Spades IO error

[kohei-108]

Running Shovill in Linux server, the following error occurred.

[spades]   0:10:33.510    17G / 18G   ERROR   General                 (kmer_splitters.hpp        : 145)   I/O error! Incomplete write! Reason: No space left on device. Error code: 28
[spades]
[spades]
[spades] == Error ==  system call for: "['/mnt/gpfsA/home/t15089ki/miniconda3/share/spades-3.12.0-1/bin/spades-core', '/mnt/gpfsA/home/t15089ki/2018-05-04-6station-metagenome-fasta/shovill_results/shovill_results_1/spades/K51/configs/config.info']" finished abnormally, err code: 255
[spades]
[spades] ======= SPAdes pipeline finished abnormally and WITH WARNINGS!
[spades]
[spades] === Error correction and assembling warnings:
[spades]  * 0:23:37.309    11G / 26G   WARN    General                 (kmer_coverage_model.cpp   : 327)   Valley value was estimated improperly, reset to 38
[spades]  * 0:35:06.206     9G / 26G   WARN    General                 (simplification.cpp        : 479)   The determined erroneous connection coverage threshold may be determined improperly
[spades] ======= Warnings saved to /mnt/gpfsA/home/t15089ki/2018-05-04-6station-metagenome-fasta/shovill_results/shovill_results_1/spades/warnings.log
[spades]
[spades] === ERRORs:
[spades]  * 0:10:33.510    17G / 18G   ERROR   General                 (kmer_splitters.hpp        : 145)   I/O error! Incomplete write! Reason: No space left on device. Error code: 28
[spades]  * system call for: "['/mnt/gpfsA/home/t15089ki/miniconda3/share/spades-3.12.0-1/bin/spades-core', '/mnt/gpfsA/home/t15089ki/2018-05-04-6station-metagenome-fasta/shovill_results/shovill_results_1/spades/K51/configs/config.info']" finished abnormally, err code: 255
[spades]
[spades] In case you have troubles running SPAdes, you can write to spades.support@cab.spbu.ru
[spades] or report an issue on our GitHub repository github.com/ablab/spades
[spades] Please provide us with params.txt and spades.log files from the output directory.
[shovill] Assembly failed - spades.fasta has zero contigs!

Server has enough space. $ df -h gpfsA 814T 251T 564T 31% /mnt/gpfsA

$ df -i gpfsA 3360606208 242354921 3118251287 8% /mnt/gpfsA

shovill.log

[tseemann]

Can you please paste the output of shovill.log ?

What was the exact command you ran? What version? shovill --version.

Also, did you set --tmpdir on Shovill ? I suspect you have run out of space on /tmp on your HPC node.

[kohei-108]

I sent shivill.log to the previous message.

$ shovill --version shovill 1.0-pre1

I ran the following scripts.

for i in {1..18}
do
        shovill --assembler spades  --ram 500 --cpus 32  --outdir /home/t15089ki/2018-05-04-6station-metagenome-fasta/shovill_results/shovill_results_${i} --R1 /home/t15089ki/2018-05-04-6station-metagenome-fasta/fastq/trimmed-fastq/trimmed_${i}_R1.fq --R2 /home/t15089ki/2018-05-04-6station-metagenome-fasta/fastq/trimmed-fastq/trimmed_${i}_R2.fq
done

[tseemann]

1. You only sent me the Spades part of the log. I need the whole log file to be able to diagnose properly. This will tell me specific versions of all the tools you are using etc.

2. Are you assembling a metagenome? Shovill only works on bacterial isolates. "Illumina whole genome sequencing data of bacteria and other small microbes" I will make this clearer in the documentation.

3. Does your computer have 500 GB of RAM available?

[kohei-108]

I send shovill.log.

I'm assembling metagenomic data.

My computer has more than 500GB of RAM available, and the same error occurred when specifying RAM as 250 GB.

shovill.log

[tseemann]

Sorry, but Shovill can NOT be used with metagenomic data. It only works with a single pure bacterium sample. I do things like estimate the genome size, and use that to make important decisions. You don't have a single genome.

You should use minia 3.x or megahit or spades.py --meta

[kohei-108]

Thanks.