Update GFSv16.2.0 tag to use hpc-stack on Hera and Orion #629

GeorgeGayno-NOAA · 2022-02-08T16:07:30Z

To support the global-workflow group.

Also, unload the first build module at the end of the loop. Fixes #629.

Fixes #629

GeorgeGayno-NOAA · 2022-02-09T16:12:37Z

@KateFriedman-NOAA - Try my hotfix branch at ba7efbd. I don't know what default library versions you plan to use. I can adjust those later.

KateFriedman-NOAA · 2022-02-09T20:26:51Z

@KateFriedman-NOAA - Try my hotfix branch at ba7efbd. I don't know what default library versions you plan to use. I can adjust those later.

@GeorgeGayno-NOAA So global_cycle builds without an error but the emcsfc build fails. If I run the emcsfc build standalone it builds. I looked at the diffs between building it standalone and via global-workflow. I see you are using the hpc-intel/2020.2 module as the default but via global-workflow I builds with hpc-intel/2018.4 (and I override your setting when I build). The g2 library path is missing when I build on Orion because of the mismatch I think. Can you use the 2018 stack on Orion/Hera?

From the new global-workflow $target.ver files:

orion.ver:

export hpc_ver=1.1.0
export hpc_intel_ver=2018.4
export hpc_impi_ver=2018.4

hera.ver:

export hpc_ver=1.1.0
export hpc_intel_ver=18.0.5.274
export hpc_impi_ver=2018.0.4

kgerheiser · 2022-02-09T20:28:07Z

Yes, I would use 2018.4 on Orion.

GeorgeGayno-NOAA · 2022-02-09T20:32:24Z

Yes, I would use 2018.4 on Orion.

Will do.

Fixes #629.

GeorgeGayno-NOAA · 2022-02-11T18:59:26Z

@KateFriedman-NOAA - Try my hotfix branch at ba7efbd. I don't know what default library versions you plan to use. I can adjust those later.

@GeorgeGayno-NOAA So global_cycle builds without an error but the emcsfc build fails. If I run the emcsfc build standalone it builds. I looked at the diffs between building it standalone and via global-workflow. I see you are using the hpc-intel/2020.2 module as the default but via global-workflow I builds with hpc-intel/2018.4 (and I override your setting when I build). The g2 library path is missing when I build on Orion because of the mismatch I think. Can you use the 2018 stack on Orion/Hera?

From the new global-workflow $target.ver files:
orion.ver:

export hpc_ver=1.1.0
export hpc_intel_ver=2018.4
export hpc_impi_ver=2018.4

hera.ver:

export hpc_ver=1.1.0
export hpc_intel_ver=18.0.5.274
export hpc_impi_ver=2018.0.4

@KateFriedman-NOAA Try 3c5a350

KateFriedman-NOAA · 2022-02-15T19:02:38Z

@KateFriedman-NOAA Try 3c5a350

@GeorgeGayno-NOAA Thanks, it builds for me both 1) standalone (as-is) and 2) via global-workflow if I set export libpng_ver=1.6.35 in global-workflow's versions/orion.ver file. Please see these logs and confirm they indeed built ok:

/work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/build_cycle.log
/work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/build_emcsfc.log

Note: @kgerheiser opened NOAA-EMC/hpc-stack#387 to rename the png modules to libpng to match the WCOSS2 install. Once that happens you'll need to update your modulefiles to use libpng instead of png. I'm also waiting for this issue (NOAA-EMC/hpc-stack#388) to get completed and some library versions to get installed too, I'm using alternative versions at the moment just to test build/version file functionality. Given the needed stack updates happening on Hera/Orion we won't make the self-imposed 2/16 deadline to get the components updated. Will hopefully not happen too long after though.

GeorgeGayno-NOAA · 2022-02-15T20:05:42Z

@KateFriedman-NOAA Try 3c5a350

@GeorgeGayno-NOAA Thanks, it builds for me both 1) standalone (as-is) and 2) via global-workflow if I set export libpng_ver=1.6.35 in global-workflow's versions/orion.ver file. Please see these logs and confirm they indeed built ok:

/work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/build_cycle.log /work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/build_emcsfc.log

Kate, I am using 1.6.35 as the default for Orion. Should I use another version as default?

Note: @kgerheiser opened NOAA-EMC/hpc-stack#387 to rename the png modules to libpng to match the WCOSS2 install. Once that happens you'll need to update your modulefiles to use libpng instead of png. I'm also waiting for this issue (NOAA-EMC/hpc-stack#388) to get completed and some library versions to get installed too, I'm using alternative versions at the moment just to test build/version file functionality. Given the needed stack updates happening on Hera/Orion we won't make the self-imposed 2/16 deadline to get the components updated. Will hopefully not happen too long after though.

