backported kernel metapackage or Liquorix?
I would lean towards Liquorix because they update *very* quickly and their kernel is "desktop tuned".
Do I follow your instructions there to a -T- ??
They're old instructions. I rage-quit the Debian forums after that fat ugly bastard Steve McIntyre criticised me for being rude. Wanker ![]()
For Devuan add this line to /etc/apt/sources.list:
deb http://deb.devuan.org/merged chimaera-backports mainThen update the package database and install the metapackage:
# apt update
# apt install -t chimaera-backports linux-image-amd64^ That presumes a 64-bit system; for 32-bit use linux-image-686-pae or linux-image-686 for really old systems that don't support PAE.
During boot-up, I get this screen below, and it stays stuck on it for about 30-60 seconds, after which it then continues to the boot-up.
See https://bbs.archlinux.org/viewtopic.php?id=266085 & https://bugzilla.kernel.org/show_bug.cgi?id=213023
tl;dr: your motherboard firmware ("BIOS") is somewhat crappy and this is interfering with the new kernel's ACPI interface.
You could try updating the firmware but that can be risky (as seen in the Arch forums thread) and you could also try adding acpi=off as a kernel command line parameter but that might break other stuff. Other acpi= options are available: https://github.com/torvalds/linux/blob/ … ers.txt#L1
The kernel bug report suggests this has been fixed in later versions so perhaps try the backported kernel metapackage and see if that works better. The Liquorix kernel is another option (they have Debian bullseye packages that should be compatible) and that is updated faster than backports so will be better in respect of security.
mine? didn't it count ...?
Not to me. Goodbye.
I do not understand the reason why I have to do something in swap
Because the resume device was set in the initramfs image when you installed Devuan but when you installed another distribution the installer re-formatted the swap partition which changed the UUID that the initramfs image was using to find the swap.
If the initramfs image can't find the resume device (swap) during boot it waits 30 seconds just in case it is slow to attach, hence the delay.
Just to confirm: do you see the message "gave up waiting for suspend/resume device" during the boot delay?
Going back to post#11, command:
sudo dmesg | grep -i "error||warning|fail|segfault|fatal".
A firmware is missing but fails to install because that package does not exist:
firmware: failed to load ath10k/pre-cal-pci-0000:02:00.0.bin (-2)
Do you not know how to use a search engine?
I entered that exact message into the search box at startpage.com and this was the top result:
https://forums.debian.net/viewtopic.php?p=703288
And arochester's answer gives this link which explains exactly why that message is totally irrelevant:
Report the driver crash to whoever wrote the module. Maybe they will be able to fix the problem.
https://lists.debian.org/debian-user/20 … 00866.html
You will have to repeat those steps whenever the swap partition is re-formatted, which most distributions do when they're installed.
EDIT: and if you had actually answered this question the thread would be a lot shorter and you wouldn't have wasted other people's time:
What do you actually see during the boot process?
The "gave up waiting for suspend/resume device" message should have been shown during the delay and I would have known exactly what it meant and how to fix it.
Yes, I know. My suggestion was for you to create them yourself.
The configuration in xfce4-pulseaudio-plugin is not kept after shutdown/startup
No idea about Xfce but systemd runs alsactl store on shutdown and alsactl restore on boot.
/etc/rc.local can be used to run alsa-restore so try that and see if it's enough. Perhaps Xfce runs alsactl store itself.
If the state isn't stored automatically you'll have to add scripts to /ect/rc6.d/ & /etc/rc0.d/ to run alsactl store at reboot and shutdown, respectively. Or create an init script in /etc/init.d/ to add the commands to the appropriate runlevels then run update-rc.d $name_of_script deafults to activate it.
The fancontrol package supplies an init script that should start the /usr/sbin/fancontrol daemon automatically.
Did you edit /usr/sbin/fancontrol directly or did you copy it out, edit that and run the copy from a terminal? The former method should be preferred but note that any alterations will be over-written if the fancontrol package is updated.
Is it possible to setup a shell script to run at startup and stay running?
I think pwmconfig just needs to be run once to generate the configuration file (hence the name). So you don't need to start it automatically or keep it running at all.
when it exits it sets everything to MAX. This is bizarre. Why?
To cool down the machine after it has operated with the fans disabled by pwmconfig when it attempts to test if it has control over them.
It's not a problem because the /usr/sbin/fancontrol daemon should set the speed of the fans.
Disclaimer: I've never used fancontrol so perhaps wait for steve_v to correct me.
Proton is supposed to be good, but can you use it without signing up for Steam?
A quick search suggests perhaps. But I've never tried it.
I only claim true love to wind up the h8ers. Sorry to disappoint.
EDIT: oh, you're being ironic. D'oh!