conda-lock hangs with a combination of python 3.11 and dsub 0.4.9 #619
Comments
|
I see this as well with just: name: my_env
channels:
- conda-forge
dependencies:
- python
- pip
- pip:
- poetryRunning with |
|
This was fixed upstream in Poetry here: python-poetry/poetry#5881 |
|
name: myenv
channels:
- conda-forge
dependencies:
- pip=24.0
- python=3.11.8
- pip:
- dsub==0.4.8
- google-api-core[grpc]>=2.7.3,<=2.11.0It appears that the vendored Poetry has issues handling pip package extras, and the Poetry solver infinitely loops if both I haven't tested, but #613 may help. |
|
I think I'm experiencing the same thing with name: servicex-conda-lock-failure
channels:
- conda-forge
dependencies:
- python=3.11
- pip
- pip:
- servicex>=3.0.0As seen in this minimal failing example below, I'm able to create a new environment from $ docker pull condaforge/miniforge3:latest
$ docker run --rm -ti -v ./example.yml:/tmp/example.yml condaforge/miniforge3:latest
(base) root@4db92033a607:/# mamba init
(base) root@4db92033a607:/# . /root/.bashrc
(base) root@4db92033a607:/# mamba --version
mamba 1.5.9
conda 24.7.1
(base) root@4db92033a607:/# time mamba env create --yes --file /tmp/example.yml
...
real 0m25.827s
user 0m19.124s
sys 0m2.730s
(base) root@4db92033a607:/# mamba run --name servicex-conda-lock-failure python -c 'import servicex; print(servicex)'
<module 'servicex' from '/opt/conda/envs/servicex-conda-lock-failure/lib/python3.11/site-packages/servicex/__init__.py'>
(base) root@4db92033a607:/# mamba create --name lock --yes conda-lock
(base) root@4db92033a607:/# mamba activate lock
(lock) root@4db92033a607:/# conda-lock --version
conda-lock, version 2.5.7
(lock) root@4db92033a607:/# timeout 120 conda-lock lock --mamba --platform linux-64 --kind lock --file /tmp/example.yml
Locking dependencies for ['linux-64']...
INFO:conda_lock.conda_solver:linux-64 using specs ['python 3.11.*', 'pip *']
(lock) root@4db92033a607:/# echo $?
124I guess this is fixed(?) given PR #637 and PR #678, but there is no post |
|
Hi @matthewfeickert, it works for me from the |
|
If you'd be willing to test out pipx install --force git+https://github.com/conda/conda-lock@main |
@maresb, yeah the minimal failing example that I had now works on root@e49221b89d44:/# pipx install --force git+https://github.com/conda/conda-lock@main
...
root@e49221b89d44:/# timeout 120 conda-lock lock --platform linux-64 --kind lock --file /tmp/example.yml
Locking dependencies for ['linux-64']...
INFO:conda_lock.conda_solver:linux-64 using specs ['python 3.11.*', 'pip *']
- Install lock using: conda-lock install --name YOURENV conda-lock.yml
root@e49221b89d44:/# grep servicex conda-lock.yml
- name: servicex
url: https://files.pythonhosted.org/packages/61/d2/27ed7c70c289b9238266e25809cc4ef75a39808da8813dc6d92e293d432d/servicex-3.0.0-py3-none-any.whland if I extend that name: iris-hep-py311
channels:
- conda-forge
dependencies:
- python 3.11.*
- uproot >=5.0.0
- hist >=2.7.0
- mplhep >=0.3.50
- cabinetry >=0.6.0
- coffea >=2024.1.0
- func-adl >=3.3.0
- xrootd >=5.7.0
- dask-awkward >=2024.1.0
- pip
- pip:
- func-adl-servicex>=2.2
- servicex>=3.0.0
- fastjet>=3.4.0.0the ...
root@e49221b89d44:/# timeout 120 conda-lock lock --platform linux-64 --kind lock --file ./environment.yml
Locking dependencies for ['linux-64']...
INFO:conda_lock.conda_solver:linux-64 using specs ['python 3.11.*', 'uproot >=5.0.0', 'hist >=2.7.0', 'mplhep >=0.3.50', 'cabinetry >=0.6.0', 'coffea >=2024.1.0', 'func-adl >=3.3.0', 'xrootd >=5.7.0', 'dask-awkward >=2024.1.0', 'pip *']
- Install lock using: conda-lock install --name YOURENV conda-lock.yml |
|
Many thanks for reporting back @matthewfeickert, I really appreciate it!!! This sort of report makes me more confident in cutting the release. |
Checklist
What happened?
When I run
conda-lock -p osx-arm64 -f environment.yml -k envwith the following environment.yml file, it hangs indefinitely.If I change the python version to 3.10.13, it completes. If I change the dsub version to 0.4.7 or earlier, it completes. 0.4.8 or later fails. If I run
conda env update -f environment.yml(conda version24.1.2) it builds the environment successfully.Conda Info
Conda Config
Conda list
Additional Context
No response
The text was updated successfully, but these errors were encountered: