Provide a default fall-through driver for the multi scheduler
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Fix Released
|
Low
|
David Scannell |
Bug Description
We are building an Openstack Compute extension that has its own high level manager process, similar to nova-compute or nova-volume: https:/
We want to use the common nova-scheduler to schedule requests to our manager but unfortunately the multi scheduler is hard-codeded for only compute and volume. Basically we want to have a default fall-through option so that our extension, and potentially other ones, can make use the multi scheduler. In other words, if the scheduling request is not destined for the compute or volume topic, then it will use this default driver (e.g. ChangeScheduler) to handle the request.
Attached is a simple patch that defines a new config option "default_
Changed in nova: | |
importance: | Undecided → Low |
Changed in nova: | |
milestone: | none → folsom-2 |
status: | Fix Committed → Fix Released |
Changed in nova: | |
milestone: | folsom-2 → 2012.2 |
This seems reasonable, although I was considering removing the multi-scheduler completely once the move to volumes is complete and switching the default to the filter scheduler.