Live data from Hacker News

OpenZFS 2.0

github.com

41–50 of 151 posts

Re: OpenZFS 2.0

#41
post #24

I'd love to get rid of my FreeNAS VM and run ZFS directly on my Linux desktop, but having to mess with the kernel has kept me from attempting it so far. Maybe I'm worrying about nothing. btrfs seems like the main alternative if you want native kernel support, but when I checked a couple years ago there seemed to be a lot of concerns about the stability. Is that still the case?

I personally use ZFS on Arch Linux. The DKMS package works almost out of the box and I haven't had any troubles. It takes a long time (but not too much) to compile though. Or you can use the latest Ubuntu that is shipped with ZFS.

Be aware of possible incompatibilities with the regular Arch kernel. I switched my NAS to linux-lts when a kernel point-release proved incompatible with openzfs.

Re: OpenZFS 2.0

#42
post #31

Sadly dRAID (parity Declustered RAIDz) just missed the cut-off for 2.0, but it looks like it will be in 2.1: * https://openzfs.github.io/openzfs-docs/Basic%20Concepts/dRAI... * https://www.youtube.com/watch?v=jdXOtEF6Fh0

dRAID looks really fascinating, but presentation is pretty abstract. Would it allow to add/remove drives from a pool, and allow ZFS to rebalance itself? Would be great for home use, where I have a lot of drives that I collected over the years that are not the same size. EDIT: The more I read into this, it still seems assume that all drives must be of the same size.

Maybe this presentation by Mark will help?

dRAID, Finally![0]

[0]: https://www.youtube.com/watch?v=jdXOtEF6Fh0

Re: OpenZFS 2.0

#43
post #31

Earlier quoted context omitted.

dRAID looks really fascinating, but presentation is pretty abstract. Would it allow to add/remove drives from a pool, and allow ZFS to rebalance itself? Would be great for home use, where I have a lot of drives that I collected over the years that are not the same size. EDIT: The more I read into this, it still seems assume that all drives must be of the same size.

You can already do that with btrfs.

I currently use btrfs with RAID1 at home, and it works great. But btrfs also does not have the track record for being the most stable filesystem as compared to ZFS.

[1] https://lore.kernel.org/linux-btrfs/20200627032414.GX10769@h... [2] https://lore.kernel.org/linux-btrfs/20200627030614.GW10769@h... [3] https://lore.kernel.org/linux-btrfs/20200520013255.GD10769@h...

Re: OpenZFS 2.0

#44
post #32

Earlier quoted context omitted.

ZFS is no extra work with NixOS! You just declare the filesystem type like any other in the config and it takes care of kernel modules and what-not.

Is there a guide for this? Sounds interesting!

For instance, I have:

     fileSystems."/zfs/media" =
        { device = "tank/media";
          fsType = "zfs";
        };
in my hardware-configuration.nix. tank/media is defined as using a legacy mount-point or whatever the ZFS terminology is. Done.

ETA: I mean, I had to do all the gruntwork to get the pool built, yeah. But once it was defined, getting it mounted and all the kernel bits and bobs set was trivial like that.

Re: OpenZFS 2.0

#45

I'd love to get rid of my FreeNAS VM and run ZFS directly on my Linux desktop, but having to mess with the kernel has kept me from attempting it so far. Maybe I'm worrying about nothing. btrfs seems like the main alternative if you want native kernel support, but when I checked a couple years ago there seemed to be a lot of concerns about the stability. Is that still the case?

ZFS is no extra work with NixOS! You just declare the filesystem type like any other in the config and it takes care of kernel modules and what-not.

Sure - after you figure out NixOS lol

Re: OpenZFS 2.0

#46

I'd love to get rid of my FreeNAS VM and run ZFS directly on my Linux desktop, but having to mess with the kernel has kept me from attempting it so far. Maybe I'm worrying about nothing. btrfs seems like the main alternative if you want native kernel support, but when I checked a couple years ago there seemed to be a lot of concerns about the stability. Is that still the case?

On the btrfs mailing list [1], there are still sporadic reports of unrecoverable FS corruption for whatever reason. See [2], [3] for some recent examples.

[1] https://lore.kernel.org/linux-btrfs/

[2] https://lore.kernel.org/linux-btrfs/CAD7Y51i=mTDnEWEJtSnUsq=...

[3] https://lore.kernel.org/linux-btrfs/CAMXR++KUj2L7qpR7QZeiM2T...

Re: OpenZFS 2.0

#47
post #20
post #5

Will OpenZFS on Linux ever be integrated with the Linux page cache?

No, they have ARC and ARCL2, if you want the traditional thing go to NILFS2 or BTRFS or in the future XFS (when they have full check-summing).

>in the future XFS (when they have full check-summing).

Is this actually planned?

Re: OpenZFS 2.0

#48
Congratulations - it's great to see the code unification on the two key ZFS platforms, and continuing to add useful features, especially around at-rest encryption.

Many thanks to the various OpenZFS contributors.

Re: OpenZFS 2.0

#49
post #29

Earlier quoted context omitted.

> well the commands are terrible compared to ZFS Really? I don’t think so, I find btrfs usage extremely straightforward and easy to grok. ZFS on the other hand has all that confusing lingo about vdevs, etc... I get that this is subjective but I disagree.

Hey everyone has a different taste, but vdevs, datasets, and pool are for me much more logical than lv's and lg's (pun was NOT intended).

but thats not really btrfs, thats LVM. I use BTRFS directly on physical disks and dont use pvs, vgs or lvs.

Re: OpenZFS 2.0

#50

I'd love to get rid of my FreeNAS VM and run ZFS directly on my Linux desktop, but having to mess with the kernel has kept me from attempting it so far. Maybe I'm worrying about nothing. btrfs seems like the main alternative if you want native kernel support, but when I checked a couple years ago there seemed to be a lot of concerns about the stability. Is that still the case?

> but having to mess with the kernel has kept me from attempting it so far. Maybe I'm worrying about nothing.

For the most part, yes. Occasionally a kernel developer who seems to be bitter about a company that doesn't exist any more tries to break compat with ZFS, but it's generally smooth sailing on Fedora, Debian, and CentOS, with dkms handling the building of modules seamlessly.

Post reply on HN