Setting the migration record status to error is easy, so I'll start with that. Refactoring the code to handle errors but still cleanup allocations is a bit more complicated, so that can be split into a separate change to close the bug.
Also note related bug 1793569 with some scripts for cleaning up orphaned allocations.
IRC discussion here:
http:// eavesdrop. openstack. org/irclogs/ %23openstack- nova/%23opensta ck-nova. 2019-03- 25.log. html#t2019- 03-25T13: 07:57
Setting the migration record status to error is easy, so I'll start with that. Refactoring the code to handle errors but still cleanup allocations is a bit more complicated, so that can be split into a separate change to close the bug.
Also note related bug 1793569 with some scripts for cleaning up orphaned allocations.