Springe direkt zu Inhalt

sheldon cluster down due to maintenance

sheldon is running again.

News from Mar 11, 2024

The cluster sheldon and all its corresponding systems as well as the storage system data05 will be switched off on 11.3. at 20:00 Uhr in preperation of ZIB's maintenance day. After we have been informed of its end we will restart the systems.

Update (13.03.): While switching sheldon back on we noticed that one of our InfiniBand switches is not talking to us. Unfortuantely it is needed for the proper operation of the scratch filesystem. We will investigate this further tomorrow, when we have physical access to the switch. data05 continues to be offline because we are moving it to a new system, which is unrelated and has been planned beforehand.

Update 2 (13.03.): data05 is available again.

Update (14.03.): sheldon is running again. We are still trying to get some nodes back on, which didn't like the reboot, but we will bring them back online as we fix them.

3 / 60