Maintenance FAQ

What are maintenances?

An outage where jobs are prevented from being scheduled while system-wide updates and improvements are implemented.

 

Why are maintenances important?

Supercomputers are a complex arrangement of dozens of hardware and software components. As issues arise with updates in firmware, OS patches, and software that affect security, performance, and compatibility improvements, the systems may need to be updated all at once. These maintenance outages are scheduled with Researchers in mind and during some at least one of our supercomputers will
remain available to preserve continuity.

 

How are maintenances decided?

Research Computing understands that these maintenances are impactful to our faculty's and researchers' work. However, these maintenances are critical to keeping the supercomputers operational. To mitigate the impact these maintenances have on our community, we first identify which dates we feel will cause the least impact. We then approve these maintenances through our governance board.

How are maintenances communicated?

We communicate these maintenances through the following communication channels:

  • our maintenance schedule page

  • our web portals and login prompts  

  • our #rc-support Slack channel 

  • our user mailing list with a three-week announcement and a one-week reminder

 

My work cannot wait for maintenance to complete. What options do I have?

[need description]

We understand that these measures may not alleviate all frustrations, and we welcome any suggestions from our community for further improving our processes. Your feedback is invaluable in helping us better serve our community and ensure a smoother experience for all users.

Please feel free to reach out if you have any questions or concerns. We are here to support you throughout this process.

Additional Help