Cluster down

by **Admin** Wed Aug 13, 2014 10:56 am

StarExec is down today unexpectedly due to a power outage. We had been told it would be a brief outage, and so we took down all2.q since those nodes are not on a UPS. We expected n033 to n192 to stay up thank to their UPS. Unfortunately, there was some miscommunication from the data center folks to our IT folks, and the outage is 4 hours, not 2 minutes. So the whole cluster is down now, and will be until the afternoon.

Hopefully this will not inconvenience anyone too much. For running jobs, you may just have to re-run a few job pairs that might have gotten lost when the cluster was taken down. This can be done from the details page for the job pair, or via StarExecCommand.

Sorry for the inconvenience and the short notice (we just found out about this minutes ago),
Aaron