Description of problem: Cinder quotas break refresh when registering an Openstack cloud provider using Keystone v3 domains [----] E, [2016-06-01T05:09:35.453581 #25175:1121988] ERROR -- : <Fog> excon.error #<Excon::Errors::InternalServerError: Expected(200) <=> Actual(500 InternalServerError) excon.error.response :body => "{\"computeFault\": {\"message\": \"The server has either erred or is incapable of performing the requested operation.\", \"code\": 500}}" :cookies => [ ] :headers => { "Content-Length" => "128" "Content-Type" => "application/json; charset=UTF-8" "Date" => "Wed, 01 Jun 2016 09:09:35 GMT" "X-Compute-Request-Id" => "req-2de8c3a5-de00-4ec5-8ede-0549a50441da" "X-Openstack-Request-Id" => "req-2de8c3a5-de00-4ec5-8ede-0549a50441da" } :host => "172.16.18.25" :local_address => "172.16.17.143" :local_port => 57915 :path => "/v2/98f1116da56f487db1ea1ff156743c8e//os-quota-sets/98f1116da56f487db1ea1ff156743c8e" :port => 8776 :reason_phrase => "Internal Server Error" :remote_ip => "172.16.18.25" :status => 500 :status_line => "HTTP/1.1 500 Internal Server Error\r\n" > Version-Release number of selected component (if applicable): 5.6.0.8-rc1-nightly.20160531084619_1bff101 How reproducible: 100% Steps to Reproduce: 1. Register Openstack cloud provider by using a Keystone v3 non default domain and the admin user of the domain 2. Refresh cloud provider Actual results: Refresh fails Expected results: Refresh completes fine.
This is an actual OpenStack BUG: https://bugs.launchpad.net/cinder/+bug/1544888 So fix for this will be getting closer to 'ignore all OpenStack errors'
To workaround this Cinder quota issue I had to edit on the controller node, /etc/cinder/cinder.conf, keystone_authtoken section: auth_uri = http://10.0.0.11:5000/v2.0 to auth_uri = http://10.0.0.11:5000/v3