Would you mind sharing with us how did you get it working ?
Did you do “grub-update” or something ?
However, looks like Kernel 6.11.2 is up and coming shortly…
1 Like
Are you using Nvidia graphics?
I haven’t used nvidia in quite a few years, (so perhaps my memory is shaky), but I think nvidia can cause problems with a kernel downgrade. If you’re using the proprietary nvidia driver, does this not require downgrading and freezing the nvidia driver version in some cases as well.
Freezing multiple core components is never a good idea IMO. Others opinions may differ on this issue, but to me using the LTS kernel is the better option in cases such as this.
Well I use Intel IGPU so I don’t have to deal with driver shenanigans thankfully, last time I froze my kernel packages was because I was completely fine and stable with the vanilla linux package and didn’t have WiFi access , I did have Cellular Data tho so I did that to save some data.
I did update my system and not doing the kernel updates did not harm me , saved me some data and the system was also pretty stable afterwards.
Using dkms versions of nvidia drivers should be fine regardless of kernel versions, right ?
NOTE FOR NEW USERS:
Check my reply here if you do not want to break your system ! :
I would think the dkms versions would be OK, but in some cases the driver in use may not be dkms.
@BluishHumility knows far more than myself about the intricacies of dealing with nvidia drivers, perhaps he might be kind enough to share his knowledge of the subject with us.
ME?! I only know what is in the wiki, or what I read in the forums. I have never owned Nvidia hardware…and probably never will.
Technically, yes the DKMS package “should” work. But it is obviously a different package, so likely has differences that may affect stability or performance. If it works, it works 
but in general I would recommend sticking to the specific package which has been recommended for the hardware/kernel combination in use.
https://wiki.archlinux.org/title/NVIDIA#Installation
GPU Family Driver Status Turing (NV160/TUXXX) and newer nvidia-open for linux
nvidia-open-dkms for any other kernel(s)Current, Supported1 Maxwell (NV110/GMXXX) through Ada Lovelace (NV190/ADXXX) nvidia for linux
nvidia-lts for linux-lts
nvidia-dkms for any other kernel(s)Current, Supported Kepler (NVE0/GKXXX) nvidia-470xx-dkms AUR Legacy, Supported2 Fermi (NVC0/GF1XX) nvidia-390xx-dkms AUR Legacy, Unsupported2,3 Tesla (NV50/G80-90-GT2XX) nvidia-340xx-dkms AUR Curie (NV40/G70) and older No longer packaged
3 Likes
So I just recovered from yet another Kernel 6.11.2 broke everything symptom.
I saw the Kernel has been updated so I tried.
It broke everything the same as before.
I couldn’t launch apps with the launcher, not even from a terminal…
Bluetooth was not on, so I turned it on, and wiped out all the paired items.
I couldn’t even restore snapshots nor shut down from the KDE main menu.
I had to turn off manually and the booted to 6.05-amd snapshot, and restored the last working snapshot again by using Snapshot Tool.
Do I have to give up on this 6.11 series kernel ?
Please help me what’s causing these problems to me and some other people and some other people are doing just fine ?
Thanks,
Isao
Try setting the kernel parameters mentioned in this thread from another forum:
1 Like
Sorry for putting you on the spot @BluishHumility . From all your great answers on Nvidia help requests I figured you were the go to guy for any graphics issues. Kudo’s for just saying no to nvidia.
2 Likes
Thanks, but my graphics card is AMD Radeon 7900 XTX.
Looks like the article is for Nvidia drivers…
My mistake, it seems there are a couple different things going on in this thread.
For your thing, try temporarily switching to the LTS kernel.
sudo pacman -S linux-lts linux-lts-headers
That way you can keep your other packages fully up to date, but avoid using the problematic kernel version until a fix comes down. Give it a week or so, then test the Zen kernel again.
3 Likes
Thanks, I’m currently just enabling IgnorePackage= in /etc/pacman.conf, and getting by. I can update other packages without affecting linux-zen.
Also, I just tried the kernel update to zen-6.11.2 on my Framework 16 laptop, and everything works just fine.
How do I look into log files and what log files when I can’t even use Konsole on my desktop PC after 6.11.2 update ? I really wish I could compare at least dmesg output…
Isao
Beware, updating your system with ignored packages can put you in a partially upgraded state. This is because libraries and other packages which depend on (or are dependencies of) ignored packages will continue to get updates. If you must ignore packages, it is better if they are packages that don’t have dependency relationships. Not the Linux kernel, which crucial core libraries, utilities, and drivers depend on.
It is generally safer to not update any packages at all if you cannot update them all together for some reason.
You can still inspect the journal after restoring a snapshot, since the logs are stored in a separate subvolume from root (in other words, the journal is not rolled back with the rest of the system). Another option would be to switch to a TTY from the broken system and access the journal from there.
1 Like
Thank you for pointing out those info, BluishHumility.
Sorry for asking questions after questions, but how do I look into those journals in a TTY ?
Please help,
Isao
You can switch between TTYs with Ctrl+Alt and the F-keys (Ctrl+Alt+F2, Ctrl+Alt+F3, and so on).
You can access the journal with journalctl. A lot of messages are written to the journal, so typically you will wish to narrow your focus by searching for a priority level, a specific time frame, messages related to a specific service, et cetera. You can learn more about how to read through the journal in this wiki post: systemd/Journal - ArchWiki
1 Like
Thanks.
