Possible race condition?

Hello list

I was experincing random hangs of my code (using netCDF as IO format)
when running several jobs from a batch queue system on a cluster, I
was wondering whether that could be related to netCDF or not, since
such hangs did not exist before introducing netCDF interfaces. And the
hang does not happen when only one single job is running.

Some more details:

Platform: Linux 2.4.20-28.9smp #1 SMP i686 i686 i386 GNU/Linux
Compiler:  Intel ifc Version 7.1   Build 20030307Z
netCDF:  3.5.0

The code does not use MPI, and uses openMP speedup, each job is
assigned one node with 2CPU.

I know this is a very vague message and I might be looking at
something totally unirrelevant, just hope somebody has similar
experience or know a bit of possible cause.

Best Regards!
-- 
Yongtao Yang

email: yongtao.yang@xxxxxxxxx

  • 2005 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the netcdfgroup archives: