Skip to content

Update README with testing status for boot layouts - #1

Merged
doug445 merged 1 commit into
mainfrom
doug445-patch-2
Sep 14, 2026
Merged

doug445 merged 1 commit into
mainfrom
doug445-patch-2

Conversation

@doug445

@doug445 doug445 commented Sep 14, 2026

Copy link
Copy Markdown
Owner

Clarified testing status for various boot layouts and filesystems.

Clarified testing status for various boot layouts and filesystems.
@doug445
doug445 merged commit c84f2e8 into main Sep 14, 2026
10 checks passed
doug445 added a commit that referenced this pull request Sep 15, 2026
…ext4 on LUKS2, systemd-boot Type #1 entries

Restored from the installed system onto a blank USB SSD with testbed/testbed.sh
and booted into a fully working system (Wi-Fi, DNS, KDE Plasma desktop with its
panel, launcher and terminals; system state running, no failed units):
testbed.sh collect → VERDICT: PASS, 99.96% of 583,535 files restored with the
archived size (every missing file Firefox site storage the booted session
cleared), the source disk and SD card identical before and after. Machine: 2014
ASUS X750JN, the same 3-SSD triple boot as the Mint and Fedora passes; the
EndeavourOS SSD restored. borg-restore.sh and restore-rebuild-boot.sh needed no
change.

Test bed (both faults found by the first test boot, fixed, re-verified by a
second restore):
- the functional test archive kept ~/.config but not ~/.local/share: Plasma
  logged in to no panel, no launcher and no terminal (panel theme, plasmoids,
  color schemes, icons, konsole profiles missing). TB_HOME_KEEP now keeps
  desktop state for Plasma, GNOME, Cinnamon and XFCE, ~/.local/state,
  ~/.themes, ~/.icons, ~/.fonts, and what ~/.local/bin and the shell load
  (Claude Code, pipx, ble.sh).
- the restored crypttab unlocked the host's SD card on the test boot (a nofail
  entry, its keyfile restored with /etc/luks-keys) and the boot report FAILed
  "source-machine containers open". finish sets noauto on the test drive's
  crypttab/fstab entries for disks other than the test drive, keeps the
  originals in /root/restore-test, and never touches the boot chain's own
  entries (root/boot containers, /, /boot, the ESP, /usr, /var, /home) — one
  of those still naming another disk is a restore failure and is warned.
- the restore summary matched borg's file list (MEMORY_FAILURE,
  CURLOPT_FAILONERROR.3) as failures; only the verification's FAIL: lines count.

README: EndeavourOS row expanded with the restore evidence and the findings;
Arch / Manjaro / EndeavourOS and systemd-boot Type #1 bare-metal rows ✅; the
test bed section documents the desktop state kept and the other-disk isolation.
CONTRIBUTING: EndeavourOS / Type #1 moved out of the under-test row.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
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.

1 participant