On 1/22/07 9:39 AM, "Greg Watson" <gwat...@lanl.gov> wrote:

> I tried adding '-mca btl ^sm -mca mpi_preconnect_all 1' to the mpirun
> command line but it still fails with identical error messages.
> 
> I don't understand the issue with allocating nodes under bproc. Older
> versions of OMPI have always just queried bproc for the nodes that
> have permissions set so I can execute on them. I've never had to
> allocate any nodes using a hostfile or any other mechanism. Are you
> saying that this no longer works?

Turned out that mode of operation was a "bug" that caused all kinds of
problems in production environments - that's been fixed for quite some time.
So, yes - you do have to get an official "allocation" of some kind. Even the
changes I mentioned wouldn't remove that requirement in the way you
describe.

> 
> Greg
> 
> On Jan 22, 2007, at 8:45 AM, Ralph H Castain wrote:
> 
>> Actually, I just looked at my notes on bluesteel and see that it is an
>> unscheduled system - i.e., no LSF to allocate the nodes. That's a
>> problem
>> with the current code.
>> 
>> I assume you must be feeding it a hostfile or something?? How are you
>> telling mpirun which nodes to use?
>> 
>> Brian and I just talked about the more general bproc allocator
>> question and
>> I can commit a change later today to fix the situation for
>> bluesteel. Until
>> then, I fear that we may not run on that system, though you could
>> give it a
>> try anyway.
>> 
>> 
>> On 1/22/07 8:14 AM, "Ralph H Castain" <r...@lanl.gov> wrote:
>> 
>>> Oh yeah - Galen noted that you also have to do a preconnect, so
>>> what you
>>> need to add to your command line is:
>>> 
>>> -mca btl ^sm -mca mpi_preconnect_all 1
>>> 
>>> Ralph
>>> 
>>> 
>>> 
>>> On 1/22/07 8:02 AM, "Greg Watson" <gwat...@lanl.gov> wrote:
>>> 
>>>> 
>>>> On Jan 19, 2007, at 4:39 PM, Li-Ta Lo wrote:
>>>> 
>>>>> On Fri, 2007-01-19 at 14:42 -0700, Greg Watson wrote:
>>>>>> 
>>>>>> The libraries required by the program are:
>>>>>> 
>>>>>> $ ldd x
>>>>>>          librt.so.1 => /lib64/tls/librt.so.1 (0x00002aaaaabc1000)
>>>>>>          libbproc.so.4 => /usr/lib64/libbproc.so.4
>>>>>> (0x00002aaaaacdb000)
>>>>>>          libdl.so.2 => /lib64/libdl.so.2 (0x00002aaaaade2000)
>>>>>>          libnsl.so.1 => /lib64/libnsl.so.1 (0x00002aaaaaee5000)
>>>>>>          libutil.so.1 => /lib64/libutil.so.1 (0x00002aaaaaffc000)
>>>>>>          libm.so.6 => /lib64/tls/libm.so.6 (0x00002aaaab100000)
>>>>>>          libpthread.so.0 => /lib64/tls/libpthread.so.0
>>>>>> (0x00002aaaab286000)
>>>>>>          libc.so.6 => /lib64/tls/libc.so.6 (0x00002aaaab39b000)
>>>>>>          /lib64/ld-linux-x86-64.so.2 (0x00002aaaaaaab000)
>>>>>> 
>>>>>> These all appear to be available on the nodes.
>>>>>> 
>>>>> 
>>>>> I tried a recent (today/yesterday?) svn trunk. It works but it is
>>>>> very slow (I am using tcp now).
>>>>> 
>>>>> Ollie
>>>> 
>>>> Does that mean that 1.2 does not support bproc?
>>>> 
>>>> Greg
>>>> _______________________________________________
>>>> devel mailing list
>>>> de...@open-mpi.org
>>>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>>> 
>>> 
>>> _______________________________________________
>>> devel mailing list
>>> de...@open-mpi.org
>>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>> 
>> 
>> _______________________________________________
>> devel mailing list
>> de...@open-mpi.org
>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
> 
> _______________________________________________
> devel mailing list
> de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/devel


Reply via email to