KateFriedman-NOAA · 2022-02-15T20:12:40Z

Kate, I am using 1.6.35 as the default for Orion. Should I use another version as default?

We're going to be using libpng/1.6.37 once it's available in the stacks on Hera/Orion so the default should be 1.6.37. I'm using the available png/1.6.35 at the moment to test functionality while awaiting the module updates.

build. Use variables from the netcdf module. Fixes #629.

Fixes #629.

previously, but would not run without a system error. This fixes the problem. Fixes #629.

GeorgeGayno-NOAA · 2022-03-04T21:55:08Z

The branch at 423c704 was tested on Hera and Orion using these scripts:

Hera - /scratch1/NCEPDEV/da/George.Gayno/ufs_utils.git/UFS_UTILS.gfsv16
Orion - /work/noaa/da/ggayno/save/ufs_utils.git/gfsv16_tag

All tests passed.

Fixes #629.

GeorgeGayno-NOAA · 2022-03-04T22:33:22Z

The same tests were repeated using Intel 2022 (6195f80). All tests passed.

kgerheiser · 2022-03-04T22:59:14Z

@GeorgeGayno-NOAA no problems with Intel MPI 2022?

GeorgeGayno-NOAA · 2022-03-04T23:08:57Z

@GeorgeGayno-NOAA no problems with Intel MPI 2022?

No. The global_cycle program uses MPI. And it ran fine for me.

GeorgeGayno-NOAA · 2022-03-07T16:20:29Z

@KateFriedman-NOAA Would you like to try 6195f80? It compiles and runs. But I am not sure I am using the correct library and compiler versions (or environment variable names).

KateFriedman-NOAA · 2022-03-07T16:27:39Z

@KateFriedman-NOAA Would you like to try 6195f80? It compiles and runs. But I am not sure I am using the correct library and compiler versions (or environment variable names).

Yes, thanks @GeorgeGayno-NOAA ! I'll check it out on Hera and Orion and report back.

KateFriedman-NOAA · 2022-03-07T16:41:20Z

@GeorgeGayno-NOAA The updated hash built without error on both Hera and Orion with the version file overrides in play. Please see the following logs to confirm the build success within g-w:

Hera: /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*
Orion: /work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

The g-w version settings used (build.ver & hera.ver/orion.ver):

Hera: /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/versions/
Orion: /work/noaa/global/kfriedma/git/dev_v16/versions/

GeorgeGayno-NOAA · 2022-03-07T19:31:22Z

@GeorgeGayno-NOAA The updated hash built without error on both Hera and Orion with the version file overrides in play. Please see the following logs to confirm the build success within g-w:

Hera: /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/* Orion: /work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

The g-w version settings used (build.ver & hera.ver/orion.ver):

Hera: /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/versions/ Orion: /work/noaa/global/kfriedma/git/dev_v16/versions/

Looks like it worked. And I am using the same library/compiler versions as you are.

KateFriedman-NOAA · 2022-03-07T19:35:33Z

Looks like it worked. And I am using the same library/compiler versions as you are.

Great, thanks for confirming @GeorgeGayno-NOAA ! Once I get a few more of the updated components available I will test this updated UFS_UTILS branch execs in some cycled tests of the system on Hera/Orion. I'll report any issues regarding those execs. Hoping do that in the next week or two, need to wrap this up this month before the old modules get removed. Thanks!

Fixes #629.

GeorgeGayno-NOAA · 2022-03-10T17:02:35Z

@KateFriedman-NOAA I reverted back to Intel 2018. I also removed some unused utilities. See 75c74cd.

KateFriedman-NOAA · 2022-03-10T18:09:53Z

I reverted back to Intel 2018. I also removed some unused utilities. See 75c74cd.

Thanks @GeorgeGayno-NOAA ! Need to work on a WCOSS2 issue today so I'll try to test that hash tomorrow.

KateFriedman-NOAA · 2022-03-11T15:22:34Z

I updated my ufs_utils copies on Hera/Orion, reverted the g-w hera.ver back to the 2018 intel, and ran the ufs_utils build. Appear to have built without issue, please confirm, thanks:

Hera: /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*
Orion: /work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

GeorgeGayno-NOAA · 2022-03-11T19:26:55Z

I updated my ufs_utils copies on Hera/Orion, reverted the g-w hera.ver back to the 2018 intel, and ran the ufs_utils build. Appear to have built without issue, please confirm, thanks:

Hera: /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*
Orion: /work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

The Orion build looks good. On Hera, the global_cycle build used 2018, but the emcsfc build used 2022. That is odd.

