Does stand alone ribopicker have a max # of sequences allowed - trying to explain segmentation fault

[Dallas Thomas]

Wondering if there is a way to determine what is causing the segfault in bwa64. I have attempted to run the bwa command with different flags and am coming up with the same error.

This is what I see:

bwa64 bwasw -A -f tmp.tsv -z 1 /data/ribodb/rrnadb LA1224mod_1.fastq
[bsw2_aln] read 99010 sequences (10000010 bp)...
Segmentation fault

Memory should not be an issue as the machine I am running the command on is a 48 core server with 1TB of RAM.

I have tested with the -S command setting the value quite high and all that happens is the segmentation fault occurs after more sequences read in.

Is this version of bwa having issues with number of sequences? This should not be an issue I would think?

[Robert Schmieder]

Does the issue occur after processing the same number of sequences from other datasets? If not, then the dataset might be causing the segmentation fault. If yes, then you might not have enough memory (even if the system has 1TB, you might only have access to a few GB).

[Dallas Thomas]

Hello Robert,

I do not believe it has anything to do with memory as I am the system administrator for the system - unless bwa64 has something written in that would limit the RAM it is using.

As the issue was only occurring with bwa64 I decided to run a quick test. I re-indexed the rrnadb.fasta file with both bwa64 and bwa (version 0.7.5) and then started the alignment again with both bwa64 and bwa.  I removed the -A flag of course.  The result: bwa64 seg faulted and bwa ran without any problems.

Any other ideas?

Dallas

[Dallas Thomas]

Robert,

bwa64 is segfaulting on every dataset I test it with.