Healing time frame

Registered by Hiroaki Kobayashi on 2018-02-15

Current resource monitor tries to heal all of related reservations when it detects resource failure. However, some reservations will not suffer from the failure because the failed resource will recover sometime in the future. So, the healing should consider the recovery of failed resources.

This blueprint is for introducing time frame for healing. The monitor will try to heal only active reservations and reservations which will start in the time frame. Rest of reservations will be healed little by little until the failed resource recovers.

This is a remaining task of the resource-monitoring blueprint.

Blueprint information

Status:
Complete
Approver:
Masahito Muroi
Priority:
Medium
Drafter:
Hiroaki Kobayashi
Direction:
Needs approval
Assignee:
Hiroaki Kobayashi
Definition:
Approved
Series goal:
Accepted for rocky
Implementation:
Implemented
Milestone target:
milestone icon rocky-2
Started by
Hiroaki Kobayashi on 2018-03-16
Completed by
Masahito Muroi on 2018-05-10

Related branches

Sprints

Whiteboard

See the discussion in https://etherpad.openstack.org/p/blazar-ptg-rocky

Gerrit topic: https://review.openstack.org/#q,topic:bp/healing-time,n,z

Addressed by: https://review.openstack.org/529999
    Add periodic healing

Addressed by: https://review.openstack.org/561512
    Fix initialization of the PhysicalHostMonitorPlugin

(?)

Work Items

This blueprint contains Public information 
Everyone can see this information.

Subscribers

No subscribers.