KateFriedman-NOAA · 2022-03-15T16:35:25Z

@GeorgeGayno-NOAA I found a couple issues in the modulefiles for global_cycle and emcsfc (for Hera, didn't check others yet) but after fixing them I ran into another issue (the real issue).

A couple of the version variables in the following modulefiles were incorrect. I fixed them before hitting the next error I'll describe below:

-bash-4.2$ git diff modulefiles/*
diff --git a/modulefiles/fv3gfs/global_cycle.hera.lua b/modulefiles/fv3gfs/global_cycle.hera.lua
index 49fcf0d5..9c515f49 100644
--- a/modulefiles/fv3gfs/global_cycle.hera.lua
+++ b/modulefiles/fv3gfs/global_cycle.hera.lua
@@ -10,7 +10,7 @@ load(pathJoin("hpc", hpc_ver))
 hpc_intel_ver=os.getenv("hpc_intel_ver") or "18.0.5.274"
 load(pathJoin("hpc-intel", hpc_intel_ver))
 
-hpc_intel_ver=os.getenv("hpc_impi_ver") or "2018.0.4"
+hpc_impi_ver=os.getenv("hpc_impi_ver") or "2018.0.4"
 load(pathJoin("hpc-impi", hpc_impi_ver))
 
 ip_ver=os.getenv("ip_ver") or "3.3.3"
diff --git a/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua b/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua
index 140c759e..b0574ee0 100644
--- a/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua
+++ b/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua
@@ -7,7 +7,7 @@ prepend_path("MODULEPATH", "/scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gf
 hpc_ver=os.getenv("hpc_ver") or "1.2.0"
 load(pathJoin("hpc", hpc_ver))
 
-hpc_ver=os.getenv("hpc_intel_ver") or "18.0.5.274"
+hpc_intel_ver=os.getenv("hpc_intel_ver") or "18.0.5.274"
 load(pathJoin("hpc-intel", hpc_intel_ver))

Not sure if you also need to load hpc-impi for the emcsfc builds, I didn't add that yet given the other error I hit.

When attempting to load hpc-intel/18.0.5.274 I get the following error:

+ module load global_cycle.hera
++ /apps/lmod/8.5.2/libexec/lmod bash load global_cycle.hera
Lmod has detected the following error: Cannot load module "hpc-intel/18.0.5.274" without these module(s) loaded:
   intel/18.0.5.274

While processing the following module(s):
    Module fullname       Module Filename
    ---------------       ---------------
    hpc-intel/18.0.5.274  /scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gfsv16/modulefiles/core/hpc-intel/18.0.5.274.lua
    global_cycle.hera     /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/modulefiles/fv3gfs/global_cycle.hera.lua

+ eval false

My guess it the build hits that error and then loads the hpc-intel default (which is 2022). The hpc-intel/18.0.5.274 module shows a prereq/load for intel/18.0.5.274:

-bash-4.2$ module show hpc-intel/18.0.5.274
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------
   /scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gfsv16/modulefiles/core/hpc-intel/18.0.5.274.lua:
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------
help([[]])
family("MetaCompiler")
conflict("hpc-intel")
conflict("hpc-gnu","hpc-gcc")
load("intel/18.0.5.274")
prereq("intel/18.0.5.274")
prepend_path("MODULEPATH","/scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gfsv16/modulefiles/compiler/intel/18.0.5.274")
setenv("FC","ifort")
setenv("CC","icc")
setenv("CXX","icpc")
setenv("SERIAL_FC","ifort")
setenv("SERIAL_CC","icc")
setenv("SERIAL_CXX","icpc")
whatis("Name: hpc-intel")
whatis("Version: 18.0.5.274")
whatis("Category: Compiler")
whatis("Description: Intel Compiler Family and module access")

@Hang-Lei-NOAA Why would we have to load intel/18.0.5.274 on Hera before being able to load hpc-intel/18.0.5.274?

Fixes #629.

GeorgeGayno-NOAA · 2022-03-15T19:17:02Z

@GeorgeGayno-NOAA I found a couple issues in the modulefiles for global_cycle and emcsfc (for Hera, didn't check others yet) but after fixing them I ran into another issue (the real issue).

