On 8/9/2018 1:30 PM, Graham Hayes wrote:
You could just send the deprecation notice, and we could deprecate designate-sink if no one came forward to update it - that seems fairer to push the burden on to the people who actually use the feature, not other teams maintaining legacy stuff. Does that seem overly harsh?
It's harsh depending on my mood the day you ask me. :) Nova has done long-running deprecations for things that we know we don't want people building on, like nova-network and cells v1. And then we've left those around for a long time while we work on whatever it takes to eventually drop them. I think we could do the same here, and designate-sink could log a warning once on startup or something that says, "This service relies on the legacy nova notification format which is deprecated and being replaced with the versioned notification format. Removal is TBD but if you are dependent on this service/feature, we encourage you to help work on the transition for designate-sink to use nova versioned notifications." I would hold off on doing that until after we've actually agreed to deprecate the legacy notifications at the PTG.
-- Thanks, Matt __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
