Updates


Event Date Summary
Today's update included a surprise outage of cooling water, so some additional user-visible hardware (VDI nodes, some login nodes) has been shut down to reduce heat generation.
There will be a power outage for Graham on Saturday March 20 starting at 7:20am until 5:30pm. This affects all compute nodes, but not cloud, storage, networking and other infrastructure. The scheduler will only start jobs that are expected to end before the outage. Note that as previously mentioned, Graham's scratch filesystem will be erased (all files will be deleted!) and updated at this time.

Incident description

System Incident status Start Date End Date
Graham Closed
Created by Mark Hahn on

Title


/scratch Planned Outage - Arrêt planifié


Summary


The update to scratch and the power outage has completed. During the outage, the water cooling loop was unexpectedly cut, forcing us to shut down many additional systems. Please report any issues you observe via the ticket system.


Updated by Mark Hahn on