Dear meep users, I have installed meep on a large cluster consisting of some 100 nodes of which I can use some from Torque/MIPCH. Meep installs fine and can be started on as many processors as I like, but after a while the process dies sporadic with the error p23_602: p4_error: interrupt SIGx: 15 p24_11906: (441.871094) net_send: could not write to fd=5, errno = 32
Googleing this tells me that some process can't write to some disc (https://wiki.mst.edu/numerical/nic/faq) but I have no idea of how to remedy the problem or where to start looking? Anyone have seen this problem and can give me a hint where to start? Meep is compiled with hdf5 parallelised and my PBS-script looks as # Arguments to qsub can be submitted via the script as well by starting # the line with #PBS # Set your mail address #PBS -M [EMAIL PROTECTED] # # Mail on abort #PBS -m a # # Specify time for job #PBS -l walltime=00:02:00 # # Request 1 processor (node) #PBS -l nodes=2:ppn=4 # #PBS -q ada # #PBS -N bend_0406 # #PBS -v GUILE_WARN_DEPRECATED=no # #PBS -A Physics # # End of arguments to qsub #Preparation work cd ~/job/bended_pc_x_guided_new common="N=10 mtrl=0 r=0.025 a=0.5 fcen=1.5 df=2 res=192 save_field=0 save_eps=0 comp=Ey fiber_res=0.05 air_frame1=10" # Go! mpiexec -verbose /c3se/users/e9ravn/opt/bin/meep-mpi dx=0.00 with_pc=0 $common main.ctl #End of script (make sure line before this gets run) Regards, Robert -- Robert Rehammar PhD-Student Applied Physics, Chalmers University of Technology Department of Physics, Göteborg University SE-421 96 Göteborg Sweden Tel +46 (0)31 772 3156 Fax +46 (0)31 416 984 Cel +46 (0)738 328834 Web fy.chalmers.se/~e9ravn _______________________________________________ meep-discuss mailing list meep-discuss@ab-initio.mit.edu http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss