Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

Error when running PTS_MODE

C

chen1718@umn_edu

Member
Dear all:

I was testing a single PTS_MODE. Here is the commend I created the case:

./create_newcase -case testsingle -res f19_g16 -compset I -mach generic_linux_pgi -scratchroot /($RUNDIR) -din_loc_root_csmdata /($DATAROOT) -pts_lat 40.0 -pts_lon -105 -max_tasks_per_node 1

After configure and build, I submitted the job using qsub. And I got this error message in lnd.log.*


Successfully initialized variables for accumulation

reading restart file
I2000CN_f19_g16_c100503.clm2.r.0001-01-01-00000.nc
Reading restart dataset
ERROR - setlatlon.F:Cant get variable dim for lat or lsmlat
ENDRUN: called without a message string

Where is the file setlatlon.F? Do I need to specify variable dimension?

Besides, I wonder with compet I (which is used in the User's Guide and has no specified year), why it is using a restart file I2000CN_f19_g16_c100503.clm2.r.0001-01-01-00000.nc ?

Thanks!
 
slevis

slevis

Moderator
Staff member
Try starting with arbitrary initial conditions (finidat=' ') to see if the problem is with the restart file.

Sam

airscolor said:
Dear all:

I was testing a single PTS_MODE. Here is the commend I created the case:

./create_newcase -case testsingle -res f19_g16 -compset I -mach generic_linux_pgi -scratchroot /($RUNDIR) -din_loc_root_csmdata /($DATAROOT) -pts_lat 40.0 -pts_lon -105 -max_tasks_per_node 1

After configure and build, I submitted the job using qsub. And I got this error message in lnd.log.*


Successfully initialized variables for accumulation

reading restart file
I2000CN_f19_g16_c100503.clm2.r.0001-01-01-00000.nc
Reading restart dataset
ERROR - setlatlon.F:Cant get variable dim for lat or lsmlat
ENDRUN: called without a message string

Where is the file setlatlon.F? Do I need to specify variable dimension?

Besides, I wonder with compet I (which is used in the User's Guide and has no specified year), why it is using a restart file I2000CN_f19_g16_c100503.clm2.r.0001-01-01-00000.nc ?

Thanks!
Click to expand...
 
C

chen1718@umn_edu

Member
Thanks a lot, Sam. Could you specify where to add this arbitrary initial condition? I tried to search finidat in the case directory, but did not find it.

Besides, I found that in the file env_conf.xml, "RUN_TYPE" was set to "hybrid", I changed it to "startup". When building it, it downloaded another file: clmi.BCN.2000-01-01_1.9x2.5_gx1v6_simyr2000_c100309.nc
But after I submitted the job I got similar error:

Successfully initialized variables for accumulation

reading restart file
/data/04/chen1718/inputdata/CESM/lnd/clm2/initdata/clmi.BCN.2000-01-01_1.9x2.5_gx1v6_simyr2000_c100309.nc
Reading restart dataset
ERROR - setlatlon.F:Cant get variable dim for lat or lsmlat
ENDRUN: called without a message string

That's what I can find right now. If I figure out how to add the non-inidat condition, I would like to test. Thanks!
 
erik

erik

Erik Kluzek
CSEG and Liaisons
Staff member
It turns out this is a Known problem that is in the KnownBugs file. See the KnownBugs file in your distribution (under models/lnd/clm/doc/KnownBugs) or see the documentation on the web under:

http://www.cesm.ucar.edu/models/cesm1.0/clm/models/lnd/clm/doc/KnownBugs

Here's the description there:


Bug Number: 1017 and 1025
PTS_MODE can NOT restart or use a global finidat file

Single column mode (or PTS_MODE turned on using the -pts_lat and -pts_lon options in
scripts/create_newcase) can NOT read restart files or global finidat initial condition
files.

Here's the message that happens when you try to run from a restart file.

(OPNFIL): Successfully opened file ./rpointer.lnd on unit= 14
Reading restart data.....
------------------------------------------------------------
(GETFIL): attempting to find local file
c40b19+.I.pts.01.clm2.r.0001-01-06-00000.nc
(GETFIL): using c40b19+.I.pts.01.clm2.r.0001-01-06-00000.nc in current working
directory
Reading restart dataset
ERROR - setlatlon.F:Cant get variable dim for lat or lsmlat
ENDRUN: called without a message string
 
erik

erik

Erik Kluzek
CSEG and Liaisons
Staff member
If you need to use restarts you'll need to create datasets to run with rather than using global datasets and PTSMODE. Note, that with PTSMODE you can start from arbitrary initial conditions and run for a very long time. And if you set it up with USE_MPISERIAL you can run interactively, and won't have time limits from the queue. Which is why you might not need to have restart files. Oh, by the way, to be clear, you can't run from a restart you create in the middle of your simulation either. No restart or initial files work with PTSMODE.

