index : archinstall32 | |
Archlinux32 installer | gitolite user |
summaryrefslogtreecommitdiff |
Age | Commit message (Collapse) | Author | |
---|---|---|---|
2021-09-18 | Preparing to split up ./lib/disk.py | Anton Hvornum | |
And added some sane(?) defaults, which I'll massage into code some how. | |||
2021-09-18 | Adding in options for BTRFS subvolumes | Anton Hvornum | |
2021-09-18 | Removing hardcoded btrfs for suggested partition layouts. | Anton Hvornum | |
2021-09-17 | Removing /tmp from fstab | Anton Hvornum | |
Since systemd takes care of mounting /tmp as a tmpt, there's no point in having a manual entry in fstab. | |||
2021-09-15 | Making sure the drive paths are in the JSON structure, and not the class ↵ | Anton Hvornum | |
object, as it won't work seamlessly to access for instance storage['disk_layouts'][ClassInstance()] if it's not the identical mem copy of the object we're accessing, so strings are better for storage/comparisons. | |||
2021-09-14 | Added a 'hidden' variable called --disk-sleep for delaying disk partition up ↵ | Anton Hvornum | |
time before continuing after a format. This is an ugly hack to get around some disk issues, for now. | |||
2021-09-14 | Moved --noconfirm at the end of the pacstrap command to get piped properly ↵ | Anton Hvornum | |
to pacman inside pacstrap. | |||
2021-09-13 | Merging latest changes | Anton Hvornum | |
2021-09-13 | Added --noconfirm to pacstrap and increased logging verbosity when packages ↵ | Anton Hvornum | |
fail, and adding a exception so that the installation doesn't continue silently. This causes too much confusion (even tho I liked the idea that people could recover the missing parts without having to re-run the installer for subsequent steps. | |||
2021-09-12 | Removed legacy need to add_bootloader | Anton Hvornum | |
`add_bootloader` no longer needs to have a harddrive given as a argument. It will (and should) auto-detect what's mounted in the `self.target` (aka mountpoint) of the installation. | |||
2021-09-06 | Merge pull request #623 from conqp/clean_hardware_imports | Anton Hvornum | |
Remove unused import of json | |||
2021-09-06 | Merge pull request #626 from conqp/refactor_meminfo | Anton Hvornum | |
Refactor meminfo() to allow direct key access | |||
2021-09-06 | Merge pull request #625 from conqp/fix_tab_indentation | Anton Hvornum | |
Fix indentation with tabs instead of spaces | |||
2021-09-06 | Fix indentation with tabs instead of spaces | Richard Neumann | |
2021-09-06 | Refactor meminfo() to allow direct key access | Richard Neumann | |
2021-09-06 | Remove unused import of json | Richard Neumann | |
2021-09-06 | Merged in master | Anton Hvornum | |
2021-09-06 | Merged in master. | Anton Hvornum | |
2021-09-06 | Refactor cpu_model() to use cpuinfo() | Richard Neumann | |
2021-09-06 | Refactor cpu_vendor() to use cpuinfo() | Richard Neumann | |
2021-09-06 | Merge branch 'master' into native_cpu_detection | Anton Hvornum | |
2021-09-06 | Merge pull request #610 from luckspt/patch-2 | Anton Hvornum | |
Typo | |||
2021-09-06 | Merge pull request #605 from CYPT71/patch-1 | Anton Hvornum | |
Update hardware.py | |||
2021-09-06 | Added exception handling to check_output | Anton Hvornum | |
I tweaked the optimized return of check_output. Worth mentioning that `check_output()` will raise an exception `subprocess.CalledProcessError: Command 'lscpu | grep AMD' returned non-zero exit status 1.`. | |||
2021-09-06 | No need for try/catch any longer. | Anton Hvornum | |
The old behavior of SysCommand was that exit codes raised an exception, which needed to be handled by each individual caller. We now utilize `.exit_code` instead to manually detect faulty commands and raise exceptions where needed. | |||
2021-09-06 | Merge branch 'master' into master | Anton Hvornum | |
