[openstack-dev] [ironic] FFE request for node traits

Dmitry Tantsur dtantsur at redhat.com
Tue Jan 23 10:04:34 UTC 2018

+1 on keeping moving forward with it. that's important for future nova work, as 
well as our deploy steps work.

On 01/22/2018 10:11 PM, Mark Goddard wrote:
> The node traits feature [1] is an essential priority for ironic in Queens, and 
> is an important step in the continuing evolution of scheduling enabled by the 
> placement API. Traits will allow us to move away from capability-based 
> scheduling. Capabilities have several limitations for scheduling including 
> depending on filters in nova-scheduler rather than allowing placement to select 
> matching hosts. Several upcoming features depend on traits [2].
> Landing node traits late in the cycle will lead to less time being available for 
> testing, with a risk that the feature is release with defects. There are changes 
> at most major levels in the code except the drivers, but these are for the most 
> part fairly isolated from existing code. The current issues with the grenade CI 
> job mean that upgrade code paths are not being exercised frequently, and could 
> lead to additional test/bug fix load on the team later in the cycle. The node 
> traits code patches are all in review [3], and are now generally getting 
> positive reviews or minor negative feedback.
> rloo and TheJulia have kindly offered to review during the FFE window.
> [1] 
> http://specs.openstack.org/openstack/ironic-specs/specs/not-implemented/node-traits.html
> [2] 
> https://review.openstack.org/#/c/504952/7/specs/approved/config-template-traits.rst
> [3] https://review.openstack.org/#/q/topic:bug/1722194+(status:open)
> Thanks,
> Mark (mgoddard)
> __________________________________________________________________________
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

More information about the OpenStack-dev mailing list