Dear slurm developer, I'm running SLURM on a cluster with 24 computation node. Several nodes are frequently at DOWN state. When I brought them up, they went to DOWN state again. I followed the troubleshooting guidelines on line, but still couldn't figured it out.
I went to the DOWN node, do scontrol show slurmd Active Steps = NONE Active CPUs = 8 ----------------------------- Last slurmctld msg time = NONE slurmd Debug = 3 I went to the slurmd.log file. the error message is: error: Unable to register; Zero Bytes were transmitted or received error: slurm_receive_msg: Zero Bytes were transmitted or received I also check munge, it looks working well. Would you please give me some suggestions of how to proceed to debug? I really appreciate your help. Thanks. Blessings, Binghui -- Binghui Deng, PH.D Dept. Of Material Science & Engineering Rensselaer Polytechnic Institute