2021-09-06 | Merge pull request #600 from monosans/comprehension | Anton Hvornum | |
Replace for loops with comprehensions | |||
2021-09-06 | Corrected some minor modifications mentioned in #595 as well as merged in ↵ | Anton Hvornum | |
latest changes from master. For future PR's do create a patch-branch for upstream patches to avoid changes to your local repo. | |||
2021-09-04 | Merge branch 'master' of github.com:archlinux/archinstall into ↵ | Anton Hvornum | |
torxed-rework-partitioning | |||
2021-09-04 | Fixed edge case where size of disks could be exactly 40GB and a /home would ↵ | Anton Hvornum | |
never be created on single devices. | |||
2021-09-04 | Added a PARTUUID -> parted index, this in order to properly set the boot ↵ | Anton Hvornum | |
flag on the correct partition. Perhaps there's a smarter way. I suspect parted can operate on a given partition ID, but haven't found the docs for it yet. | |||
2021-09-04 | Simplified lambda and made it filter out non relevant disks for the ↵ | Anton Hvornum | |
selection process. | |||
2021-09-04 | Fixed filtering if largest disk selection | Anton Hvornum | |
2021-08-20 | Remove excess newline | Richard Neumann | |
2021-08-20 | Rewrite CPU vendor detection functions | Richard Neumann | |
Use cpuinfo() function rather than a subprocess. | |||
2021-08-20 | Add cpuinfo() | Richard Neumann | |
2021-08-18 | Reworking logic that selects / and /home for multi-disk configurations. Also ↵ | Anton Hvornum | |
added some more debugging | |||
2021-08-17 | Fixed a prompt error in one of the parted calls. Also started on a more ↵ | Anton Hvornum | |
reliable size-conversion that isn't limited to Gigabytes in free_space(). | |||
2021-07-28 | Typo | Lucas Pinto | |
Fix a tiny typo | |||
2021-07-22 | Update hardware.py | CYPT71 | |
Rewrite some function if condition is True then return true else return false, transform in return condition directly Also I don't understand why we need a try/except at line 151 and why we not write return condition ?? | |||
2021-07-19 | Mention GNOME and KDE's soft dependency on NetworkManager in the install | j-james | |
2021-07-18 | fix for #602, running mklabel before formatting | Yash Tripathi | |
2021-07-18 | created load_config() to load configuration | Yash Tripathi | |
2021-07-16 | Replace for loops with comprehensions | monosans | |
2021-07-07 | Fix encryption + grub both in UEFI and BIOS systems. | Hamad Al Marri | |
We need to have two partitions in BIOS one for boot (grub) and the other for root (/). The format of the boot partition is ext2 (so it is added). If disk is chosen to be encrypted, `then /etc/default/grub` is edited as the followings: https://wiki.archlinux.org/title/Dm-crypt/Encrypting_an_entire_system#Configuring_GRUB_2 Issue: https://github.com/archlinux/archinstall/issues/586 This work is done while working on CachyOS which is Arch based and it is using customized archinstaller. To test these changes you can try CachyOS installer which supports encrypted disk with grub (https://wiki.cachyos.org/). Hamad | |||
2021-07-04 | Removed debugging | Anton Hvornum | |
2021-07-04 | Added a filesystem check when marking for formatting, this should ensure ↵ | Anton Hvornum | |
that encrypted volumes get a proper filesystem without having to go through an extra step of selecting filesystem. | |||
2021-07-03 | Added options to mark/unmark partitions for formatting (useful when re-using ↵ | Anton Hvornum | |
partitions, and fine tune which data to save and which to wipe). Setting a desired filesystem for a partition (both new ones and the ones being re-used). | |||
2021-07-03 | Saving partitioning layout in a layout file (JSON format) | Anton Hvornum | |
2021-07-03 | Missing .format() on 'use entire disk' step | Anton Hvornum | |