You’ve posted this in Support > Kernel, and the quoted part of your other post also shows kernel messages.
In other words, this has nothing to do with systemd. In fact, it’s the verbosity of the kernel’s own ring buffer — which is what you can look at by way of the dmesg command — and its output is hard-coded in the kernel sources themselves, since this output comes from the kernel’s address space.
The only way to suppress that is by adding quiet on the kernel’s command line in /etc/default/grub and running update-grub. You don’t need to add splash, because that’s only needed if you want to use plymouth or another boot splash.
Note: If at any time during the boot process you still want to see the boot-up messages, then you can press Esc.
I’m not even sure that quiet would fix the errors to pop during boot.
Wouldn’t a message classified as ‘error’ instead of ‘warning’ during the boot process, override the quiet boot parameter? I think I recall something like that.
Real fix for me is to work on the errors themselves if you have some, and if some errors shouldn’t be classified as errors I guess it should be fixed upstream?
//EDIT: yeah I think if log level is 3, which by default it is in Manjaro if I recall correctly, ERROR messages will override it.
If you really want the system to shut up, add this to grub: loglevel=0 udev.log_priority=0 systemd.show_status=false rd.udev.log_level=0 vt.global_cursor_default=0
If they are serious errors, they will still show, but this is also dependent on the udev.log_priority, which is another boot-time parameter.
Of course. If it concerns errors with regard to systemd services — which does not appear to be the topic here — then these should be looked into and corrected by the system administrator. But if they are kernel warnings, then either something’s amiss with the hardware, or they might require correction from upstream.
Note: Some of those spurious warnings/errors may actually be innocent, but nevertheless a result of patches applied to the kernel sources by the Manjaro developer responsible for compiling the kernels.
The holy wiki has an article on silent boot, if you want a black screen instead of animation or messages. Basically what pwx wrote adapted with loglevel 0.
I tried 6.18 LTS and the latest 7.1 Kernel but no changes. Since i have NEW Hardware, i should use the newest Kernel.
I still want to see if the boot process is fine or if something hanging, which includes the Green Status Checks for the System…
Nah i don’t want my system to entirely shut up
I just don’t want to see the unneeded false flag boot errors… or what ever they are.
But maybe reducing one of the log prioritys could help? Is what i need to find out then… thanks.
I normally would prefer to see all the green system checks, specially important if my system hangs… which sometimes happends specially when i shutdown the system.
For example I remember USER ID 1000 is hanging more often on my Laptop or my older PC, i think for this situation a QUIET command in Grub would be a big downgrade
If the red warnings/errors you see are appearing during the green “[OK]” parade, then these are not kernel messages but systemd service messages. This is an entirely different thing, and then you should examine why those messages are appearing, and remedy the problem.
Quite possibly you have some services starting at boot time which you don’t need.
Yeah i would normally made a photo from that, but the initialition procress is so quick,
i can’t even read it and the green lines moved the red info’s so fast down at my screen… its like a big blur.
With my other systems my info’s are entirely green… where i don’t care much… since Green Stays for Fine and Red for a problem.
You can’t hide red and display green, since red is higher priority. Every new level up you show/hide includes all the previous levels. It is just how it works.
These are mutually exclusive demands. One either has boot messages or not.
The errors are not unneeded, just de feedback, eg the code expects x gets y, report: I get y but expect x.
Fixing this will either be writing (your own) kernel patches or changing something about your system.
Okay i just went to my new PC… i was on my old PC before, here are the fresh logs now :
$ journalctl -p 3 -xb
Aug 20 20:59:22 koboldx-x870e kernel: virt/tdx: TDX not supported by the host platform
Aug 20 20:59:22 koboldx-x870e kernel: integrity: Problem loading X.509 certificate -22
Aug 20 20:59:22 koboldx-x870e kernel: integrity: Problem loading X.509 certificate -22
Aug 20 20:59:23 koboldx-x870e kernel: thunderbolt 0000:1c:00.0: probe with driver thunderbolt failed with error -110
Aug 20 20:59:24 koboldx-x870e kernel: Bluetooth: hci0: Malformed MSFT vendor event: 0x00
Aug 20 20:59:24 koboldx-x870e bluetoothd[760]: Failed to set mode: Failed (0x03)
Aug 20 20:59:29 koboldx-x870e bluetoothd[760]: Failed to set mode: Failed (0x03)
I think this 2 Lines (Services) showing up and annoying me in the boot procress:
Aug 20 20:59:23 koboldx-x870e kernel: thunderbolt 0000:1c:00.0: probe with driver thunderbolt failed with error -110
Aug 20 20:59:24 koboldx-x870e kernel: Bluetooth: hci0: Malformed MSFT vendor event: 0x00
Well, i also live with all these except thunderbolt one. Actually i have one more in the bluetooth and if i have headphones connected it spams the log every couple of seconds. Annoying.
At least the tdx will disappear in the next kernel versions, i think it was accepted as a bug to display it on non server CPUs.
TDX was new introduced, but at least TDX was showing up at the first message which i could easy ignore… but this red stuff between the green boot checks are really annoying.
I mean its okay, when i see it in journal… but not at a boot progress and the system is running still okay and are unaffected from this errors/warnings.