Comment by gavinsyancey
7 months ago
My btrfs filesystem has been slowly eating my data for a while; large files will find their first 128k replaced with all nulls. Rewriting it will sometimes fix it temporarily, but it'll revert back to all nulls after some time. That said, this might be my fault for using raid6 for data and trying to replace a failing disk a while ago.
raid 5/6 is completely broken and there's not much interest in fixing it — nobody who's willing to pay for its development (which includes Facebook, SUSE, Oracle, and WD) uses raid 5/6; you shouldn't have been running it in the first place. I understand it's basically blaming the victim, but doing at least some research on the filesystem before starting to use it is a good idea in any case.
https://btrfs.readthedocs.io/en/latest/Status.html
edit: just checked, it says the same thing in man pages — not for production use, testing/development only.
iirc, btrfs has fixed the issues with raid 5/6 but it requires a breaking change to the on disk format which means you have to create an entirely new partition and copy the data over (you cannot update an existing partition to it). This new on disk format also needs its own testing.
Your point raid 5/6 not being tested heavily by actual users is entirely on point, those enterprise heavy users are only running RAID 10 like configurations.
If you want RAID 5/6, just use ZFS as they have solved all of these issues. I don't know if its due to sheer luck or maybe the fact is that Sun at its time was actually running RAID 5/6 in production (hard drives were not as cheap back then as they are now)?