ERROR ABORT in subroutine FLCHECK on PE 005 Inconsistent header data on input files

Dear all,
I’m finding an error running the CCTM module of CMAQ5.2.1. I attached to this message the full log of the run but basically the error is the following:

**** ERROR ABORT in subroutine FLCHECK on PE 005*

  • Inconsistent header data on input files*
  • PM3EXIT: DTBUF 1:00:00 Jan. 2, 2016*
  • Date and time 1:00:00   Jan. 2, 2016   (2016002:010000)*

MPI_ABORT was invoked on rank 5 in communicator MPI_COMM_WORLD
with errorcode 538976288.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
[bber0501u17a.bb2.cluster:06338] 1 more process has sent help message help-mpi-api.txt / mpi-abort
[bber0501u17a.bb2.cluster:06338] Set MCA parameter “orte_base_help_aggregate” to 0 to see all help / error messages
0.115u 0.211s 0:05.35 5.9% 0+0k 0+16408io 6pf+0w

Do you have any suggestion about the type of error? Following similar topics, if I correctly understood, the error could be connected with the time dimension, but I checked all the time steps in the input files and they are correct for the date and also for the number of time steps.

Thanks in advance for the help.

Kind Regards

Andrea Mazzeo
(upload://tH5qD2P7f3cIXUkvh4lR2BwyGq1.txt) (360.1 KB)

Routine FLCHECK (and routine GRDCHECK, which it calls) do not do an adequate job of logging what they find – in particular, the reason for any mismatch and the logical name of the file involved. Moreover, the tests in GRDCHECK demand an unreasonable degree of precision.

It would have been better had the codes used the standard I/O API routines FILCHK3 and GRDCHK3, which do perform such logging, and which have checks that are carefully tuned for appropriate numerical precision.