[torqueusers] Bugs/glitches in torque 2.3.0?

Prakash Velayutham prakash.velayutham at cchmc.org
Wed Apr 9 05:46:12 MDT 2008


Hi,

On Apr 9, 2008, at 4:17 AM, Chris Samuel wrote:

>
> ----- "Roy Dragseth" <roy.dragseth at cc.uit.no> wrote:
>
>> On Tuesday 08 April 2008, Steve Snelgrove wrote:
>>> Sorry, I generated the first patch out of the trunk and because of
>>> a previous attempt at fixing the ncpus problem, the patch failed  
>>> when
>>> I ran it against 2.3.  Here is a better one for 2.3.
>>
>> Yes, this works as expected. Both the ncpus and size reports are now
>> correct.
>
> Sounds good!
>
>> Any plans for a 2.3.1 release anytime soon?  I'm working on a new
>> release of the pbs-roll roll for Rocks and need this fix.   If
>> 2.3.1 is just around the corner I'll wait for that, if not, I'll
>> just include the patch in the release.
>
> It would be really nice to have cpusets support fixed
> so that the vnode sets were optional, dunno if I'll have
> time to look at that in the near future though.
>
> Perhaps for 2.3.2..
>
> The documentation should warn against using the current
> Torque cpusets with OpenMPI at least though.

Could you explain how these 2 fit together (other than the obvious)  
and why it won't work in this case? I am using Open MPI with Torque  
(2.2.1) , but was planning on enabling cpusets with 2.3.0 as an  
upgrade in the near future.

And also my compile of Open MPI automatically enables support for  
Torque as TM.

> cheers,
> Chris

Thanks,
Prakash


More information about the torqueusers mailing list