abba at soungui.cm
2022-Jul-13 13:23 UTC
[Gluster-users] Geo replicated VM disks corrupted
Hello, I have a Gluster volume on which I store virtual machines disks. I've created a geo replication session for disaster recovery and as suggested in many documentations and blogs on Internet: - I create a snapshot of all disks before starting the replication , to make sure that the base image is consistent. - I set a checkpoint. - I start the geo replication. - When the checkpoint is reached, the geo replication stops and the snapshots are deleted on the primary volume. However, most of time (not always), when I try to use the replicated base images, they are unusable (XFS partitions on virtual disks are generally corrupted). I wanted to know if I am doing something wrong and if someone ever managed to use the geo replication feature successfuly for disaster recovery of virtual machines. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20220713/9a326d7e/attachment.html>