A number of users of our self-managed “DHCP” subnet experienced problems accessing the network starting some time on Saturday 20th up until 14:45 on Monday 22nd. We believe that this was caused by a rogue DHCP server, and that disabling its network access restored the service for everyone else.
Please be careful when adding devices to the network. In most cases these will cause no problem, but depending on how they are configured they do bring the possibility of disrupting the service for everyone else on the subnet. In particular, please check your settings when bringing in devices from a home environment, as what may be valid (indeed required) there may be completely inappropriate in our larger-scale situation.
We do use the network switches’ built-in features to protect the network and its users where possible. Unfortunately not all our switches have as much capability in this regard as we would like. As part of our rolling upgrade policy they will be replaced in due course by more capable switches, but given budgetary constraints it is likely to be a couple of years before this is complete.