harvardinformatics / snpArcher

Snakemake workflow for highly parallel variant calling designed for ease-of-use in non-model organisms.
MIT License
63 stars 30 forks source link

split envs in qc and r-base to 4.1.3 #82

Closed cademirch closed 1 year ago

erikenbody commented 1 year ago

for posterity, here is the explanation for this:

The environment building started to fail for the QC module.

Running QC test like so:

snakemake -d .test/qc --use-conda --cores 10 -s workflow/modules/qc/Snakefile

resulted in:

Building DAG of jobs...
Your conda installation is not configured to use strict channel priorities. This is however crucial for having robust and correct environments (for details, see https://conda-forge.org/docs/user/tipsandtricks.html). Please consider to configure strict priorities by executing 'conda config --set channel_priority strict'.
Creating conda environment /data/snpArcher/workflow/modules/qc/envs/qc.yml...
Downloading and installing remote packages.
CreateCondaEnvironmentException:
Could not create conda environment from /data/snpArcher/workflow/modules/qc/envs/qc.yml:
Command:
mamba env create --quiet --file "/data/snpArcher/.test/qc/.snakemake/conda/9ec289de8e88dc89e086c8afaeb5b696_.yaml" --prefix "/data/snpArcher/.test/qc/.snakemake/conda/9ec289de8e88dc89e086c8afaeb5b696_"
Output:
Encountered problems while solving:
  - package r-base-4.1.1-hb67fd72_0 requires libxml2 >=2.9.12,<2.11.0a0, but none of the providers can be installed

I ran the github tests and received the same error.

It is very confusing, this should not happen but as near as I can tell the r-base 4.1.1 on conda-forge became non-functional. We were past time to split up conda environments per rule which QC especially needs, so I have done that in this commit.

I saw that 4.1.3 had recently been patched on conda as it was the most recently updated: https://anaconda.org/conda-forge/r-base/files

So I switched to this and it passed test