LZ4 For Btrfs Arrives While Its FSCK Remains M.I.A.

[phoronix]

Phoronix: LZ4 For Btrfs Arrives While Its FSCK Remains M.I.A.

The proper fsck utility for the Btrfs file-system remains M.I.A. while a contribution from an independent developer introduces LZ4 compression support to this next-generation Linux file-system...

http://www.phoronix.com/vr.php?view=MTA1OTQ

[stikonas]

And what is that: https://git.kernel.org/?p=linux/kern...gerdonteveruse

[mgmartin]

And what is that: https://git.kernel.org/?p=linux/kern...gerdonteveruse

For me, a branch that says "danger..dont ever use"...and a branch that doesn't compile on the latest stable kernel ( 3.2.6 ) .

[Delgarde]

And what is that: https://git.kernel.org/?p=linux/kern...gerdonteveruse

Going by the URL, it's dangerous and shouldn't ever be used.

Hopefully that's a temporary thing.

[jwilliams]

And what is that: https://git.kernel.org/?p=linux/kern...gerdonteveruse

Thanks for the link!

That is interesting. Unless I'm missing something, those commits do not add full repair capability, but they do add a repair option that can fix extent allocation errors:

"This also includes a new --repair btrfsck option. For now it can only fix errors in the extent allocation tree."

[bulletxt]

I won't be using BTRFS on any of my servers before mid 2013. I really can't trust it and most importantly I don't trust this "rush for the fsck utility". We don't need rush, we need stable things on servers, there is nothing to play when you manage data.

[LLStarks]

Fsck Btrfs. Seriously.

I can't believe it's still not ready.

[bastiaan]

Compression is all nice and dandy, but I don't see Btrfs outperforming ext4 on most benchmarks even if they shave off a couple more percentage points from Btrfs+snappy.

(That is not to deny there may be other features Btrfs provides that ext4 doesn't.)

[jwilliams]

I don't trust this "rush for the fsck utility". We don't need rush, we need stable things on servers, there is nothing to play when you manage data.

It is hardly a rush. IIRC, btrfsck (with full repair capability) has been "coming soon" for more than a year now.

[F.Ultra]

Compression is all nice and dandy, but I don't see Btrfs outperforming ext4 on most benchmarks even if they shave off a couple more percentage points from Btrfs+snappy.

(That is not to deny there may be other features Btrfs provides that ext4 doesn't.)

BTRFS has never been about speed, it's more about durability (data integrity), snapshotting and the capability to handle very large data sets. So IMHO it's not that much of a client FS and more of a stable server FS.