Common GEOS-Chem error messages

From Geos-chem
Revision as of 15:49, 15 November 2010 by Bmy (Talk | contribs) (A3 met fields not found)

Jump to: navigation, search

Here is a list of some commonly-encountered GEOS-Chem error messages.

File I/O errors

A3 met fields not found

If routine OPEN_A3_FIELDS (in a3_read_mod.f) returns a "file not found error" very near to the beginning of a GEOS-Chem simulation, i.e.:

 $$ Finished Reading Linoz Data $$
 
===============================================================================
GEOS-CHEM I/O ERROR    29 in file unit    73
Encountered at routine:location open_a3_fields:1

Error   29: File not found
===============================================================================
     - CLEANUP: deallocating arrays now...

then this could indicate one of the following conditions:

Error caused by MEGAN biogenic emissions

MEGAN keeps a 10-day running average of temperature, and therefore requires that the the met field files for the 10 days prior to the start of the GEOS-Chem simulation be present on disk.

You can solve this error in one of two ways:

  1. Make sure you have the previous 10 days (or better yet, the entire previous month!) of data prior to your GEOS-Chem simulation's starting date
  2. Start your GEOS-Chem simulation at a later date

Error caused by not recompiling cleanly

This error may also indicate that you have attempted to change the resolution of GEOS-Chem without recompiling cleanly. Some parts of the GEOS-Chem code (object files, module files) may still be compiled for a different grid resolution.

You can usually fix this doing a

make realclean

command before recompiling the code.

--Bob Y. 10:47, 15 November 2010 (EST)

I/O Error #29

This error indicates that GEOS-Chem cannot find the proper A3 met field file.

NOTE: Error #29 is specific to the IFORT compiler. If you are using a different compiler, then the I/O error number may differ.

--Bob Y. 15:30, 3 November 2010 (EDT)

Problem reading binary punch file

If you are having problems reading a binary punch file into GEOS-Chem, make sure that you have the correct endian setting in your makefile. These are:

  • Intel Fortran compiler (IFORT): -convert big_endian
  • PGI compiler: -byteswapio
  • Sun Studio compiler: -xfilebyteorder=big16:%all

Most machines that use an Intel or AMD chipset are little-endian machines. A few of the older architectures (e.g. Cray, SGI Origin) are big-endian. Binary punch files are always big-endian (due to historical reasons), so you will need to tell your compiler to do the byte swapping manually.

The symptoms of such an error can be as follows:

Daewon Byun wrote:

In the SUBROUTINE READ_BPCH2, It reads the FTI = CTM bin 02 fine, but then fails to read anything after. I dumped the IOUNIT and IO error code -- as you see TMP_TITLE is empty....
   IUNIT, IOS, TMP_TITLE =            98           -1
Then the program stops at the
   IF ( IOS /= 0 ) THEN
      PRINT*, 'open_bpch2_for_read:2'
      STOP
   ENDIF
If I force to read further removing the "STOP", then I get (again, I tried to dump..)
   MODELNAME, LONRES, LATRES, HALFPOLAR, CENTER180 =  0.000000    0.000000       0         0

--Bob Y. 09:57, 24 July 2008 (EDT)

Module file cannot be read

If you should encounter this type of error:

ifort -cpp -w -O2 -auto -noalign -convert big_endian -openmp -Dmultitask -c time_mod.f
fortcom: Error: time_mod.f, line 259: This module file was generated for a different 
platform or by an incompatible compiler or compiler release. It cannot be read.   [JULDAY_MOD]
      USE JULDAY_MOD, ONLY : JULDAY, CALDATE 

Then this means that you are trying to link to previously-created *.mod files that were generated by a different compiler. Making clean and re-compiling from scratch should solve this problem.

