Scheduled Bell Maintenance
Link to update at October 6, 2022 6:26pm EDT UPDATE:
As of 6:26pm EDT, engineers have completed the maintenance and have returned the Bell cluster back to normal service. All queues have been enabled and jobs have resumed scheduling.
Please note that Bell continues to operate in degraded capacity due to power and cooling issues. You may experience longer than usual scheduling delays. Please rest assured that we are fixing and replacing affected nodes as soon as hardware becomes available. We appreciate your patience and understanding in this situation.
Please report any issues to rcac-help@purdue.edu
Link to update at October 6, 2022 12:15pm EDT UPDATE:
Engineers continue bringing Bell back to normal operation. We anticipating everything to be completed today and will provide another update by 10pm or sooner.
Link to update at October 5, 2022 10:47pm EDT UPDATE:
Work continues on completing all maintenance tasks and bringing Bell back to normal operation. Bell login nodes and services are back and available for users, but scheduling remains paused.
We will provide another update by noon tomorrow, October 6th, 2022.
Link to original posting ORIGINAL:
The Bell cluster will be unavailable for use October 4, 2022 8:00am - October 5, 2022 11:59pm EDT while we perform scheduled maintenance to expand Bell's scratch storage capabilities and make upgrades to Bell's AMD GPU drivers and other system software. This maintenance will also allow work to be done on the cooling systems for the upcoming deployment of the Negishi cluster.
Any SLURM job which requests walltime extending past Tuesday, October 4th, 2022 at 8:00am EDT will not begin until after maintenance concludes.