error message while running job parallely
 

Dear friends,

I have a 3d problem with 80,000 meshes generated. It has no problem to run under one PC or several PCs running interactively. However, as long as I ran the job parallely on background. I always got the error message as following:

> Halting due to end of file on input. node 0 retrying on zero socket read..... node 0 retrying on zero socket read.....

0 (mpsystem.c@1228): mpt_read: failed: errno = 0

0: mpt_read: error: read failed trying to read 8 bytes: Success [bt] Execution path: [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(Process_Stackframe+0x14) [0x8719086] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(mpt_error+0x9d) [0x86f45f9] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(mpt_read+0xd7) [0x86f74cb] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(mpt_tcpip_crecv_raw+0x30) [0x86f8f58] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(mpt_tcpip_crecv_all+0x27) [0x86fb103] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(MPT_crecv_char+0xd5) [0x86e4c49] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(PRF_Command_Start+0xbe) [0x81b116e] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(PRF_Node_repl+0x5f) [0x81b4fbf] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(main+0xde) [0x8621cfe] [bt] /lib/i686/libc.so.6(__libc_start_main+0xa1) [0x556a7261] [bt] /opt/fluent/Fluent.Inc/fluent6.2.16/lnx86/3ddp_node/fluent_nmpi.6.2.16(sinh+0x6d) [0x8090101] P4 procgroup file is /tmp/p4pg-lzhan003-31310-orville-1-3.local.

-driver: Command not found.
:

I tried to regenerate the grid but the same error occured. Do you know what's the problem?

Thanks,

LYZ