Open zhuchcn opened 7 months ago
What I don't understand is rm -r
should remove the directory even if it is not empty. So why does it complain that the directoy isn't empty? Maybe the files were being used (for example being moved by publishDir
)?
The publishDir rules shouldn't have any race conditions since the L55 doesn't run until the nextflow run
finishes and that only finished after all publishDir have finished; I'm not sure why the complaint is there about rm -r
This is a repeated issue: /hot/project/process/MissingPeptides/MISP-000132-MissingPeptidesPanCanP1/CPTAC_PAAD/WXS/work/d6/18e638e93144f211ed12a14280164c/slurm.out
convert-BAM2FASTQ failed.
But the pipeline actually finished successfully, except that the working directory cannot be removed.
I'm guessing maybe this is because Ln 55 exceuted too early, before the publishDir finishes moving files to the destination directory, because the working directory that it failed to remove is indeed empty now.
https://github.com/uclahs-cds/metapipeline-DNA/blob/59b96630180f0dca2cb9963478fc8cf507a10a57/module/convert_BAM2FASTQ/call_convert_BAM2FASTQ.nf#L41-L55