After the update you can install the 4.18 kernel without problems. At least in my case.
I keep getting this error:
error: failed to commit transaction (conflicting files) lib32-ncurses: /usr/lib32/libtinfo.so exists in filesystem (owned by lib32-libtinfo) Errors occurred, no packages were upgraded.
No idea ho to fix this.
I figured that out but thanks. Had to reinstall Steam afterwards however.
I cannot boot after last update. On 4.9.124 kernel it just hangs on “Loading initial ramdisk”. I have old 3.16 kernel just in case, but when I try to boot, it says that it failed to start simple desktop display manager. Here’s more details in pictures because I’m writing this from other PC.
lib32-libtinfo is no longer in the repos (official or AUR). I would remove this package.
Fixed via 4.9.124-2. I assume you use an Intel CPU. Happy L1TF to you
I end up removing lib32-ncurses and then doing the updates, rebooted, and having to reinstall ncurses for Steam.
Actually I am on 4.9.124-2, and no, I’m using AMD APU A8-6600K.
I wonder why there is no file conflict message when reinstalling lib32-ncurses through Steam?
What is the output of this command?
pacman -Qo /usr/lib32/libtinfo.so
You should still clean orphaned packages like lib32-libtinfo. It might cause more problems in the future.
/usr/lib32/libtinfo.so is owned by lib32-ncurses 6.1-2
Yesterday when i applied this latest Stable Update to all my Tower’s VMs [KDE, Cinnamon, Deepin, Xfce, Enlightenment], to my Lappy’s real KDE, & to my Tower’s real KDE, everything seemed fine. Sadly this morning however i discovered an irritating bug/regression afflicting only my Tower’s [naturally, sigh] real KDE… my Lappy’s real KDE does not suffer this.
Hours wasted this morning fighting a return of the dreaded no-network-after-Suspend-Resume misbehaviour [same symptoms per my older No Ethernet Network Connection after Resume; very recent problem]. Something in the latest Stable Update 2018-08-25 did this. The bug occurs with latest kernel packages
linux417-4.17.18-1-x86_64.pkg.tar.xz … but fortunately
Kernel: x86_64 Linux 4.14.66-1-MANJARO still behaves properly.
Once i isolated the new problem to only two of my three installed kernels, I Downgraded 418 to package
Kernel: x86_64 Linux 4.18.3-1-MANJARO. Happily after this, proper Suspend-Resume network availability returned [ie, just like had been occurring prior to doing yesterday’s update].
Given that my 418 is now good again, & i already know that my 414 stayed good, i have not bothered also downgrading my 417.
Are these latest 418 & 417 regressions an external/upstream problem, or internal /within the control of our Manjaro Devs?
Hi. I have one issue. No key for phonon-qt4. Solution is: gpg --keyserver pgp.mit.edu --recv-key B92A5F04EC949121. Thanks for update!
Please check if 4.9.124-3 fixes the issue for you:
sudo pacman -Syyu
Latest issues are due to the L1TF problem. Always when there is an issue with security most patches got not reviewed properly as they get developed without the public to avoid too early knowledge to be spread. Sometimes this happens. Either we wait until it is stabilized and therefore have the system vulnerable or we go with the flow.
I encounter a problem with the settings of my touchpad with my Lenovo T470.
The settings are set to some default frequently e.g. inverse scrolling and accelleration. What exactly is causing these problems?
Update went fine, but the qt4 removal hit hard, had to removed quite a few pacakges:
sqliteman pybitmessage m4baker-git syncplay plasma5-themes-menda kfritz
I’m hoping that at least some of these, or their dependencies, pop up on the AUR.
Thank for help. This solved a problem, but first I had to resolve networking issue I faced after this update, and I managed to fix it using this forum post.
After the upgrade I can’t start riot-desktop.
FATAL: error reading `/proc/sys/crypto/fips_enabled' in libgcrypt: Invalid argument /usr/bin/riot-desktop: Zeile 3: 6112 Abgebrochen (Speicherabzug geschrieben) electron /usr/lib/riot/ "$@"
linux44 is installed.
Any ideas? Couldn’t find a corresponding bug-report anywhere.