[Unstable Update] 2023-02-17 - Plasma 5.27 LTS, GNOME, Python

thanks @Yochanan, you save my all day life. btw: Is there any harm in changing to nvidia-dkms / virtualbox-host-dkms ?

The DKMS modules are the same as the precompiled modules… well, except for they’re not precompiled. It may take some time to compile them on potatoes :stuck_out_tongue_winking_eye:

4 Likes

Cannot open and decompress .zst files in xarchiver with the latest zstd (1.5.4-1) Downgrade to zstd 1.5.2.8 works fine. I didn’t try via CLI.

This looks to be an xarchiver issue as it works okay with File Roller(Archive Manager)

reported to xarchiver developer:

Update: the bug has been fixed. I was able to (crudely)build/ patch my xarchiver and it now will decompress .zst files with zstd 1.54-1

1 Like

I had recently to erase “.cache” entirely otherwise #plasma couldn’t function. I checked the usual suspect under “.config/plasma-org.kde.plasma.desktop-appletsrc” but it seemed fine.

I’m sharing this here in case this type of bug reaches stable. If it is an isolated case, the fix is doable.

2 Likes

A post was split to a new topic: Cups produces an error and will not install the printer

Kernel 5.15, 6.1 or 6.2 forces my PC (full AMD system) to shut down after waking up from sleep/suspend because the temperature is critical, but it is not true.

$ journalctl --no-pager -p 3 -b -1 --output=cat
thermal thermal_zone1: acpitz: critical temperature reached, shutting down
reboot: HARDWARE PROTECTION shutdown (Temperature too high)
reboot: Hardware protection timed-out. Trying forced poweroff

But I checked the temperature is actually not high. My guess is that the Kernel has this bug.

Can you confirm if you have the same problem when using AMD CPU and Kernel 6.1 or 6.2?


Solution:

My trick is to use kernel parameter thermal.crt=-1 to disable critical thermal. (But not recommend)

Kernel 6.1, AMD CPU and Nvidia card here, I do systemctl suspend wait 10 second, wake the computer, and write this post. All seem OK here.

Thanks for the answer.

Can you show the output of $ sudo dmesg | grep thermal ?

Mine is:

[    0.430394] thermal_sys: Registered thermal governor 'fair_share'
[    0.430394] thermal_sys: Registered thermal governor 'bang_bang'
[    0.430394] thermal_sys: Registered thermal governor 'step_wise'
[    0.430394] thermal_sys: Registered thermal governor 'user_space'
[    0.430394] thermal_sys: Registered thermal governor 'power_allocator'
[    2.652159] thermal LNXTHERM:00: registered as thermal_zone0
[    2.652160] ACPI: thermal: Thermal Zone [TZ10] (17 C)
[    2.652230] thermal LNXTHERM:01: registered as thermal_zone1
[    2.652231] ACPI: thermal: Thermal Zone [PCT0] (17 C)
[    2.652295] thermal LNXTHERM:02: registered as thermal_zone2
[    2.652296] ACPI: thermal: Thermal Zone [UAD0] (17 C)

But why Thermal Zone shows the default value 17 °C that is too small. :man_facepalming:

 omano  ~/Scripts/GitHub/dittoMenuKDE $  systemctl suspend

 omano  ~/Scripts/GitHub/dittoMenuKDE $  sudo dmesg | grep thermal
[sudo] password for omano: 
[    0.241781] thermal_sys: Registered thermal governor 'fair_share'
[    0.241781] thermal_sys: Registered thermal governor 'bang_bang'
[    0.241781] thermal_sys: Registered thermal governor 'step_wise'
[    0.241781] thermal_sys: Registered thermal governor 'user_space'
[    0.241781] thermal_sys: Registered thermal governor 'power_allocator'
[    2.475349] thermal LNXTHERM:00: registered as thermal_zone0
[    2.475350] ACPI: thermal: Thermal Zone [TZ10] (17 C)
[    2.475453] thermal LNXTHERM:01: registered as thermal_zone1
[    2.475454] ACPI: thermal: Thermal Zone [UAD0] (17 C)

 omano  ~/Scripts/GitHub/dittoMenuKDE $  

I see, this is why you do not have Thermal Zone [PCT0].


Update:

Adding the kernel parameter thermal.crt=30 to fix the issue. :+1:

[    0.430622] thermal_sys: Registered thermal governor 'fair_share'
[    0.430622] thermal_sys: Registered thermal governor 'bang_bang'
[    0.430622] thermal_sys: Registered thermal governor 'step_wise'
[    0.430622] thermal_sys: Registered thermal governor 'user_space'
[    0.430622] thermal_sys: Registered thermal governor 'power_allocator'
[    2.652337] ACPI: thermal: Critical threshold 30 C
[    2.652545] thermal LNXTHERM:00: registered as thermal_zone0
[    2.652546] ACPI: thermal: Thermal Zone [TZ10] (17 C)
[    2.652572] ACPI: thermal: Critical threshold 30 C
[    2.652618] thermal LNXTHERM:01: registered as thermal_zone1
[    2.652619] ACPI: thermal: Thermal Zone [PCT0] (17 C)
[    2.652638] ACPI: thermal: Critical threshold 30 C
[    2.652683] thermal LNXTHERM:02: registered as thermal_zone2
[    2.652683] ACPI: thermal: Thermal Zone [UAD0] (17 C)

Lately I observe strange behaviour of energy profiles on my laptop, the respective Plasma widget fails to switch a profile to any other than Balanced. It’s been like this for a week or so. At first I thought it’s kernel 6.2 related but now the same is haunting me on 6.1, too. Anyone having the same?

Latest updates introduced hard lockups on reboot.

EDIT: kernel updates resolved this.

After the last pamac update, I’m getting some strange behaviour between pamac and polkit1 in KDE. Even with a wrong password, pamac can install/remove/update packages when executed via the terminal. GUI doesn’t allow it without a valid password being entered. All is well with it in /etc/passwd

polkitd:x:102:102:PolicyKit daemon:/:/sbin/nologin
rtkit:x:133:133:RealtimeKit:/proc:/sbin/nologin
pamac update
Preparing...
==== AUTHENTICATING FOR org.manjaro.pamac.commit ====
Authentication is required to install, update, or remove packages
Authenticating as: George (george)
Password: 
polkit-agent-helper-1: pam_authenticate failed: Authentication failure
==== AUTHENTICATION FAILED ====
Error: Authentication failed
Nothing to do.
Transaction successfully finished.

I just tried pamac upgrade used the wrong password and got the same message as you.I tried it again with the right password and it showed I needed updates.

My wired Ethernet isn’t working under 6.2.6-1.

I checked the output with lshw -C network and got an *-network UNCLAIMED for RTL8111/8168/8411.

driver / module was not loaded (probably r8169), you can try manually (# modprobe r8169 and watch journalctl) but the question remains why it didn’t on startup.
Same controller / kernel here, all is fine.

1 Like

Found the issue! I missed an error on pamac stating “bad return status for module build on kernel: 6.2.6-1-MANJARO”. So I rebuild r8168-dkms and everything is back on track. Thanks, @dragan, for checking, it saved me some debugging.

I’m having trouble with linux510 5.10.175-1. It hangs and won’t continue booting. 5.10.174-1 works fine.

Looks like it’s the same as here: [Testing Update] 2023-03-18 - Kernels, Thunderbird, Cinnamon, KDE-git, Python, Haskell - #3 by Oldhabbits

When update mkinitcpio:
“WARNING: Possibly missing firmware for module: ‘csiostor’”

I attempted a fix with mkinitcpio 35.1-3.

2 Likes