Move mrms_pull_topofhour.py from scripts/ to ush/ [develop]#571
Conversation
…er environment loaded. Fixed a tab issue in the py script as well.
jwolff-ncar
added
enhancement
jwolff-ncar
requested review from
BenjaminBlake-NOAA and
JulieSchramm
as code owners
JeffBeck-NOAA
left a comment
JeffBeck-NOAA
left a comment
There was a problem hiding this comment.
Should ush/load_modules_run_task.sh have changed?
|
As I mentioned in the description of changes, I had to add the set +u to get the environment right for the python script. I had worked with @mkavulich on this so perhaps he wants to also take a look for approval. |
OK, if it's just to ensure the correct environment, it's likely only necessary for debugging, correct? |
| @@ -331,7 +332,6 @@ if [ -n "${SRW_ENV:-}" ] ; then | |||
| conda activate ${SRW_ENV} | |||
| fi | |||
|
|
|||
There was a problem hiding this comment.
If we want to be entirely consistent with previous behavior, we should include a "set -u" line here. But I do not think it's necessary
There was a problem hiding this comment.
I am pretty sure I tested with that and it didn't work so I would say let's not add that if it isn't deemed necessary.
|
@JeffBeck-NOAA The |
|
The way we approached it for the changes we introduced for Jet/Hera in RRFS_dev1 is here starting at line 332. I suggest wrapping only the single line that needs it (and is out of our control) in the |
|
Thanks @mkavulich, @christinaholtNOAA. I'm still a bit confused, since I've not had any problems on Hera with the load_modules_run_task script before this PR. |
|
@JeffBeck-NOAA you would have only run into this problem if you were pulling/processing MRMS data from the HPSS rather than using staged observations. It was not an issue when the capability was first added but noticed about 1.5 months ago so something changed and it is now required. |
|
@mkavulich Were you going to make a change to include the the set +u and set -u updates for me to test? |
… the end to restore the unset variable flag
|
@jwolff-ncar I made the change and tested on Hera, both for the GST and your sample MET case you sent me. Everything is functioning as expected with the updated set flags. |
…unity#571) * Moved mrms_pull_topofhour.py to ush/ rather than scripts/ * Updates to get the mrms_pull_topofhour.py script to run with the proper environment loaded. Fixed a tab issue in the py script as well. * Move "set +u" to the place it is absolutely needed, add a "set -u" at the end to restore the unset variable flag Co-authored-by: Michael Kavulich, Jr <kavulich@ucar.edu>
* Save stdout for EnKF radardbz analysis Remove Jet local directories from config.sh.RRFS_CONUS_3km_ens Remove specification of "NET" from config.sh.RRFS_CONUS_3km and config.sh.RRFS_CONUS_3km_ens, and change the file dependencies of the ensctrl grib2 files for ensemble_maps and enspost * Update exregional_run_prepstart.sh
4 participants
DESCRIPTION OF CHANGES:
Moved the mrms_pull_topofhour.py from scripts/ into ush/.
Made a small change to the script to remove a 'tab' and make it 'spaces' instead.
Added a set +u in the load_modules_run_task.sh to ensure the correct environment for running the python script.
TESTS CONDUCTED:
Tested one case on hera including the verification step. The verification tasks are not yet supported on other platforms.
DEPENDENCIES:
None.
DOCUMENTATION:
No updates needed.
ISSUE (optional):
Fixes issue mentioned in https://github.com/NOAA-EMC/regional_workflow/issues/486
CONTRIBUTORS (optional):