Update 2026-05-21: The 2000A power rail was successfully rebuilt, electrically checked/measured and has been re-powered.
Due to the recent security flaws, currently the OS images are being renewed (takes time for testing) – after that we will begin bringing up the cluster.

_____________________________

This electrical take-down is confined to the “normal power” only, ie. affecting only the non-UPS circuits.

We expect the maintenance to last the whole Wednesday. Therefore, we plan to bring the cluster back online only the following day.

Your batch jobs

You do not need to do anything with respect to your (running or pending) batch jobs.

The scheduler knows about the downtime and will

• start pending jobs only if these will be safely finished before the downtime
  and
• hold all others until after the downtime and recommencing of the normal scheduling.

Thus, unusual long pending times are to be expected (particularly in the “long” queue/partition), even though more and more compute nodes become idle.

Details

Due to the power consumption of the next expansion stage NHR1 of Lichtenberg II, the main power rail needs to be extended in the midst, and two new large connection trays need to be mounted on this new rail part.

To insert this new rail section, the existing 2000A power rail needs to be splitted at the right position, and the outer part needs to be moved away in its remaining length.