Following an internal network problem this morning (Friday 16th March) many services at the RAL Tier1 are back in operation. We still have one network stack giving problems that means that a batch of disk servers and worker nodes are not available. This is still affecting the Atlas, CMS and LHCb Castor instances as there are a significant number of disk servers not available for each of these instances. The Castor 'GEN' instance is unaffected by this and is now running OK (this is Alice and the non-LHC VOs.).
We are working to resolve this final part of the problem and get the remaining Castor services back along with restarting batch services.