Could you give us more details on the version of Open MPI you are
using ? I suppose from your message this is one of the 1.2 series, but
mode details would be greatly helpful.
An utility called orte-clean can be used to try to remove all the
garbage left, should something go wrong.
We have fixed a number of bugs in the MPI-2 dynamics recently. The
forthcoming 1.3 should be more robust to that aspect. You can have a
preview of it by using the svn version of Open MPI. The concepts have
changed a little since then as --persistent --seed have been replaced
by an external name server called ompi-server. I can give you more
details if you want to try the svn version.
* Dr. Aurélien Bouteiller
* Sr. Research Associate at Innovative Computing Laboratory
* University of Tennessee
* 1122 Volunteer Boulevard, suite 350
* Knoxville, TN 37996
* 865 974 6321
Le 10 déc. 08 à 10:28, Bernard Secher - SFME/LGLS a écrit :
> Hi everybody
> I want to use MPI_publish_name function to do communicaztion between
> two independant code.
> I saw on the web i must use the orted daemon with the following
> orted --persistent --seed --scope public --universe foo
> The communication success, but when i close the communication I
> have a dead-lock at the following function: MPI_Comm_disconnect();
> I have a second question:
> How can i stop the orted daemon?
> users mailing list