finished abnormally, OS return value: 12

[yhezema]

Hello,

I have tried to run SPAdes to assemble illumina short reads, I have use default parameters and used mem=240G, CPUS=16 for 16 hours. However, the high computetional resources I used, I received the error message "finished abnormally, OS return value: 12" and the run stopped at "Read error correction" step. I have tried using SPAdes though running Unicycler as well. However, it run for more that 7 houres it stucks at K27 assembly step with the error "finished abnormally, OS return value: -9". However, the high resources I used, the run was finished before using all resources. Please find the analysis.

Please note that I am using paired end reads and the size of R1 and R2.fastq.gz files is around 15G and has around 176,817,872 sequencing reads.

I am wondering if you could help.

spades_RG11.log Unicycler_RG11.log

Thanks, Yasmine

[azeezoe]

We're having the exact same issue while running this using Nextflow, despite allocating more than enough compute resources (tried all the way up to 600GB memory) on our cluster

[asl]

@yhezema How much free RAM is available for your SPAdes job?

Overall the error message is quite clear:

  0:27:03.628     1M / 16G   ERROR   General                 (mmapped_reader.hpp        :  52)   mmap(2) failed. Reason: Cannot allocate memory. Error code: 12

So, the OS is unable to fulfill SPAdes' request to allocate memory. There is nothing that SPAdes could do to overcome this problem.