The same storage system that gave us problems on Wednesday hung again Friday evening. This time we did not lose any nodes. We know what is causing the issue now and are working with the vendor on a software fix. Running jobs had time added to compensate for the hang.