We had a hypervisor hardware failure at Birmingham at about 1530 on 23/02/2018. All eight instances were migrated from this hypervisor to functioning ones and are now running again, but will have been restarted in the process.
Apologies for any inconvenience caused.