Has anybody experienced issues with cheese after the update? Video flickers just like if the app was working on a persistent burst mode or something …
Unfortunately, I’m not able to reproduce the issue anymore since I decided to roll back to my previous snapshot. There should be many journal entries related to cheese with a gstreamer error, if somebody is willing to reproduce it.
Just decided to step up and try the update again.
This is what I get so far whenever I run cheese:
(cheese:1901): cheese-WARNING **: 13:16:10.322: Internal GStreamer error: code not implemented. Please file a bug at https://gitlab.freedesktop.org/gstreamer/gstreamer/issues/new.: ../gst-plugins-base/gst-libs/gst/video/gstvideofilter.c(296): gst_video_filter_transform (): /GstCameraBin:camerabin/GstViewfinderBin:vf-bin/GstVideoConvert:vfbin-csp:
invalid video buffer received
Smooth update, the only thing I’ve noticed is that when clicking on Show Applications (GNOME) the number of columns of the application grid changed from 6 to 7. This was an issue because the first row was below my application dock (in the left side of the screen).
This might be in preparation for GNOME40, however I don’t want it, and the fix in this case is installing App Grid Tweaks (GNOME extension) and changing the amount of rows and columns manually.
Update ran on my Raspberry Pi 4 but now X only runs for about 30 seconds then my monitor looses the signal and goes to sleep. Both xfce and i3 experience the same problem. If I’m quick I can login but makes no difference. Now stuck using tty console only.
There’s still an issue with resume from suspend and amdgpu at least on some Renoir (Ryzen 4xxx) systems, leading to the system hanging up when resuming from suspend. It works on 5.11.6 and it seems to work again on 5.10.32 (5.10 was affected for some subversions, too), but neither 5.11.16 nor 5.12.0 work, so I went back to 5.11.6 again.
See my topic about this:
After installing kernel 5.10.32, I checked whether it worked and it did. I then installed 5.12. When I rebooted into 5.11.6, I got an error message (“Failed to load Kernel modules…” or similar) and was thrown into fallback mode. I then installed 5.11.16 (which has the suspend bug) and rebooted into 5.10.32 after checking 5.11.16 - now 5.10.32 gave the same warning.
Currently I’m on 5.11.6 again after downgrading.
Ran into a problem of black screen at boot (after plymouth screen).
I’m using KDE and sddm seems to have a little bug now that it doesn’t open in the right tty, so some users might be able to switch tty (most like to the second with Ctrl+Alt+F2) to go into graphical interface.
For my part, I couldn’t switch ttys. It simply stayed a black screen.
Disabling sddm-plymouth and enabling sddm didn’t help, but when I downgraded sddm to one version back, it finally booted for me (as the new bug has been introduced in the now-upstream version).
I did this with chrooting into my installation and execute these commands inside my manjaro-chroot:
pacman -U sddm-0.19.0-4-x86_64.pkg.tar.zst
Then exiting chroot and rebooting into my installed system. Worked fine then. I currently have work to do, so I can’t further investigate this issue (whether it really solved it or it was just a lucky boot). I might investigate more these days.
I hope I could help another user with the same problem.
This was the site which gave me the idea to downgrade this package: bugs . archlinux.org /task/70528
KDE with kernel 5.10.32 LTS, systemd-boot and SDDM. Everything seems to work, but I noticed some strange things during startup after upgrade (rebooted multiple times, the issue exists):
After entering volume decryption password at boot (got encrypted main drive with custom partitioning + encrypted RAID1), boot manager starts displaying all the usual messages, including UUID of disk drives. And now one of UUID is displayed incorrectly - a question mark is added instead of one of characters. Before update:
After update: UUID=a1b92429-9eda-?260-8099-204a151d8e3f (notice ? instead of “normal” character).
both fstab and crypttab contain valid entries and RAID is automatically decrypted during boot just fine. It’s just that glitch with ? mark…
When boot ends and it’s time to start GUI (I have auto-login enabled in SDDM) instead of displaying KDE desktop immediately, system displays black window that has blinking _ symbol at the top. Just like command line. After a second or two, it disappears and desktop is displayed.
Not sure if this is issue with Manjaro or KDE/SDDM itself, but I have it for 2 months already, and it’s still not fixed. If there are multiple keyboard input languages installed and lock screen is displayed (either after screensaver or manual locking), keyboard icon is displayed at the bottom as usually, but language icon is missing; therefore, it’s impossible to know which language you use to enter password. The only way to find out - enter password and “reveal” it by clicking the corresponding icon in password field. Not a very big problem, but language selector worked for years and it’s now broken.
Upgraded today with “pacman -Suy”, without issues, and after reboot the system seems to be fine, BUT:
when I try to access (i.e. mount) a partition on an internal HDD (which is not used right-away, but is just a destination for occasional backups) Thunar refuses with error message
“Der Datenträger >xxx< konnte nicht eingehängt werden. No object for D-Bus interface.”
Mounting partionitions on this HDD manually with mount -t ext4 /dev/sdc1 /mnt it works, without errors.
Apparently no log entries relating to this issue (tried journalctl -f), no boot issues (tried dmesg)
Any ideas?? Thx.