Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: Re: [OMPI users] FRQ: Warn if components specified but missing
From: Bernhard Hermann (hermann_at_[hidden])
Date: 2008-08-04 10:05:36


Jeff Squyres wrote:
> Thanks for the feedback! Sorry the lack of a clear error message caused
> you pain. :-(
>
> FWIW, we do have this feature in the upcoming v1.3 release. For example:

That's very good to know. Thanks for the prompt and informative reply.

>
> [6:46] svbu-mpi:~/mpi % mpirun -np 2 --mca btl openibbbb,self hello
> --------------------------------------------------------------------------
> A requested component was not found, or was unable to be opened. This
> means that this component is either not installed or is unable to be
> used on your system (e.g., sometimes this means that shared libraries
> that the component requires are unable to be found/loaded). Note that
> Open MPI stopped checking at the first component that it did not find.
>
> Host: svbu-mpi.cisco.com
> Framework: btl
> Component: openibbbb
> --------------------------------------------------------------------------
>
> Hopefully, this will help such situations in the future.
>
>
>
> On Aug 4, 2008, at 9:23 AM, Bernhard Hermann wrote:
>
>> Hi,
>>
>> We would like to request a warning/error message to be printed whenever
>> a component (like "openib") is specified, but not actually loaded.
>>
>> Our case: Running a job via "tcp" over IB went OK, trying to use
>> "openib" failed. It turned out that while compiling the "openib" module
>> wasn't made (missing path), so OpenMPI croaked when trying to access it.
>>
>> This was the error that was of little help to us in tracing the problem:
>> --------------------------------------------------------------------------
>>
>> Process 0.1.7 is unable to reach 0.1.0 for MPI communication.
>> If you specified the use of a BTL component, you may have
>> forgotten a component (such as "self") in the list of
>> usable components.
>> --------------------------------------------------------------------------
>>
>> --------------------------------------------------------------------------
>>
>> It looks like MPI_INIT failed for some reason; your parallel process is
>> likely to abort. There are many reasons that a parallel process can
>> fail during MPI_INIT; some of which are due to configuration or
>> environment
>> problems. This failure appears to be an internal failure; here's some
>> additional information (which may only be relevant to an Open MPI
>> developer):
>>
>> PML add procs failed
>> --> Returned "Unreachable" (-12) instead of "Success" (0)
>> --------------------------------------------------------------------------
>>
>> *** An error occurred in MPI_Init
>> *** before MPI was initialized
>> *** MPI_ERRORS_ARE_FATAL (goodbye)
>>
>>
>>
>>
>> Unfortunately we can't provide a patch and would be glad if someone
>> could implement and include this.
>>
>> Thanks for your attention,
>> best regards,
>> Bernhard HERMANN
>>
>> init.at informationstechnologie GmbH.
>>
>> Mariahilfer Str. 61/1/10
>> 1060 Wien/Vienna
>> Austria - EU
>>
>> Tel: +43 1 522 53 77 - 35
>>
>> http://www.init.at
>> hermann_at_[hidden]
>>
>> _______________________________________________
>> users mailing list
>> users_at_[hidden]
>> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
>

-- 
Bernhard HERMANN
init.at informationstechnologie GmbH.
Mariahilfer Str. 61/1/10
1060 Wien/Vienna
Austria - EU
Tel: +43 1 522 53 77 - 35
http://www.init.at
hermann_at_[hidden]