search

marbl / canu

A single molecule sequence assembler for genomes large and small.
http://canu.readthedocs.io/
660 stars 179 forks source link

Failed to submit compute jobs. #2341

Closed yt55 closed 2 months ago

yt55 commented 2 months ago

Hello,

I tried running Canu using the command 

canu -p genome_773m_hf -d genome_773m_hf_assembly genomeSize=773m  -pacbio-hifi sandai.fq.gz gridOptions=--partition=smp02 useGrid=true

on a slurm grid cluster, but got this "abnormal" error:

CRASH: canu 2.2
CRASH: Please panic, this is abnormal.
CRASH:
CRASH:   Failed to submit compute jobs.
CRASH:
CRASH: Failed at /public/software/env01/share/canu-2.2/bin/../lib/site_perl/canu/Execution.pm line 1259.
CRASH:  canu::Execution::submitOrRunParallelJob("genome_773m_hf", "cns", "unitigging/5-consensus", "consensus", 1, 2, 3, 4, ...) called at /public/software/env01/share/canu-2.2/bin/../lib/site_perl/canu/Consensus.pm line 352
CRASH:  canu::Consensus::consensusCheck("genome_773m_hf") called at /public/software/env01/bin/../share/canu-2.2/bin/canu line 1135
CRASH:
CRASH: Last 50 lines of the relevant log file (unitigging/5-consensus/consensus.jobSubmit-01.out):
CRASH:
CRASH: sbatch: error: AssocMaxSubmitJobLimit
CRASH: sbatch: error: Batch job submission failed: Job violates accounting/QOS policy (job submit limit, user's size and/or time limits)

Please ask for help. Thank you very much.

skoren commented 2 months ago

It seems the number of consensus jobs is too large for your grid, what's in unitigging/5-consensus/consensus.jobSubmit-01.sh? Can you also post the full genome_773m_hf.report file to give more info on the assembly up to this point?

yt55 commented 2 months ago

unitigging/5-consensus/consensus.jobSubmit-01.sh The contents are as follows

#!/bin/sh

sbatch \
  --cpus-per-task=8 --mem-per-cpu=64m --partition=smp02 -o consensus.%A_%a.out \
  -D `pwd` -J "cns_genome_773m_hf" \
  -a 1-1000 \
  `pwd`/consensus.sh 0 \
> ./consensus.jobSubmit-01.out 2>&1
yt55 commented 2 months ago

I put the information of the whole genome_773m_hf.report file in the attachment. genome_773m_hf.report.txt

Looking forward to your answer.Thank you very much.

skoren commented 2 months ago

From the report, you've got essentially no assembly. The data is very short (all reads are 1-3kb), is that expected? Typically HiFi reads are 15+kb in length. The histogram also shows a higher than expected error rate (2-copy k-mers) and the peak is only 12-20x coverage. I wouldn't bother finishing this assembly as it will essentially give you back your reads. Given the low coverage and very short reads, I don't think hicanu will work on this sample very well. You could try the suggestions from the FAQ (trimming and/or increasing correctedErrorRate to 0.025 or similar) but I don't expect you'll get much of an assembly anyway.

yt55 commented 2 months ago

Thank you very much for your answer！