LAM/MPI logo

LAM/MPI General User's Mailing List Archives

  |   Home   |   Download   |   Documentation   |   FAQ   |   all just in this list

From: Jeff Squyres (jsquyres_at_[hidden])
Date: 2004-05-22 09:39:35


Sorry -- I was traveling all this past week and wasn't able to reply to
your followup from Tuesday until now.

As usual, Bogdan raises excellent points. :-) I also just replied (see
http://www.lam-mpi.org/MailArchives/lam/msg07952.php) to the original
thread; see if that helps as well.

On Sat, 15 May 2004 amzm_at_[hidden] wrote:

> Bogdan, thanks for your answer.
>
> On 21 May 2004, at 19:20, Bogdan Costescu wrote:
>
>> On Fri, 14 May 2004 amzm_at_[hidden] wrote:
>>
>>> I have to write an article about this subject, because that, this is
>>> important to me.
>>
>> Coming up with reasons why this is important to _you_ doesn't matter.
>> You should come up with reasons that show how important it is to _us_
>> the developers and users of LAM-MPI that could potentially answer your
>> question.
>
> Ok, I agree.
>
>>
>>> With LAM/MPI 6.5.9, I have verified that can be used different users
>>> in different hosts to create a LAM cluster, but I can't use it to
>>> execute my program (perhaps I am doing something wrong).
>>
>> Are the two users sharing the same home directory ?
> No, each user has a different home directory
>
>> Are the binaries that have to be executed located in the same place on
>> all nodes ?
> Yes, in each user's home directory
>
> Do their 'executable' bit(s) allow execution by the
>> different users ?
> Do the programs try to read/write files for which
>> the different users might not have the appropriate permissions ?
> All permissions (chmod 777)
>
>
>>
>>> (I search in google and altavista and I don't find nothing about
>>> mpirun and multiple users).
>>
>> 'man bhost' (or 'man 5 bhost')
>>
>> It's amazing how much information you can find on _your_ system,
>> without need for Internet search...
> :D ok, I will see it.
>>
>> --
>> Bogdan Costescu
>>
>> IWR - Interdisziplinaeres Zentrum fuer Wissenschaftliches Rechnen
>> Universitaet Heidelberg, INF 368, D-69120 Heidelberg, GERMANY
>> Telephone: +49 6221 54 8869, Telefax: +49 6221 54 8868 E-mail:
>> Bogdan.Costescu_at_[hidden]
>>
>>
>>
>> _______________________________________________
>> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
>
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>

-- 
{+} Jeff Squyres
{+} jsquyres_at_[hidden]
{+} http://www.lam-mpi.org/