We are aware of issues affecting connectivity to certain servers on our backbone currently.
Engineers are working on diagnosing the problem which started at approx 15:20 today.
Whilst we initially believed the issue was resolved by 16:30 we are still seeing issues from certain hosts and are continuing to investigate the problem in conjunction with Juniper
We have identified a rogue switch which appears to be polluting the CAM / Forwarding tables and we have effected a replacement of the faulty hardware
Services will have flapped during the replacement and also while the switch was active
Please accept our apologies for the disruption caused by this unusual hardware failure. It was of a nature which was not easy to identify initially and once identified required replacement components to be sourced and express delivered to the Docklands