Status of System

Status of System

Lise in NHR@ZIB

timebar.png

Aug 13, 2025 GPU partitions OS upgrade to Rocky Linux 9

Our GPU-based "A100" and "PVC" partitions will undergo a major operating system (OS) upgrade from Rocky Linux 8 to Rocky Linux 9. To facilitate this upgrade, the GPU login nodes and GPU compute nodes will be drained and rebooted on that day. (Note: The CPU partitions "CLX" and "Genoa" are not affected.)

Because this is a major upgrade, some of the new OS libraries may cause incompatibilities with software compiled under the previous OS, so users may need to rebuild or relink their software. To ensure a smooth transition, we encourage all users of our GPU partitions to verify in advance that their application software stacks and workflows are compatible with the new OS. For this purpose, the GPU login nodes bgnlogin2.nhr.zib.de and bgilogin2.nhr.zib.de are already available under the new OS, along with an updated set of environment modules. Additionally, GPU A100 test jobs can be submitted to the temporary gpu-a100:el9 SLURM partition, which also provides the new OS environment.

The SLURM partition names for the GPU nodes will remain unchanged after the OS update.

If you have further questions or software requests, please contact the NHR@ZIB support team (support@nhr.zib.de).

 

2025-07-11 Maintenance of PERM

There will be a short maintenance of the PERM filesystem starting Friday at 10am. PERM will be unavailable.

 

2025-07-04 power outage

There was a power outage last night. Recovery took a while, but the system is back online. Please report any issues.

 

2025-07-02 gpu-pvc power save mode

We have enabled power saving for the gpu-pvc partition. Unused nodes are powered off to save energy. Thus, you may need to wait a few minutes for nodes to start up at the beginning of your job. Please report any problems.

 

2025-07-02 cpu-clx updated to Rocky 9.6

cpu-clx has been rebooted into new Rocky 9.6 images. Please report any problems.

 

2025-06-30 bgnlogin2 is a Rocky 9 test node

We plan to update the gpu-a100 nodes to Rocky 9. bgnlogin2 is already running Rocky 9 for tests.

 

2025-06-26 cpu-genoa updated to Rocky 9.6

Today, cpu-genoa has been rebooted into new Rocky 9.6 images. Please report any problems to support@nhr.zib.de