Engineers have been investigating the issue and can confirm that: As load rose on the servers handling voice client connections, the normal load-balancing procedures did not take over as expected, and the servers started to run out of memory. Engineers were able to spot the issue within minutes and restore service to normal. We have since found a weak point in the load balancing and the procedures around the maintenance of the same, steps have been taken to avoid a situation like this in the future. We apologize for the inconvenience caused.

We have investigated the issue and believe that the issue was due to a memory issue on a server, restarting the server has resolved the issue. We are monitoring the dialler closely to ensure that there is no further degraded performance.