Contractors will be be performing an extended maintenance on our chiller system from April 11th at 7:00AM to April 15th at 5:00PM. It is necessary to complete this maintenance prior to the arrival of warm weather, and scheduling is subject to contractor availability. 

The chiller will be offline for a large portion of the maintenance and so all compute resources for both the Bluecrab and Rockfish clusters (compute, bigmem, and gpu nodes) will be shut down and offline for the duration of the outage. Users will not be able to schedule jobs during this timeframe. Home directories, DTNs, and storage directories will be available, but access will be intermittent. Though users will still have some access, we recommend planning as if the system were going to be unavailable.

We will send additional reminders as we get closer. Please email help@rockfish.jhu.edu if you have any questions. You can follow twitter.com/marcc_systems for updates during the outage.