On Thu, Feb 27, 2014 at 09:16:10PM +0000, Dafna Ron wrote:
wipe = writing zero's on the space allocated to that disk to make sure any data once written will be deleted permanently.
so it's a security vs speed decision on using this option - since we zeroing the disk to make sure any information once written will be overwritten, a delete of a large disk can take a while.
To be more precise: if not using "Wipe on delete" on block storage, the same storage used by the deleted disks is to be used for a future vDisk. If that disk is to be allocated as raw storage, and it is attached to another VM, that VM would have access to the original data. If you trust your VMs, or use qcow, or mkfs before running guests, you do not need this io-eating flag. Dan.