This page is intended to give a short overview on known issues and to provide potential solutions and workarounds to the issues seen.

Last update: 2021-01-12

To stay informed, please refer to the News page. Also, please pay attention to the information contained in the "Message of the day" displayed when logging onto the system.

Detected HW and node issues

CM nodes

• dp-cn25: FW issues (#2495)

DAM nodes

• dp-dam02: node currently reserved for special use case (#2554)
• dp-dam03: node currently reserved for special use case (#2242)

ESB nodes

• dp-esb[01-25]: currently not avialable due to Fabri3 installation
• dp-esb73: GPU ECC error (#2523)
• dp-esb75: node currently reserved for special use case (#2568)

SDV nodes

• several nodes have been taken offline:
  • deeper-sdv[11-16]
  • deeper storage system: (deeper-fs[01-03], deeper-raids)
• deeper-sdv[01-10]: currently not available: configuration change needed
• knl01: NVMe issues (#2011)

Software issues

SLURM jobs

• due to introduction of accounting with the start of the early access program there is some re-configuration of user accounts needed within SLURM to assign the correct QOS levels and priorities for the jobs
• this might lead to (temporary) failing job starts for certain users
• if you cannot start jobs via SLURM, please write an email to the support list: sup(at)deep-est.eu

GPU direct usage with IB on ESB

• only available via Developer stage, for testing load:

module --force purge
module use $OTHERSTAGES
module load Stages/Devel-2019a
module load GCC/8.3.0 or module load Intel
module load ParaStationMPI

• use PSP_CUDA=1 and PSP_UCP=1

GPU direct usage with Extoll on DAM

• new Extoll driver for GPU direct over Extoll still shows low performance on the DAM nodes
• available via Developer stage, for testing load:

module --force purge
module use $OTHERSTAGES
module load Stages/Devel-2019a
module load GCC/8.3.0 or module load Intel
module load ParaStationMPI

• expect performance and stability issues