Segmentation fault

14 views
Skip to first unread message

Zihao Zhao

unread,
Jul 1, 2022, 5:41:25 AM7/1/22
to LotuS rRNA pipeline
Dear Falk,

I got the 'Segmentation fault' while running lotus.

"sh: line 1: 700199 Segmentation fault      (core dumped) /apps/lotus/2.21/bin//sdm -i_path reads/ -o_fna /tmp/slurm-4381304/demulti.1.fna,/tmp/slurm-4381304/demulti.2.fna -o_fna2 /tmp/slurm-4381304/demulti.1.add.fna,/tmp/slurm-4381304/demulti.2.add.fna -sample_sep ___ -log euk_18s_125bp/LotuSLogS//demulti.log -map euk_18s_125bp/primary/in.map -options sdm_hiSeq.txt -o_dereplicate /tmp/slurm-4381304/derep.fas -dere_size_fmt 0 -min_derep_copies 8:1,4:2,3:3 -suppressOutput 1 -o_qual_offset 33 -paired 2 -maxReadsPerOutput 8000000 -oneLineFastaFormat 1 -threads 8 >> euk_18s_125bp/LotuSLogS/LotuS_progout.log 2>&1
"

This error message poped up again when I re-run my scripts, however the difference between this two errors is that lotus stopped at the 86th sample for the first time and 106th sample for this run.

I have 334 samples in one run, is this too much for the run?

Here I also paste my scripts with this issue:

#!/bin/bash
#
#SBATCH --job-name=lotus
#SBATCH --cpus-per-task=8
#SBATCH --mem=60GB
#SBATCH --mail-type=ALL
#SBATCH --mail-user=zihao...@univie.ac.at
#SBATCH --error=file-%x-%A_%a.err
#SBATCH --constraint=array-8core


module load lotus
module load ncbiblast
module load ncbiblastplus

lotus2 -tax_group fungi -taxAligner blast  -i reads/ -o euk_18s_125bp -m euk_18s_map.txt  -t 8 -tmp $TMPDIR -s sdm_hiSeq.txt -p hiSeq

Thanks in advance for your help

Zihao



Falk Hildebrand

unread,
Aug 1, 2023, 9:29:42 AM8/1/23
to LotuS rRNA pipeline
Hey Zihao,
apologies, I check this group only rarely these days (please use github if possible).
334 is not too many, i use sometimes 1000's of samples in a single run. Due to the run breaking at different samples, this is likely a threading error. please ensure that you use the latest sdm (lotus2 github) and if that doesn't work, use "-t 1" to deactivate multiple cores.
hth, Falk

Reply all
Reply to author
Forward
0 new messages