You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From what I've seen, a specialized driver for a new filesystem doesn't really add much advantage as we've already experienced with BTRFS.
The overlay storage backend we've got now works with many different setups and offers features that will be difficult (if not impossible) to replicate in a different backend for bcachefs, like for example its integration with partial pulls, composefs that are we currently working on right now, as well as additional image stores.
I don't think anyone in the core team will consider such effor, but if someone in the community is interested to play with it and find it useful, it can be considered for inclusion as we did for BTRFS.
With bcachefs officially mainlined into linux 6.7, has anybody looked into this new filesystem?
It is similar in features to BTRFS and ZFS, so potentially a libcontainers storage driver could be written for bcachefs.
What would the Container Storage Library need in order to consider support for bcachefs?
Would you need mainstream support of bcachefs in popular distributions first?
Packaging of the bcachefs-tools available in supported distros?
Is this something the team could consider developing?
If not, would an outside pull request be considered/accepted or are new storage drivers not welcomed?
Just wanted to get the discussion started around this new filesystem that is getting some attention now.
The text was updated successfully, but these errors were encountered: