This issue was caused by a PSU (power supply) fault to a core router, along with a related hardware fault simultaneously, preventing normal failover behaviour, and leaving core routing in an abnormal state.
The core routers at fault are due for replacement in the next few weeks, and our hosting provider will be assessing this incident carefully, to reduce the likelihood of something similar occurring again.
Although the affected equipment is currently 'at risk' because the failed PSU has not yet been replaced, we're marking the incident as resolved as we don't expect any further problems.
Sorry for the issue this afternoon, and thanks for your patience whilst we fixed it.
Posted Sep 27, 2016 - 13:31 BST
Monitoring
Service is restored, the hosting provider will continue to investigate and resolve the issue. Sorry for the problems this afternoon.
Posted Sep 27, 2016 - 12:59 BST
Update
The affected equipment is currently being looked at. We'll update here again within 30 minutes or when we have more news for you. Thanks for your patience.
Posted Sep 27, 2016 - 12:54 BST
Identified
Our network provider has confirmed that they are suffering a core router failure which is affecting us. We'll update here when we have more news or in 30 minutes, whichever comes sooner.