On Mon, Jun 29, 2020 at 11:33:40AM +0200, Florian Weimer wrote:
Just to be clear here, the choice of XFS here is purely based on
performance, not on the reliability of the file systems, right?
(So it's not “all the really important data is stored in XFS”.)
Be careful about overloading quite a few definitions into the single
You seem to be referring to btrfs features like file checksumming that
can detect silent corruption, and automagically fix things if you've
enabled the equally automagic RAID1-like features. (Which, for the
record, I think are really frickin' awesome!)
But what good is btrfs' attestation of file integrity when it craps
itself to the point where it doesn't even know those files even _exist_
anymore? How can we brag about robustness in the face of cosmic rays or
recovery from the power cord getting yanked when it couldn't reliably
_remount_ a lightly-used, cleanly unmounted filesystem?
By that "reliability" metric, for me XFS has been infinitely better than
btrfs; sure XFS can't automagically tell me if an individual file got
corrupted (much less fix it) but it's also never eaten entire
filesystems across clean unmount/mount cycles. Whereas btrfs has done
I realize this is several-years-out-of-date anectdata, but it's the sort
of thing that has given btrfs (quite deservedly) a very bad reputation.
(BTW, I didn't try using XFS until after my second bad btrfs experience.
It hasn't so much as hiccupped me since, proving to be the robust
filesytem I've ever used...)
I concede that my experience is outdated, and am willing to take the
btrfs authors at their word that the bugs that led to my filesystems
eating themselves have been fixed, but.. sure, the btrfs proponents say
it's "ready for production" now, but they also said that back then, too.
So. Instead of making btrfs the default for F33, perhaps a better
approach is is to plan to make it the default for F34, and use the F33
cycle to encourage folks (eg release notes, installer prompting?) to try
The point here is not F32 vs F33 or whatever, but that of _time_ -- I
don't think there's enough time between now and the F33 go/no point for
folks like me to set up and sufficiently burn-in F32 btrfs systems to
gain confidence that btrfs is indeed ready. (In any case, the
traditional beta period is _way_ too short for something like this!)
Solomon Peachy pizza at shaftnet dot org (email&xmpp)
@pizza:shaftnet dot org (matrix)
High Springs, FL speachy (freenode)