GT Home : : Campus Maps : : GT Directory

Author Archive

Infiniband Interruption

Posted by on Monday, 21 March, 2011

Approximately 10:15am this morning, we received a trouble alert with the Infiniband fabric. We found the fabric subnet manager (SM) unresponsive and restarting the system proved ineffective. A backup system was brought online and a secondary SM brought online as well.

Service was restored by 11:15. A secondary system with an auto-failover SM is now running to protect against this outage. A service request has been logged with the vendor. As always, please let us know if you observe any lingering difficulties.