This is a reminder that all of the HTC cluster is unavailable until Wed evening due to the scheduled extended maintenance. Furthermore, the H2P cluster will be undergoing maintenance this Tues. Jobs that show "(ReqNodeNotAvail, Un" as outputs to the crc-squeue.pl command is due to a system-wide reservation of all compute nodes for maintenance. Upon completion of maintenance, your jobs should start running.
Posted: August 22, 2018
Our quarterly maintenance for the MPI, SMP, and GPU clusters is scheduled for Tues Sept 11 from 8:30AM to 6PM. Users will not be able to access CRC compute and storage resources during this time.
Also, there will be an extended maintenance for the HTC cluster from Mon Sept 10 to Wed Sept 12. This extended schedule is needed in order to update the Mobydisk storage array, update the compute node images, and rebuild user applications. Users will not have access to HTC compute and storage resources during this time.
Please plan your computing needs accordingly. As always, questions and issues can be submitted to our ticketing system, https://crc.pitt.edu/tickets.