RRStor File System Issue Resolved

Hi all,

Regarding this recent post, the issue that developed on our main file system server in RRStor has been resolved. We are working with our support vendor to understand the details, and an abridged report will be shared as an edit to this post once we have it.

No connections were interrupted and we have seen no jobs terminate unexpectedly as a result of this issue.

As always, please communicate any issues/questions to the Research Computing RT Queue (hpcf.umbc.edu > User Support > Request Help).

Thanks,
Roy Prouty
Assistant Director for Research Computing, UMBC DoIT

UPDATE

The reported performance issue was caused by a vendor-supplied update that made an unexpected update to the network file system (NFS) configuration. The update caused the NFS service to restart and the update was invalid. The performance issue was a direct result of the service being unable to restart cleanly due to the invalid unexpected update. Once RCD sytsem administrators determined the cause and removed the update, the NFS service was able to cleanly restart and performance returned to normal.