NFCORE_HLATYPING:HLATYPING:SAMTOOLS_COLLATEFASTQ - OutOfMemory?

[dansteiert]

I am running the hlatyping workflow with a given file and am running into an OutOfMemory issue.

In other workflows, I used thus far, there was a retry option, to increase memory and runtime incase of such an insidents:
looking trough the log file, I got a message error 137:
terminated with an error exit status (137) -- Execution is retried (1)
Which in principal seems to work (for the normal samples)
But not for my two associated tumor samples. They get an exit status 1. but looking into my slurm execution, they are also out of memory.
Failed to read header for "-"

Looking into the associated folder, I see 64 bam files (..00{00-63}.bam) with a size of 3.7GB, the logs and error files, as well as 4 empty files:

A samtools quickcheck showed for the inputfile as well as the newly generated files:

checking ..0063.bam
opened ..0063.bam
..0063.bam is sequence data
..0063.bam has 87 targets in header.
..0063.bam has good EOF block.

nextflow run nf-core/hlatyping -profile singularity --input /PATH_TO_SAMPLESHEET/samplesheet_HLA_typing_test.csv --outdir /PATH_TO_OUTPUT/HLA_Typing/ --genome GRCh37

Your help and insights are highly appreciated!
Best

here the log:
nextflow.log