Sisu and Taito: Unplanned service interruption in /wrk file system 2019-04-16 14:25 - 15:20
Dear Sisu and Taito users, There was an unplanned service interruption in the parallel file system of Sisu and Taito which caused file system operations to /wrk hang between 14:25 - 15:20. This caused eight GPU and seven CPU nodes to crash and jobs that were running on those nodes failed. Please check you jobs and re-submit as needed. The acute problem has been solved and we are analyzing the root cause. We are sorry for any inconvenience this problem may have caused.