[maker-devel] Segmentation Fault

[Carson Holt]

Seg faults will be the result of one of the programs used by MAKER or MPI itself.

If using OpenMPI, add '-mca btl ^openib’ to your command line.
Example: mpiexec -mca btl ^openib -n 20 maker

Also you nY need to do this before running MAKER —> export LD_PRELOAD=/location/of/openmpi/lib/libmpi.so

Both steps are explained in the ../maker/INSTALL file.

MAKER is not compatible with MVAPICH2 or Intel MPI.


For external programs that segfault. Just reinstall them or pick different versions.


Finally the error from the forks module. See the following from the devel list —> https://groups.google.com/forum/#!topic/maker-devel/2kYQ90TCuXA
It can be fixed with a small edit to the forks.pm file from CPAN (or a change in perl version)

—Carson



> On Jun 28, 2016, at 7:24 PM, Madolyn Macdonald <mmacd at udel.edu> wrote:
> 
> Hi Maker,
> I am trying to run Maker and got the following in my stderr.log file for one of my scaffolds.
> 
> sh: line 1: 34865 Segmentation fault      (core dumped) /usr/local/exonerate-2.2.0/bin/exonerate -q /tmp/maker_kqNxYt/1/NP_001233663%2E1.for.11692140-11693047.1.fasta -t /tmp/maker_kqNxYt/1/picr50.11692140-11693047.1.fasta -Q protein -T dna -m protein2genome --softmasktarget --percent 20 --showcigar > /tmp/maker_kqNxYt/1/picr50.11692140-11693047.NP_001233663%2E1.p.exonerate
> Use of each() on hash after insertion without resetting hash iterator results in undefined behavior, Perl interpreter: 0x1846010 at /usr/lib64/perl5/vendor_perl/forks.pm <http://forks.pm/> line 1736.
> Use of each() on hash after insertion without resetting hash iterator results in undefined behavior, Perl interpreter: 0x1846010 at /usr/lib64/perl5/vendor_perl/forks.pm <http://forks.pm/> line 1736.
> ERROR: Exonerate failed
> --> rank=1, hostname=biomix2.dbi.local
> ERROR: Failed while polishing proteins
> ERROR: Chunk failed at level:10, tier_type:3
> FAILED CONTIG:picr50
> 
> I am running exonerate-2.2.0, anyone know if updating to 2.4.0 would help with this issue?
> Any other thoughts?
> Thanks
> 
> -- 
> Madolyn Stinner (formerly Madolyn MacDonald)
> UDel Bioinformatics and Systems Biology, PhD student
> RIT Alumnus 13'
> _______________________________________________
> maker-devel mailing list
> maker-devel at box290.bluehost.com
> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20160630/8069358d/attachment-0003.html>