You are not logged in.
Not exactly sure about the OS name "daedalus/ceres"
That is taken from /etc/os-release, which is provided by the base-files package. That line always lists either stable or testing/unstable (by names), it never distinguishes between the development branches.
EDIT: one of the reasons for this is that most experienced users of the development branches will have both testing and unstable in their sources (and perhaps even stable as well), with the appropriate pinning.
https://bugs.launchpad.net/ubuntu/+sour … ug/1752194
Does reverting those GRUB & sysctl changes remove the error?
Btw I'm pretty sure kernel.dmesg_restrict is already enabled by default.
From the gist of it I get the distinct impression he's suffering from a bad case of the Dunning-Kruger effect. But seriously his facial hair is so irritating I literally can't watch the cunt without wanting to smash the screen. Sorry.
Debian instructions here: https://wiki.debian.org/PipeWire#Debian_11
Needs some tweaking though because the pipewire systemd unit files don't have an /etc/init.d/ alternative and it relies on socket activation; not such if that's supported at all outside of systemd.
Check https://dev1galaxy.org/viewtopic.php?id=4865 for a way to make init scripts and see if you can make it work. Post back here if you have any problems.
Thank you for the extended reply. I always enjoy reading your posts.
For the record I do agree that the GPL is superior to the BSD-style licenses but that really has nothing to do with blobs. Their code is still open even if you don't like the terms of redistribution.
I'll leave the licensing argument itself to blackhole (if they can be bothered). I'm not very good with that sort of thing.
The Flaws of Academic Statistics: the Null Ritual - https://piped.kavin.rocks/watch?v=an0RFLzJ5Yo
Hold on, that's Luke Smith. He's an idiot. I'm not wasting an hour of my life listening to his drivel. He doesn't offer refunds. I've tried before with that dickwad.
I studied statistics in medical school back in the 90s. He is speaking out of his ass. IMO. Not that I've bothered listening
It's not grey at all. The FAQ and installation guides make clear what will happen. And if the user is so against non-free firmware why do they have hardware for which it is a requirement?
Any hardware that doesn't load firmware blobs from the operating system has them implanted at the factory instead so it's not like blobs can be avoided with modern components like hard drive & motherboard controllers.
get over it
Never!
No, Kelsoo is not correct. The FAQ clearly states that firmware is *not* included with OpenBSD, and hence the kernel is blob-free.
And my comment was in respect of the spelling rather than any philosophical argument.
From the FAQ:
For licensing reasons, some firmware cannot be directly distributed with OpenBSD. The fw_update(8) tool will automatically download and install any missing firmware, but this requires a working internet connection.
https://www.openbsd.org/faq/faq4.html#WifiOnly
EDIT: and please sed 's/SystemD/systemd/' in your signature. TIA.
You don't listen to lectures and academic podcasts while you are bike riding
Dear me, no. I need to have all my senses available and at full disposal to maximise my chances of survival.
If you want to speed-read through the 9 studies that are referenced in the podcast, you can find them listed on the podcast website, notrelated.xyz.
Thanks.
He did another one a few months later that's of interest in the same vein - "Against Method and For 'Pseudoscience'".
Hmm. They do know that pseudo is latin for "false", right?
Only driver installed is firmware-amd-graphics
Firmware != driver.
Did you install the relevant Realtek non-free firmware package as well?
If you have to configure your phone connection manually after first boot then the automatic method would not have had a connection available to download the firmware.
The Flaws of Academic Statistics: the Null Ritual
Is there a transcript? I'm very impatient and I can read faster than I can listen.
it doesnt even try to connect
Did you remove the Tomas Pinho driver version? It might conflict or override the mainline version.
Otherwise check the logs.
If you have seatd installed and running then consolekit should not be needed. For seatd the user needs to be in the audio, video & seat groups. Or at least in Alpine that's the case. Have you asked about this on the antiX forums? The community there will probably have a better grasp of the situation.
I followed this guide but it would have to be adapted. This is under sway, I can't remember what antiX uses. Is it fluxbox or something similar?
EDIT: and do you really need pipewire? I find that everything on my system works without a sound server. IIRC it's only remote conferencing that needs one.
One for golinux:
Why Most Published Research Findings Are False
It is old though. We're much better now. I hope.
An unreferenced quote talking about "firmware images" in a vague and general sense doesn't really seem to support a claim that the OpenBSD kernel contains blobs.
Perhaps try OpenBSD yourself? The system is installed in a pristine, blob-free state but on first boot the hardware is checked and any firmware that is needed is downloaded and installed. It is an automatic process but it only works if the machine has an internet connection. OpenBSD is pragmatic about non-free firmware because if the user didn't want blobs then why do they have hardware that requires it?
So the fact remains that the OpenBSD kernel does not contain blobs and nor does an installed system. The only time blobs are added is if required and if the machine is connected to the internet.
Debian will include non-free firmware in the official bookworm release, which is not something that will ever happen with OpenBSD (IMO).
You should have installed the metapackage rather than an individual kernel. 5.18 is old now.
# apt install -t stable-backports linux-image-amd64
I posted here asking about timing of the update of these two packages
Those packages are no more. The last versions were for Debian oldstable.
Have you tried seatd with pipewire? That works for my Alpine system.
Why do you think you need those two packages? I've just installed the consolekit package in a daedalus system with no problems. Those packages are not dependencies of consolekit.
After what, exactly? What did you change?
You can use a static IP address instead, that doesn't require a running dhcp daemon or DHCP router support. Just make sure you don't pick an address that's already in use.
I don't use NVIDIA but shellcheck.net says that script is a pile of buggy shite. Don't use it. Have you tried nouveau?
My system works with efivarfs as fs_spec under sysvinit, runit & openrc. It's using the daedalus kernel also. How bizarre.
Perhaps your machines are haunted
DHCPDISCOVER on eth0 to 255.255.255.255 port 67 interval
many times at booting and no network after booting
I entered "dhclient syslog DHCPDISCOVER" into a search engine and found this:
https://serverfault.com/questions/32715 … r-of-times
how can i choose networking? from upper config file or NetworkManager?
If an NIC is configured in /etc/network/interfaces then NetworkManager will ignore it.
I always remove NM. It's useless bloat IMO.
Tried with other kernels
Devuan's stable-backports repository has >6, have you tried that? Newer is probably better for that card, upstream is on v6.0.10 at the moment.
There are also some driver parameters that can be tweaked to aid performance:
# tee /etc/modprobe.d/rtw8821ce.conf >/dev/null <<!
options rtw88_pci disable_aspm=y
options rtw88_core disable_lps=y
!
# update-initramfs -u -k all
# reboot
^ The tee command creates a file at /etc/modprobe.d/rtw8821ce.conf that disables ASPM & Deep PowerSaving, which might help maintain the connection. Delete the file & run update-initramfs -u -k all again if it doesn't help.