Hector Martin

I disagree. ZFS has been rock solid for me and withstood multiple disk failures without data loss. Not even ext4 or btrfs can keep up.

I'm not talking about disk failures, I'm talking about something going wrong with ZFS itself. Of course btrfs is worse. ext4 and XFS are better, they can actually be understood and debugged by mere mortals.

To wit: I had a SNAFU with RAID6 during some disk exchanges that I ended up fixing, but I wound up with a FS that was a mix of disk state from a span of about 1 hour. XFS happily fixed everything with fsck. ZFS would've probably died badly with such an unconventional issue.

The lack of fsck being the first problem, of course. The rule with ZFS is that when something goes wrong, you restore from backup. I *do* have backups, but it don't trust a FS with that mentality.

I've been running ZFS in critical production for over 6 years and I haven't seen a single file system failure even when disks decided to die.

On top of that, ZFS don't need fsck.

Disks dying is a normal failure mode. A hardware glitch corrupting RAID6 stripes during disk replacement necessitating a roll back to the old disk and thus a patchwork of old and new disk state isn't. ZFS works great until it doesn't and then you can't fix it.

"ZFS doesn't need fsck" is what they want you to think. Then something happens they don't plan for and you're screwed. Go look around for ZFS data loss stories, they exist.