This site is migrating to a new forum software on Tuesday, September 24th 2019, you may experience a short downtime during this transition

Main menu

Navigation

Problem running FSDCHM compset on evergreen

1 post / 0 new
f_garcia@...
Problem running FSDCHM compset on evergreen

Hi,

I am trying to run CESM 1.2 on the supported machine evergreen.

I am using the following options to create a new case:
-mach evergreen
-compset FSDCHM
-res f19_f19

However each time attempt run the default case (cam4_chem_radpsv_geos5), the run is terminated with the following message:

Mon Dec 9 18:14:36 EST 2013 -- CSM EXECUTION BEGINS HERE
=>> PBS: job killed: node 2 (compute-2-36.evergreen.umd.edu) requested job terminate, 'EOF' (code 1099) - received SISTER_EOF attempting to communicate with sister MOM's
Terminated

The model does appear to begin running but the job is killed shortly after (about 20 tsteps into the simulation). Log files are generated for each component, although I cannot identify the error within. Towards the end of the cesm.log file the following is written out before the process is killed:

dpcoup cant adjust 89 458 3 2.899606928142302E-039
3.031820011645065E-040 1.007020596805401E-036
dpcoup cant adjust 66 142 4 1.541481934267371E-040
0.000000000000000E+000 1.015449337949024E-036
QNEG3 from convtran2/ISOP:m= 53 lat/lchnk= 803 Min. mixing ratio violated at 1 points. Reset to 1.0E-36 Worst =-1.2E-12 at i,k= 14 54
--------------------------------------------------------------------------
mpirun noticed that process rank 0 with PID 10782 on node compute-6-10.evergreen.umd.edu exited on signal 9 (Killed).
--------------------------------------------------------------------------
forrtl: error (78): process killed (SIGTERM)

Would anyone know if the error is machine-related, or what might be causing it? Any suggestions on how to address it?

Your help is greatly appreciated,

Fernando

Who's new

  • jwolff
  • tinna.gunnarsdo...
  • sarthak2235@...
  • eolivares@...
  • shubham.gandhi@...