Option needed for ceph rbd erasure coding support

Registered by Mike Lowe on 2018-07-05

Erasure coded pools for rbd have been supported since the ceph
luminous release. To use this feature the erasure coded pool where
the data will be held needs to be specified during rbd device creation.
The images_rbd_pool option still needs to point to a replicated pool where
a small amount of metadata will be stored in OMAP. Currently the
driver implicitly uses the default value of None.

Blueprint information

Status:
Not started
Approver:
None
Priority:
Undefined
Drafter:
Mike Lowe
Direction:
Needs approval
Assignee:
Mike Lowe
Definition:
New
Series goal:
None
Implementation:
Unknown
Milestone target:
None

Related branches

Sprints

Whiteboard

We discussed this blueprint in the nova meeting today and we agreed to hold approval for the blueprint contingent on proof-of-concept code being uploaded to gerrit. The rationale was that this work should only be a small patch to the ceph imagebackend code and as we're well past spec freeze (June 7) and only 3 weeks from feature freeze (July 26), we need to see the code ASAP to have any expectation of this making it into Rocky. So, please upload a proof-of-concept patch and if the change is as simple as we expect, we can approve this blueprint. -- melwitt 20180705

Gerrit topic: https://review.openstack.org/#q,topic:bp/rbd-erasure-coding,n,z

Addressed by: https://review.openstack.org/581055
    Option needed for ceph rbd erasure coding support

(?)

Work Items

This blueprint contains Public information 
Everyone can see this information.

Subscribers

No subscribers.