CHPC DOWNTIME: CHPC Network Upgrade, Dec 15, 2020 starting at 8am until early hours on Dec 16th - Center for High Performance Computing

Update December 16th:
The network upgrade yesterday ran into issues that required additional time for CHPC staff to work with the switch vendor to resolve.

Access to CHPC systems was restored in stages late yesterday evening and into the early hours this morning. Reservations on the clusters were released and jobs are now running as expected.

Please report any remaining issues by submitting a ticket at helpdesk@chpc.utah.edu

Original Post on November 30th, 2020:

CHPC DOWNTIME - December 15, 2020 all day

CHPC will be having a downtime to replace (upgrade) the CHPC core network switch in the Downtown Data Center on December 15th, starting at 8am. The switches are being replaced with new hardware obtained via CARES act funding. The new hardware will both provide the capability of increased network speeds as well as adding redundancy at this core switch.

During this downtime there will be NO ACCESS to any CHPC resource. This includes resources in both the general and the protected environment, and includes the compute clusters, the Windows servers, stand-alone servers, all storage, all virtual machines, and hosted websites.

There are reservations in place on all of the clusters to empty the slurm batch queue of running jobs by 7:30am on Dec 15th.

If all goes as anticipated, the resources in the general environment will start to become accessible in early afternoon, starting with the storage and VMs, and then to the clusters interactive nodes and the Windows server Beehive.

Access to the resources in the protected environment will be restored late afternoon.

Please log out of all CHPC resources before the start of the downtime (including closing FastX sessions). For users on the Windows servers, please save your work before the start of the downtime.

During the downtime, we will send updates as CHPC resources become accessible.