Additional fixes for downstream jobs

[DavidHuber-NOAA]

Description

This fixes a number of issues with the postsnd and arch jobs discovered during testing of HR4.

Resolves #3019

Type of change

• Bug fix (fixes something broken)

Change characteristics

• Is this a breaking change (a change in existing functionality)? NO
• Does this change require a documentation update? NO
• Does this change require an update to any of the following submodules? NO

How has this been tested?

• Extended test on Hera
• Extended test on WCOSS2

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• My changes generate no new warnings
• New and existing tests pass with my changes
• This change is covered by an existing CI test or a new one has been added
  • Portions of this change are covered by existing CI, but others are not (e.g. archiving, which is disabled on WCOSS2)
  • Updates will need to be made to run the AWIPS jobs in the existing C96_atm3DVar_extended test

[DavidHuber-NOAA]

All post jobs now run for the C96_atm3dvar_extended test when DO_AWIPS=YES on WCOSS2. I have thus enabled AWIPS for this test to ensure the AWIPS jobs and linked downstream jobs are always run to avoid lapses in support.

The AWIPS jobs still cannot be enabled on other machines due to the fbwinds executable which currently only runs on WCOSS2. That will be an issue for another time.

Also, during this testing, I found that the MOS jobs, when enabled, do not run on WCOSS2. I opened issue #3196 to look into this, but I believe the MOS package will need an overhaul to work with the new COM structure.

Marking this PR ready for review.

[aerorahul]

Looks good to me. Just one observation. Approve.

[aerorahul]

lgtm.

[aerorahul]

@WalterKolczynski-NOAA
Can you please run the ci on wcoss2?

[JessicaMeixner-NOAA]

@JessicaMeixner-NOAA Thank you for tagging me on this issue. Following our conversation you mentioned you will add those points to the file. Am I missing anything?

@AminIlia-NOAA - This is unrelated to the point list changing. This is fixing a bug so that the expected output points are produced when using an unstructured grid. The output here: /lfs/h2/emc/couple/noscrub/jessica.meixner/GWPR3190upwav/WaveProdTest01/COMROOT/WaveProdTest01/gfs.20200913/00/products/wave/station uses a WW3 with the bug fix so the upstream issue is resolved and the downstream point jobs should succeed.

[AminIlia-NOAA]

@JessicaMeixner-NOAA Thank you for the clarification. I'm a little behind on the wave side, however, please let me know if I can help with anything.

[DavidHuber-NOAA]

@RuiyuSun I've finished fixing the wavegempak and waveawipsgridded jobs and removed the a couple of jobs from the forecast-only configuration. I believe all jobs should now run with the one exception of waveawipsbulls job. I will continue on with this job based on Jessica's input directory, but it will take some time before the job will run correctly in the workflow as we will need a new UFS hash. Thus, I am marking this PR ready for review.

[RuiyuSun]

Thanks @DavidHuber-NOAA. I will make a HRv4 like experiment to test this.

[JessicaMeixner-NOAA]

Thanks @DavidHuber-NOAA. I will make a HRv4 like experiment to test this.

@RuiyuSun if you need the wave post jobs to work - you'll need a branch of ufs - I can give that to you if you'dlike

[RuiyuSun]

Thanks @DavidHuber-NOAA. I will make a HRv4 like experiment to test this.

@RuiyuSun if you need the wave post jobs to work - you'll need a branch of ufs - I can give that to you if you'dlike

Sure. Thanks @JessicaMeixner-NOAA

[DavidHuber-NOAA]

@aerorahul Who should I make codeowner(s) of these new Perl/bash scripts moved from gfs-utils?

[aerorahul]

I have no idea. @yangfanglin Do you know where these came from:

ush/make_NTC_file.pl
ush/make_ntc_bull.pl
ush/make_tif.sh
ush/month_name.sh

[DavidHuber-NOAA]

@JessicaMeixner-NOAA Thanks for the heads up! I will be happy to jump back on this when the test case is available.

@DavidHuber-NOAA @AminIlia-NOAA This has the point output:

/lfs/h2/emc/couple/noscrub/jessica.meixner/GWPR3190upwav/WaveProdTest01/COMROOT/WaveProdTest01/gfs.20200913/00/products/wave/station That should contain the info that you need for the wave product jobs. This doesn't have the boundary pionts, which was an oversight on my part, but I don't think those are needed for any downstream jobs so that should be okay.

This: /lfs/h2/emc/couple/noscrub/jessica.meixner/GWPR3190upwav/WaveProdTest01/COMROOT/WaveProdTest01/gfs.20200913/00/products/wave/gridded also should have the global.0p50 which is needed for the downstream scripts/exgfs_wave_prdgen_gridded.sh as well.

@JessicaMeixner-NOAA The file /lfs/h2/emc/couple/noscrub/jessica.meixner/GWPR3190upwav/WaveProdTest01/COMROOT/WaveProdTest01/gfs.20200913/00/products/wave/station/gfswave.t00z.cbull_tar appears to also not have some of the buoy data in it that the gfs_waveawipsbulls job is looking for. Running tar -tvf on the file shows, for example, no file for buoy number 46001.

[RuiyuSun]

All CI tests completed successfully. However, I ran a C48_S2SW test case with DO_AWIPS, DO_GEMPAK, DO_GOES, DO_BUFRSND, and DO_NPOESS all set to YES, and discovered a number of failures in gempak and awips jobs. Converting back to draft while I resolve these issues.

@ChristopherHill-NOAA @DavidHuber-NOAA Do we still need DO_GOES? Synthetic GOES should be removed if it haven't been removed upon successful PNS process.

[aerorahul]

code changes look good to me.
Thank you for taking care of downstream product generation jobs.

[aerorahul]

All CI tests completed successfully. However, I ran a C48_S2SW test case with DO_AWIPS, DO_GEMPAK, DO_GOES, DO_BUFRSND, and DO_NPOESS all set to YES, and discovered a number of failures in gempak and awips jobs. Converting back to draft while I resolve these issues.

@ChristopherHill-NOAA @DavidHuber-NOAA Do we still need DO_GOES? Synthetic GOES should be removed if it haven't been removed upon successful PNS process.

@RuiyuSun
I don't think there has been a final decision on synthetic GOES products. There hasn't been a PNS for it that I know of.
@WenMeng-NOAA has there been further discussions on discontinuing synthetic GOES product generation?

[ChristopherHill-NOAA]

All CI tests completed successfully. However, I ran a C48_S2SW test case with DO_AWIPS, DO_GEMPAK, DO_GOES, DO_BUFRSND, and DO_NPOESS all set to YES, and discovered a number of failures in gempak and awips jobs. Converting back to draft while I resolve these issues.

@ChristopherHill-NOAA @DavidHuber-NOAA Do we still need DO_GOES? Synthetic GOES should be removed if it haven't been removed upon successful PNS process.

@RuiyuSun I don't think there has been a final decision on synthetic GOES products. There hasn't been a PNS for it that I know of. @WenMeng-NOAA has there been further discussions on discontinuing synthetic GOES product generation?

@RuiyuSun @aerorahul According to notes I received from Gwen in July 2024, the options of DO_GOES=YES and DO_NPOESS=YES options are intended to be retained with GFSv17. Otherwise, I am not presently aware of the status of synthetic satellite products for GFSv17.