There will be a scheduled downtime of all RRZE HPC systems this week, currently scheduled for Wednesday, 22.09. starting at 06:00. As usual, jobs that would collide with the downtime will be postponed until it is over. Cluster frontends and fileservers will also be down shortly, but available most of the time.
In case you’re interested: the reason is welding work for connecting our next cluster to the cooling water supply, which means the cooling water for the whole building will have to be turned off. We will therefore shut down all cluster nodes to allow the rest of the computing center to keep on running. We will also take this opportunity to do necessary maintenance work on the cluster fileservers and frontends.
Update Sept 22, 18:00 – TinyBlue is back in business. The other clusters (Woody, Cluster32) won’t be back before tomorrow (Sept 23).
Update Sept 23, 18:00 – Woody and Cluster32 are back in business.