CMAQv5.4 stops after the first hour is written to output files, while code seems to run in an infinite loop

CMAQ
1

Greetings:
CMAQv5.4 has been running without issues for first PM2.5 modeling case. But when another PM2.5 case is used for a similar simulation, I got this problem of CCTM immediately stopping after a few seconds of run only giving the first hour output, but without any notable error message. The interesting part is the 12 processors are busy with the “CCTM_v54.exe” occupying all the memory of the server in an endless loop, as if it is executing the actual job.
I have even tried editing a file from the VDIFF subdirectory per one modeler’s tip, following his quote here=> {{{I finally found that CCTM is running instead of halting but never reaching to the convergence criteria, and then CCTM actually gets into a infinite loop at Line 281, module SEDIMENTATION.F dealing with sedimentation of aerosols:
281 do while ( adts .lt. dtsec )
I changed this line into the following after I referred to the same module in CMAQ 5.3,
281 do while ( adts .lt. dtsec .and. iter .le. 20). ! you can test different criteria (10, 20, …)
282 iter = iter + 1.}}}

But there is no change any change at all. I have tried to attach the CCTM-logs and *.csh file, but this page refused to allow me saying “new users can’t upload file”.
Please, your help is highly requested.
Thank you,
Tes

2

Tes,
Does this second modeling case use different emissions, different meteorology, or both?
In giving the first hour of output, do you mean the hour that is a copy of the initial conditions, or do you mean the model has performed calculations for an hour of simulation?
The first thing I would try is compiling in debug mode (set Debug_cctm in bldit_cctm.csh) and seeing if that leads to a crash. Sometimes the model can hang due to a data error.

3

Thank you for the quick reply, cgnolte!
Yes, they are different cases in time, 2013 vs 2019 (different in emis & met). Last night, before I left my desk I loaded the job with the following command:
/] nohup ./run_cctm.csh >& cctm_Oct2019nudg.log &
And I am very surprised to see this morning the model is actually running normally and producing outputs for 8 days already (it is supposed to run for a month and half). In the past several days, I didn’t use the “nohup - run a command immune to hangups” Unix command. Perhaps, the “&” symbol at the end of the command without the “nohup” was suspending/killing the job execution before.
I will update if I encounter a different/new situation.
I wish you a wonderful weekend!
Tes