[torqueusers] mixed 32/64 bit cluster

Jeff DeReus jdereus at gmail.com
Fri Oct 24 14:14:21 MDT 2008

i have compiled the 32 bit mom on the 64  bit arch.  everything seemed to
complete properly.  however, now the pbs_mom segfaults with an error 7 when
i try to start it up.  also,  i am running a 32 bit server on the head.
will this cause communication problems with the 64 bit moms on the other
nodes?  i read that they could not communicate with each other properly.

 the 64 bit compilation of moms are up and running on the other nodes but no
jobs are pushed out to them.  i am assuming from your response that this is
because of some library dependencies?  does torque check job dependencies or
is the job pushed out and then rejected when the job itself determines that
it does not have the required libs?  if that is indeed the case, how does
one find the library requirements for jobs that i do not compile or submit?

Thank you,
Jeff D

On Fri, Oct 24, 2008 at 1:39 PM, Torsten Rohlfing
<torsten at synapse.sri.com>wrote:

> We've done this for several years, with no problems ever. We're even
> running a mixed setup with 32bit mom on 32bit machines and 64bit mom on
> 64bit systems. Head node is running 64bit server and mom. We're using the
> torque RPMs from the Fedora repos, currently on FC9.
> The only thing you need to make sure is that your jobs only run 64bit
> binaries on 64bit systems, and if they run 32bit executables on 64bit
> systems you may need to install extra 32bit shared libraries.
> Other than that, piece of cake.
> TR
>  Jeff DeReus said...
>> |/As a joint project here we are attempting to integrate a mixed 32/64 bit
>> /|/cluster.  I have not been able to find any documentation as to the
>> success
>> /|/of this in the past.
>> /|/
>> /|/While the current 32 bit head can see the proper aspects of the 64 bit
>> /|/nodes, no jobs are successfully pushed out.  Everything seems to be
>> /|/communicating properly between nodes and ssh is properly set up.
>> /|/
>> /|/Has anyone had any success with this type of project?  We would like to
>> be
>> /|/able to offer dual functionality and the possibility for users to have
>> /|/access to the entire joint cluster.
>> /
>> We did that for a while.  I believe we ran 32 bit
>> moms on all systems, including 64 bit systems.
>> Have you tried running a 32 bit mom on a 64 bit
>> system?  (I'm assuming there's no other 32/64 bit
>> clash, such as jobs requesting a 32 bit property
>> that isn't on the 64 bit clients.)
> --
> Torsten Rohlfing, PhD          SRI International, Neuroscience Program
> Research Scientist             333 Ravenswood Ave, Menlo Park, CA 94025
>  Phone: ++1 (650) 859-3379      Fax: ++1 (650) 859-2743
>  torsten at synapse.sri.com        http://www.stanford.edu/~rohlfing/<http://www.stanford.edu/%7Erohlfing/>
>    "Though this be madness, yet there is a method in't"
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20081024/166b9d39/attachment.html

More information about the torqueusers mailing list