[Testing Update] 2020-07-31 - Kernels, Plasma 5.19.4, Firefox 79, Deepin, UKUI, Systemd, Python

bootloader is updated on command grub-install ( binaries )
file config cfg is renew and update with 2 commands after

if not EFI , it should be

sudo grub-install /dev/sdX ( or /dev/nvmeX) --recheck --verbose
sudo grub-mkconfig -o /boot/grub/grub.cfg
sudo update-grub
1 Like

Isn't sudo grub-mkconfig -o /boot/grub/grub.cfg doing exactly the same as sudo update-grub in a Manjaro install?

1 Like

For those who are not sure if grub update is necessary or not: during the update my Ubuntu installation did a grub-install operation as a last step of grub package update process, right before the update-grub command, which was invoked in the end, as always.
So yes, it looks like grub-install is a necessary thing, as least Ubuntu maintainers think so.

All fine on: :manjaro: :penguin: XFCE + BTRFS + Linux57
Only Vim get's an downgrade to an older Version

And I do not have to trigger update-grub the reboot goes well without issues

Xfce on linux 5.7 standard Intel chips. All fine after reboot

Does this apply for those of us that had our kernel(s) broken? Please see my post 5 post above yours and the below linked thread. Thanks

Exakt meine Beobachtung = Exactly my observation. :innocent:

One should really investigate. Will test this asap.
grub.mo is the jumping point, but not really?!
After "grub-install" and "update-grub" procedure as adviced by stephane
-- I suddenly had German menu - what ? Miracle happens...
Oh - the complete installation got finished, after I edited Uefi-Bios, disabled "Bootscreen";
boot 1x - then the Grub-Resolution was like before update (after: other Resolution)
Then enabled "Bootscreen" again - and everything stayed o.K.

Everything went fine (Manjaro KDE).
Now I can see UEFI Firmware Settings in Grub but haven't tested it yet. I just got the periodical bug with doubled Grub entries but that's easy fixable.

Yes it is:

$ cat $(which update-grub)
#! /bin/sh
grub-mkconfig -o /boot/grub/grub.cfg

Yea noticed this on my roommates machine, thought it a great touch.

That is not a valid path, because /dev/sda1 is a file, not a directory. Also, the physical device is irrelevant here, because it will look for the directory /boot/efi, regardless of what underlying device is mounted to that directory.

In essence, you don't even have to specify any parameters, other than for disabling the Reed-Solomon error codes ─ --no-rs-codes ─ because they're not needed in UEFI installations with a GPT partition table, but the default install in Manjaro leaves them enabled.

Everything else will be autodetected, based upon the contents of /etc/default/grub. :wink:

1 Like

I just switched branches because the double-giga stable update created a lot of annoying bugs for me and other people, and so I thought waiting for nearly two months for such a pretty buggy update isn't worth it anymore.

This update installed fine. My grub menu is working perfectly fine after updating, but no change with the issues I got in KDE so far, among all other things. I like that there's now an entry for UEFI firmware settings in grub.

Switched branches from stable to testing for the first time. Did grub-install and update-grub after reboot, like some have mentioned here, and did a second reboot.
Not casting my vote just yet, but saw no errors in any of the steps I took.

1 Like

Thanks for the reply. I've reading about it and yes, you right. Unfortunately didn't work in my system, I did it 4 times without errors but on reboot "grub callout missing" or some like that. I did it with a live iso, from the system itself. Everytime it went wrong, I restored my system and try again. Now I'm with the last testing update again reading about that error and how I can fix it.

I tried grub-install again this morning, all went right like any other time, then update the system and reboot. This time the booting was normal, I can see grub, see the UEFi firmware option on it. The only different step that I did was update the system before reboot, I don't know it update make any different in the process or maybe was doing it with fresh eyes.
Thanks everyone!!!

1 Like

Udpated and rebooted using grub without any problem.

I received warning about vim downgrade:

warning: vim: local (8.2.1260-0) is newer than extra (8.2.0814-3)
warning: vim-runtime: local (8.2.1260-0) is newer than extra (8.2.0814-3)

Someone actually praising for systemd-related software.

This is rare. :laughing:

VAAPI is failing
[alberto@rocio-82c5 ~]$ vainfo
libva error: /usr/lib/dri/i965_drv_video.so init failed
vaInitialize failed with error code -1 (unknown libva error),exit

I think it's here since the last update, can somone confirm it?

I'm not sure if I understand what's your problem is but if you're talking about grub prompt which appears instead of loading of a kernel then you can workaround it with booting to live mode, chrooting and doing pacman -Syu to update again. Also I advice to add an UEFI boot entry for direct kernel boot in accordance with instructions described in my forum profile. This could be a good secondary way to boot in case of similar grub failure in future.
Oh, and the main thing: I encourage everyone NOT to reboot if you faced some obvious issues in the terminal output during the update. Often it is enough just to wait a bit, then do pacman -Syu again to receive the next batch of updated packages with a fix of the potential problem. Patience is golden, not silence :upside_down_face:


Forum kindly sponsored by