I have a workflow with the containerize directive (I have one large docker container with all Conda envs for the whole workflow).

When I run this workflow on my desktop computer with the following profile:

cores: 40
latency-wait: 20
use-conda: True
keep-going: False
rerun-incomplete: True
printshellcmds: True
cache: False
show-failed-logs: True
use-apptainer: True

It correctly applies the image and activates all the Conda envs within.

However when I run the same workflow on a slurm-based HPC system (using Snakemake 8.11.3), it will not use Apptainer but instead is making the separate Conda envs based on the yaml files in the workflow directory. This is the profile I use on the HPC:

executor: slurm
jobs: 100
rerun-incomplete: True
printshellcmds: True
use-conda: True
use-apptainer: True
default-resources:
        slurm_partition: icelake
        slurm_account: XXX

I don’t see why it would not not apply the image as use-apptainer is True?