Add csawmg debug test, add to rt_gnu.conf #552
Labels
enhancement
New feature or request
Comments
13 tasks
|
@climbfuji It looks like intel has both prod and debug tests now, but this still hasn't been done for GNU. I tried adding both a prod and debug to the rt_gnu.conf. Both segfault. The control_csamwg_debug test produces this starting traceback: See /scratch1/NCEPDEV/stmp2/Denise.Worthen/FV3_RT/rt_304214/control_csawmg_debug |
|
Update 08/05/2021: PR #724 fixes the aerosol options that led to recent problems with the csawmg based regression tests. The two tests See also the discussion in #724 for what was tried so far. |
pjpegion
pushed a commit
to NOAA-PSL/ufs-weather-model
that referenced
this issue
Apr 4, 2023
… temperatures (ufs-community#552) * typo caused potential temperature calculation to use exponentiation instead of multiplication
epic-cicd-jenkins
pushed a commit
that referenced
this issue
Apr 17, 2023
## DESCRIPTION OF CHANGES: 1. Add if condition in tests/run_experiments.sh so to get informative error message when MET and MET paths are not available on a machine. 2. Remove MET/MET+ paths in tests/baseline_configs/config.verification.sh ## TESTS CONDUCTED: A test run was conducted on Hera and has been finished successfully. Another test was run on Jet, and it failed in the expected way. ## DEPENDENCIES: To have MET verification run successfully, the observational data (e.g., CCPA, MRMS, NDAS) must be available. ## DOCUMENTATION: N/A ## ISSUE (optional): This is a follow up PR to complete the previous one in https://github.com/NOAA-EMC/regional_workflow/pull/537 ## CONTRIBUTORS (optional): @gsketefian contributed the revision.
epic-cicd-jenkins
pushed a commit
that referenced
this issue
Apr 17, 2023
…onf. (#552) Co-authored-by: Linlin.Pan <Linlin.Pan@noaa.gov>
|
@jkbk2004 @junwang-noaa Is this still needed? |
|
@yangfanglin @Qingfu-Liu May I ask if there is some person in physics groupwho are maintaining the csawmg and csawmgt tests? |
|
Moorthi was the one who were working on CSAW convection and MG MP. |
|
Fanglin, sure. I can take over those RT tests.
…On Wed, Dec 20, 2023 at 9:49 AM Fanglin Yang ***@***.***> wrote:
Moorthi was the one who were working on CSAW convection and MG MP.
@AnningCheng-NOAA <https://github.com/AnningCheng-NOAA> Anning, please
take a look and see if you can still provide the support to maintain these
RT tests. Otherwise, it is time to remove them
—
Reply to this email directly, view it on GitHub
<#552 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ALQPMIPM55FG4EL5NYSID5DYKL3IXAVCNFSM434W4FN2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOBWGQ3DANBUGA3A>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
What is the status of this issue? |
|
The debugging option is rt.conf and working. Please let me know if there is
any problem.
…On Thu, Feb 1, 2024 at 2:27 PM Dom Heinzeller ***@***.***> wrote:
What is the status of this issue?
—
Reply to this email directly, view it on GitHub
<#552 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ALQPMIMWAAFLU45ZQYLX2M3YRPUABAVCNFSM434W4FN2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOJSGIYDOMJTGYYQ>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
Sorry. I just noticed that a gnu debug needs to be added. Does gnu control
need too?
On Thu, Feb 1, 2024 at 2:39 PM Anning Cheng - NOAA Affiliate <
***@***.***> wrote:
… The debugging option is rt.conf and working. Please let me know if there
is any problem.
On Thu, Feb 1, 2024 at 2:27 PM Dom Heinzeller ***@***.***>
wrote:
> What is the status of this issue?
>
> —
> Reply to this email directly, view it on GitHub
> <#552 (comment)>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ALQPMIMWAAFLU45ZQYLX2M3YRPUABAVCNFSM434W4FN2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOJSGIYDOMJTGYYQ>
> .
> You are receiving this because you were mentioned.Message ID:
> ***@***.***>
>
|
|
Dom and Fanglin:
Do you have a link to issue #552? I need to understand what is really
needed.
On Thu, Feb 1, 2024 at 2:43 PM Anning Cheng - NOAA Affiliate <
***@***.***> wrote:
… Sorry. I just noticed that a gnu debug needs to be added. Does gnu control
need too?
