hello all, I faced an incident for one of my very important rbd volumes with 5TB data, which is managed by OpenStack. I was about to increase the volume size live but I shrinked the volume unintentionally by running a wrong command of "virsh qemu-monitor-command". then I realized it and expand it again, but obviously I lost my data. can you help me or give me hints on how I can recover data of this rbd volume? unfortunately, I don't have any backup for this part of data, and it's really important (I know I made a big mistake). also, I can't stop the cluster since it's under heavy production load. It seems qemu has been using the old set of APIs that allows shrinking by default without any warnings even in the latest version. (All other standard ways of resizing volume, does not allow shrinking) * Ceph: https://sourcegraph.com/github.com/ceph/ceph@luminous/-/blob/src/librbd/librbd.cc#L815 * Qemu: https://sourcegraph.com/github.com/qemu/qemu/-/blob/block/rbd.c#L832 * Libvirt: https://sourcegraph.com/github.com/libvirt/libvirt/-/blob/src/storage/storage_backend_rbd.c#L1280 _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |