Did you mount root partition?
I have mounted using this command. Thanks for the reply
/dev/nvme0n1p1 is your efi system partition. You need to mount your root partition (where your sustem resides) to /mnt and your efi system partition to /mnt/boot/efi. Then you can chroot.
Can you please list the specific commands I am a complete noob. Here is my file system:
see this link
https://discovery.endeavouros.com/system-rescue/arch-chroot-for-efi-uefi-systems/2021/03/
if you are in “partial chroot” , exit and redo commands
You will need to read and follow instructions here.
I am hoping assuming nvme0n1p2 is your root partition.
sudo umount /mnt
sudo mount /dev/nvme0n1p2 /mnt
sudo mount /dev/nvme0n1p1 /mnt/boot/efi
sudo arch-chroot /mnt
sudo grub-install --target=x86_64-efi --efi-directory=/boot/efi
sudo grub-mkconfig -o /boot/grub/grub.cfg
Obviously, UEFI needs to be enabled, to follow along that mentioned path. In your case (with UEFI disabled) a simple, non-UEFI way of reinstalling grub should have done the job? I think your trouble may have resulted from complexity, due to BTRFS snapshots you previously had in Grub?
I would have though that was sufficient.
Yes, I think the BTRFS snapshots may have been the problem.
I am wondering if we can still use BTRFS Assistant and snapshots using rEFInd. This grub thing gave me a scare, because all my 1500 snapshots did not help me in this situation.
At least I have excellent backups. Of course I lost some SSH keys, but they can be rebuilt without losing access to the remote servers.
I really don’t think that is related.
Two things:
- They would have helped you. If you simply booted of the ISO and restored a recent snapshot that would have got you up and going. People are getting too obsessed with booting off a snapshot being the only way to recover them.
- 1500 snapshots is probably too many on Btrfs. That is likely making your btree a mess.
You certainly can because you don’t need to “boot off a snapshot” to recover or use snapshots.
However, if you want to be able to boot off a snapshot, I believe it is possible using refind-btrfs
. I have not personally tried that though.
According to this man-page, it shouldn’t be a problem regarding rEFInd, as it should work with btrfs out-of-the-box.
So I boot off the ISO, but the grub menu does not appear, so how do I restore from there? Sorry if it’s a dumb question.
Can you post a suggested snapshot configuration page so I can set to them?
I will wait until I read of some other members’ experiences with it.
Thanks for all you do Dalto.
Since you mentioned you were using btrfs-assistant, it is easy. Install btrfs-assistant
, mount any part of your btrfs partition, restore in btrfs-assistant.
That is actually the normal way of doing it.
Err…it depends what data you are trying to protect and what the partition in question is. Personally, I don’t set them all the same. Although, to get to 1500 snapshots you either have a lot of subvolumes or you didn’t have the cleanup timer enabled.
Its out of the box BTRFS installed by Calamares. All default.
But that would get you to 90 snapshots max after the cleanup. How did you get to 1500?
EDIT: For snapshots of /
I probably wouldn’t save any yearly snapshots.
I normally choose the snapshot number from the grub menu and in btrfs assistant, restore the same number snapshot and reboot. Isn’t that the normal way?
So Instead of choosing the snapshop from the grub menu (which in that previous case, did not display) you are saying I can just install btrfs assistant on the new ISO and it willdetect all the snapshots, which I choose and hot restore?
I am not sure there is a “normal way”. Some people do it that way but that certainly isn’t how I do it.
Yes.
I know, the maintenace was cleaning them up but it was continuing to count. I only would have had 90 as you said, but the previous 1410 were cleaned up, but the snapshots were still sequential.
Oh, so you didn’t have 1500 snapshots. You just meant that the numbering system had gone to 1500. That is fine.
Awesome! I will look into refind-btrfs soon.
Do you think we can continue to rely on grub after this incident?