Upd:Linux-Cluster: All partitions available again

LRZ aktuell publish at lrz.de
Do Mai 3 17:37:04 CEST 2012

 Changes to this message: Update
Update: The cluster and visualization systems are available again. In
 the Myrinet segment, batch processing with SLURM should now work.
 Please note that jobs using more than 32 cores should be submitted to
 the Infiniband MPP cluster. The login nodes lx64ia2 and lx64ia3 are
 available for submission of SGE jobs; however Parastation MPI will not
 run on lx64ia2 any more. Please login to the MPP cluster and build and
 run your programs from there as indicated in the cluster docs.
 Dear users of the Cluster systems at LRZ,
 as announced in http://www.lrz.de/aktuell/ali4261/ user operation on
 the cluster will be interrupted due to an electrical power shutdown
 test which affects all storage systems.
 In order to cleanly shut down all accesses to the storage systems, we
 have already started to drain the SLURM queues and all running jobs
 will be removed from 4:30 pm onwards.
 Please also note that after this interruption, the SGE queues mpi_8,
 shm_8, mpi_32 and shm_32 will become permanently unavailable because
 the systems will be upgraded and moved to SLURM. The SLURM
 documentation will be appropriately updated as soon as this upgrade is
 We will keep you up-to-date about the status of the cluster via this

 This information is also available on our web server

 Reinhold Bader

Mehr Informationen über die Mailingliste aktuell