On 10/16/2016 11:52 AM, Duncan Thomas wrote: > On 14 October 2016 at 23:55, Jay Pipes <jaypi...@gmail.com > <mailto:jaypi...@gmail.com>> wrote: > > The primary thing that, to me at least, differentiates rolling > upgrades of distributed software is that different nodes can > contain multiple versions of the software and continue to > communicate with other nodes in the system without issue. > > In the case of Glance, you cannot have different versions of the > Glance service running simultaneously within an environment, > because those Glance services each directly interface with the > Glance database and therefore expect the Glance DB schema to look > a particular way for a specific version of the Glance service > software. > > > Cinder services can run N+-1 versions in a mixed manner, all talking > to the same database, no conductor required.
That's true, keeping DB schema compatible is possible even if all the services are talking to the DB, but this requires very careful programming (and reviewing) of the DB migrations. That's why our DB migrations guidelines are quite complicated [1]. I think conductor-based design is superior but definitely isn't the only way. [1] http://docs.openstack.org/developer/cinder/devref/rolling.upgrades.html#database-schema-and-data-migrations __________________________________________________________________________ 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
