Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

From: Manjunath GV (manjugv_at_[hidden])
Date: 2006-04-24 13:48:51


Thanks, problem got solved. I had installed old version of open mpi
as root for all users and now
installed Open MPI version 1 for myself. Though the previous version
was not in the path, seems like this was causing the problems.
The combination of removing previous libraries, runtimes solved the
problem.

thanks,
-Manjunath

On Apr 22, 2006, at 11:12 AM, Ralph Castain wrote:

> Another thing to try - go to your installation location's lib
> subdirectory (at $prefix/lib) and delete everything that is there.
> Then go back to the directory where you put the software and do a
> "make install" again.
>
> Sometimes, especially if you are upgrading to a new version, you
> can be burned by stale shared libraries. This sounds like it could
> be the problem here. We don't remove any old libraries when you do
> an installation, so if you change versions, you really should do
> this procedure to avoid picking up "old stuff".
>
> Alternatively, you could build and run without shared libraries to
> avoid this problem altogether - just reconfigure with "--enable-
> static --disable-shared" and then do "make clean all install".
>
> Ralph
>
>
> Brian Barrett wrote:
>> Well, so much for the easy one :(.
>>
>> Is it possible that you have two versions of Open MPI in your path
>> somewhere and that you might be getting different versions on
>> different nodes? The errors below generally indicate that data was
>> received in a totally different format than expected, so I'm just
>> kind of guessing as to how one could get to that situation...
>>
>> Brian
>>
>> On Apr 21, 2006, at 5:01 PM, Manjunath G Venkata wrote:
>>
>>
>>> On Thu, 20 Apr 2006, Brian Barrett wrote:
>>>
>>>
>>>> Are these both identical architecture? Those look suspiciously
>>>> like what happens when you're trying to mix 32/64 bit or little
>>>> endian / big endian.
>>>>
>>>>
>>> - Both my nodes are Intel Xeons and run linux 2.4.26.
>>>
>>> -Manjunath
>>>
>>>
>>>> Brian
>>>>
>>>> On Apr 20, 2006, at 8:53 PM, Galen M. Shipman wrote:
>>>>
>>>>
>>>>> Hey Guys,
>>>>> Not sure what is going on here, has anyone seen this before?
>>>>> - Galen
>>>>>
>>>>>> Hi Galen,
>>>>>> Sorry to bother you.
>>>>>> I have installed latest stable version of Open MPI(1.0) on two
>>>>>> of spider
>>>>>> nodes(s7,s4) for some experiments, but there seems to be
>>>>>> configuration
>>>>>> error or something else which I don't understand. After
>>>>>> installing, as
>>>>>> a test I ran an simple MPI program but it complains with
>>>>>> following
>>>>>> errors.
>>>>>> [s4:10685] [0,0,0] ORTE_ERROR_LOG: Pack data mismatch in file
>>>>>> dps_unpack.c at line 121
>>>>>> [s4:10685] [0,0,0] ORTE_ERROR_LOG: Pack data mismatch in file
>>>>>> dps_unpack.c at line 95
>>>>>> Further digging with gdb prints following errors
>>>>>> [s7:07005] ERROR: A daemon on node s4 failed to start as
>>>>>> expected.
>>>>>> [s7:07005] ERROR: There may be more information available from
>>>>>> [s7:07005] ERROR: the remote shell (see above).
>>>>>> [s7:07005] The daemon received a signal 5.
>>>>>> [s7:07005] [0,0,0] ORTE_ERROR_LOG: Pack data mismatch in file
>>>>>> dps_unpack.c at line 121
>>>>>> [s7:07005] [0,0,0] ORTE_ERROR_LOG: Pack data mismatch in file
>>>>>> dps_unpack.c at line 95
>>>>>> [s7:07005] [0,0,0] ORTE_ERROR_LOG: Pack data mismatch in file
>>>>>> dps_unpack.c at line 121
>>>>>> [s7:07005] [0,0,0] ORTE_ERROR_LOG: Pack data mismatch in file
>>>>>> dps_unpack.c at line 95
>>>>>> any clue on what I am doing wrong ?
>>>>>> thanks,
>>>>>> -Manjunath
>>>>>>
>>>>> _______________________________________________
>>>>> devel mailing list
>>>>> devel_at_[hidden]
>>>>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>>>>>
>> _______________________________________________
>> devel mailing list
>> devel_at_[hidden]
>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>>
>>
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel