Scheduler race condition

[Brian Elliott]

The scheduler is subject to a race condition which can cause it to incorrectly identify available resources on a particular compute host. The problem occurs if multiple scheduler instances/threads concurrently issue an instance build request (i.e. run_instance) to the same compute host. This situation may oversubscribe the given compute host and cause one or more run_instance requests to fail.

See blueprint & associated wiki page:

https://blueprints.launchpad.net/nova/+spec/scheduler-resource-race

[Brian Elliott (belliott)]

https://review.openstack.org/9540 added retries to deal with resolution when races occur. (and other possibly retry-able build failures)