NIH HPC News & Announcements
* Emergency Biowulf/HPC maintenance window starting today at 3 pm *
Date: 10 March 2023 12:03:50
From: Susan Chacko
The HPC storage administrators have detected a small number (less than
25) of corrupted files on one of the storage systems, and have been
engaged in investigation and ongoing discussions with the storage
vendor. To repair the corrupted files and to reduce the risk of any
additional damage, a consistency check and upgrade to the storage
systems is needed.
Unfortunately, this requires that all Biowulf jobs be stopped while the
filesystem checks and upgrades are performed. Job scheduling on Biowulf
has been stopped as of now, and jobs that are still running in 2 hrs
will be terminated. Users with corrupted files will be contacted
individually. The Biowulf login node, Helix, Globus and hpcdrive will
also be unavailable to prevent any file changes while these operations
are performed.
We apologize for the disruption to your work, but it is vital that this
problem be resolved to prevent future problems. You can check the status
of the system at https://hpc.nih.gov/systems/status/ at any time. We
estimate that the process will take about 24 hrs, and we will keep you
informed via email.
NIH HPC Staff.
########################################################################
Please contact staff@hpc.nih.gov with any questions about the NIH HPC Systems
[Last 12 months of HPC announcements]