Picard "--TMP_DIR" arguments are hardcoded

Description of the bug

An error "No space left on device" occurred by process NFCORE_ATACSEQ:ATACSEQ:MERGED_LIBRARY_MARKDUPLICATES_PICARD:PICARD_MARKDUPLICATES (E145I_REP1). Possible reason "--TMP_DIR" arguments are hardcoded in modules.config (example https://github.com/nf-core/atacseq/blob/1a1dbe52ffbd82256c941a032b0e22abbd925b8a/conf/modules.config#L358).

Caused by:
Process NFCORE_ATACSEQ:ATACSEQ:MERGED_LIBRARY_MARKDUPLICATES_PICARD:PICARD_MARKDUPLICATES (E145I_REP1) terminated with an error exit status (1)

Command executed:

picard \
    -Xmx29491M \
    MarkDuplicates \
    --ASSUME_SORTED true --REMOVE_DUPLICATES false --VALIDATION_STRINGENCY LENIENT --TMP_DIR tmp \
    --INPUT E145I_REP1.mLb.sorted.bam \
    --OUTPUT E145I_REP1.mLb.mkD.sorted.bam \
    --REFERENCE_SEQUENCE Mus_musculus.GRCm38.dna.primary_assembly.fa \
    --METRICS_FILE E145I_REP1.mLb.mkD.sorted.MarkDuplicates.metrics.txt

cat <<-END_VERSIONS > versions.yml
"NFCORE_ATACSEQ:ATACSEQ:MERGED_LIBRARY_MARKDUPLICATES_PICARD:PICARD_MARKDUPLICATES":
    picard: $(echo $(picard MarkDuplicates --version 2>&1) | grep -o 'Version:.*' | cut -f2- -d:)
END_VERSIONS

Command error:

Runtime.totalMemory()=4311744512
  To get help, see http://broadinstitute.github.io/picard/index.html#GettingHelp
  Exception in thread "main" htsjdk.samtools.SAMException: Error loading new map from disk.
        at htsjdk.samtools.CoordinateSortedPairInfoMap.ensureSequenceLoaded(CoordinateSortedPairInfoMap.java:143)
        at htsjdk.samtools.CoordinateSortedPairInfoMap.remove(CoordinateSortedPairInfoMap.java:86)
        at picard.sam.markduplicates.util.DiskBasedReadEndsForMarkDuplicatesMap.remove(DiskBasedReadEndsForMarkDuplicatesMap.java:61)
        at picard.sam.markduplicates.MarkDuplicates.buildSortedReadEndLists(MarkDuplicates.java:560)
        at picard.sam.markduplicates.MarkDuplicates.doWork(MarkDuplicates.java:270)
        at picard.cmdline.CommandLineProgram.instanceMain(CommandLineProgram.java:289)
        at picard.cmdline.PicardCommandLine.instanceMain(PicardCommandLine.java:104)
        at picard.cmdline.PicardCommandLine.main(PicardCommandLine.java:114)
  Caused by: java.io.IOException: No space left on device
        at java.base/java.io.FileOutputStream.writeBytes(Native Method)
        at java.base/java.io.FileOutputStream.write(FileOutputStream.java:349)
        at java.base/java.io.BufferedOutputStream.flushBuffer(BufferedOutputStream.java:81)
        at java.base/java.io.BufferedOutputStream.flush(BufferedOutputStream.java:142)
        at java.base/java.io.FilterOutputStream.close(FilterOutputStream.java:182)
        at htsjdk.samtools.CoordinateSortedPairInfoMap.ensureSequenceLoaded(CoordinateSortedPairInfoMap.java:119)

Command used and terminal output

Slurm declaration

export TMPDIR=$PWD
export SINGULARITY_TMPDIR=$PWD
export SINGULARITY_CACHEDIR=$PWD
unset XDG_RUNTIME_DIR

Relevant files

No response

System information

• Nextflow version - 24.01.0-edge
• Hardware - HPC
• Executor - slurm
• Container engine: - Singularity
• OS - Linux
• Version of nf-core/atacseq 2.1.2

Start by inspecting the referenced conf/modules.config entry around line 358 and the Slurm environment variables shown in the report. Check how the Picard MarkDuplicates process selects its temporary directory and reproduce the failure on an HPC job if possible. Done means the hardcoded temporary location no longer causes the reported device-space failure.