Open MPI logo

MTT Devel Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all MTT Users mailing list

From: Jelena Pjesivac-Grbovic (pjesa_at_[hidden])
Date: 2007-09-17 23:26:12


Hello,

I am trying to set up mtt client for collectives on UTK clusters and I
received the following error

---
> ./client/mtt --scratch /home/pjesa/mtt/scratch/ \
    --file /home/pjesa/mtt/collective-bakeoff/samples/ompi-core-perf-testing.ini \
    --verbose --print-time
*** MTT: ./client/mtt --scratch /home/pjesa/mtt/scratch/ --file
   /home/pjesa/mtt/collective-bakeoff/samples/ompi-core-perf-testing.ini
   --verbose --print-time
*** WARNING: Unable to read hostfile: /home/pjesa/machinefile  -- ignoring
*** Reporter initializing
>> Initializing reporter module: MTTDatabase
    Set HTTP credentials for realm "OMPI"
*** WARNING: >> Failed test ping to MTTDatabase URL:
     https://www.open-mpi.org:443/mtt/submit/
*** WARNING: >> Error was: 501 Protocol scheme 'https' is not supported
     (Crypt::SSLeay not installed)
     LWP will support https URLs if the Crypt::SSLeay module is installed.
     More information at <http://www.linpro.no/lwp/libwww-perl/README.SSL>.
*** ERROR: Module aborted: MTT::Reporter::MTTDatabase:Init: *** ERROR: >>
     Do not want to continue with possible bad submission URL --
         aborting
---
Is there a way to run MTT without SSL?
Thanks,
Jelena
On Fri, 14 Sep 2007, Jeff Squyres wrote:
> It has taken a while, but we are now finally open for business on the
> MTT collective bakeoff.  Please start running your tests so that we
> can have some results to discuss in Paris.  I have put up a wiki page
> with the scope of what we are doing (largely taken from the notes
> from the ABQ meeting) and instructions on how to setup MTT for the
> bakeoff at your site:
>
>     https://svn.open-mpi.org/trac/mtt/wiki/CollPerformanceBakeoff
>
> Some brief notes:
>
> 1. The configuration file that is provided is a *template*.  You
> *WILL* need to customize it for your site.  Those with SLURM may have
> to do very little.  ;-)
>
> 2. I did not test MPICH-MX, Scali MPI, or Cray MPI.  Those are are
> interested, please contact me.  A small amount of work needs to be
> done to get MTT to understand these MPI's.
>
> 3. Please address all questions about MTT and the bakeoff to the mtt-
> users list (http://www.open-mpi.org/mailman/listinfo.cgi/mtt-users).
>
> -- 
> Jeff Squyres
> Cisco Systems
>
> _______________________________________________
> devel-core mailing list
> devel-core_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel-core
>
--
Jelena Pjesivac-Grbovic, Pjesa
Graduate Research Assistant
Innovative Computing Laboratory
Computer Science Department, UTK
Claxton Complex 350
(865) 974 - 6722 
(865) 974 - 6321
jpjesiva_at_[hidden]
"The only difference between a problem and a solution is that
  people understand the solution."
                                       -- Charles Kettering