- cross-posted to:
- linux@programming.dev
- cross-posted to:
- linux@programming.dev
There’s been some Friday night kernel drama on the Linux kernel mailing list… Linus Torvalds has expressed regrets for merging the Bcachefs file-system and an ensuing back-and-forth between the file-system maintainer.
Can someone say why bcachefs is interesting? Btrfs I can sort of understand. I haven’t much kept track of most others.
bcachefs is way more flexible than btrfs on multi-device filesystems. You can group storage devices together based on performance/capacity/whatever else, and then do funky things like assigning a group of SSDs as a write-through/write-back cache for a bigger array of HDDs. You can also configure a ton of properties for individual files or directories, including the cache+main storage group, amount of data replicas, compression type, and quite a bit more.
So you could have two files in the same folder, one of them stored compressed on an array of HDDs in RAID10 and the other one stored on a different array of HDDs uncompressed in RAID5 with a write-back SSD cache, and wouldn’t have to fiddle around with multiple filesystems and bind mounts - everything can be configured by simply setting xattr values. You could even have a third file which is striped across both groups of HDDs without having to partition them up.
Thank you. w o w
Now that’s what I call serious over-engineering.
Who in the world wants to use that?
And does that developer maybe have some spare time? /s
This is actually a feature that enterprise SAN solutions have had for a while, being able choose your level of redundancy & performance at a file level is extremely useful for minimising downtime and not replicating ephemeral data.
Most filesystem features are not for the average user who has their data replicated in a cloud service; they’re for businesses where this flexibility saves a lot of money.
I’ll also tac on that when you use cloud storage, what do you think your stuff is stored on at the end of the day? Sure as shit not Bcachefs yet, but it’s more likely than not on some netapp appliance for the same features that Bcachefs is developing.
Simple example: my Steam library could be RAID0 and unencrypted but my backups I definitely want to be RAID1 and compressed, and encrypted for security. The media library doesn’t need encryption but maybe want it in RAID1 because ripping movies takes forever. I may also want to have the games on NVMe when I play them, and stored on the HDDs when I’m not playing them, and my VMs on the SATA SSD array as a performance middleground.
Wonderful.
But these are libraries. Not single files.
Yes, which is why these settings can also be configured per-directory as well as per-file.
This probably meets some extreme corporate usecase where they are serving millions of customers.
It’s not that obscure - I had a use case a while back where I had multiple rocksdb instances running on the same machine and wanted each of them to store their WAL only on SSD storage with compression and have the main tables be stored uncompressed on an HDD array with write-through SSD cache (ideally using the same set of SSDs for cost). I eventually did it, but it required partitioning the SSDs in half, using one half for a bcache (not bcachefs) in front of the HDDs and then using the other half of the SSDs to create a compressed filesystem which I then created subdirectories on and bind mounted each into the corresponding rocksdb database.
Yes, it works, but it’s also ugly as sin and the SSD allocation between the cache and the WAL storage is also fixed (I’d like to use as much space as possible for caching). This would be just a few simple commands using bcachefs, and would also be completely transparent once configured (no messing around with dozens of fstab entries or bind mounts).
Is there a reason for bind mounting and not just configuring the db to point at a different path?
I mean… If you have a ton of raw photos in one directory, you can enable the highest compression rate with zstd to it. Every other directory has lz4 with the fastest compression. Your pics take much less space, but the directory will be slower to read and write.
My company serves millions, too. We don’t have such needs.
Congrats.
Sounds like zfs with extra steps
But is GPL-compatible, unlike ZFS.
How do you define GPL compatible?
Do your own research, that’s a pretty well-discussed topic, particularly as concerns ZFS.
I’m all over ZFS and I am not aware of any unresolved “licence issues”
License incompatibility is one big reason OpenZFS is not in-tree for Linux, there is plenty of public discussion about this online.
Like this that states there is no issue https://opensource.stackexchange.com/questions/2094/are-cddl-and-gpl-really-incompatible
Your lack of awareness is fine with me.
Okay thanks for your comment?
Not under a license which prohibits also licensing under the GPL. i.e. it has no conditions beyond what the GPL specifies.
Not true
The only condition is that CCDL and GPL don’t apply to the same file. Wifi works just fine and the source code isn’t GPL yet wifi drivers are in the kernel…
https://opensource.stackexchange.com/questions/2094/are-cddl-and-gpl-really-incompatible
…because they are incompatible licenses.
There’s no requirement for them to apply to the same file? There’s already blobs in the kernel the gpl doesn’t apply to the source of
ZFS doesn’t support tiered storage at all. Bcachefs is capable of promoting and demoting files to faster but smaller or slower but larger storage. It’s not just a cache. On ZFS the only option is really multiple zpools. Like you can sort of do that with the persistent L2ARC now but TBs of L2ARC is super wasteful and your data has to fully fit the pool.
