Hydra is Off-Line for Maintenance

Hydra is CURRENTLY under maintenance.

In the best possible scenario eventually an existing node will take over the role of f1n1 (hydra.tamu.edu) which had a critical hardware failure on 2012-01-01. We are doing the best that is possible given the fact that the system has neither hardware nor software support by the vendor in the last 2 years and we operate with limited resources for these types of repairs.

The next level of outcome is that the system will come back up with only three out of the four GPFS I/O nodes, but with its operating state cleaned up of any non-persistent hardware or software errors.

The least desirable scenario includes the following three possibilities:

  1. GPFS and underlying sub-systems will have to be re-installed from scratch. This may take up to a couple of weeks to accomplish.
  2. The entire cluster will have to be re-installed from scratch. This may take a much longer amount of time than the item above.
  3. Hardware is not stable enough so neither GPFS nor the cluster can be reinstalled.

Please contact the SC for more information or clarifications on this issue and consult previous announcements on our web page.

Posted on: 10:17 AM, February 2, 2012