We already run unit tests only using real Postgresql. But this still doesn't answer the question how we should test migrations. On Fri, Mar 6, 2015 at 5:24 PM, Boris Bobrov <bbobrov at mirantis.com> wrote: > On Friday 06 March 2015 16:57:19 Nikolay Markov wrote: > > Hi everybody, > > > > From time to time some bugs appear regarding failed database migrations > > during upgrade and we have High-priority bug for 6.1 ( > > https://bugs.launchpad.net/fuel/+bug/1391553) on testing this migration > > process. I want to start a thread for discussing how we're going to do > it. > > > > I don't see any obvious solution, but we can at least start adding tests > > together with any changes in migrations, which will use a number of > various > > fake environments upgrading and downgrading DB. > > > > Any thoughts? > > In Kyestone adding unit tests and running them in in-memory sqlite was > proven > ineffective.The only solution we've come to is to run all db-related tests > against real rdbmses. > > -- > Best regards, > Boris Bobrov > > __________________________________________________________________________ > 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 > -- Best regards, Nick Markov -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150306/f527ccc4/attachment.html>