I've been getting inconsistent answers to this through various sources, so I figured I'd give this group a shot as a more legitimate authority. I'm using a set of EBS volumes on AWS in an mdadm RAID0 array. In order to do an EBS (block-level) snapshot, we are currently stopping the service, freezing the FS, disassembling the RAID and snapshotting once the disks can't possibly be talking to each other anymore, for safety. Is all that necessary? If I stop writes from the service and sync + xfs_freeze the filesystem, will those snapshots be point-in-time consistent without having to tear it down and rebuild it every time? Or should we be doing something else to further protect it, like adding an LVM layer? Thanks, Jon -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
