btrfs / btrfs-todo

An issues only repo to organize our TODO items
21 stars 2 forks source link

btrfs redundancy report #59

Closed CHerSun closed 6 months ago

CHerSun commented 6 months ago

I'd really appreciate some way to cleanly get blocks redundancy state.

Example:

lsblk - reports 3 drives now. btrfs device scan - reports 3 devices with no other notices btrfs device usage ... - reports all 4 drives with no marks really, only device size = 0 btrfs filesystem show ... - shows all 4 drives, but marks drive as missing. btrfs filesystem usage -T ... - shows all devices without marking them in any way (with missing device value > 0). btrfs filesystem df ... - may report or may not report *some devices are missing. Looks like that depends on which drive went missing, not sure.

Judging by that - btrfs knows it does miss a device, but it doesn't do the next step - to notify about redundancy problem in a clear way, so that end user could take a proper action, if needed (in my case btrfs device delete <devid> until I can replace). What I mean is that I'd really like to see something like btrfs filesystem status ... with statements like:

Or:

Or (hypothetically):

Is that possible?

kdave commented 6 months ago

For things that can be implemented purely in btrfs-progs please open the issue at https://github.com/kdave/btrfs-progs, here on the btrfs-todo the issue can get lost.

CHerSun commented 6 months ago

For things that can be implemented purely in btrfs-progs please open the issue at https://github.com/kdave/btrfs-progs, here on the btrfs-todo the issue can get lost.

Thank you. I've duplicated this issue in btrfs-progs https://github.com/kdave/btrfs-progs/issues/791 .

Should I close this one, or keep it till there's some kind of reply in any of repos?

kdave commented 6 months ago

Thanks, we better keep just one open for any dicussions, so closing this one.