Problems encountered after setting 'CONTINUE_RUN=TRUE'

[zjulin]

Hi, there was an error while I was running a program.

It was interrupted while running the BHIST compset program, and then I set CONTINUE_RUN=TRUE in env_run.xml, with no other settings changed. But when the program was re-run for a month, there was an error called 'WARNING: Rearr optional argument is a pio2 feature, ignored in pio1'.

I don't know how to solve the problem. It appears to be an error when writing to an 'nc file' called 'myBHIST.cam.h0.1870-01.nc'. I will attach my log files for your perusal.

Sincerely hope to get your help! Best wishes!

 

[katec]

Hi, there was an error while I was running a program.

It was interrupted while running the BHIST compset program, and then I set CONTINUE_RUN=TRUE in env_run.xml, with no other settings changed. But when the program was re-run for a month, there was an error called 'WARNING: Rearr optional argument is a pio2 feature, ignored in pio1'.

I don't know how to solve the problem. It appears to be an error when writing to an 'nc file' called 'myBHIST.cam.h0.1870-01.nc'. I will attach my log files for your perusal.

Sincerely hope to get your help! Best wishes!

Hi, this is kind of difficult, because there's not a ton of details about what you are doing with this run. But looking at the log files, it's the errors before that are more important. The
NetCDF: Variable not found
NetCDF: Invalid dimension ID or name
Errors are usually the result of one of two issues. Either you are trying to read beyond the actual data on a forcing file or trying to write out a bad field. It seems more like the latter here to me. As if the model is crashing and writing bad values to the NetCDF history file. You may have corrupted restart files. Can you try restarting from an earlier set of restart files?