segmentation fault using wrf-chem4.1.3 or 4.4.1

[Lily_Zh]

Hi,

I ran my case using WRF-Chem 3.9.1 on Cheyenne, and now try to same inputs (or newly generated v4.0 input files using modis land use data) to run the new version of wrf-chem to check if it has better performance. However, it always has a segmentation fault around the 18th or 21st hour wrfout file. I attached my namelist and the rsl.error file.

Thank you in advance.
Lily

 

[jordanschnell]

Hi Lily,

I don't immediately see anything from your log or namelist - as for debugging though, first try and increase the "debug_level" to 300, and hopefully that will give us a better idea of where the error is coming from.

Do you see any unrealistic concentrations of aerosol or trace gas species in your output?

As a last step, you can turn off groups of processes to narrow it down (e.g., turn off aero/gas chem, turn off emissions, etc.)

Jordan

 

[Lily_Zh]

Hi Lily,

I don't immediately see anything from your log or namelist - as for debugging though, first try and increase the "debug_level" to 300, and hopefully that will give us a better idea of where the error is coming from.

Do you see any unrealistic concentrations of aerosol or trace gas species in your output?

As a last step, you can turn off groups of processes to narrow it down (e.g., turn off aero/gas chem, turn off emissions, etc.)

Jordan

Hi Jordan,

Thanks for your reply.
I tried to configure with debug mode, it can generate several days' results slowly. Without debug mode, it can just generate several hours' results. I cannot understand the reason. Do you have any idea about that?

Thank you in advance.
Lily

 

[jordanschnell]

Hi Lily,

Don't configure/compile with FCDEBUG, just incerase the debug_level parameter in your namelist.

Do you see any unrealistic concentrations of aerosol or trace gas species in your output?

As a last step, you can turn off groups of processes to narrow it down (e.g., turn off aero/gas chem, turn off emissions, etc.)

Jordan