All work has completed, and Summit has been returned to service.
Posted Jan 06, 2021 - 17:42 MST
Update
The previously-announced electrical fault at the HPCF was fixed around 4:00 PM, and we have been bringing up affected Blanca and Summit nodes. Blanca has been returned to full operation and its PM reservations released. Core components of Summit have been powered on and are working, and compute nodes are being powered on now.
Posted Jan 06, 2021 - 17:09 MST
Update
Today's planned maintenance activities have completed. However, an electrical fault occurred in the HPCF power distribution system during the Summit bringup. We have halted the Summit bringup process until a facilities electrician is able to assess and respond to the fault.
In the mean time, we intend to bring up the Blanca nodes that were affected by today's HPCF outage, as they should be unaffected by this electrical fault.
Posted Jan 06, 2021 - 15:16 MST
Update
Summit, including Summit storage, and Blanca nodes in the HPCF, have been shut down as scheduled. This supports powering off the HPCF cooling system for regular maintenance.
We will be monitoring the facilities work throughout the day, and will restore Summit and Blanca to full production as soon as possible once the work has been completed.
Posted Jan 06, 2021 - 07:19 MST
In progress
Scheduled maintenance is currently in progress. We will provide updates as necessary.
Posted Jan 06, 2021 - 07:00 MST
Scheduled
Research Computing will perform regularly-scheduled planned maintenance Wednesday, 6 January 2021. January's activities include
- HPCF tower cleaning and maintenance
Maintenance is scheduled to take place between 07:00 and 19:00, though service will be restored as soon as all activities have concluded. During the maintenance period no jobs will run on Summit resources, or Blanca resources that reside in the HPCF. This includes all "bhpc" nodes and a single Blanca GPU node. Summit storage will also be unavailable.