On Sun, Jun 12, 2011 at 10:43 PM, Simon Tian <[email protected]> wrote: > 2011/6/13 Yehuda Sadeh Weinraub <[email protected]>: >> Rollback is (currently) is in the order of the number of object of >> that image. It needs to go through all the objects and execute an >> explicit request of each. For very large images it can be quite long. > > When a VM is running based on rbd image, and need to rollback to S1. > During rollback, seem like that data written by the VM is not allowed? > Is there any way to rollback without affecting the user experience of the VM? > Data written by the vm to the image is not recommended as it'll most likely corrupt it. It is recommended that you unmount any relevant partition while doing it. Currently there's not much to do to avoid the relatively long rollback. We do have on our roadmap the layering feature that will allow having some sort of writable snapshots, but we're not there yet.
Yehuda -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