--Bob Y. 13:39, 1 July 2008 (EDT)

Error when reading the "restart_gprod_aprod" file

Eric Leibensperger wrote:

I am trying to run GEOS-Chem and have encountered and error. The log file gives me this:
  ===============================================================================
  GEOS-CHEM ERROR: No matches found for file restart_gprod_aprod.2001070100!
  STOP at READ_BPCH2 (bpch2_mod.f)!
  ===============================================================================
I have the aerosol restart file (with the same name) in my ~/testrun/runs/run.v7-04-12/ folder. Is it looking for it elsewhere? I get an additional message in the log.error file, but I think that it is possibily the result of not being able to find the file above:
  ******  FORTRAN RUN-TIME SYSTEM  ******
  Error 1183:  deallocating an unallocated allocatable array
  Location:  the DEALLOCATE statement at line 4933 of "carbon_mod.f"
  Abort
Any thoughts would be appreciated. Sorry to bother you with this!
Eric

Philippe Le Sager replied:

You must rewrite your restart_gprod_aprod.YYYYMMDDHH so that the date in the filename is the same as the one in the datablock header.
I wrote a routine to do that: ~phs/IDL/dvpt/various_rewrite/rewrite_agprod.pro
-Philippe

NOTE: The file rewrite_agprod.pro will be released in the next GAMAP version.

--Bmy 15:59, 9 May 2008 (EDT)

For GEOS-Chem v8-03-01 and higher

In GEOS-Chem v8-03-01 and higher, the restart_gprod_aprod.YYYYMMDDhh file has been renamed to soaprod.YYYYMMDDhh. The above-described error can occur with the soaprod.YYYYMMDDhh file if the date in the file does not match the starting date of your simulation.

--Bob Y. 13:10, 4 November 2010 (EDT)

File ann_mean_trop.geos5.* not found

If you are running a GEOS-5 simulation and get an error that says that GEOS-Chem cannot locate the ann_mean_trop.geos5.2x25 or ann_mean_trop.geos5.4x5 file, then make sure that the following option is set in your input.geos file.

Use variable tropopause?: T

Starting in version GEOS-Chem v7-04-12, GEOS-Chem can now use a variable tropopause (i.e. chemistry is done up to the location of the actual tropopause as diagnosed from the met fields at any given timestep). You cannot use the annual mean tropopause for GEOS-5.

--Bob Y. 15:18, 7 July 2008 (EDT)

Problem reading GEOS-4 TROPP files

Please see this wiki post for more information about a common problem that can occur if you using GEOS-4 meteorology with the dynamic tropopause.

Crashes or abnormal exits

KPP "Step size too small" error

The following abnormal exit from the KPP chemical solver:

    - PHYSPROC: Trop chemistry at 2006/07/01 00:00
Forced exit from Rosenbrock due to the following error:
--> Step size too small: T + 10*H = T or H < Roundoff
T=   3044.21151383269      and H=  1.281206877135470E-012

...

JLOOP, I, J, L         7231          31          51           1
Forced exit from Rosenbrock due to the following error:
--> Step size too small: T + 10*H = T or H < Roundoff
T=   3044.21151383269      and H=  1.281206877135470E-012
failed twice !!!

indicates that the chemistry could not converge to a solution in the given grid box. Possible reasons for this could be:

  1. A particular tracer has numerically underflowed or overflowed. This can happen especially in the aerosol chemistry and equilibrium routines, where many exponentials and logarithms are used in the algorithms.
  2. The restart file is not appropriate for the given simulation. For example, if the restart file was created using the Synoz O3 flux boundary condtion, but you have turned on the Linoz stratospheric O3 chemistry, then this mismatch can cause the solver not to converge. You can try switching to a restart file generated from a simulation with the same input options as the simulation that you wish to perform.

You may have to manually adjust the convergence criteria in the GEOS-Chem code to fix this condition.

--Bob Y. 11:30, 9 November 2010 (EST)

Permission denied error

If you receive this error:

Thu Nov  4 11:03:57 EDT 2010
run.geos: Permission denied. 

after having submitted a run script to a queue system (such as SGE), then doublecheck the Unix permissions of your script. If the script does not have the Unix "execute" permission then the queue system will not be able to run it.

Use the Unix chmod command to make your script executable

chmod 755 run.geos

and then re-submit the script to the queue system.

--Bob Y. 10:26, 9 November 2010 (EST)

Error computing F_OF_PBL

NOTE: The same error described below can cause GEOS-Chem to die elsewhere, and not just in the F_OF_PBL computation.

