replace esmf_logmsg_error with calls to shr_abort #532
Conversation
There was a problem hiding this comment.
@jedwards4b thanks so much for taking this on! This is great to see come in quickly.
I saw a couple things to handle and some questions that I bring up. I also saw some additional instances of this pattern to handle. I think doing...
git grep ESMF_LOGMSG_ERRORand removing all instances of that is the best way to go. In some cases it's now just in "use" statements, but there is no reason to keep it around there either.
|
I'm personally happy with this change, though want to get feedback from others in the ESMF team tomorrow in case someone sees issues with this that I may not be seeing. One thing I'm wondering is whether it's safe to introduce this dependence on shr_sys_mod - for the sake of systems other than CESM. I see that some other uses of shr modules are protected by ifdefs, though not all, so I'm not sure what the general rule is. The one pre-existing use of shr_sys_abort was actually one that I added in #471 , but I suspect I didn't think it through carefully at the time. |
|
@billsacks this (actually the same changes in cdeps) has already been reviewed and tested by @DeniseWorthen. So the changes are fine for UFS. |
Ah, sounds great - thanks! |
|
@ekluzek This is ready to merge now with your review. |
|
OK, approved and ready to launch! |
Description of changes
Replace calls to ESMF_LogWrite that preceed an error exit with calls to shr_sys_abort which will include the error message in both the ESMF pet file and
the component log for the appropriate component. This requires an update to the cesm_share code as well.
Specific notes
Fixes #527
Fixes #531
Fixes #455
Contributors other than yourself, if any:
CMEPS Issues Fixed (include github issue #):
Are changes expected to change answers? (specify if bfb, different at roundoff, more substantial)
Any User Interface Changes (namelist or namelist defaults changes)?
Testing performed
Please describe the tests along with the target model and machine(s)
If possible, please also added hashes that were used in the testing