Will look into these in some time. But to make sure we have conversation going, will track everything here: https://github.com/gluster/glusterfs/issues/145
Regards,On Wed, Mar 22, 2017 at 1:19 PM, Mark Ferrell <major@xxxxxxxxxxxxxxxxxx> wrote:
Both of these branches are currently in-flux, and I have been performing regular rebasing against the GlusterFS master branch.This code is branched from my lvm-snapshot-cleanup branch and is directly dependant on the changes contained there-in.I currently have 2 branches related to snapshots on GlusterFS.https://github.com/major0/glusterfs/tree/lvm-snapshot- cleanup
Based code from Sriram that I found on GitHub, though I removed the ZFS code from Sriram's patches and created this LVM-only branch. Subsequent changes to this branch where done i response to changes necessary to implement support for btrfs snapshots (and cleanup a memory leak in the missed-snapshot code)...
https://github.com/major0/glusterfs/tree/btrfs-snapshots Btrfs snapshots currently work and have been tested on Ubuntu 16.04, though the process for configuring btrfs for gluster snapshots has not yet been documented.Required interfaces to support hard-coded snapshot support can be found:
https://github.com/major0/glusterfs/blob/lvm-snapshot- cleanup/xlators/mgmt/glusterd/ src/snapshot/glusterd-lvm- snapshot.h It should be relatively straight-forward to package up these interfaces into a structure of function pointers and allow glusterd to itterate the array firing off the glusterd_is_<type>_brick() function at the path to figure out which subsystem to use.
_______________________________________________
Gluster-devel mailing list
Gluster-devel@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-devel
--
Amar Tumballi (amarts)
_______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-devel
