This pipeline is minimal reproducible test case for running different versions of tb-profiler in parallel using conda/mamba.

The core problem with v4.1.1 was that, when there were many tasks (using tb-profiler) being run in parallel, the import of custom DB gave rise to a race condition as these tasks were trying to use the tb-profiler load_library functionality to add the custom database.

picard.PicardException: /home/xbs-nf-EarlyMGIT/xbs-nf/work/conda/xbs-nf-env-1-d99876e5fea88a1c4bd18887d111ae27/share/tbprofiler/resistance_db_who.dict already exists. Delete this file and try again, or specify a different output file.
at picard.sam.CreateSequenceDictionary.doWork(CreateSequenceDictionary.java:220)
at picard.cmdline.CommandLineProgram.instanceMain(CommandLineProgram.java:308)
at

This problem can be reproduced locally by relying upon the following command (assuming nextflow and conda) are installed.

$ nextflow run https://github.com/abhi18av/parallel-tbprofiler-conda-test-nf -profile tbprofiler411