Repository navigation
RGI_MAIN completes with exit 0 and truncated output after sub-process OOM kill #551
Description
Description of the bug
On large metagenome assemblies (2.3–3.5 million contigs), RGI_MAIN tasks completed with exit status 0 but produced severely truncated output (e.g. 11 hits vs 600–1,400 in comparable samples). SLURM's .command.log for each affected task shows:
slurmstepd: error: Detected 1 oom_kill event in StepId=… Some of the step tasks have been OOM Killed.
Because the parent process still exits 0, the default errorStrategy retry is never triggered, and the truncated results are published and passed to hAMRonization without warning. 9 of 10 large assemblies were affected; the reduction ranged from near-total to partial and was only detectable by comparing against read-based results.
version used nf-core/funcscan 4.0.0
Command used and terminal output
Relevant files
No response
System information
No response
Activity
jfy133 commentedon Oct 15, 2026
Hi @dedewantstocode thanks for the report and sorry for the slow reply.
If I understand correctly (which I might not be - sorry if not), RGI is reporting a 0 despite an internal OOM kill (captured by SLURM). If RGI is reporting the 0, there is not much we can do unfortunately as the Nextflow only really has that one way of capturing a job failure :(.
This would need a a change by RGI to correctly report the exit code correctly at the top-level RGI step.