HomePC relatedLinuxBTRFS Blues and Backup Basics Log in


BTRFS Blues and Backup Basics — 3 Comments

  1. Did you have a redundant setup? It is normal for power outages, hard drive phantom writes, or writing in the wrong spot, or perhaps cosmic rays, to kill a drive (rarely, but normal nonetheless). But ZFS was designed to correct these using redundancy and checksums to find which of the 2 copies is correct, and repair the damaged one, and BTRFS is supposed to be the same. Is it, or did you just not use redundancy? (raid1, raid10, and not yet in btrfs: raid5, raid6)

  2. I hope you at least reported the bug because the devs need to know what they need to fix. Also Chris Mason is still working on btrfsck and it’ll be wise to report corruptions to him so he can add code to fix those corruptions.

    It might not be immediately useful for you if you report the corruption, but it’d be nice if btrfs users in the future have a working btrfsck

    • I didn’t consider it a bug and unfortunately I don’t have the error code anymore.

      All file systems get corrupted from time to time, but the file checker program usually fixes it without data loss. In this case btrfsck was not yet able to fix it. It was totally my fault for using it in the first place for my /home partition. I was burning a DVD with K3b and it hung and the system became unresponsive. I hit the restart button. That in itself could have caused the partition to become corruption.