If you should encounter this error (which occurs in routine COMPUTE_PBL_HEIGHT from pbl_mix_mod.f)

 bad sum at:            1          70  -1.00000000000000     
 bad sum at:            1          81  -1.00000000000000     
===============================================================================
GEOS-CHEM ERROR: Error in computing F_OF_PBL!
STOP at COMPUTE_PBL_HEIGHT ("pbl_mix_mod.f")
===============================================================================
  
===============================================================================
GEOS-CHEM ERROR: Error in computing F_OF_PBL!
STOP at COMPUTE_PBL_HEIGHT ("pbl_mix_mod.f")
===============================================================================
     - CLEANUP: deallocating arrays now...
 bad sum at:            1          48  -1.00000000000000     
 bad sum at:            1          59  -1.00000000000000     
 bad sum at:           73          46  -1.00000000000000     

you should check to see if your code was not recompiled cleanly. Often this is a result of not doing a

make realclean

before trying to switch from 2° x 2.5° to 4° x 5° or vice versa in the define.h header file. Often this type of error can be deduced by looking at the GEOS-Chem log file. When this error occurs the resolution indicated at the top of the GEOS-Chem log file, e.g.:

*************   S T A R T I N G   2 x 2.5   G E O S--C H E M   *************

will not match the longitudes and latitudes as reported a little further down in the log file:

Grid box longitude centers [degrees]: 
-180.000 -175.000 -170.000 -165.000 -160.000 -155.000 -150.000 -145.000
 ...

Grid box latitude centers [degrees]:
 -89.000  -86.000  -82.000  -78.000  -74.000  -70.000  -66.000  -62.000 
 ... 

--Bob Y. 12:13, 25 October 2010 (EDT)

Severe(174) SIGSEGV error

If you compiled GEOS-Chem with the IFORT compiler, you may encounter the following error message:

forrtl: severe (174): SIGSEGV, segmentation fault occurred

Most often, this indicates an array out-of-bounds condition. To find out more information about where this error is occurring, recompile GEOS-Chem with the following Makefile options:

make realclean
make BOUNDS=yes TRACEBACK=yes

The BOUNDS=yes option will turn on Array Out-of-Bounds error checking. The TRACEBACK=yes option will print out the Error Stack, or a list of routines that were called, and the line at which the error occurred. These options will provide more detailed error output.

After recompiling, you should receive an error message such as:

forrtl: severe (408): fort: (3): Subscript #1 of the array PBL_THICK has value -1000000 which is less than the lower bound of 1

This tells you that there is a problem with a certain array. Use the Unix grep command to search for all instances of this array in the GEOS-Chem source code:

grep -i PBL_THICK *.f*

and search for the problem.

NOTE: In the above example, we manually forced an out-of-bounds error with this line of code:

        !### FORCE OOB error for testing
        PBL_THICK(-1000000,J)   = BLTHIK

Removing this line will fix the error.

--Bob Y. 13:31, 10 November 2010 (EST)

Segmentation fault encountered after TPCORE initialization

On the IFORT Compiler, there is a known issue with the the glibc library that can make your code appear to have run out of memory even though there is plenty of memory available. The symptom is that your code will die with a segmentation fault right after the following text is printed:

NASA-GSFC Tracer Transport Module successfully initialized

This error may not occur if you are running GEOS-Chem at 4 x 5 resolution. Users have reported that this error typically occurs when running 2 x 2.5 simulation.

The workaround is to reset your stacksize limit to a large positive number instead of "unlimited". See the following discussion on our Intel Fortran Compiler wiki page.

--Bob Y. 10:55, 27 October 2010 (EDT)

Negative tracer found in WETDEP

If your simulation encounters negative (or NaN) tracer concentrations in the WETDEP routine, then this can be an indication of a problem further upsteam, perhaps in the aerosol routines (highly probable if the tracer is SO4, SO4s, HNO3, SO2, or NH3). We have fixed some of these bugs by making the code more robust. Patches are at ftp://ftp.as.harvard.edu/pub/geos-chem/patches/v8-01-01/. Please see the following links for more information:

If the fixes above do not solve your problem, you will need to debug. The first step is to use few calls to CHECK_STT (from tracer_mod.f) to isolate the part of the code where negative tracers are created. This can be done quite fast if the code dies early enough in the run.

