I would not trust in the quality of any ZFS features in any implementation that were not developed by Sun or at least thought out by Sun before it was acquired.

Care to elaborate?

ZFS on Linux is very buggy.

And when something breaks (maybe because one of those numerous bugs) there are no proper tools for repair. Your ZRAID and ZLOG probably won't matter. The to-go tool for removing corrupt superblocks is a 10 year old hacky python script, it's ridiculous.

Not been my experience at all.

I have seen folks get in over their heads though. Like -- "OMG I've somehow broken boot for ZFS on root and I don't know how to fix!!" or "I'm using a custom kernel and OMG things don't work!!"

Canonical has even been complicit. When you download an HWE kernel, you don't also download a newer/matching versions of the tools. This should pretty clearly be considered a bug[0], and fixed but Canonical perhaps doesn't have the bandwidth?

I just can't lay any of these problems at the feet of ZFS though. Right now I trust ZFS a hell of a lot more than I trust Linux to Do The Right Thing. When ZFS seems broken (though I have little experience with native encryption so this may not apply to it), I am far more likely to think the issue is Linux, my distribution, or me, than any problem with ZFS.

I'm a pretty serious ZFS lover though.[1]

[0]: https://bugs.launchpad.net/ubuntu/+source/zfs-linux/+bug/193... [1]: https://github.com/kimono-koans/httm