RESOLVED - All nodes affected during maintenance are back into service. Some nodes in the HPCC are still offline, but this is due to typical hardware issues and is in line with usual compute node availability.

After yesterday’s maintenance, some compute nodes are not available for job scheduling. This is primarily affecting nodes with GPUs. Jobs requesting GPUs will likely queue for longer than usual or, depending on constraints, will stay in the queue until nodes are returned to service.

Our system administrators are working on this and will return nodes to service as soon as possible. Updates will be posted here.