[mpich-discuss] Error Running MPICH for Photochemical Modeling

[Abhishek Bhat]

Okay,

I might have confused you (or I might be confused). 

If I am running the same larger scale run with 1 master and 6 nodes (1 process each so NUMPROCS = 7) there are no issues.  But if I change that to anything more than 7 then I get the error.  

In another situation, for a smaller scale run, I can go as much as 64 processes without any error.


Thank You for assistance.

Abhishek

………………………………………………………………………………………………….
Abhishek Bhat, PhD, EPI,
Senior Consultant


-----Original Message-----
From: Balaji, Pavan [mailto:balaji at anl.gov] 
Sent: Wednesday, September 17, 2014 1:04 PM
To: Abhishek Bhat
Cc: discuss at mpich.org
Subject: Re: [mpich-discuss] Error Running MPICH for Photochemical Modeling


On Sep 17, 2014, at 1:00 PM, Abhishek Bhat <abhat at trinityconsultants.com> wrote:
> The application works when there is a less resource intensive runs (larger grid with larger grid spacing).  The issue occurs when we have nested grid runs.  Also the application works without any issues for less than 7 processes (1 I/O and 6 nodes).

That’s not an indication that the application is correct with more processes.  Many applications work at smaller scales but fail at larger scals.  From the error, the indications point to the application.  I’d recommend digging into the application and figuring out what’s breaking.  The -print-all-exitcodes that Sangmin suggested, or attaching a debugger might be useful for this.

  — Pavan

--
Pavan Balaji  ✉️
http://www.mcs.anl.gov/~balaji


-- 
_________________________________________________________________________

The information transmitted is intended only for the person or entity to
which it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you received
this in error, please contact the sender and delete the material from any
computer.
_________________________________________________________________________
_______________________________________________
discuss mailing list     discuss at mpich.org
To manage subscription options or unsubscribe:
https://lists.mpich.org/mailman/listinfo/discuss