There will be a scheduled downtime of all the HPC systems of NHR@FAU
starting at 06:30 on Tuesday, June 9.
We expect normal operation to resume by the evening of Wednesday, June 10.
As usual, Jobs that would collide with the downtime will automatically be postponed until the downtime is over. Frontends and most fileservers will be available through most of the downtime with some short interruptions. The exception is the workspaces under /hnvme on Helma, which will NOT be available throughout the downtime.
Reason for the downtime is various maintenance work, including but not limited to Firmware and Software updates on central servers, and welding work on the cooling water circuit for the Fritz and Helma clusters.
We will keep this post updated with progress reports.
Updates
Update 2026-06-01: csnhr and tinyx are now more strict and only accept SSH keys uploaded to the HPC portal, i.e. they follow the policy of the more recent clusters and ignore keys that you put into ~/.ssh/authorized_keys.
Update 2026-06-01 12:30: Batch processing on Woody has been resumed at 11:35
Update 2026-06-10 12:30: Batch processing on Alex has been resumed at 12:25
Update 2026-06-10 12:30: Batch processing on TinyGPU&TinyFAT has been resumed at 12:25
Update 2026-06-10 14:15: Batch processing on Testcluster has been resumed at 14:10
Update 2026-06-10 14:30: Cooling water for Fritz and Helma is back, and we’ll now start to bring them back online
Update 2026-06-10 16:20: Batch processing on Fritz has been resumed at 16:00
Update 2026-06-10 17:45: Batch processing on Helma-GPU has been resumed around 16:45, and batch processing on Helma-CPU has been resumed at 17:45.
Update 2026-06-10 17:50: We are mostly done. While we still have some minor cleanup to do, things should work normally again.