Storage Problems - GPFS services (Project / Scratch)
OSC Project and Scratch file systems have resumed normal operations.
OSC Project and Scratch file systems have resumed normal operations.
Owens is experiencing a disruption of GPFS availability. At about 4:17PM today (January 6th), OSC monitoring noticed a problem with mounts of Project on the Owens supercomputer. Jobs may have been impacted. Normal service has resumed, and we are still investigating the root cause.
We are having issues with GPFS filesystem (both project and scratch) on all clusters. OnDemand is also down.
The scratch filesystem has been returned to service.
Data on /fs/scratch is accessible now.
We fixed the problem with both project and scratch filesystem and the service has been restored.
We fixed the problem with both project and scratch filesystem.
Updated on 2:30pm Feb 1st:
Scratch filesystem is back. OnDemand is also available now.
Original Post:
Scratch filesystem is down now. Users can not access OSC OnDemand as well. Please use other methods to connect to OSC systems. We are working to fix this issue now and apologize for any inconvenience. Please contact OSC Help if you have any questions.
NFS outage on Thursday, Jan 17 from 7 am to 8 am
OSC is experiencing some problems with the Project and Scratch filesystems that are resulting in some jobs seeing "stale file handles". We are investigating the problem and will provide updates as we have them. Job scheduling is currently paused on all filesystems.
We are working with vendor engineers to resolve the issue, but you can currently still log in via SSH and access your files in your home directories. OnDemand logins will likely hang, and attempts to access files on Project or Scratch will likely fail.