Skip to content

Update installer.py (#1)#793

Merged
Torxed merged 1 commit intoarchlinux:masterfrom
OneLongneck:master
Jan 26, 2022
Merged

Update installer.py (#1)#793
Torxed merged 1 commit intoarchlinux:masterfrom
OneLongneck:master

Conversation

@OneLongneck
Copy link
Copy Markdown
Contributor

Added "--removable" after "--bootloader-id=GRUB" on Line 669, because it would throw:
"Could not prepare Boot variable:
Interrupted system call
grub-install: error efibootmgr failed to register the boot entry: Input/output error"
without it on my laptop (with every distro I tried installing)

🚨 PR Guidelines:

New features (v2.2.0)

All future work towards v2.2.0 is done against master now.

Any patch work to existing versions will have to create a new branch against the tagged versions.

Describe your PR

If the changes has been discussed in an Issue, please tag it so that we can backtrace from the issue later on.

If the PR is larger than ~20 lines, please describe it here unless described in an issue.

Testing

Any new feature or stability improvement should be tested if possible. Please follow the test instructions at the bottom of the README or use the ISO built on each PR.

Added "--removable" after "--bootloader-id=GRUB" on Line 669, because it would throw an input/output error without it on my laptop
@OneLongneck OneLongneck requested a review from Torxed as a code owner December 10, 2021 16:17
@Torxed
Copy link
Copy Markdown
Member

Torxed commented Jan 2, 2022

There's a warning on the arch wiki:

If dual booting with Windows, be aware Windows usually places an EFI executable there, but its only purpose is to recreate the UEFI boot entry for Windows.

We've had similar issues on systemd-boot (#350, systemd/systemd#13603) and in most cases it's a firmware issue. The way we solved it was similar as well, we added --no-variables. But that in of itself is just a bandaid and causes other issues. So I'm wondering if perhaps we can find a way to fix the reason why you need --removable instead?

For instance, have you updated your bios firmware since purchase? Can we detect the need for this and only do it when necessary?

@Torxed
Copy link
Copy Markdown
Member

Torxed commented Jan 26, 2022

Me saying this is an understatement, bu you are awesome!
Spent 18 hours trying to figure out grub+uefi+[btrfs/subvolumes/luks] issues.

All I had to do, was add this. So I'm merging this, and pulling it into v2.3.1-dev as well! Thank you!

@Torxed Torxed merged commit f0886f3 into archlinux:master Jan 26, 2022
Torxed pushed a commit that referenced this pull request Jan 26, 2022
Added "--removable" after "--bootloader-id=GRUB" on Line 669, because it would throw an input/output error without it on my laptop
Torxed added a commit that referenced this pull request Jan 26, 2022
* Added missing import

* Brought back the mounting of subvolumes, or at least the beginning of it. Improved some logging and decreased the usage of try/except where unessecary (we can catch it higher up without catching and raising the same thing.

* Trying to flatten the mount information to allow for sorted subvolume mounting. Still got issues.

* Re-worked mountpoints logic to be a flat dictionary structure, where all subvolumes are mountpoint targets, but with relevant information. Had to work that into Partition().mount to support re-mounting a partition multiple times. That will have to be cleaned up.

* Changed the default timeout to 15 as per the ISO

* Corrected a path when creating loopdev name.

* Added some debugging/warning

* Fixed a path lookup issue

* Added some debugging

* Added more debugging

* Removed excessive luks add_key() and crypttab() entries being generated by subvolumes.

* Added more debugging

* Avoiding excessive keyfile usage for partitions. Also added some debugging for BlockDevice.

* Added more debugging

* Added debugging and a hack to find /dev/sda from /dev/sda2

* Fixed ugly hack to use /dev/sdX instead of just sdX

* Inconsistency in lsblk -no pkname made sorting a bit tricky, resorting to shortest string wins.

* Tweak lookup

* Tweak lookup

* Added more debugging

* Had to filter lsblk output because some entries are empty rows

* Syntax issue

* Added a comment clarifying the many steps of partition->blockdevice logic.

* Added debug information

* Added debug information

* Added debug information

* Fixed issue #908

* Added subvolume support for grub. Also added check to see if grub is installed before installing it (to avoid multiple calls to pacstrap even tho it doesn't hurt)

* Modified 'sed' to use % as divider, since the replacement string can contain '/'

* Added debugging

* Testing GRUB with /boot/efi instead (might revert)

* Reverted /boot/efi for GRUB. Did no difference.

* Flake8 fix

* Testing --removable from #793 if it solves the UEFI issue

* Removing --remoable and pulling it from #793 for tracability

* Update installer.py (#1) (#793)

Added "--removable" after "--bootloader-id=GRUB" on Line 669, because it would throw an input/output error without it on my laptop

* Remove debug information

Co-authored-by: OneLongneck <95297344+OneLongneck@users.noreply.github.com>
@Torxed Torxed mentioned this pull request Mar 7, 2024
1 task
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants