Skip to content

Comment on Case study: recovery of a corrupted 12 TB multi-device pool

Comments

Btrfs allows migration from ext4 with a rather good rollback strategy...

Post-migration, a complete disk image of the original ext4 disk will exist within the new filesystem, using no additional disk space due to the magic of copy-on-write.

Why isn't the repair process the same? Fix the filesystem to get everything online asap, and leave a complete disk image of the old damaged filesystem so other recovery processes can be tried if necessary.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.