On 14th August, a multiple hardware failure in one of the core network routers resulted in extended and widespread network and Internet access issues. This affected services across the whole of the University, primarily affecting the Whiteknights and London Road campuses. Senior staff from IT Services and our network support partner were on site until midnight diagnosing and replacing the failed hardware.
The likelihood of a repeat incident of this kind is very small; however as due diligence, IT Services will be asking for the other core network router to be checked to see if the components are from the same manufacturing batch, and pre-emptively replaced if required.
IT Services apologises to all the service users for the inconvenience caused by this failure yesterday.
Initial analysis shows that a faulty batch of capacitors were used in the manufacture of the chassis backplane and two switch fabric cards within the router, and over time these have degraded and are no longer performing to within tolerance. The equipment manufacturer are considering whether they wish to analyse the failed cards and backplane in their hardware engineering lab in order to ensure the effect these failed components had on the equipment is fully understood.