But, one way to create datasets is to run the getregional script to give you files you can use. Instructions to do that are in the UsersGuide...

http://www.cesm.ucar.edu/models/ces...UsersGuide/x8335.html#getregional_datasets.pl
 
C

chen1718@umn_edu

Member
Erik:

Thanks a lot for your help. I modified the model to start from an arbitrary initial condition by modifying the file env_conf.xml:

!changed to "startup"
!changed to "off"

After submitting the job, I checked lnd.in in $RUNDIR/run. Here is the name list:

co2_ppmv = 368.9
co2_type = 'constant'
dtime = 1800
fatmgrid = '$DATAROOT/lnd/clm2/griddata/griddata_1.9x2.5_060404.nc'
fatmlndfrc = '$DATAROOT/lnd/clm2/griddata/fracdata_1.9x2.5_gx1v6_c090206.nc'
finidat = ' '
fpftcon = '$DATAROOT/lnd/clm2/pftdata/pft-physiology.c100226'
fsnowaging = '$DATAROOT/lnd/clm2/snicardata/snicar_drdt_bst_fit_60_c070416.nc'
fsnowoptics = '$DATAROOT/lnd/clm2/snicardata/snicar_optics_5bnd_c090915.nc'
fsurdat = '$DATAROOT/lnd/clm2/surfdata/surfdata_1.9x2.5_simyr2000_c091005.nc'
hist_crtinic = 'NONE'
hist_mfilt = 1
hist_nhtfrq = 0
outnc_large_files = .true.
urban_hac = 'ON_WASTEHEAT'
urban_traffic = .false.

This is the end of the output log file：

(seq_mct_drv): =============== SUCCESSFUL TERMINATION OF CPL7-CCSM ===============
cp: No match.
Archiving ccsm output to UNSET
Calling the short-term archiving script st_archive.sh

st_archive.sh: start of short-term archiving
st_archive.sh: restart files from end of run will be saved,
interim restart files will be deleted
st_archive.sh: short-term archiving completed successfully
mv: cannot stat `$CASEROOT/timing/ccsm_timing.testPTS_MODE.100721-152227': No such file or directory

While the file
"/$CASEROOT/timing/ccsm_timing.testPTS_MODE.100721-152227" does exist.

I attached the whole log file from the latest run in case you would like to have a look. Thanks again for your time and help!
 
erik

erik

Erik Kluzek
CSEG and Liaisons
Staff member
Well it ran fine, but it had trouble moving the timing file. I think it moves it to the log sub-directory in your case directory. Presumably this has to do with some type of permission problem. Technically it could be a filesystem problem, and a disk failure as well.

If you don't need the timing file -- I wouldn't worry about it and continue with your work. If you need the timing file -- you could just try to copy it by hand.
 
C

chen1718@umn_edu

Member
Hi, Erik:

Thanks a lot. Now I can run the PTS_MODE and parallel mode (global) with the compset I_2000. But after one year of simulation, I only got one output history file at the end of the simulation. Here is the file clm.buildnml.csh under $CASEROOT/Buildconf/:

&clm_inparm
co2_ppmv = 368.9
co2_type = 'constant'
dtime = 1800
fatmgrid = '$DIN_LOC_ROOT/lnd/clm2/griddata/griddata_1.9x2.5_060404.nc'
fatmlndfrc = '$DIN_LOC_ROOT/lnd/clm2/griddata/fracdata_1.9x2.5_gx1v6_c090206.nc'
finidat = ' '
fpftcon = '$DIN_LOC_ROOT/lnd/clm2/pftdata/pft-physiology.c100226'
fsnowaging = '$DIN_LOC_ROOT/lnd/clm2/snicardata/snicar_drdt_bst_fit_60_c070416.nc'
fsnowoptics = '$DIN_LOC_ROOT/lnd/clm2/snicardata/snicar_optics_5bnd_c090915.nc'
fsurdat = '$DIN_LOC_ROOT/lnd/clm2/surfdata/surfdata_1.9x2.5_simyr2000_c091005.nc'
hist_crtinic = 'NONE'
hist_mfilt = 1
hist_nhtfrq = 0
outnc_large_files = .true.
urban_hac = 'ON_WASTEHEAT'
urban_traffic = .false.
/


I think hist_nhtfrq=0 should lead to monthly output history file? Please correct me if I'm wrong or if I should modify anything else. Thanks!

Ming
 
slevis

slevis

Moderator
Staff member
Ming,

You are right, you should get 12 monthly files. Do you get only one file in the global run, too?

If not, then try changing hist_mfilt to 12 in the PTS_MODE run so that all months get put in one file.

Sam
 
C

chen1718@umn_edu

Member
OK...I modified the file clm.buildnml.csh in the directory $CASEROOT/Buildconf as follows:

hist_mfilt = 12

Then I cleaned build, and rebuild.
./*.clean.build
./*.build
After that I resubmitted the job
qsub *.run

And I still got one history file with one time dimension:

netcdf testPTS_MODE.clm2.h0.0002-01 { // format variant: 64bit
dimensions:
gridcell = 1 ;
landunit = 2 ;
column = 6 ;
pft = 22 ;
levgrnd = 15 ;
levlak = 10 ;
numrad = 2 ;
lon = 1 ;
lat = 1 ;
lonatm = 1 ;
latatm = 1 ;
time = UNLIMITED ; // (1 currently)
hist_interval = 2 ;
string_length = 8 ;
Now I really have no idea what could possibly cause this...
 
C

chen1718@umn_edu

Member
Dear Sam:

I was not looking at the right output directory- and there is one last output file that was not moved to the output directory. So I take that as the only output file... There is no problem now.

Ming
 
slevis

slevis

Moderator
Staff member
Dear Ming,

I'm glad that your problem was as simple as not knowing the existence of the archive directory. I recommend reading through all the cesm documentation, so that you may familiarize yourself with the model. If you still encounter problems after that, you may use this bulletin board to ask questions.

Sam
 
You must log in or register to reply here.
Top