Now my desktop PC spontaneously frozen while I was watching 4K anime, and I got the following “jornalctl -b -1” after a fresh reboot.
Oct 08 21:36:53 isao-dsktp05 systemd[1]: Started Daily Cleanup of Snapper Snapshots.
░░ Subject: A start job for unit snapper-cleanup.service has finished successfully
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ A start job for unit snapper-cleanup.service has finished successfully.
░░
░░ The job identifier is 6667.
Oct 08 21:36:53 isao-dsktp05 systemd[1]: Starting DBus interface for snapper...
░░ Subject: A start job for unit snapperd.service has begun execution
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ A start job for unit snapperd.service has begun execution.
░░
░░ The job identifier is 6790.
Oct 08 21:36:53 isao-dsktp05 systemd[1]: Started DBus interface for snapper.
░░ Subject: A start job for unit snapperd.service has finished successfully
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ A start job for unit snapperd.service has finished successfully.
░░
░░ The job identifier is 6790.
Oct 08 21:36:53 isao-dsktp05 systemd-helper[364229]: Running cleanup for 'root'.
Oct 08 21:36:53 isao-dsktp05 systemd-helper[364229]: Running number cleanup for 'root'.
Oct 08 21:36:53 isao-dsktp05 systemd-helper[364229]: Running timeline cleanup for 'root'.
Oct 08 21:36:53 isao-dsktp05 systemd-helper[364229]: Running empty-pre-post cleanup for 'root'.
Oct 08 21:36:53 isao-dsktp05 systemd-helper[364229]: Running 'btrfs qgroup clear-stale /.snapshots'.
Oct 08 21:36:53 isao-dsktp05 systemd-helper[364229]: 'btrfs qgroup clear-stale /.snapshots' failed.
Oct 08 21:36:53 isao-dsktp05 systemd[1]: snapper-cleanup.service: Deactivated successfully.
░░ Subject: Unit succeeded
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ The unit snapper-cleanup.service has successfully entered the 'dead' state.
Oct 08 21:37:53 isao-dsktp05 systemd[1]: snapperd.service: Deactivated successfully.
░░ Subject: Unit succeeded
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ The unit snapperd.service has successfully entered the 'dead' state.
Looks like snapper is causing my system hung ?
When this happened, my kernel is still 6.10.10.
As you’re saying, the system is getting unstable, I think…
I would suspect that WD HDD you got there. Try to disconnect it and test without it.
I’m doing gpu passthrough it was stuff related to vfio, so i just cleaned the grub and left as default and generated it again. Working like a charm now
And yes, after editing grub you need to use update-grub
Thanks.
I disconnected those as well as another USB device and installed Kernel 6.11.2, and it is almost working.
Only issue now I am having is the Bluetooth not working as before, as with 6.10.10.
Here is my journalctl output:
╭─isao@isao in ~ took 30ms
╰─λ journalctl -p 3 -x -b -1 | grep -i blue
Oct 09 22:40:12 isao-dsktp05 bluetoothd[1099]: Failed to add UUID: Authentication Failed (0x05)
Oct 09 22:40:12 isao-dsktp05 kernel: Bluetooth: hci0: Opcode 0x0c24 failed: -110
Oct 09 22:40:12 isao-dsktp05 kernel: Bluetooth: hci0: command 0x0c24 tx timeout
Oct 09 22:40:14 isao-dsktp05 bluetoothd[1099]: Failed to add UUID: Authentication Failed (0x05)
Oct 09 22:40:14 isao-dsktp05 kernel: Bluetooth: hci0: Opcode 0x0c24 failed: -110
Oct 09 22:45:09 isao-dsktp05 dbus-broker-launch[1090]: Activation request for ‘org.bluez’ failed.
Oct 09 22:45:09 isao-dsktp05 dbus-broker-launch[1090]: Activation request for ‘org.bluez’ failed.
Do I just restart bluez at this point ?
Isao
Edit: I spoke a bit too soon. On the first boot when I wrote the above post, I had a spontaneous freeze right after I plugged in a USB dongle, but after the 2nd boot with 6.11.2, my bluetooth (that is built-in on the mother board) started to work. Looks like everything is back to normal now. I really don’t get it…
No packages require the linux kernel explicitly as a dependency, and until a super crucial update which overhauls things and makes certain system libraries incompatible with older kernels(which is very unlikely to happen) is pushed upstream, this “partial upgrade” is very safe IMO, even if issues to arise from a old kernel (very unlikely) then a simple sudo pacman -Syu should fix things.
It isn’t as dangerous as you might think, though again, partial upgrades are not advisable but this can be considered an exception.
NOTE FOR NEW USERS:
Check my reply here if you do not want to break your system ! :
Is this a typo? If this is what you meant to say, it is false.
You can see the packages which explicitly require the kernel as a dependency by running pacman -Sii linux (see the Required By line). Basically any package that includes out-of-tree modules will require the kernel as a dependency.
Additionally, you can see that the kernel itself depends on several core packages (see the Depends On line). If these packages are updated while the kernel is being held back, it could cause compatibility issues or instability.
This is also false. A partial upgrade is never “safe”. Barring any major security vulnerabilities that need immediate patching, it is far safer to just not update your system at all.
I insist, there are no exceptions to the fact that partial upgrades are not advisable.
In general, your whole post is problematic because there is the potential for inexperienced users to read it and mistakenly think what you are saying is true. It has already been stated in this thread, but perhaps it bears repeating: partial upgrades are not supported, and should be avoided at all costs.
1 Like