On Tue, Oct 16, 2018 at 10:08:59PM +0200, Michael Biebl wrote: > Am 16.10.18 um 21:36 schrieb Adam Borowski: > > Systemd's algorithm for btrfs RAID is: > > So your complaint is specific to btrfs RAID which afaik is still > considered unstable?
Care to specify what's unstable with btrfs RAID? Any non-experimental level (experimental include 5/6, 3-way mirroring, etc) has no known stability issues; what's left are missing optimizations or features that'd be nice to have. But Ian is right -- this has turned into advocacy again; I'd be happy to discuss filesystem issues with you privately or on linux-btrfs if you wish. My point earlier in this subthread was that mounting RAIDs doesn't fit well within systemd's event scheme and would require at least some rethinking -- even using static ordering would work better. Meow! -- ⢀⣴⠾⠻⢶⣦⠀ ⣾⠁⢰⠒⠀⣿⡁ 10 people enter a bar: 1 who understands binary, ⢿⡄⠘⠷⠚⠋⠀ 1 who doesn't, D who prefer to write it as hex, ⠈⠳⣄⠀⠀⠀⠀ and 1 who narrowly avoided an off-by-one error.