From: Yinglong Miao (yimiao_at_indiana.edu)
Date: Tue Aug 26 2008 - 16:22:21 CDT
Thanks for your reply!
Normally the return code for executing the Fortran code is 0 as no error
message would come up upon using catch.
For the memory usage, I did see that there is a steady increase during the
run. However, I just found it's not because of calling the fortran code, but
some other parts in my simulation. Let me see what I can find out finally.
On Tue, Aug 26, 2008 at 3:38 PM, Peter Freddolino <petefred_at_ks.uiuc.edu>wrote:
> Hi Yinglong,
> what return code does your fortran code give on normal exit? If it isn't 0,
> this is why exec thinks there is an error, and the only things you can do
> are return 0 on normal completion or use catch, as you have done.
> As far as the memory issues... have you watched the memory usage on the
> head node during execution? Does it steadily increase during the run?
> Yinglong Miao wrote:
>> Dear NAMD users,
>> In my NAMD simulations, I have been calling "exec" to execute an MPI
>> Fortran code. The MPI code is tested to run properly given the input needed
>> before I put it into the NAMD run. But an error message "child process
>> exited abnormally" keeps coming up in NAMD simulation, even though the MPI
>> code seems to run well giving proper output. I have tried to suppress this
>> warning error message by implementing "catch", but after a certain number of
>> calls to execute the MPI code, the program runs into crash with "couldn't
>> fork child process: not enough memory."
>> Does anybody have experiences with handling such errors with calling exec?
>> Your help will be greatly appreciated!
>> Yinglong Miao
>> Ph.D. Candidate
>> Center for Cell and Virus Theory
>> Chemistry Department, Indiana University
>> 800 E Kirkwood Ave Room C203A, Bloomington, IN 47405
>> Tel: 1-812-856-0981
-- Yinglong Miao Ph.D. Candidate Center for Cell and Virus Theory Chemistry Department, Indiana University 800 E Kirkwood Ave Room C203A, Bloomington, IN 47405 Tel: 1-812-856-0981
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:21:15 CST