Plant-Food-Research-Open / assemblyqc

A Nextflow pipeline for evaluating assembly quality
https://plant-food-research-open.github.io/assemblyqc/
MIT License
26 stars 4 forks source link

DNADIFF is killed by out of memory signal on SLURM #141

Closed GallVp closed 2 weeks ago

GallVp commented 1 month ago

Description of the bug

DNADIFF is killed by out of memory signal on SLURM

Command used and terminal output

No response

Relevant files

No response

System information

plant-food-research-open/assemblyqc Version 2.1.1

GallVp commented 1 month ago

The OOM kill does not always result in exit code 12 or 137. It might be process or context specific. For example, the DNADIFF process from Plant-Food-Research-Open/assemblyqc/2.1.1 has an OOM exit code of 2. Whereas, MUMMER's OOM exit code is 137. The pipeline uses the nf-core template's default retry on error strategy which covers error code 137 but does not cover error code 2. Error code 2 is for No such file or directory so perhaps it is not a good idea to retry on 2. For now, I am going to bump up the memory requirement for DNADIFF from 6.GB to 12.GB.