Cluster update: new software and known issues

The following page lists known software issues following the cluster update to SLES12 (Jan 2017), with plan for fixes and current state.

https://www.pik-potsdam.de/services/it/hpc/user-guides/hlrs-update-2017-known-issues-software

Performance of parallel filesystems degraded

Performance of the parallel file systems will be impaired for at least until early February. The reason for this is that in order to use the new physical storage blocks all data has to be migrated from the existing storage blocks. Existing storage need to be reconfigured once that's completed and all data will then be redistributed over all storage blocks. This is being done on-line but puts a significant extra load on the system.