Skip to content

lalburst:conda failing in nightly

The lalburst:conda job has been failing in the nightly build for some time, e.g. https://git.ligo.org/lscsoft/lalsuite/-/jobs/4206477. The last part of the build log is:

ERROR: Failed to get package records, max retries exceeded.
ExplainedDependencyNeedsBuildingError: Unsatisfiable dependencies for platform linux-64: {MatchSpec("liblalsimulation==6.2.0.1.dev20250808=py39h12de6ab_9999"), MatchSpec("python[version='>=3.9,<3.10.0a0']")}
Encountered problems while solving:
  - package liblalsimulation-6.2.0.1.dev20250808-py39h12de6ab_9999 requires python >=3.9,<3.10.0a0, but none of the providers can be installed
Could not solve for environment specs
The following packages are incompatible
├─ python-lalsimulation >=6.2.0a0 * is installable with the potential options
│  ├─ python-lalsimulation 6.2.0.1.dev20250808 would require
│  │  └─ liblalsimulation ==6.2.0.1.dev20250808 py39h12de6ab_9999, which requires
│  │     └─ python >=3.9,<3.10.0a0 *, which can be installed;
│  ├─ python-lalsimulation 6.2.0.1.dev20250808 would require
│  │  └─ python >=3.10,<3.11.0a0 *, which can be installed;
│  ├─ python-lalsimulation 6.2.0.1.dev20250808 would require
│  │  └─ python >=3.11,<3.12.0a0 *, which can be installed;
│  ├─ python-lalsimulation 6.2.0.1.dev20250808 would require
│  │  └─ python_abi =3.12 *_cp312, which requires
│  │     └─ python =3.12 *_cpython, which can be installed;
│  └─ python-lalsimulation 6.2.0 conflicts with any installable versions previously reported;
└─ python =3.13 *_cp313 is not installable because it conflicts with any installable versions previously reported.
Leaving build/test directories:
  Work:
 /builds/lscsoft/lalsuite/conda-bld/work 
  Test:
 /builds/lscsoft/lalsuite/conda-bld/test_tmp 
Leaving build/test environments:
  Test:
source activate  /builds/lscsoft/lalsuite/conda-bld/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_p 
  Build:
source activate  /builds/lscsoft/lalsuite/conda-bld/_build_env

The strange thing is that this job works perfectly fine when run as part of the MR pipeline, e.g. https://git.ligo.org/lscsoft/lalsuite/-/jobs/4206249, so it must be triggered by some differences between how the job is run in each pipeline (MR vs nightly).