Case Study: Recovery of a Corrupted 12 TB Btrfs Multi-Device Pool

After a hard power cycle, a 12 TB multi-device Btrfs pool was severely corrupted, leaving the extent tree and free space tree in a state beyond native repair capabilities. The btrfs check --repair command entered an infinite loop, making no progress and rotating backup_roots slots past any pre-crash rollback point. Recovery was achieved using 14 custom C tools built against the internal btrfs-progs API, resulting in minimal data loss of 7.2 MB out of 4.59 TB. The case study outlines the environment, timeline, root cause, and proposes nine upstream changes that could prevent similar issues. These include improvements in progress detection, symmetric handling of delayed references, prechecks for sibling safety, and new rescue subcommands. The custom tools and a patch are shared for reference, not as direct upstream proposals, emphasizing the need for design discussions with subsystem experts.