Fatal error in IFORT

The following error, which resulted on the Altix platform using Intel "ifort" v9.1 compiler:

 ifort: error: /opt/intel/fc/9.0/bin/fpp: core dumped
 ifort: error: Fatal error in /opt/intel/fc/9.0/bin/fpp, 
 terminated by unknown signal(139)
 make: *** [transport_mod.o] Error 1

was caused by an omitted " in an #include declaration, i.e.

 #     include "CMN_DIAG

Adding the closing " fixed the problem.

Too many levels in photolysis code

Please see this discussion about the "Too many levels in photolysis code" error that can sometimes happen in the FAST-J photolysis code.

--Bob Y. 11:09, 12 January 2010 (EST)

General types of errors

Exit Status

If you submit a job to the PBS Queue, the exit status will tell you a little bit about what happened. The exit status code is in the email that you will get back from the PBS server:

 PBS Job Id: 6015.altix
 Job Name:   test.sh
 Execution terminated
 Exit_status=1
 resources_used.cpupercent=0
 resources_used.cput=00:00:00
 resources_used.mem=2304kb
 resources_used.ncpus=1
 resources_used.vmem=4544kb
 resources_used.walltime=00:00:00

Jobs that finish normally should have an exit status of 0. In the above example, we have an exit status of 1, which means that the job encountered some kind of error (either a compilation error or a run-time error).

A common exit status number is 143. This means that your job has exceeded the wall-clock time limit of the queue. The solution is to restart your job in a queue with a longer time limit.

Segmentation Fault

A segmentation fault may mean:

  • You have gone outside the declared bounds of an array
  • You are trying to access an ALLOCATABLE array that hasn't been allocated yet
  • You are trying to read data from a file into an array of the wrong dimensions

If you encounter a segmentation fault error, recompile GEOS-Chem with the following options:

make realclean
make BOUNDS=yes TRACEBACK=yes

The BOUNDS=yes option will turn on Array out-of-bounds error checking. The TRACEBACK=yes option will print out the Error stack, or a list of routines that were called, and the line at which the error occurred.

Example 1

Here is segmentation fault caused by the code trying read data from a file into an array which is too small to contain the data. This error was detected on the Altix platform. The output comes from the IDB debugger.

 Thread received signal SEGV
 stopped at [<opaque> for_read_seq_xmit(...) 0x40000000006b6500] 
 
 Information:  An <opaque> type was presented during execution of 
 the previous command.  For complete type information on this symbol,
 recompilation of the program will be necessary.  Consult the compiler
 man pages for details on producing full symbol table information using   
 the '-g' (and '-gall' for cxx) flags.

Example 2

The same error message as in Example 1 has also been known to occur on Altix when a variable that has not been initialized is passed to a subroutine or function:

 CALL MYSUB( I )   ! I is uninitialized!
 DO I = 1, IIPAR 
    ... 
 ENDDO

The solution to the problem is as follows:

 DO I = 1, IIPAR 
    CALL MYSUB( I )  ! Put the MYSUB call in the I-loop!
    ... 
 ENDDO

Example 3

A segmentation fault may indicate array out-of-bounds error (i.e. an array index is pointing outside of the upper or lower bound a given dimension). This condition may manifest itself with an error message such as SIGSEGV error. (The exact error message will depend on which compiler you are using.

If you suspect an out-of-bounds error, recompile with the following Makefile options:

  make realclean
  make BOUNDS=yes TRACEBACK=yes

The BOUNDS=yes option will turn on Array out-of-bounds error checking. The TRACEBACK=yes option will print out the Error stack, or a list of routines that were called, and the line at which the error occurred.

Using these options should give you a more detailed error message, from which you should be able to deduce the source of teh error.

Bus Error

A bus error usually means that you are trying to call a subroutine with the wrong number of arguments.

On SGI, you usually will only get a bus error if you call a subroutine with too many, rather than too few arguments. Other behaviors are platform and compiler dependent.

--Bob Y. 13:01, 4 November 2010 (EDT)