A couple of the version variables in the following modulefiles were incorrect. I fixed them before hitting the next error I'll describe below:

-bash-4.2$ git diff modulefiles/*
diff --git a/modulefiles/fv3gfs/global_cycle.hera.lua b/modulefiles/fv3gfs/global_cycle.hera.lua
index 49fcf0d5..9c515f49 100644
--- a/modulefiles/fv3gfs/global_cycle.hera.lua
+++ b/modulefiles/fv3gfs/global_cycle.hera.lua
@@ -10,7 +10,7 @@ load(pathJoin("hpc", hpc_ver))
 hpc_intel_ver=os.getenv("hpc_intel_ver") or "18.0.5.274"
 load(pathJoin("hpc-intel", hpc_intel_ver))
 
-hpc_intel_ver=os.getenv("hpc_impi_ver") or "2018.0.4"
+hpc_impi_ver=os.getenv("hpc_impi_ver") or "2018.0.4"
 load(pathJoin("hpc-impi", hpc_impi_ver))
 
 ip_ver=os.getenv("ip_ver") or "3.3.3"
diff --git a/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua b/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua
index 140c759e..b0574ee0 100644
--- a/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua
+++ b/modulefiles/modulefile.global_emcsfc_ice_blend.hera.lua
@@ -7,7 +7,7 @@ prepend_path("MODULEPATH", "/scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gf
 hpc_ver=os.getenv("hpc_ver") or "1.2.0"
 load(pathJoin("hpc", hpc_ver))
 
-hpc_ver=os.getenv("hpc_intel_ver") or "18.0.5.274"
+hpc_intel_ver=os.getenv("hpc_intel_ver") or "18.0.5.274"
 load(pathJoin("hpc-intel", hpc_intel_ver))

Not sure if you also need to load hpc-impi for the emcsfc builds, I didn't add that yet given the other error I hit.

When attempting to load hpc-intel/18.0.5.274 I get the following error:

+ module load global_cycle.hera
++ /apps/lmod/8.5.2/libexec/lmod bash load global_cycle.hera
Lmod has detected the following error: Cannot load module "hpc-intel/18.0.5.274" without these module(s) loaded:
   intel/18.0.5.274

While processing the following module(s):
    Module fullname       Module Filename
    ---------------       ---------------
    hpc-intel/18.0.5.274  /scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gfsv16/modulefiles/core/hpc-intel/18.0.5.274.lua
    global_cycle.hera     /scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/modulefiles/fv3gfs/global_cycle.hera.lua

+ eval false

My guess it the build hits that error and then loads the hpc-intel default (which is 2022). The hpc-intel/18.0.5.274 module shows a prereq/load for intel/18.0.5.274:

-bash-4.2$ module show hpc-intel/18.0.5.274
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------
   /scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gfsv16/modulefiles/core/hpc-intel/18.0.5.274.lua:
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------
help([[]])
family("MetaCompiler")
conflict("hpc-intel")
conflict("hpc-gnu","hpc-gcc")
load("intel/18.0.5.274")
prereq("intel/18.0.5.274")
prepend_path("MODULEPATH","/scratch2/NCEPDEV/nwprod/hpc-stack/libs/hpc-stack-gfsv16/modulefiles/compiler/intel/18.0.5.274")
setenv("FC","ifort")
setenv("CC","icc")
setenv("CXX","icpc")
setenv("SERIAL_FC","ifort")
setenv("SERIAL_CC","icc")
setenv("SERIAL_CXX","icpc")
whatis("Name: hpc-intel")
whatis("Version: 18.0.5.274")
whatis("Category: Compiler")
whatis("Description: Intel Compiler Family and module access")

@Hang-Lei-NOAA Why would we have to load intel/18.0.5.274 on Hera before being able to load hpc-intel/18.0.5.274?

@KateFriedman-NOAA I fixed the environment variable names for the Hera build. See 2f19a61. But I am not getting the build errors you are.

KateFriedman-NOAA · 2022-03-15T19:32:56Z

I fixed the environment variable names for the Hera build. See 2f19a61. But I am not getting the build errors you are.

Thanks @GeorgeGayno-NOAA , I updated my copy to that new hash and opened a fresh Hera terminal to try the build again. It built without issue and emcsfc used 2018 intel. Please see these logs to confirm:

/scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

So when I open a new Hera terminal I get intel/18.0.5.274 and impi/2018.0.4 loaded by default (I'm not loading them in my .bashrc). With a fresh terminal the prereqs for the intel 2018 module is met and it loads the stack ones without issue at build-time. I had done a module purge to test the build cleanly but apparently my environment was too clean(?).