On Thu, Feb 1, 2024 at 2:39 PM Anning Cheng - NOAA Affiliate <
***@***.***> wrote:
> The debugging option is rt.conf and working. Please let me know if there
> is any problem.
>
> On Thu, Feb 1, 2024 at 2:27 PM Dom Heinzeller ***@***.***>
> wrote:
>
>> What is the status of this issue?
>>
>> —
>> Reply to this email directly, view it on GitHub
>> <#552 (comment)>,
>> or unsubscribe
>> <https://github.com/notifications/unsubscribe-auth/ALQPMIMWAAFLU45ZQYLX2M3YRPUABAVCNFSM434W4FN2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOJSGIYDOMJTGYYQ>
>> .
>> You are receiving this because you were mentioned.Message ID:
>> ***@***.***>
>>
>
|
|
@AnningCheng-NOAA I was just checking on old issues. If you think what is there at the moment is good enough, then we can close this as not planned. |
|
Just find the thread and will work on it, It might take some time. An issue that ocurred years ago even when Moorthi was here. |
|
@dominikus Heinzeller - NOAA Affiliate ***@***.***> I just
found the issue and will work on it. CSAW and MG might be useful in the
future if GNU works.
…On Thu, Feb 1, 2024 at 3:00 PM Dom Heinzeller ***@***.***> wrote:
@AnningCheng-NOAA <https://github.com/AnningCheng-NOAA> I was just
checking on old issues. If you think what is there at the moment is good
enough, then we can close this as not planned.
—
Reply to this email directly, view it on GitHub
<#552 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ALQPMIK52RSXOULGGJF2TYDYRPX3XAVCNFSM434W4FN2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOJSGIYTGMBZHEYA>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
A major update for CSAW was missed a few years ago. Here is a piece of
Email from Don at CSU. I am working on the update and should be done in a
few days.
On Thu, Oct 24, 2019 at 3:01 PM Donald Dazlich - NOAA Affiliate <
***@***.***> wrote:
Hi Moorthi, Anning,
Dave and I have been using the GFS version of CS to redo our
implementation of AW more completely and consistently. Now, you will find
that the fix_form implemetation of the tendencies match up very closely
with the traditional form. Also you will notice that it is the fluxes for
each cloud type (and the downdraft) that are saved and the reduction is
done on the fluxes, not the tendency computed from the flux divergence.
Since sigma is a function of height the latter approach is incorrect. I’ve
also dimensioned several cloud type variables nctp+1 - the nctp+1 slot has
downdraft output. Another addition is the sorting of the lamdas after they
are computed for all cloud types, then computing the sigma. This approach
is detailed in a manuscript Dave is working on. Also, sigma is not applied
to the inputs to the downdraft routine, but to the outputs, which are now
mass fluxes not tendencies. The water and thermodynamic budgets remain
satisfied now.
I’ve attached my version below. It should be pretty close to your version
with regards to the interface so you shouldn’t have to do much work with it.
Regards,
Don
On Thu, Feb 1, 2024 at 3:08 PM Anning Cheng - NOAA Affiliate <
***@***.***> wrote:
… @dominikus Heinzeller - NOAA Affiliate ***@***.***> I just
found the issue and will work on it. CSAW and MG might be useful in the
future if GNU works.
On Thu, Feb 1, 2024 at 3:00 PM Dom Heinzeller ***@***.***>
wrote:
> @AnningCheng-NOAA <https://github.com/AnningCheng-NOAA> I was just
> checking on old issues. If you think what is there at the moment is good
> enough, then we can close this as not planned.
>
> —
> Reply to this email directly, view it on GitHub
> <#552 (comment)>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/ALQPMIK52RSXOULGGJF2TYDYRPX3XAVCNFSM434W4FN2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOJSGIYTGMBZHEYA>
> .
> You are receiving this because you were mentioned.Message ID:
> ***@***.***>
>
|
This was referenced Feb 12, 2024
This was referenced Mar 7, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Description
csawmg-based jobs never ran with GNU, crashed with segmentation faults. Now that we have assumed-size arrays in CCPP (#527), we should add a DEBUG test tort.confand one PROD+DEBUG test tort_gnu.conf.When testing #527, I ran
fv3_csawmgwith Intel in DEBUB mode on Jet, and it worked.The text was updated successfully, but these errors were encountered: