[Stable Update] 2026-02-23 - Kernels, Mesa, Wine, COSMIC, GNOME, KDE Frameworks

Announcements Stable Updates
45

After this update my rootless Docker failed to start:

failed to create NAT chain DOCKER
iptables failed: modprobe: ERROR: could not insert 'ip_tables': Operation not permitted

Cause: rootless mode tried to use iptables, but cannot load kernel modules.

Fix was to disable iptables in ~/.config/docker/daemon.json:

{
  "iptables": false,
  "ip6tables": false
}

Then:

systemctl --user daemon-reload
systemctl --user restart docker

Rootless Docker starts normally now.

Hope this can helps someone.

2 Likes
46

I have a Pascal NVidia graphics card and wanted to follow this advice:

Install the official linuxXXX-nvidia-575xx, linuxXXX-nvidia-570xx, or related DKMS packages.

But alas, it looks like its not available for kernel 6.19?

47

Arch recently added nftables as a required dependency to docker v29:

This page might be helpful if you need to make any adjustments to your configuration:

1 Like
48

I had to reinstall keyring packages:

sudo pacman -Syy archlinux-keyring manjaro-keyring

The described solution corrupts pgp keys.

1 Like
49

By the way it is “Arbeitsflächenanimationen” not “Arbeisflächenanimationen”. The “t” is missing.

image
image1035×589 103 KB

2 Likes
50

Yesterday I updated my Dell latitude 5530, Lenovo LOQ 15IAX9I and desktop PC (MSI Tomahawk with RTX 2060 Ti)

Regular NTB’s / PC’s with some office work, browsing and some media playerz

all without any issues and on KDE plasma and also installed 6.18 kernels; had no issues with key-signatures

have not switched to new login manager for KDE Plasma
Gr8 job Team !

1 Like
51

plasmashell running 1 core at 100% after this update, which I also try to replace sddm with plasma login manager. At first, I though it’s due to the latter, but going back to sddm doesn’t resolve the issue, so it should be something within plasmashell. htop shows this happens at the root process, not any of the child processes.

EDIT:
One of my USB4 connected external SSD connection is faulty, it makes the kernel keep trying to enumerate the device over and over again. Replugging the device seems to have fixed it. It may be triggered due to the reboot after update.

EDIT 2:
That’s only half of the story. While neither dmesg nor journalctl shows any more spammy log, but plasmashell is still hovering around 50% single core usage.

EDIT 3:
It’s this manually installed plasmoid. I never updated it and it seems to require one, everything is good now.

52

5 posts were split to a new topic: Following Stable Update 2026-02-23, default DE and home directory changed?

54

That’s a plasmoid for Plasma 5. No wonder it misbehaves. :wink:

55

No, it’s for Plasma 6, the Plasma 5 version is on a different page as the description says (they back link each other for clarity, I guess). I missed the v0.2.7 update, probably 1 of the 3 changelogs (I’ll put my bet on the 3rd).

1 Like
57

A post was split to a new topic: pCloud FUSE filesystem client stopped working

59

Looks like issue is solved now: newer versions landed to stable.
Some older version than current fixed package conflict and 16.39.11-11 mitigates cursor change in userspace level: Comparing 16.39.10...16.39.11 ¡ manjaro-sway/desktop-settings ¡ GitHub

Also brezee-cursors now fixes default cursor theme issue in system level too: [pkg-upd] 5.27.12-3 (d41e80f2) ¡ Commits ¡ Packages / extra / breeze-cursors5 ¡ GitLab

60

The update went smoothly (:+1: developers!), but now I have a suspicious warning:

 sudo pacman -Syu 
:: Synchronizing package databases...
 core is up to date
 extra is up to date
 multilib is up to date
:: Starting full system upgrade...
warning: btrfsmaintenance: local (1:0.5.2-1) is newer than extra (0.5.2-3)
there is nothing to do

Seems like a change of version number name that is confusing pacman?

I fixed it with sudo pacman -Syuu (allowing “downgrades”). I hope it was the correct way…

1 Like
61

Since version 0.41 vo=gpu-next is the default. If you’re still using vo=gpu it might be buggy.

62

I think you mean

not “Arbeisflächenamimationen” :sweat_smile:

You might want to report it to https://bugs.kde.org/enter_bug.cgi

1 Like
63

You’re right, I overlooked the second mistake.

64

I updated and everything is fine except that the nextcloud system tray icon and nextcloud desktop theme are different (and only for my account). I did some digging and cannot figure out how to change it back to the way it was. This is a QT app, isn’t it?

65

Hi, I had the same problem with LXQt.
I ignored the updates and I removed manjaro-lxqt-extra-settings and other 2 packages that it found out: manjaro-lxqt-base-settings and manjaro-lxqt-skel-20171029. This removed the old package xcursor-breeze, then i ran the full update.
I don’t know if this is a good solution but it worked for me.
You can wait for confirmation of Manjaro Team here.

66

System is unbootable after the update. The last 3 times this happened was Nvidia shenanigans, but from what I could see they didn’t update this time… I read something about new systemd hooks, I don’t know. This was on kernel 5.15, I didn’t try a new one, just Timeshift’ed back.

67

Let us guess… Your kernel was gone or something?