We are getting regular refresh failures with the following message content: Last refresh failed: Tue Oct 10 2017 16:01:43 GMT+0100 (GMT Daylight Time) Hide details Cluster URI: WABI-NORTH-E...
Still have it from time to time. I think it helps to understand your network and make sure jobs don't clash. Upgrading the server to the recommended spec helped performance.