[mpich-discuss] Problems with running make

Gustavo Correa gus at ldeo.columbia.edu
Sun Mar 2 17:53:51 CST 2014


Hi Ron
On Mar 2, 2014, at 6:27 PM, Ron Palmer wrote:

> Reuti,
> thanks for you suggestions... I did the ldd cpi (in the examples directory) on three computers, and I have attached three screenshot. I am using sargeant as the master, where I start these jobs from.
> 
> Looking at these, I installed gcc-gfortran.x86_64 on the other two computers as well, and they now have identical outputs as the first one (sargeant). Thanks for pointng this out and giving me help in how to check this.
> 

The screenshots show that gfortran is installed only on sargeant, but not on constable and
gainsborough.  
Are you telling us that you took the corrective action after seeing these results,
and installed gfortran on constable and gainsborough as well?
Just to make sure.


I am still rather confused that cpi is linked with the gfortran library.
Did you compile the cpi.c *C* program, or some Fortran version of it?
The C program (cpi.c) should compile with mpicc and I don't see any reason for
it to be linked to the gfortran library.
That is unclear so far.

Gus Correa

> I tried:
> mpiexec -machinefile cluster.txt -np 6 ./cpi were the machine file has only two members, each with 4 cores. I received the same error message as before but that could be for many reasons, and I will now respond to Gus's email with a few clarifications.
> 
> I appreciate your efforts!
> R
> 
> <fiaecjcj.png>
> 
> 
> <ddjdacfd.png>
> 
> 
> <dhchcjgg.png>
> 
> 
> 
> On 3/03/2014 00:28, Reuti wrote:
>> Am 02.03.2014 um 14:01 schrieb Ron Palmer:
>> 
>> 
>>> I am running centos on two of these, sargeant and constable, and RHEL on the third, gainsborough.
>>> 
>>> By runtime libraries, do you mean for the three compilers included, ie gcc, fortran77 and fortran90?
>>> 
>> Yes. But there should be a package with the runtime libraries, i.e. so that it's not necessary to install the complete compiler suite (I assume in your situation that they are installed, but with different versions). Maybe you can check for older/newer libs in the repository of the distributions.
>> 
>> Which libraries are necessary can be listed by issuing:
>> 
>> $ ldd cpi
>> 
>> on each of the machines and whether they are found.
>> 
>> -- Reuti
>> 
>> NB: With commercial compilers it's for this reason often allowed in the license to give away the runtime libraries (and only these) with your compiled application, as otherwise your customers couldn't run your applications.
>> 
>> 
>> 
>>> Please use one syllable words, I am the edge of my understanding :-(.
>>> 
>>> Thanks
>>> Ron
>>> 
>>> 
>>> On 2/03/2014 22:55, Reuti wrote:
>>> 
>>>> Hi,
>>>> 
>>>> Am 02.03.2014 um 13:20 schrieb Ron Palmer:
>>>> 
>>>> 
>>>>> Gus, Reuti, Rajeev and others,
>>>>> I do have mpif77 and mpif90 in mpi/bin, so I conclude that it has compiled ok.
>>>>> 
>>>>> I have checked reuti's suggestions, indicating that the environment is ok. I had added 'PATH=/my/mpi/bin:$PATH' to each of the .bashrc files (for the user I am running on all three)
>>>>> 
>>>>> I am using the same username on all three computers - is that ok?
>>>>> 
>>>>> I have installed MPI on to a NFS drive, all accessing the same; 'which' confirms. Are there any libraries that needs to be installed elsewhere? do  need to run make install as root?
>>>>> 
>>>> No.
>>>> 
>>>> 
>>>> 
>>>>> I have added the 'LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:/my/mpi/lib', though made no difference. (should those "{" and "}" be included? I entered it exactly like I have typed)
>>>>> 
>>>>> I have mpicc the cpi program as suggested by Gus and tried it, as well as the originally provided cpi (part of the downloaded mpi) which I called cpi_orig, see the screen shot below. It looks like there is an issue with libgfortran.so.3...
>>>>> 
>>>> Yep.
>>>> 
>>>> Unless you build static applications, the runtime libraries (of the compiler) need to be available on all machines.
>>>> 
>>>> There should be package with the runtime libraries in your Linux distribution.
>>>> 
>>>> -- Reuti
>>>> 
>>>> 
>>>> 
>>>>> Perhaps I have not got the various compilers installed correctly? I would appreciate if someone could give me a suggestion on how to properly remove and re-install these (and select the correct ones).
>>>>> 
>>>>> By the way, in the runs of cpi below, I reduced the number of cores to be used on the first pc listed in the machinefile, from 8 to 4, in the case there was an issue with the 9th one. As can be seen, it is the first node (or core) on a remote computer where the problems start to be listed which is node 5 in this case...
>>>>> 
>>>>> If I run cpi on only one computer, it works. And yes, hostname works even if all computers are included in the machinefile.
>>>>> 
>>>>> Clear as mud, hey?
>>>>> Thanks for your help.
>>>>> 
>>> _______________________________________________
>>> discuss mailing list     
>>> discuss at mpich.org
>>> 
>>> To manage subscription options or unsubscribe:
>>> 
>>> https://lists.mpich.org/mailman/listinfo/discuss
>> _______________________________________________
>> discuss mailing list     
>> discuss at mpich.org
>> 
>> To manage subscription options or unsubscribe:
>> 
>> https://lists.mpich.org/mailman/listinfo/discuss
> 
> -- 
> Ron Palmer MSc MBA.
> Principal Geophysicist
> ron.palmer at pgcgroup.com.au
> 0413 579 099
> 07 3103 4963
> 
> 
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss




More information about the discuss mailing list