submodules do not update with latest develop code #194
Labels
bug
Something isn't working
Comments
|
Dom is going to fix this in his PR #191
<#191> (fv3 PR 164
<NOAA-EMC/fv3atm#164>). We plan to get it committed
today.
…On Mon, Aug 31, 2020 at 2:22 PM Jessica Meixner ***@***.***> wrote:
Description
'ccpp/framework' returns fatal: reference is not a tree
when updating submodules for ufs-weather-model commit fd5c310
<fd5c310>
To Reproduce:
What compilers/machines are you seeing this with? WCOSS-Dell (I assume all)
Give explicit steps to reproduce the behavior.
git clone https://github.com/ufs-community/ufs-weather-model
cd ufs-weather-model
git submodule update --init --recursive
Output
$ git submodule update --init --recursive
Submodule 'FMS' (https://github.com/NOAA-GFDL/FMS) registered for path 'FMS'
Submodule 'FV3' (https://github.com/NOAA-EMC/fv3atm) registered for path 'FV3'
Submodule 'NEMS' (https://github.com/NOAA-EMC/NEMS) registered for path 'NEMS'
Submodule 'WW3' (https://github.com/NOAA-EMC/WW3) registered for path 'WW3'
Submodule 'stochastic_physics' (https://github.com/noaa-psd/stochastic_physics) registered for path 'stochastic_physics'
Cloning into 'FMS'...
remote: Enumerating objects: 15444, done.
remote: Total 15444 (delta 0), reused 0 (delta 0), pack-reused 15444
Receiving objects: 100% (15444/15444), 17.56 MiB | 16.04 MiB/s, done.
Resolving deltas: 100% (10946/10946), done.
Submodule path 'FMS': checked out 'f61416fef691d9ba39a40df1ce72aa574f54c390'
Cloning into 'FV3'...
remote: Enumerating objects: 11, done.
remote: Counting objects: 100% (11/11), done.
remote: Compressing objects: 100% (11/11), done.
remote: Total 4149 (delta 0), reused 2 (delta 0), pack-reused 4138
Receiving objects: 100% (4149/4149), 7.58 MiB | 10.43 MiB/s, done.
Resolving deltas: 100% (2956/2956), done.
Submodule path 'FV3': checked out '0b2d3ec4ad172d718abdb05493e9fe6e5e9c272b'
Submodule 'atmos_cubed_sphere' (https://github.com/NOAA-EMC/GFDL_atmos_cubed_sphere) registered for path 'atmos_cubed_sphere'
Submodule 'ccpp/framework' (https://github.com/NCAR/ccpp-framework) registered for path 'ccpp/framework'
Submodule 'ccpp/physics' (https://github.com/NCAR/ccpp-physics) registered for path 'ccpp/physics'
Cloning into 'atmos_cubed_sphere'...
remote: Enumerating objects: 2784, done.
remote: Total 2784 (delta 0), reused 0 (delta 0), pack-reused 2784
Receiving objects: 100% (2784/2784), 2.88 MiB | 572.00 KiB/s, done.
Resolving deltas: 100% (1864/1864), done.
Submodule path 'FV3/atmos_cubed_sphere': checked out '8b59ebc039dafe1c20ed6dd21cb38ca564852b98'
Cloning into 'ccpp/framework'...
remote: Enumerating objects: 44, done.
remote: Counting objects: 100% (44/44), done.
remote: Compressing objects: 100% (33/33), done.
remote: Total 5577 (delta 13), reused 27 (delta 11), pack-reused 5533
Receiving objects: 100% (5577/5577), 10.96 MiB | 13.79 MiB/s, done.
Resolving deltas: 100% (3834/3834), done.
fatal: reference is not a tree: f5d4cd2bf7752ebf1e4ed16dfdfae71dbfabfb76```
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#194>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AI7D6TNDGITCQDRFBE5GVF3SDPS6TANCNFSM4QQZUAOQ>
.
|
|
Can not reproduce: |
|
@DusanJovic-NOAA What version of git are you using? On Dell I was using git version 1.8.3.1 and got the error I showed above (I checked out again to confirm), but on hera I am using git version 2.18.0 and get a similar output to what you show. Thanks for the info @junwang-noaa, I guess this issue can be closed then. |
|
|
The submodule pointer is definitely wrong, it's pointing to some hash in @ClaraDraper-NOAA's fork. |
|
How did it work for me then? Did git fetch ' |
|
I don't understand this at all either. Look here: this was a recursive checkout on hera. If you go to NCAR ccpp-framework and look at the commits for branch "master", the latest is 209f1c92d99b7d4cc63e0d41c652fcfd730bd9fa. The commit f5d4cd2bf7752ebf1e4ed16dfdfae71dbfabfb76 is from a closed (!) PR that was never merged in Clara's fork, see here: https://github.com/NCAR/ccpp-framework/pull/316/commits. Maybe, but just maybe, newer git versions are able to grab it, because they can find it in the closed PR history. Sounds odd to me, though. |
|
Jessica, would you please check again if you still have problem on git submodule update? I can't repeat the issue on dell either. |
|
worked with the latest develop branch today. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Description
'ccpp/framework' returns fatal: reference is not a tree
when updating submodules for ufs-weather-model commit fd5c310
To Reproduce:
What compilers/machines are you seeing this with? WCOSS-Dell (I assume all)
Give explicit steps to reproduce the behavior.
git clone https://github.com/ufs-community/ufs-weather-model
cd ufs-weather-model
git submodule update --init --recursive
Output
The text was updated successfully, but these errors were encountered: