Does it calm down if you stop the agents? If so, check the agent logs for exceptions because one may be stuck in a sync loop because it's encountering an error. If you still don't see anything, try reducing the report interval for the agents (and increase agent_down_time on the server accordingly) and see if that helps. On Feb 10, 2017 03:15, "David Riedl" <david.riedl at wingcon.com> wrote: Hello Everyone, I just upgraded my Openstack Installation from Liberty to Mitaka. It is a pretty small cluster with only 3 nodes. Since the upgrade, the neutron-server on my control node produces a very high cpu load. Unfortunately the log files throw no errors and google does not give me any answers either, so I am a bit lost. This is my neutron.conf http://pastebin.com/cQbgeP4H Please tell me if you need any log files or config files. Regards and thanks for any help David _______________________________________________ OpenStack-operators mailing list OpenStack-operators at lists.openstack.org http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20170210/d6a4bf12/attachment.html>