Data Depot Filesystem issue: Scheduling Resumed
Link to update at February 13, 2026 10:00am EST UPDATE:
We have completed the needed steps to return all clusters to service and resume job scheduling. Please note that some users may still encounter read errors for impacted segments of Data Depot.
If you experience these errors, please fill out the following form to ensure you get assistance as quickly as possible: Data Depot Assistance Form
You may also reach out to rcac-help@purdue.edu for additional support.
Link to update at February 13, 2026 8:52am EST UPDATE:
Dear Depot users,
Yesterday, during routine vendor-driven replacement of a hard drive, there was a failure of a RAID “adapter group” within the Depot filesystem, leading to a small section of the underlying storage no longer being accessible. The net result is that parts of files that happen to be stored on these storage targets within the Depot filesystem are unavailable, and any attempts to read those will return error messages. New writes to the filesystem continue to function without issue.
After consultation with vendor engineers, and component manufacturers Dell and Seagate, it has been determined that the RAID adapter group is not recoverable. Any files with data on those storage targets will need to be restored from our disaster recovery copy.
To enable that restore, vendor engineers are currently generating a list of the affected files. We anticipate this process to run into Friday, Feb. 13, or longer due to the large total number of files on the filesystem.
What this means to you if you encounter a read error:
• If you utilize Fortress as part of a best-practices data management strategy, the quickest way to move forward is to refresh affected files from your Fortress space into your Depot.
• If your data management process does not include Fortress, you can fill out the form at https://forms.office.com/r/tiA2SawNC2 and we will refresh the file from the disaster recovery copy.
• The Disaster Recovery copies are generated regularly and will have a copy of the data that is no more than one month old. Data written to the Depot within the last month may not have been backed up yet.
• While we do utilize snapshots on the Data Depot, they cannot protect against this particular failure mode of the underlying storage system, so we cannot easily recover data from the snapshots.
Otherwise, we will replace files behind the scenes from the disaster recovery copy once the file list is generated.
Finally, this issue is unrelated to the now complete migration to the new, faster storage system and that the overall platform and research data held on it is not at risk.
Thank you for your understanding and continued partnership,
Link to update at February 12, 2026 10:27am EST UPDATE:
We are still working with the vendor to restore the affected portion of the Data Depot filesystem, and the situation remains unchanged since yesterday
Impact to you
Scheduling remains paused, and you may continue to see error messages in some cases when you try to read data.
We will provide our next update when have new information to share.
Link to update at February 11, 2026 3:45pm EST UPDATE:
We are continuing to work on restoring the affected Data Depot filesystem and expect this work to extend into tomorrow.
Impact to you Scheduling remains paused, and you may still see error messages in some cases when you try to read data.
We will provide our next update by 11:00 a.m. tomorrow.
Link to original posting ORIGINAL:
An internal portion of the Data Depot filesystem is currently offline, as a result, all scheduling has been paused until this issue is resolved.
Impact to you Attempts to read files that are on the affected storage may result in error messages
Our IT team is actively working with the vendor to restore service as quickly as possible. We will send an update as soon as more information is available.