Red Hat Appears To Be Abandoning Their Btrfs Hopes

[pal666]

That is blatantly not true. Sure, many open-source drivers make no promises (because who can you point fingers at?) but closed source drivers aren't betas or "use at your own risk" unless specified otherwise.

they all specify no warranty. you are living in fantasy world http://www.nvidia.com/content/Driver...ce.php?lang=us

Red Hat focuses on enterprises, not gamers and home users, so they're more likely to use closed drivers.

all redhat products are opensource and closed drivers disclaim all liabilities. redhat will fix bugs, but redhat will not reimburse you for losses due to bugs

Btrfs is largely backed by Oracle

its founder worked at oracle, but he changed few employers since that. wikipedia says

Facebook, Fujitsu, Fusion-IO, Intel, Linux Foundation, Netgear, Oracle Corporation, Red Hat, STRATO AG, and SUSE[1]

[pal666]

Maybe the problem is, that till today, Btrfs is still not completely stable (RAID 5/6)

no, it can't possibly be a point, because without btrfs you don't have raid 5/6 in your fs.

and it is not a general file system!

sure it is

It is not suited for databases or VMs, because the performance sucks on this scenario, because of COW.

hello? cow can be disabled per file since v3.7 (Dec 2012)

But if Red Hat really develops an own next generation file system, wouldn't it have similar problems?

redhat realy develops xfs. it is their default filesystem. it is not a secret.

[pal666]

That doesn't explicitly say all their drivers are betas or to use them at your own risk. What they're saying is to use the software they want you to use, and that you can't sue them in the event something breaks. That doesn't mean they're not stable or reliable, it's just a way for them to not get sued when you configure something wrong.

"can't sue" is exactly "use at your own risk"

[pal666]

When I was tasked to create new company Linux NAS, after research, went for FreBSD/ZFS solution

so you failed miserably

BTRFS after so many years is simply still not ready and even deprecating in process (no plugin compression, lz4 for example, which is main compression alg in ZFS now)

btrfs is ready and btrfs has zlib and lzo compression. if you think zfs' alg is better, it is your problem. while zfs after all this years can't even change its size and doesn't exist on linux and never will ( https://github.com/torvalds/linux/tree/master/fs ). zfs was obsolete before development of btrfs started and it is admitted by zfs devs

Sooner we will see proper ZFS on Linux, than proper working btrfs.

we have proper working btrfs and we will never see zfs on linux

[pal666]

Well, that's what RH is all about. They throw their weight behind open source so enterprises don't have to "use at their own risk".

all their weight is only to try to fix bugs, all risk is still yours

[timofonic]

May this be the cause?

https://www.theregister.co.uk/2017/08/01/red_hat_acquires_permabit/

Red Hat acquires Permabit to put the squeeze on RHEL

Stallman says ZFS-on-Linux is impossible ... now Red Hat has dedupe without GNU legals

By Simon Sharwood, APAC Editor 1 Aug 2017 at 05:58
SHARE ▼

Red Hat has acquired “the assets and technology of Permabit Technology Corporation”, a data-shrinking concern, for an undisclosed sum.
Permabit offers data de-duplication and compression software and recently cooked ready-to-run Linux kernel modules of its wares after previously focusing on sales to OEMs.
Red Hat has now decided that its Enterprise Linux (RHEL) needs what Permabit had. The company will add Permabit technology to RHEL and emerge ready “to better enable enterprise digital transformation through more efficient storage options.”
Red Hat thinks Permabit will make it a better platform for containers or hyperconverged infrastructure because the latter company's technology can “increase the amount of storage available to applications without increasing the amount of physical storage.”
What? Who are you? Where am I? Sorry. Dozed off there, because de-dupe and compression are technologies that storage buyers just expect to see on a feature list. They are important, but anodyne.
Another reason for the buy may be that one way to get de-dupe into Linux is with ZFS. But no less an entity than Richard Stallman has declared that tactic impossible on licensing grounds. Could Red Hat have found a way to add necessary and expected features to RHEL while also avoiding a brush with GNU legalese?
Red Hat said the transaction will have no material impact to guidance for its second fiscal quarter. ®

What a coicidence!!!

[Ruediix]

It's sad to see this go.

However, I ran into an unusual issue with BTRFS that fsck cannot tell the difference between an ext4 volume and a BTRFS volume without being explicitly told.

This is a real problem that must be addressed in the next version of the file-system, which will likely have to be drastically changed to address this issue.

[higuita]

So basically red hat do not like to not control something so important as the filesystem, specially losing against oracle... if they supported btrfs, oracle could release better/more feature/sooner than redhat and fish more clients from redhat to oracle.
So they dump it, removing support for its usage and forcing oracle more work to use if... instead they will probably extend ext4/create a ext5 filesystem or takeover/acquire/contract the NILFS or someone inside redhat had a good idea to replace it (drdb+log filesystem) to something that is not only resilient, but friendly for clusters. ZFS i do not believe, not only there is the license problem, but linus would not accept it, it would need a major rebuild to not replace most linux block code and reuse it instead.

well, lets see... i would be glad that SuSE would grow more because of this, redhat controls too much linux blocks, losing some of then would be a good thing for everyone, to avoid things like "i'm gnome/systemd dev, i know better than everyone else"

[Awesomeness]

I heard OpenSUSE is great for btrfs usage these days (hint hint).

Fedora uses vanilla upstream kernels. At worst they'll remove btrfs from Anaconda.

[johnc]

zfs was obsolete before development of btrfs started and it is admitted by zfs devs

You're talking about that one broad who never even worked on ZFS?

zfs ... doesn't exist on linux

ohhh, you're so funny.