Periodic task cause errors in _finish_resize
Bug #1321298 reported by
Lance Bragstad
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Fix Released
|
Undecided
|
Lance Bragstad |
Bug Description
In the event that an end user sets resize_
http://
One possible solution to this would be to reverse the order, and update the instance before updating the migration, in which case the migration will get updated in _confirm_resize: http://
Changed in nova: | |
assignee: | nobody → Melanie Witt (melwitt) |
Changed in nova: | |
milestone: | none → juno-1 |
status: | Fix Committed → Fix Released |
Changed in nova: | |
milestone: | juno-1 → 2014.2 |
To post a comment you must log in.
If you swap the order of the migration/instance updates in _finish_resize, then if _poll_unconfirm ed_resizes runs when the instance is updated but before the migration state is updated, this db query will not return the associated migration so it's a no-op in that case, which fixes the race bug:
http:// git.openstack. org/cgit/ openstack/ nova/tree/ nova/compute/ manager. py#n4989