Problem in running a single point

Status
Not open for further replies.

MikeYuan

tfYuan
Member
Hello, I am running CLM5 offline of CESM2.1.2 for a single point in Brazil. I choose the compset IHistClm50BgcQianGs(

res 1x1_brazil) and run, but it didn’t work and show the error.

------------------------------error message below--------------------------------------------------

ERROR: (shr_tInterp_getCosz) ERROR: lon lat cosz sizes disagree

------------------------------------------------------------------------------------------------------------


Related file are attached below, any suggestions to diagnose the problem will be really helpful.

Thank you.
Mike
 

Attachments

erik

Erik Kluzek
CSEG and Liaisons
Staff member
OK, so it's dying in the datm. I suggest looking at the datm source code and looking for that error condition and see what it describes the problem as. I do also suggest updating to the latest CESM2.1 version, which isn't likely to solve your issue, but that's the best code base to work from. We did test with the IHistClm50BgcQianGs compset. You should make sure you are only using a single processor and also using mpi-serial for MPILIB. It's also always good to try simple cases out of the box to get something that runs to compare to the cases that you actually want to run.
 

erik

Erik Kluzek
CSEG and Liaisons
Staff member
Doing

Code:
git grep "ERROR: lon lat cosz sizes disagree"

under cime you should be able to see the code that gives the error condition.
 

MikeYuan

tfYuan
Member
OK, so it's dying in the datm. I suggest looking at the datm source code and looking for that error condition and see what it describes the problem as. I do also suggest updating to the latest CESM2.1 version, which isn't likely to solve your issue, but that's the best code base to work from. We did test with the IHistClm50BgcQianGs compset. You should make sure you are only using a single processor and also using mpi-serial for MPILIB. It's also always good to try simple cases out of the box to get something that runs to compare to the cases that you actually want to run.
Thanks for your suggestion,Mr.Erik. It worked normally when I use a single processor to run it.
I appreciate it.
 
Status
Not open for further replies.
Back
Top