Batch scheduling enabled again on the Biowulf cluster

HPC Users,

HPC staff have been monitoring the internal network and storage systems 
throughout the day and have resumed the scheduling of batch jobs on the 
Biowulf cluster.

The networking issues impacted access to most HPC filesystems and also 
disrupted communication between the batch scheduling software and the 
compute nodes. Therefore, most running jobs either failed or were 
terminated by the scheduler. Please check your outputs and re-submit any 
jobs that failed or were killed.

We would like to again acknowledge how seriously we take this event, and 
are working with our HPC network equipment vendors to identify the root 
cause of yesterday's failure.

As always, please let us know if you notice any problems.

########################################################################

Please contact staff@hpc.nih.gov with any questions about the NIH HPC Systems