I really don't known how it's works, as by nature, glusterfs is done to manage file and not blocks. So maybe it need to copy the whole file instead missing blocks ?
from doc: http://www.gluster.org/community/documentation/index.php/GlusterFS_Technical_FAQ#What_about_deletion_self.2Fauto_healing.3F " What about deletion self/auto healing? With auto healing only file creation is healed. If a brick is missing because of a disk crash re-creation of files is ok but if it's a temporary network problem synchronizing deletion is mandatory. See also Gluster 3.2: Triggering Self-Heal on Replicate. " http://www.gluster.org/community/documentation/index.php/Gluster_3.2:_Triggering_Self-Heal_on_Replicate ----- Mail original ----- De: "Dietmar Maurer" <[email protected]> À: "Alexandre DERUMIER" <[email protected]> Cc: [email protected] Envoyé: Lundi 12 Août 2013 13:03:32 Objet: RE: [pve-devel] pve-storage: add Glusterfs Plugin I tested with your single 3-node setup. But I do not understand how that self-healing work. I simply installed a VM, and rebooted one node during install. After that there is a high processor use with glusterfs/glusterfsd on all nodes. But how can I display recovery status? Seems they simply copy the whole data to the node that was offline? _______________________________________________ pve-devel mailing list [email protected] http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