Tiered storage is great for VMs and games and other large files. Play a game, promote to NVMe for fast loadings. Done playing, it gets moved to the HDDs.
You’re misrepresenting L2ARC and it’s a silly comparison to claim to need TBs of L2ARC and then also say you’d copy the game to nvme just to play it on bcachefs. That’s what ARC does. RAM and SSD caching of the data in use with tiered heuristics.
I know, that was an example of why it doesn’t work on ZFS. That would be the closest you can get with regular ZFS, and as we both pointed out, it makes no sense, it doesn’t work. The L2ARC is a cache, you can’t store files in it.
The whole point of bcachefs is tiering. You can give it a 4TB NVMe, a 4TB SATA SSD and a 8 GB HDD and get almost the whole 16 TB of usable space in one big filesystem. It’ll shuffle the files around for you to keep the hot data set on the fastest drive. You can pin the data to the storage medium that matches the performance needs of the workload. The roadmap claims they want to analyze usage pattern and automatically store the files on the slowest drive that doesn’t bottleneck the workload. The point is, unlike regular bcache or the ZFS ARC, it’s not just a cache, it’s also storage space available to the user.
You wouldn’t copy the game to another drive yourself directly. You’d request the filesystem to promote it to the fast drive. It’s all the same filesystem, completely transparent.
Looks dead on arrival to me, so much complexity for “performance” but the filessystem is outclassed by everything else in existence. If there was any real performance from this complexity it could have cool niche use cases but this is very disappointing https://www.phoronix.com/review/bcachefs-linux-67/2
Brand new anything will not show up with amazing performance, because the primary focus is correctness and features secondary.
Premature optimisation could kill a project’s maintainability; wait a few years. Even then, despite Ken’s optimism I’m not certain we’ll see performance beating a good non-cow filesystem; XFS and EXT4 have been eeking out performance for many years.
Cow is an excuse for writing performance, though the read is awful too currently
It has gotten better since November of last year though, here’s a more recent benchmark showing it beating btrfs quite often: https://www.phoronix.com/review/linux-611-filesystems/2
Improvement is nice to see, still not ready for prime time
For me the reason was that I wanted encryption, raid1 and compression with a mainlined filesystem to my workstation. Btrfs doesn’t have encryption, so you need to do it with luks to an mdadm raid, and build btrfs on top of that. Luks on mdadm raid is known to be slow, and in general not a great idea.
ZFS has raid levels, encryption and compression, but doesn’t have fsck. So you better have an UPS for your workstation for electric outages. If you do not unmount a ZFS volume cleanly, there’s a risk of data loss. ZFS also has a weird license, so you will never get it with mainline Linux kernel. And if you install the module separately, you’re not able to update to the latest kernel before ZFS supports it.
Bcachefs has all of this. And it’s supposed to be faster than ZFS and btrfs. In a few years it can really be the golden Linux filesystem recommended for everybody. I sure hope Kent gets some more help and stops picking fights with Linus before that.
ZFS doesn’t have fsck because it already does the equivalent during import, reads and scrubs. Since it’s CoW and transaction based, it can rollback to a good state after power loss. So not only does it automatically check and fix things, it’s less likely to have a problem from power loss in the first place. I’ve used it on a home NAS for 10 years, survived many power outages without a UPS. Of course things can go terribly wrong and you end up with an unrecoverable dataset, and a UPS isn’t a bad idea for any computer if you want reliability.
Totally agree about mainline kernel inclusion, just makes everything easier and ZFS will always be a weird add-on in Linux.
Why involve mdadm? You can use one btrfs filesystem on a pair of luks volumes with btrfs’s “raid1” (or dup) profile. Both volumes can decrypt with the same key.
ngl, the number of mainline Linux filesystems I’ve heard this about. ext2, ext3, btrfs, reiserfs, …
tbh I don’t even know why I should care. I understand all the features you mentioned and why they would be good, but i don’t have them today, and I’m fine. Any problem extant in the current filesystems is a problem I’ve already solved, or I wouldn’t be using Linux. Maybe someday, the filesystem will make new installations 10% better, but rn I don’t care.
It’s a filesystem that supports all of these features (and in combination):
If that is meaningless to you, that’s fine, but it sure as hell looks good to me. You can just stick with ext3 - it’s rock solid.
ZFS doesn’t have Linux fsck has it is its own thing. It instead has ZFS scrubbing which fixes corruption. Just make sure you have at least raid 1 as without a duplicate copy ZFS will have no way of fixing corruption which will cause it to scream at you.
