Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: Re: [OMPI users] Problem with MPI_Send and MPI_Recv
From: Jeff Squyres (jsquyres_at_[hidden])
Date: 2008-09-30 11:06:10


This is quite the odd problem.

1. From prior mails, you do not seem to have iptables running to block
any ports -- is there any other port blocking software running,
perchance?

2. You do seem to be able to run non-MPI apps properly.

3. I assume that you would be able to run "hello world" kinds of MPI
apps ok (i.e., ones that do not include any MPI communication
functions properly). Can you test this, to be sure? There's a "hello
world" demo app in the examples/ subdirectory in the OMPI tarball.

4. What we really need to know is why OMPI's MPI TCP communication is
apparently failing to make a connection between the two nodes. That
will require attaching to the MPI processes with a debugger and seeing
why they're not connecting. We unfortunately haven't had many
problems with this part of the code, so we haven't added too much user-
visible instrumentation...

On Sep 25, 2008, at 11:20 AM, Sofia Aparicio Secanellas wrote:

> I have tried with two computers linux with the same kernel
> (2.6.22-15-generic) and I got the same problem. I do not understand
> what happens.
>
> Sofia
>
>
> ----- Original Message ----- From: "Sofia Aparicio Secanellas" <saparicio_at_[hidden]
> >
> To: "Open MPI Users" <users_at_[hidden]>
> Sent: Wednesday, September 24, 2008 5:53 PM
> Subject: Re: [OMPI users] Problem with MPI_Send and MPI_Recv
>
>
>> No , I do not have any ethernet device aliases.
>>
>> Thank you,
>>
>> Sofia
>> ----- Original Message ----- From: "Jeff Squyres"
>> <jsquyres_at_[hidden]>
>> To: "Open MPI Users" <users_at_[hidden]>
>> Sent: Wednesday, September 24, 2008 2:33 PM
>> Subject: Re: [OMPI users] Problem with MPI_Send and MPI_Recv
>>
>>
>>> You don't happen to have ethernet device aliases on either of
>>> these machines, do you?
>>>
>>> (we have a problem with this on the trunk/v1.3 series right now;
>>> we were under the impression that it was working fine in the v1.2
>>> series -- but I figured I'd ask...)
>>>
>>>
>>> On Sep 24, 2008, at 3:22 AM, Sofia Aparicio Secanellas wrote:
>>>
>>>> Hello Terry,
>>>>
>>>> I obtain the hostnames of both computers:
>>>>
>>>> pichurra
>>>> hpl1-linux
>>>>
>>>> Thank you.
>>>>
>>>> Sofia
>>>>
>>>> ----- Original Message ----- From: "Terry Dontje" <Terry.Dontje_at_[hidden]
>>>> >
>>>> To: <users_at_[hidden]>
>>>> Sent: Tuesday, September 23, 2008 6:24 PM
>>>> Subject: Re: [OMPI users] Problem with MPI_Send and MPI_Recv
>>>>
>>>>
>>>>> Hello Sofia,
>>>>>
>>>>> Very puzzling indeed. Can your try to run hostname or uptime
>>>>> with mpirun? That is something like:
>>>>>
>>>>> mpirun -np 2 --host 10.1.10.208,10.1.10.240 --mca
>>>>> mpi_preconnect_all 1 --prefix /usr/local -mca btl self,tcp -mca
>>>>> btl_tcp_if_include eth1 hostname
>>>>>
>>>>>
>>>>> --td
>>>>>
>>>>> Date: Tue, 23 Sep 2008 17:05:22 +0200
>>>>> From: "Sofia Aparicio Secanellas" <saparicio_at_[hidden]>
>>>>> Subject: Re: [OMPI users] Problem with MPI_Send and MPI_Recv
>>>>> To: "Open MPI Users" <users_at_[hidden]>
>>>>> Message-ID: <34D2F769A7C946BF915A828A9CD7F3CC_at_aparicio1>
>>>>> Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"
>>>>>
>>>>> Hello Terry,
>>>>>
>>>>> Here you can find the files.
>>>>>
>>>>> Thank you very much.
>>>>>
>>>>> Sofia
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> users mailing list
>>>>> users_at_[hidden]
>>>>> http://www.open-mpi.org/mailman/listinfo.cgi/users
>>>>>
>>>>>
>>>>>
>>>>> No virus found in this incoming message
>>>>> Checked by PC Tools AntiVirus (4.0.0.26 - 10.100.007).
>>>>> http://www.pctools.com/free-antivirus/
>>>>
>>>>
>>>>
>>>> No virus found in this outgoing message
>>>> Checked by PC Tools AntiVirus (4.0.0.26 - 10.100.007).
>>>> http://www.pctools.com/free-antivirus/
>>>> _______________________________________________
>>>> users mailing list
>>>> users_at_[hidden]
>>>> http://www.open-mpi.org/mailman/listinfo.cgi/users
>>>
>>>
>>> --
>>> Jeff Squyres
>>> Cisco Systems
>>>
>>> _______________________________________________
>>> users mailing list
>>> users_at_[hidden]
>>> http://www.open-mpi.org/mailman/listinfo.cgi/users
>>>
>>>
>>>
>>> No virus found in this incoming message
>>> Checked by PC Tools AntiVirus (4.0.0.26 - 10.100.007).
>>> http://www.pctools.com/free-antivirus/
>> _______________________________________________
>> users mailing list
>> users_at_[hidden]
>> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
>
>
> No virus found in this outgoing message
> Checked by PC Tools AntiVirus (4.0.0.26 - 10.100.007).
> http://www.pctools.com/free-antivirus/
> _______________________________________________
> users mailing list
> users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/users

-- 
Jeff Squyres
Cisco Systems