Will test with this hash and report any issues. Thanks George!

GeorgeGayno-NOAA · 2022-03-15T19:47:21Z

I fixed the environment variable names for the Hera build. See 2f19a61. But I am not getting the build errors you are.

Thanks @GeorgeGayno-NOAA , I updated my copy to that new hash and opened a fresh Hera terminal to try the build again. It built without issue and emcsfc used 2018 intel. Please see these logs to confirm:

/scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

So when I open a new Hera terminal I get intel/18.0.5.274 and impi/2018.0.4 loaded by default (I'm not loading them in my .bashrc). With a fresh terminal the prereqs for the intel 2018 module is met and it loads the stack ones without issue at build-time. I had done a module purge to test the build cleanly but apparently my environment was too clean(?).

Will test with this hash and report any issues. Thanks George!

I also fixed an Orion modulefile. Use fb34823.

GeorgeGayno-NOAA · 2022-03-15T20:38:09Z

I fixed the environment variable names for the Hera build. See 2f19a61. But I am not getting the build errors you are.

Thanks @GeorgeGayno-NOAA , I updated my copy to that new hash and opened a fresh Hera terminal to try the build again. It built without issue and emcsfc used 2018 intel. Please see these logs to confirm:

/scratch1/NCEPDEV/global/Kate.Friedman/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

According to the logs, everything built OK.

So when I open a new Hera terminal I get intel/18.0.5.274 and impi/2018.0.4 loaded by default (I'm not loading them in my .bashrc). With a fresh terminal the prereqs for the intel 2018 module is met and it loads the stack ones without issue at build-time. I had done a module purge to test the build cleanly but apparently my environment was too clean(?).

Will test with this hash and report any issues. Thanks George!

KateFriedman-NOAA · 2022-03-16T13:44:08Z

Thanks @GeorgeGayno-NOAA ! I updated my Orion copy and built it, see logs here:

/work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

GeorgeGayno-NOAA · 2022-03-16T13:52:47Z

Thanks @GeorgeGayno-NOAA ! I updated my Orion copy and built it, see logs here:

/work/noaa/global/kfriedma/git/dev_v16/sorc/ufs_utils.fd/sorc/logs/*

Looks good.

GeorgeGayno-NOAA · 2022-03-29T21:11:51Z

@KateFriedman-NOAA Please use this release: https://github.com/ufs-community/UFS_UTILS/releases/tag/ops-gfsv16.2.1

GeorgeGayno-NOAA added the maintenance Basic upkeep label Feb 8, 2022

GeorgeGayno-NOAA self-assigned this Feb 8, 2022

KateFriedman-NOAA mentioned this issue Feb 8, 2022

Convert modulefiles to lua and hpc-stack (v16) NOAA-EMC/global-workflow#639

Closed

GeorgeGayno-NOAA added a commit that referenced this issue Feb 8, 2022

Remove redundant 'module reset' from build_emcsfc.sh.

acd261a

Also, unload the first build module at the end of the loop. Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Feb 8, 2022

Point to the hpc stack on Orion.

9e32a1f

Fixes #629

GeorgeGayno-NOAA added a commit that referenced this issue Feb 8, 2022

Use the latest version of Intel on Orion.

acfed67

Fixes #629

GeorgeGayno-NOAA added a commit that referenced this issue Feb 9, 2022

Update Hera build modules.

ba7efbd

Fixes #629

GeorgeGayno-NOAA added a commit that referenced this issue Feb 11, 2022

Revert back to the 2018 version of Intel on Orion.

3c5a350

Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 3, 2022

Update netcdf environment variables for global_cycle

5433d1d

build. Use variables from the netcdf module. Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 3, 2022

Update to use new hpc stack on Orion.

d21d534

Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 3, 2022

Point to new hpc stack on Hera.

255828e

Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 4, 2022

Update wcoss2 global_cycle build module. It was compiling

423c704

previously, but would not run without a system error. This fixes the problem. Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 4, 2022

Update orion build to use intel 2022 compiler.

194022e

Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 4, 2022

Update hera build to use intel 2022.

6195f80

Fixes #629.

GeorgeGayno-NOAA mentioned this issue Mar 10, 2022

Makefiles need to support debug and clean NOAA-EMC/global-workflow#300

Closed

13 tasks

GeorgeGayno-NOAA added a commit that referenced this issue Mar 10, 2022

Revert back to Intel 2018 on Hera.

e4cdec1

Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 10, 2022

Remove unused nems utilities.

ed0aefa

Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 10, 2022

Revert to 2018 Intel on Orion.