If you just need to get data off you can disable error checking. Just use it at your own risk.
But scrub is not fsck. It just goes through the checksums and corrects if needed. That’s why you need ECC ram so the checksums are always correct. If you get any other issues with the fs, like a power off when syncing a raidz2, there is a chance of an error that scrub cannot fix. Fsck does many other things to fix a filesystem…
So basically a typical zfs installation is with UPS, and I would avoid using it on my laptop just because it kind of needs ECC ram and you should always unmount it cleanly.
This is the spot where bcachefs comes into place. It will implement whatever we love about zfs, but also be kind of feasible for mobile devices. And its fsck is pretty good already, it even gets online checks in 6.11.
Don’t get me wrong, my NAS has and will have zfs because it just works and I don’t usually need to touch it. The NAS sits next to UPS…
I have never had an issue with ZFS as long as there is a redundant copy. A bad ram might cause an issue but that’s never happened to me. I did have a bad motherboard that corrupted data on write. ZFS threw its hands up but there wasn’t any unfixable corruption
Me neither, but the risk is there and well documented.
The point was, ZFS is not great as your normal laptop/workstation filesystem. It kind of requires a certain setup, can be slow in certain kinds of workflows, expects disks of same size and is never available immediately for the latest kernel version. Nowadays you actually can add more disks to a pool, but for a very long time you needed to build a new one. Adding a larger disk to a pool will still not resize it, untill all the disks are replaced.
It shines with steady and stable raid arrays, which are designed to a certain size and never touched after they are built. I would never use it in my workstation, and this is the point where bcachefs gets interesting.
Encryption and compression don’t play well together though. You should consider that when storing sensitive files. That’s why it’s recommended to leave compression off in https because it weakens the encryption strength
How does that work? Encryption should not care at all about the data that is being encrypted. It is all just bytes at the end of the day, should not matter if they are compressed or not.
Disabling compression in HTTPS is advised to prevent specific attacks, but this is not about compression weakening encryption directly. Instead, it’s about preventing scenarios where compression could be exploited to compromise security. The compression attack is used to leak information about the content of the encrypted data, and is specific to HTTP, probably because HTTP has a fixed or guessable structure.
Looks to be an exploit only possible because compression changes the length of the response and the data can be injected into the request and is reflected in the response. So an attacker can guess the secret byte by byte by observing a shorter response form the server.
That seems like something not feasible to do to a storage device or anything that is encrypted at rest as it requires a server actively encrypting data the attacker has given it.
We should be careful of seeing a problem in one very specific place and then trying to apply the same logic to everything broadly.
It is only in TLS where you have to disable compression, not in HTTP.
https://security.stackexchange.com/questions/19911/crime-how-to-beat-the-beast-successor/19914#19914
Could you explain how a CRIME attack can be done to a disk?
There is also the BREACH which targets gzip/deflate compression on http as well. But also, don’t see how that affects disk encryption.
I can’t explain, perhaps due to my limited knowledge about the subject. I understood that compression was a weakening factor for encryption years ago when I heard about it. Always good to do your own research in the end 🙃
bcachefs is meant to be more reliable than btrfs - which has had issues with since it was released (especially in the early days). Though bcachefs has yet to be proven at scale that it can beat btrfs at that.
Bcachefs also supports more features I believe - like encryption. No need for an extra layer below the filesystem to get the benefits of encryption. Much like compression that also happens on both btrfs and bcachefs.
Btrfs also has issues with certain raid configurations, I don’t think it yet has support for raid 5/6 like setup and it has promised that for - um, well maybe a decade already? and I still have not heard any signs of it making any progress on that front. Though bcachefs also still has this on their wishlist - but I see more hope for them getting it before btrfs which seems to have given up on that feature.
Bcachefs also claims to have a cleaner codebase than btrfs.
Though bcachefs is still very new so we will see how true some of its claims will end up being. But if true it does seem like the more interesting filesystem overall.
Also because it’s meant to be an enterprise level filesystem like ZFS, but without the licensing baggage. They share a lot of feature sets.
In addition to the comment on the mentioned better hardware flexibility, I’ve seen really interesting features like defining compression & deduplication in a granular way, even to the point of having a compression algo when you first write data, and then a different more expensive one when your computer is idle.
deleted by creator
Btrfs has architectural issues that can not be fixed. It is fine for smaller raid 0/1 but as soon as you try to scale it up you run into performance issues. This is because of how it was designed.
Bcachefs is like btrfs and has all the features btrfs does. However, it also is likely to be much faster. Additionally it has some extra features like tiered storage which allows you to have different storage mediums.