Provide a hook into the HV host stats periodic_task
In large deployments, having each host periodically poll the hypervisor for instance statistics and emitting notifications would result in far too many notifications for something like Ceilometer to handle. It would be better if some logic could be pushed down into Nova via a plugin such that it could tap into the periodic_tasks that already frequently check host status. These plugins could look for anomolies and emit notifications only when set-points/limits are exceeded, thus reducing queue traffic.
Blueprint information
- Status:
- Started
- Approver:
- Russell Bryant
- Priority:
- Undefined
- Drafter:
- Sandy Walsh
- Direction:
- Needs approval
- Assignee:
- Sandy Walsh
- Definition:
- Review
- Series goal:
- None
- Implementation:
-
Slow progress
- Milestone target:
- None
- Started by
- Russell Bryant
- Completed by
Related branches
Related bugs
Sprints
Whiteboard
Resetting priority to low as the current review is abandoned, Icehouse-1 is soon, and this doesn't have any core sponsors at the moment. It would be good to have a better design doc in the blueprint, given the current push in that direction --johnthetubaguy
I'm moving most of the functionality into a RoutingNotifier under oslo.common. This branch will be dusted off and refactored once that branch lands. --Sandy
Gerrit topic: https:/
Addressed by: https:/
Metric plugin support
With Icehouse-1 just over two weeks away (December 5th), moving this to Icehouse-2 since it hasn't been started yet. If it does get into Icehouse-1, we can re-target. --jogo
Unapproved - please re-submit via nova-spec --johnthetubagy (20th March 2014)
Removed from next, as next is now reserved for near misses from the last milestone --johnthetubaguy
Gerrit topic: https:/
Addressed by: https:/
Host Metric Hook
