| Line | |
|---|
| 1 | |
|---|
| 2 | ####################################### |
|---|
| 3 | EXECUTION of : time mpirun -hostfile hosts -rankfile rankfile -np 71 ./script_lmdz.x.ksh : -np 360 ./script_opa.xx.ksh : -np 12 ./script_xios.x.ksh |
|---|
| 4 | |
|---|
| 5 | -------------------------------------------------------------------------- |
|---|
| 6 | The OpenFabrics stack has reported a network error event. Open MPI |
|---|
| 7 | will try to continue, but your job may end up failing. |
|---|
| 8 | |
|---|
| 9 | Local host: curie4677 |
|---|
| 10 | MPI process PID: 101920 |
|---|
| 11 | Error number: 3 (IBV_EVENT_QP_ACCESS_ERR) |
|---|
| 12 | |
|---|
| 13 | This error may indicate connectivity problems within the fabric; |
|---|
| 14 | please contact your system administrator. |
|---|
| 15 | -------------------------------------------------------------------------- |
|---|
| 16 | -------------------------------------------------------------------------- |
|---|
| 17 | mpirun noticed that process rank 338 with PID 101897 on node curie4677 exited on signal 7 (Bus error). |
|---|
| 18 | -------------------------------------------------------------------------- |
|---|
| 19 | [curie1664:113817] 61 more processes have sent help message help-mpi-btl-openib.txt / of error event |
|---|
| 20 | [curie1664:113817] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages |
|---|
| 21 | 2 total processes killed (some possibly by mpirun during cleanup) |
|---|
| 22 | 4.94user 22.62system 29:10.67elapsed 1%CPU (0avgtext+0avgdata 20336maxresident)k |
|---|
| 23 | 128inputs+8outputs (0major+8625minor)pagefaults 0swaps |
|---|
Note: See
TracBrowser
for help on using the repository browser.
