Hi everyone,
I am attempting to compile NEI2020 using SMOKEv5.0 and am running into issue compiling the onroad_ca_adj sectors, specifically RPD/RPHO/RPP/RPV. Every other sector, including all the other onroad ones, compile successfully.
When running onroad_ca_adj/RPHO, for example. I get the error ‘netCDF error number -40’ in the m3stat step after opening the file and reading in the pollutant variables. I am attaching one of the m3stat logs below as an example.
Untitled.txt (6.7 KB)
Thank you in advance for the help.
From the output of the onroad/RPHO I think the error is somewhere here":
Value for PROMPTFLAG: N returning FALSE
Using default 2020001
Using default 0
***Using default 0
netCDF error number -40
Where the *** should be “Using default 24000” according to onroad/RPHO. But I am unsure how to solve this.
See the I/O API “troubleshooting” page on this topic: https://cjcoats.github.io/ioapi/ERRORS.html#ncferr
I get “netCDF error -40”
Probably means you tried to read data past the last date-and-time on the file (the I/O API runs netCDF in “verbose mode”, so that netCDF will always print all error messages, including this one. Also can happen when the calling program is running in parallel, but a non-MP-enabled version of the I/O API library was linked in.
You do not appear to have any time steps in the RPHO file:
Maximum current record number 0
Can you verify that movesmrg and m3xtract ran correctly for that day?
@james.beidler Yes, I get the program to output the emis_moles files, but there is no data within them. I see the movesmrg log files, but no log files from m3extract. So it seems like none of them ran correctly. I am attaching a log file from movesmrg for reference.
Mainly just confused because the CONUS onroad sectors were totally fine when running them. I would’ve imagined the onroad_ca_adj programs would be almost the same…
movesmrg_RPHO_onroad_ca_adj_jan_2020ha2_cb6_20k_20200101_12US1_cmaq_cb6ae7.txt (1.5 MB)
@cjcoats Thank you for this advice! I think it is an issue with the dates, but I am not sure how to resolve my issue with this advice, though. I am only trying to run a single month (January) and have the dates set correctly in Monthly_onroad_ca_adj_RPHO_daily_12US1_2020ha2_cb6_20k.csh. The CONUS onroad sectors work completely fine. I am not trying to run an MP-enabled I/O API version either. Any further advice would be greatly appreciated! Thank you!
You mentioned that the Movesmrg log stops abruptly without an error message, which we assume means it gave a segmentation fault.
-
You mentioned issues with onroad_ca_adj RPD/RPHO/RPP/RPV; does that mean onroad_ca_adj RPH and RPS worked correctly and produced emissions?
-
Could you provide the January 1 Movesmrg logs for all rates (RPD, RPP, etc) for both onroad and onroad_ca_adj?
@eyth.alison I get error messages and a “Error termination: Backtrace” message as well. The output to the terminal for is full of error messages. I attached that output for onroad_ca_adj/RPV below. They all have the same failure output so one should suffice. I get “cfpro” errors, which are not fatal, and then the fatal errors later on.
ca_adj_RPV.out.txt (227.0 KB)
You are correct, onroad_ca_adj/RPH and RPS work totally fine and produced emission files with non-zero values for all variables. The onroad_ca_adj /RPD script will produce a file for Jan. 1 but all the emission values are zero and stops running after that. The other onroad_ca_asj scripts will produce emissions files for each day, but there are no variables for emissions – just header information mainly. I am attaching all those files below. I changed the names to be easier to follow.
It is almost certainly something to do with the times. It looks like the month variable is getting lost somewhere, but I do not know why.
ca_adj_RPV.out.txt (227.0 KB)
onroad_ca_adj_RPH.txt (868.7 KB)
onroad_ca_adj_RPS.txt (2.8 MB)
onroad_ca_adj_RPV.txt (2.8 MB)
onroad_ca_adj_RPP.txt (3.3 MB)
onroad_ca_adj_RPHO.txt (1.5 MB)
onroad_RPH.txt (1.1 MB)
onroad_RPHO.txt (935.5 KB)
onroad_RPP.txt (1.7 MB)
onroad_RPV.txt (1.1 MB)
onroad_RPS.txt (970.7 KB)
onroad_RPD.txt (2.3 MB)
@eyth.alison I took a look at the environment differences between the onroad_RPHO and onroad_ca_adj_RPHO, and I noticed a difference between the “USE_CONTROL_FACTORS”, where onroad has it set to “N” and onroad_ca_adj has it set to “Y”. I imagine this is the separation between the CFPRO on and off. I changed it to “N” for onroad_ca_adj_RPHO, and that seems to produce the output I needed. If this is indeed turning off the adjustment factors, then it is related to that. If not, then that change worked and I will proceed on generating the other sectors with this same change.
Yes, you are turning off the control factor application when you set that value to N, but good to know that it made output. We are wondering what version of SMOKE you are using? Can you try rerunning with the recently released v5.2? There was a bug fix included that may resolve this issue.
I am using v5.1. I will try to recompile with v5.2 and I will let you know how that goes.