DRAM v2 run fails with permission error

[shaze]

Please could you help with this problem

The workflow is being run like this

nextflow run DRAM -r dev  --rename --input_fasta  /projects/lion/qc-contigs-all/indivs -profile singularity --outdir out

I get the following error

Command error:
  /bin/bash: line 11: .command.run: Permission denied

Work dir:
  /projects/scott/DRAM/work/99/f83e21ac7f918123103773b399ca73

Container:
  /home/scott/.singularity/community.wave.seqera.io-library-bbmap-801715ef64484762.img

Looking at the logs, this seems to be where things go wrong

  mkdir -p renamed
      rename.sh in=4915_10.fa out=renamed/4915_10.fna prefix=4915_10 addprefix=t
  rename.sh in=4915_14.fa out=renamed/4915_14.fna prefix=4915_14 addprefix=t
  rename.sh in=4915_17.fa out=renamed/4915_17.fna prefix=4915_17 addprefix=t
  rename.sh in=4915_1.fa out=renamed/4915_1.fna prefix=4915_1 addprefix=t
  rename.sh in=4915_20.fa out=renamed/4915_20.fna prefix=4915_20 addprefix=t
  rename.sh in=4915_24.fa out=renamed/4915_24.fna prefix=4915_24 addprefix=t
  rename.sh in=4915_26.fa out=renamed/4915_26.fna prefix=4915_26 addprefix=t
  rename.sh in=4915_3.fa out=renamed/4915_3.fna prefix=4915_3 addprefix=t

Looking at the work directory, I don't see the renamed directory

I looked at the .command.run file and tried running .command.sh directly

apptainer exec --no-home --pid -B /projects/lion/qc-contigs-all/indivs -B /home/scott/.nextflow/assets/WrightonLabCSU/DRAM/bin -B //projects/scott/DRAM/work/99/f83e21ac7f918123103773b399ca73 /home/scott/.singularity/community.wave.seqera.io-library-bbmap-801715ef64484762.img bash .command.sh

and the script starts running corretctly

System information
*
DRAM version DRAM v2
Nextflow version 25.04.2
Hardware HPC
Executor local
Container engine: apptainer, 1.4.1-1.el9
OS Rocky 9.6

This is what my nextflow config looks like

singularity {
        singularity.enabled     = false
        singularity.autoMounts  = true
        conda.enabled           = false
        docker.enabled          = false
        podman.enabled          = false
        shifter.enabled         = false
        charliecloud.enabled    = false
        apptainer.enabled       = true
    }

Many thanks

[shaze]

An update on this, further debugging found the error here

nxf_launch() {
    set +u; env - PATH="$PATH" ${TMP:+APPTAINERENV_TMP="$TMP"} ${TMPDIR:+APPTAINERENV_TMPDIR="$TMPDIR"} ${NXF_TASK_WORKDIR:+APPTAINERENV_NXF_TASK_WORKDIR="$NXF_TASK_WORKDIR"} APPTAINERENV_NXF_DEBUG="${NXF_DEBUG:=0}" apptainer exec --no-home --pid -B /projects/lion/qc-contigs-all/indivs -B /home/scott/.nextflow/assets/WrightonLabCSU/DRAM/bin -B /projects/scott/DRAM/work/99/f83e21ac7f918123103773b399ca73 /home/scott/.singularity/community.wave.seqera.io-library-bbmap-801715ef64484762.img /bin/bash -c "cd $NXF_TASK_WORKDIR; eval $(nxf_container_env); bash
    
    set -e # Exit if a tool returns a non-zero status/exit code
    set -u # Treat unset variables and parameters as an error
    set -o pipefail # Returns the status of the last command to exit with a non-zero status or zero if all successfully execute
    set -C # No clobber - prevent output redirection from overwriting files.
     /projects/scott/DRAM/work/99/f83e21ac7f918123103773b399ca73/.command.run nxf_trace"
}

The problem seems to be that a blank line and the sets now come between the bash and the .command.run. I am not sure the definition of process.shell in the nextflow.config is robust.

I switched off all the trace, timeline options etc and then the code ran fine:

The corresponding `.command.run file now has

set +u; env - PATH="$PATH" ${TMP:+APPTAINERENV_TMP="$TMP"} ${TMPDIR:+APPTAINERENV_TMPDIR="$TMPD\
IR"} ${NXF_TASK_WORKDIR:+APPTAINERENV_NXF_TASK_WORKDIR="$NXF_TASK_WORKDIR"} apptainer exec --no-hom\
e --pid -B /dataA/lion/qc-contigs-all/indivs -B /home/scott/.nextflow/assets/WrightonLabCSU/DRAM/bi\
n -B /dataH/projects/scott/DRAM/work/43/0b08e0b713b0cb2754a5fe6b287424 /home/scott/.singularity/com\
munity.wave.seqera.io-library-bbmap-801715ef64484762.img /bin/bash -c "cd $NXF_TASK_WORKDIR; eval $\
(nxf_container_env); /usr/bin/env bash /projects/scott/DRAM/work/43/0b08e0b713b0cb2754a5fe6b287424/.command.sh"

[madeline-scyphers]

Thank you so much, sorry it took me a couple weeks to look at these, I was out of the office. I will look at the process.shell and try and get out a fix!

[madeline-scyphers]

THis should be fixed in the latest release, beta13