75c74cd

Fixes #629.

GeorgeGayno-NOAA added a commit that referenced this issue Mar 15, 2022

Fix bug in global_cycle.hera.lua build module.

211d620

Fixes #629.

GeorgeGayno-NOAA closed this as completed Mar 29, 2022

KateFriedman-NOAA mentioned this issue Mar 30, 2022

dev_v16 build and versions updates to support Hera/Orion NOAA-EMC/global-workflow#700

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update GFSv16.2.0 tag to use hpc-stack on Hera and Orion #629

Update GFSv16.2.0 tag to use hpc-stack on Hera and Orion #629

GeorgeGayno-NOAA commented Feb 8, 2022

GeorgeGayno-NOAA commented Feb 9, 2022

KateFriedman-NOAA commented Feb 9, 2022

kgerheiser commented Feb 9, 2022

GeorgeGayno-NOAA commented Feb 9, 2022

GeorgeGayno-NOAA commented Feb 11, 2022

KateFriedman-NOAA commented Feb 15, 2022

GeorgeGayno-NOAA commented Feb 15, 2022

KateFriedman-NOAA commented Feb 15, 2022

GeorgeGayno-NOAA commented Mar 4, 2022

GeorgeGayno-NOAA commented Mar 4, 2022

kgerheiser commented Mar 4, 2022

GeorgeGayno-NOAA commented Mar 4, 2022

GeorgeGayno-NOAA commented Mar 7, 2022

KateFriedman-NOAA commented Mar 7, 2022

KateFriedman-NOAA commented Mar 7, 2022

GeorgeGayno-NOAA commented Mar 7, 2022

KateFriedman-NOAA commented Mar 7, 2022

GeorgeGayno-NOAA commented Mar 10, 2022

KateFriedman-NOAA commented Mar 10, 2022

KateFriedman-NOAA commented Mar 11, 2022

GeorgeGayno-NOAA commented Mar 11, 2022

KateFriedman-NOAA commented Mar 15, 2022

GeorgeGayno-NOAA commented Mar 15, 2022

KateFriedman-NOAA commented Mar 15, 2022

GeorgeGayno-NOAA commented Mar 15, 2022

GeorgeGayno-NOAA commented Mar 15, 2022

KateFriedman-NOAA commented Mar 16, 2022

GeorgeGayno-NOAA commented Mar 16, 2022

GeorgeGayno-NOAA commented Mar 29, 2022

Update GFSv16.2.0 tag to use hpc-stack on Hera and Orion #629

Update GFSv16.2.0 tag to use hpc-stack on Hera and Orion #629

Comments

GeorgeGayno-NOAA commented Feb 8, 2022

GeorgeGayno-NOAA commented Feb 9, 2022

KateFriedman-NOAA commented Feb 9, 2022

kgerheiser commented Feb 9, 2022

GeorgeGayno-NOAA commented Feb 9, 2022

GeorgeGayno-NOAA commented Feb 11, 2022

KateFriedman-NOAA commented Feb 15, 2022

GeorgeGayno-NOAA commented Feb 15, 2022

KateFriedman-NOAA commented Feb 15, 2022

GeorgeGayno-NOAA commented Mar 4, 2022

GeorgeGayno-NOAA commented Mar 4, 2022

kgerheiser commented Mar 4, 2022

GeorgeGayno-NOAA commented Mar 4, 2022

GeorgeGayno-NOAA commented Mar 7, 2022

KateFriedman-NOAA commented Mar 7, 2022

KateFriedman-NOAA commented Mar 7, 2022

GeorgeGayno-NOAA commented Mar 7, 2022

KateFriedman-NOAA commented Mar 7, 2022

GeorgeGayno-NOAA commented Mar 10, 2022

KateFriedman-NOAA commented Mar 10, 2022

KateFriedman-NOAA commented Mar 11, 2022

GeorgeGayno-NOAA commented Mar 11, 2022

KateFriedman-NOAA commented Mar 15, 2022

GeorgeGayno-NOAA commented Mar 15, 2022

KateFriedman-NOAA commented Mar 15, 2022

GeorgeGayno-NOAA commented Mar 15, 2022

GeorgeGayno-NOAA commented Mar 15, 2022

KateFriedman-NOAA commented Mar 16, 2022

GeorgeGayno-NOAA commented Mar 16, 2022

GeorgeGayno-NOAA commented Mar 29, 2022