The vm system disk data was overided by glance image data when evacuating a vm on shared storage.
I use the function of evacuating vm to another host on a shared storage(nova evacuate instance-uuid host --on-shared-storage command), and expect the data remains after evacuate, but when I do it, the data lost acually, I see the code in the master branch, the _rebuild_default_impl function's recreate parameter is not set when called by rebuild_instance function, thus when evacuate a vm, the recreate parameter is false and vm will be always destroyed by _rebuild_default_impl .
The nova/compute/manager.py file in icehouse version:
@object_compat
@messaging.expected_exceptions(exception.PreserveEphemeralNotSupported)
@wrap_exception()
@reverts_task_state
@wrap_instance_event
@wrap_instance_fault
def rebuild_instance(self, context, instance, orig_image_ref, image_ref,
injected_files, new_pass, orig_sys_metadata,
bdms, recreate, on_shared_storage,
preserve_ephemeral=False):
**********
kwargs = dict(
context=context,
instance=instance,
image_meta=image_meta,
injected_files=files,
admin_password=new_pass,
bdms=bdms,
detach_block_devices=detach_block_devices,
attach_block_devices=self._prep_block_device,
block_device_info=block_device_info,
network_info=network_info,
preserve_ephemeral=preserve_ephemeral)
the kwargs lost parameter recreate=recreate, and it result in the data always destroyed when a vm evacuated.
Can you provide a reproduce scenario? What's the overall configuration and version of components? What series of operations did you do (preferably a set of nova command line commands)? What did you expect to happen and what actually happened?