Sorry! I feel confused and lost.
Both statements appear to be opposite of each other for me!
I trust what you say @sradjoker and I trust also what @dalto says.
This is then a technical thing beyond my humble understanding!
So. it is OK any way.
I will stick with my current systemd-boot to be on the safe side.
My statement out of context. That was in response to someone basically saying that everything would work because the grub from XeroLinux was already installed. My point was that EOS will install a separate copy of grub. I wasn’t trying to say that EOS uses a custom version of grub.
I was just hoping it can be done smoothly.
I just thought I can go back to a working Grub and just keep it without breaking again with an update. (I am not really interested in doing so and so after each update of Grub. This is not the way Linux or a rolling release should be)
But appears to me I better stick to where I am now with systemd-boot. (though not enjoying booting to earlier snapshots in cas something happens).
I went back to grub after it tried systemd-boot but i like grub and i also like rEFInd because with rEFInd i can boot from grub or the built in kernel image vmlinuz-linuz. There are no guarantees that something won’t break whether it’s grub or something else. This is a rolling release and software changes. New things get implemented and sometimes they don’t work as intended in all cases. Grub has been around since 1995. It’s a solid as anything out there.
If no arguments are provided ‘yay -Syu’ will be performed.
man yay
This command would only install (or reinstall) the grub “package” and not the bootloader itself.
In order to installing the boot loader you would need to use the grub-install command.
This is only needed whenever there is an update to the package.
No no need to (re-)install the “package” grub on every occasion when you update your system.
Also, whenever there is an update to the grub package, you would see a message like the one below:
To use the new features provided in this GRUB update, it is recommended to install it to the MBR or UEFI. Due to potential configuration incompatibilities, it is advised to run both, installation and generation
of configuration:
$ grub-install ...
$ grub-mkconfig -o /boot/grub/grub.cfg
Thanks a lot @pebcak
So, it should be something like
yay
grub-install
grub-mkconfig -o /boot/grub/grub.cfg
and that’s all to be sure I won’t by mistake skip grub-install and grub-mkconfig,this way I will be updating “normally” as before the Grub issue.
I would just run my script of the 3 commands above (whether there is a grub update or not)?
I know this for sure.
I was just thinking of finding a way to do it as we used to update before with no need to do the 2 extra commands.
Doing them in case Grub did not update will cause any problems?
(I did it once, Grub updated and I did not do them properly - tired, sleepy, distracted…)
Yes, peace of mind! I don’t want to take any risk. This way I will be just updating as we did before this Grub issue. (though doing extra unnecessary steps that won’t hurt) Am I right?
So I can just make this script, I will create a command pointing to this script assume I call it “sysupdate” that calls the script sysupdate.sh which is
yay
grub-install
grub-mkconfig -o /boot/grub/grub.cfg
Should I put sudo before “grub-install” and “grub-mkconfig” in the script? (as I learned here I should not “sudo yay”)
P.S. I do not really need to reboot after running my script whether there were update(s) or not? What if several Grub updates without reboot?
(Sorry for too many questions)
Sorry. to be sure I get it right, I should put “sudo” before the commands.
Well, another and hopefully last question.
I remember we played with kernel, efi… etc while converting to systemd-boot as per @dalto guidance.
Should I start with
yay -Rc systemd-boot
or something like that?
I think we need to do something to “reverse” what was done?
Any link or just a few commands to reverse.