Segmentation fault with leon

Question

[intipedroso]

Hi,

I am testing your software LEON to compress fastq files. I am getting a segmentation fault and I do not know how/where to start checking for the possible cause. I have made a small files with the first 250 reads and it works OK ...

Any guidance to find the cause?

BW

[guillaume.rizk]

Hi,

Can you post the output you get? It may help identify where the cause comes from.

Is it some public data, so that I can reproduce the problem?

Guillaume

[intipedroso]

Hi,

Below are the outputs. The test file is only 4M so I can send it by e-mail.

If the file is gzip compressed:

[ipedroso@jimi GATB]$ leon-0.1.1-Linux/bin/leon -c -file test.fastq.gz -abundance 1 -nb-cores 10

gatb-core-library                       
    version                                  : 0.0.0
    build_date                               : 2014-07-31 10:31:33
    build_system                             : Linux-2.6.35.13-92.fc14.x86_64
    build_compiler                           : /opt/bin/gcc  (4.4.7)
    build_kmer_size                          : 32 64 96 128
Violación de segmento
########################

If uncompressed:

[ipedroso@jimi GATB]$ leon-0.1.1-Linux/bin/leon -c -file test.fastq -abundance 1 -nb-cores 10

gatb-core-library                       
    version                                  : 0.0.0
    build_date                               : 2014-07-31 10:31:33
    build_system                             : Linux-2.6.35.13-92.fc14.x86_64
    build_compiler                           : /opt/bin/gcc  (4.4.7)
    build_kmer_size                          : 32 64 96 128

[Bank: fasta to binary                  ]  100    %     elapsed:      0 min 13    sec      estimated remaining:      0 min 0     sec 
[DSK: Pass 9/9, Step 2: counting kmers  ]  99.7   %     elapsed:      8 min 58    sec      estimated remaining:      0 min 1     sec  

dsk                                     
    bankconverter                           
        info                                    
            input                                    : test.fastq
            sequences_number                         : 167221658
            sequences_size                           : 828323679
            output_size                              : 207084184
            ratio                                    : 4.000
        time                                     : 13.133
            conversion                               : 13.133
    dsk                                     
        config                                  
            kmer_size                                : 31
            abundance                                : 1
            available_space                          : 25070
            bank_size                                : 789
            sequence_number                          : 848
            sequence_volume                          : 789
            kmers_number                             : 828296276
            kmers_volume                             : 6319
            max_disk_space                           : 789
            max_memory                               : 1000
            nb_passes                                : 9
            nb_partitions                            : 12
            nb_bits_per_kmer                         : 64
            nb_cores                                 : 10
            partition_type                           : 0
        stats                                   
            kmers_nb_valid                           : 3962464
            kmers_nb_solid                           : 3962464
            kmers_nb_weak                            : 0
            kmers_percent_weak                       : 0.0
        time                                     : 538.402
            fill_partitions                          : 509.183
            fill_solid_kmers                         : 29.219
terminate called after throwing an instance of 'std::length_error'
  what():  basic_string::_S_create
Abortado

[guillaume.rizk]

Thanks for the output. Yes, I would like to have the data, to fix this bug on a future release.

Thanks,
Guillaume