Error in MPI_File_open

[Aroma Mahendru]

Ok. I ll try that.
Thank you for helping !

Aroma


On Wed, Jun 26, 2013 at 4:47 PM, Jim Edwards <edwards.jim at gmail.com> wrote:

> Yes, the version on the cluster is old.  It has a bug that has been fixed
> in newer versions of openmpi.   Try 1.7.1 or 1.6.5
>
>
> On Wed, Jun 26, 2013 at 2:40 PM, Aroma Mahendru <
> aroma.mahendru.ece10 at iitbhu.ac.in> wrote:
>
>> Okay I get this error :
>>
>>
>> MPI_File_open: MPI_ERR_OTHER: known error not in list
>> --------------------------------------------------------------------------
>> MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD
>> with errorcode 1.
>>
>> NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
>> You may or may not see output from other processes, depending on
>> exactly when Open MPI kills them.
>> --------------------------------------------------------------------------
>>
>> So, does this mean there is an older version of openmpi ?The version on
>> the cluster is Open MPI 1.5.4.
>>
>>
>> On Wed, Jun 26, 2013 at 4:17 PM, Rob Latham <robl at mcs.anl.gov> wrote:
>>
>>> On Wed, Jun 26, 2013 at 03:53:57PM -0400, Aroma Mahendru wrote:
>>> > Actually I did compile and execute the basic mpi program given there.
>>> It
>>> > executed without any errors.  But am I supposed to see any file
>>> created as
>>> > well ?
>>>
>>> That simple test program doesn't check for errors.  It ensures your
>>> MPI library knows about MPI-IO routines.  In 2013 it would be shocking
>>> if it did not.
>>>
>>> It *should* create a zero-length file called "testfile" in the current
>>> directory.  If it did not, try again with this version with a bit more
>>> error checking/reporting:
>>>
>>> static void handle_error(int errcode, const char *str)
>>> {
>>>         char msg[MPI_MAX_ERROR_STRING];
>>>         int resultlen;
>>>         MPI_Error_string(errcode, msg, &resultlen);
>>>         fprintf(stderr, "%s: %s\n", str, msg);
>>>         MPI_Abort(MPI_COMM_WORLD, 1);
>>> }
>>>
>>> #include <stdio.h>
>>> #include <mpi.h>
>>> int main(int argc, char **argv) {
>>>     MPI_File fh;
>>>     int ret;
>>>     MPI_Init(&argc, &argv);
>>>     ret = MPI_File_open(MPI_COMM_WORLD, "testfile", MPI_MODE_CREATE |
>>> MPI_MODE_RDWR,
>>>                   MPI_INFO_NULL, &fh);
>>>     if (ret != MPI_SUCCESS) handle_error(ret, "MPI_File_open");
>>>     MPI_File_close(&fh);
>>>     MPI_Finalize();
>>>     return 0;
>>> }
>>>
>>> --
>>> Rob Latham
>>> Mathematics and Computer Science Division
>>> Argonne National Lab, IL USA
>>>
>>
>>
>>
>> --
>> Aroma Mahendru
>> Electronics Engineering
>> Indian Institute of Technology (BHU)
>> Varanasi
>>
>
>
>
> --
>
> Jim Edwards
>
>
>


-- 
Aroma Mahendru
Electronics Engineering
Indian Institute of Technology (BHU)
Varanasi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/parallel-netcdf/attachments/20130626/7efa314f/attachment.html>