Skip to main content

looking like I'll get credited as bug reporter for a Linux btrfs bug: patch and bug report

this is silver lining on how it has become harder over time for me to whole-heartedly recommend btrfs, especially to "ordinary" Linux end users... I think I'd still do it on balance? But I'd say a backup strategy is not optional. (But maybe I'd say that anyway.)

see also: my backup strategy

in reply to Ben Millwood

Ben Weinstein-Raun
Ben Weinstein-Raun
•
Are there reasons to use btrfs over zfs?
in reply to Ben Weinstein-Raun

Ben Millwood
Ben Millwood
•
zfs has a couple of its own bugs that I've run into, e.g. bad handling of unplugged usb and terrible zfs diff performance. I don't feel super opinionated between them. zfs is a bit more aggressive about managing mount points and options instead of doing it via system configuration, which you might take as a good or bad thing. I find btrfs snapshots a bit more convenient to work with (they're just there on your disk, whereas with zfs I think you need to clone them to a new FS first or something)
in reply to Ben Millwood

Ben Weinstein-Raun
Ben Weinstein-Raun
•

Looking around on the internet for other issues just now, it's also apparently pretty common to run into showstopper bugs when trying to do zfs send or zfs receive on encrypted zpools. I've never tried it but that does seem pretty bad.

And yeah, that USB issue sure looks annoying :/

in reply to Ben Weinstein-Raun

Ben Millwood
Ben Millwood
•
Do you mean only with the native encryption? I use LUKS and I imagine it can't even tell I'm doing that?
in reply to Ben Millwood

Ben Weinstein-Raun
Ben Weinstein-Raun
•
yes, sorry, specifically with native encryption. afaik you're right that it doesn't know if it's running on top of LUKS
in reply to Ben Millwood

Ben Millwood
Ben Millwood
•
⇧