Bcachefs Might Be Ready For Upstreaming In Linux This Year

[phoronix]

Phoronix: Bcachefs Might Be Ready For Upstreaming In Linux This Year

The Bcachefs file-system that was born out of the Linux kernel's block cache code has over the past few years matured greatly. Now in 2022 the core fundamentals of the file-system are "pretty close to done" and will hopefully be mainlined this calendar year into the Linux kernel...

https://www.phoronix.com/scan.php?page=news_item&px=Bcachefs-2022-Hopes

[mdedetrich]

Really looking forward to seeing the release, especially fond of Kent's attitude in developing bcachefs (i.e. get the core design principles right before releasing it).

[birdie]

Wow, wow, wow! Looking forward to it! The guy's perfectionist and he's been developing it on its own.

[babali]

Bomb! I've been waiting for it!

[uid313]

bcachefs allegedly at one time shared about 80% of the code with bcache.

Jens Axboe asked if there was still a clean separation between bcachefs and bcache. Overstreet said that there was; roughly 80% of the code is shared.

[Jannik2099]

There are still a bunch of severe bugs in bcache that haven't been fixed for years, anyone jumping on bcachefs straight away is just asking to repeat the early btrfs experience.

It's a single dev, working on a filesystem with about as many features as ZFS - calm your expectations

[amxfonseca]

It already solves one of the current limitations of btrfs. It allows to set the number of replicas on a subvolume/folder basis.

Ideally I want to have a single pool of let’s say 4 disks. And then set the replicas for for stuff like pictures and documents to 3. And for downloaded media I would use a single replica, since I can afford to lose that data during a disk failure.

[flower]

There are still a bunch of severe bugs in bcache that haven't been fixed for years, anyone jumping on bcachefs straight away is just asking to repeat the early btrfs experience.

It's a single dev, working on a filesystem with about as many features as ZFS - calm your expectations

but compared to zfs it has many benefits. esp for soho. eg different striping/mirror strategies for different folders, tiered storage (eg m2, ssd, hdd).

so well.... i am *really* looking forward to use it. i have backups anyway (they are needed with zfs too)

[darkbasic]

There are still a bunch of severe bugs in bcache that haven't been fixed for years

It depends on the reason: if they had not been fixed because a major rewrite of some related subsystem is planned that would make sense in an experimental fs.

[Deathcrow]

It already solves one of the current limitations of btrfs. It allows to set the number of replicas on a subvolume/folder basis.

Ideally I want to have a single pool of let’s say 4 disks. And then set the replicas for for stuff like pictures and documents to 3. And for downloaded media I would use a single replica, since I can afford to lose that data during a disk failure.

The usefulness of this seems to depend a lot on how gracefully the OS can handle missing or partially missing data (some extents of a file may be available and others gone after a disc replacement).

At first glance such a setup just looks like additional headaches and not particularly useful in practice.