I  believe what this is saying is that we are not finding any TCP interfaces - 
the ioctl itself is failing. So yes - miprun failing at that point is going to 
happen because we have no way to communicate for launch.

Do you see interfaces if you do an /sbin/ifconfig? Do they have valid IP 
addresses?

On Dec 1, 2009, at 5:52 PM, Jeff Squyres wrote:

> On Nov 29, 2009, at 6:15 PM, <kevin.buck...@ecs.vuw.ac.nz> 
> <kevin.buck...@ecs.vuw.ac.nz> wrote:
> 
>> $ mpirun -n 4 hello_f77
>> [somebox.ecs.vuw.ac.nz:04414] opal_ifinit: ioctl(SIOCGIFFLAGS) failed with
>> errno=6
>> 
> 
> Oy.  This is ick, because this error code is coming from horrendously complex 
> code deep in the depths of OMPI that is probing the OS to figure out what 
> ethernet interfaces you have.  It may or may not be simple to fix this.
> 
> Do you mind diving into the OMPI code a bit to figure this out?  I'm afraid 
> that none of the developers are likely to have access to NetBSD.  :-(  I can 
> point you right where to look.
> 
>> When running on a "server" machine within the grid, a machine I am told
>> should not be any different to the workstation I was using above in
>> respect of user environment, I get a different error and find that the
>> job does not run at all.
>> 
>> This case seems to producean error message that is oft reported within
>> the OpenMPI community:
>> 
>> $ mpirun -n 4 hello_f77
>> [somebox2.ecs.vuw.ac.nz:25244] [[51186,0],0] ORTE_ERROR_LOG: Error in file
>> ess_hnp_module.c at line 150
>> --------------------------------------------------------------------------
>> It looks like orte_init failed for some reason; your parallel process is
>> ...
>> 
>>  orte_rml_base_select failed
>>  --> Returned value Error (-1) instead of ORTE_SUCCESS
>> 
> 
> This could well be a side-effect of the same error as above -- OMPI may have 
> concluded that it didn't find any ethernet devices and therefore aborted.
> 
> -- 
> Jeff Squyres
> jsquy...@cisco.com
> 
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users


Reply via email to