CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   error message in parallel (https://www.cfd-online.com/Forums/fluent/63701-error-message-parallel.html)

cedm237 April 16, 2009 13:14

error message in parallel
 
I run fluent 6.3.26 in parallelm I have 2 pc [2*4core] with 2*8Go Ramm with mpich2 and I hqve this error [you know thanks for help...] my OS xp 64 bits SP2 on each pc [firewall unactivate] switch gigabit ethernet

Checking the status of SMPD for MPICH2 on the local machine ... '\\NT-1BEE049B35EB\codedol20C'

job aborted:
rank: node: exit code[: error message]
0: NT-1BEE049B35EB: 13
1: NT-1BEE049B35EB: 13
2: NT-1BEE049B35EB: 13: Fatal error in MPI_Barrier: Invalid buffer pointer, error stack:
MPI_Barrier(406)....................: MPI_Barrier(MPI_COMM_WORLD) failed
MPIR_Barrier(76)....................:
MPIC_Sendrecv(142)..................:
MPID_Isend(97)......................: failure occurred while attempting to send an eager message
MPIDI_CH3_iSend(152)................:
MPIDI_CH3I_VC_post_connect(540).....:
MPIDU_Sock_get_conninfo_from_bc(232): no space for the host description
(unknown)(): Invalid buffer pointer
3: NT-1BEE049B35EB: 13: Fatal error in MPI_Barrier: Invalid buffer pointer, error stack:
MPI_Barrier(406)....................: MPI_Barrier(MPI_COMM_WORLD) failed
MPIR_Barrier(76)....................:
MPIC_Sendrecv(142)..................:
MPID_Isend(97)......................: failure occurred while attempting to send an eager message
MPIDI_CH3_iSend(152)................:
MPIDI_CH3I_VC_post_connect(540).....:
MPIDU_Sock_get_conninfo_from_bc(232): no space for the host description
(unknown)(): Invalid buffer pointer
4: NT-B38516325C1E: 13
5: NT-B38516325C1E: 13
6: NT-B38516325C1E: 13
7: NT-B38516325C1E: 13

Interrupting...
Interrupting client...
The Parallel FLUENT process could not be started.
Done.

cedm237 April 17, 2009 07:05

I have partially resolve my problem with the environnement variable:MPICH2_CHANNEL

but now, I have a new error message, I use the model discrete ordinate and I change the DO by scheme to accelerate the calcul.
After 3 ou 4 changing of DO --> BOOM ERROR

ex: do 1 1 1 1--->ok
do 2 2 2 2--->ok
do 3 3 3 3--->ok
do 4 4 4 4 --->ok
do 5 5 5 5 ---> error

or other example:
do 2 2 2 2 --->ok
do 4 4 4 4--->ok
do 6 6 6 6--->ok
do 8 8 8 8 --->ok
do 10 10 10 10 ---> error

example for the error message:
170 2.0357e-03 6.7638e-05 6.3893e-05 1.2912e-04 2.6070e-07 2.5162e-05 0:38:56 10
Library "\\NT-1BEE049B35EB\Fluent.Inc\custom\Shared\LIBUDF\win64 \3d_node\libudf.dll" opened
999999 (..\..\src\mpsystem.c@1123): mpt_read: failed: errno = 10054

999999: mpt_read: error: read failed trying to read 4 bytes: No such file or directory

job aborted:
rank: node: exit code[: error message]
0: NT-1BEE049B35EB: 123
1: NT-1BEE049B35EB: -1073741819: process 1 exited without calling finalize
2: NT-1BEE049B35EB: 123
3: NT-1BEE049B35EB: 123
4: NT-B38516325C1E: 123
5: NT-B38516325C1E: 123
6: NT-B38516325C1E: 123
7: NT-B38516325C1E: 123
The Parallel FLUENT process could not be started.


All times are GMT -4. The time now is 05:58.