Learn how easy it is to sync an existing GitHub or Google Code repo to a SourceForge project! See Demo


#269 consensusAfterScaffolder failed with Segmentation fault

Brian Walenz

I am creating three assemblies of different yeast. All assemblies consist of a single PE Illumina library with 70x coverage. The insert size is 500 bp and the read length is 251 bp. Two out of three assemblies run without any troubble, the third assembly fails at the stage consensusAfterScaffolder with the error:
/home/jaco001/gist/Sc/wgsAssembly//assembly//8-consensus/assembly_011 failed -- no .success.

runCA failed.

Stack trace:

at /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/runCA line 1501.
main::caFailure("1 consensusAfterScaffolder jobs failed; remove /home/jaco001/"..., undef) called at /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/runCA line 5542
main::postScaffolderConsensus() called at /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/runCA line 6317

Failure message:

1 consensusAfterScaffolder jobs failed; remove /home/jaco001/gist/Sc/wgsAssembly//assembly//8-consensus/consensus.sh to try again

The logfile of the failed job (number 11) is in the attachment. In this logfile there are many errors like:

ERROR! Column ci=437 has a next pointer (553).
This logfile ends with the following backtrace:

Could (really) not find overlap between 202396 (U) and 2714106 (R) estimated ahang: 660 (ejecting frag 2714106 from contig)

Failed with 'Segmentation fault'

Backtrace (mangled):


Backtrace (demangled):

[0] /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/ctgcns::AS_UTL_catchCrash(int, siginfo, void) + 0x25 [0x40c755]
[1] /lib/x86_64-linux-gnu/libpthread.so.0::(null) + 0xfbd0 [0x7f6847547bd0]
[2] /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/ctgcns::AbacusRefine(MANode, int, int, CNS_RefineLevel, CNS_Options) + 0x42b [0x43e86b]
[3] /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/ctgcns::MultiAlignContig(MultiAlignT, gkStore, CNS_Options*) + 0xc4f [0x42b3df]
[4] /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/ctgcns::(null) + 0x4b0 [0x4096b0]
[5] /lib/x86_64-linux-gnu/libc.so.6::(null) + 0xf5 [0x7f6847190ea5]
[6] /home/jaco001/fungiAssembly/programs/wgs-8.0/Linux-amd64/bin/ctgcns() [0x408eb1]


Segmentation fault (core dumped)

This segmentation could occur because of a lack of memory. But during this step, a maximum of 15% of memory was used (70% of memory was free). Also running step 8-consensus again, leads to the same error. Running the full assembly again leads also to the same error...

2 Attachments


  • Jetse

    This bug is introduced after runCA version 7.0. I ran exactly the same data with exactly the same settings with runCA version 7.0, and I receive no errors. Running the assembly with runCA version 8.1 gives the same error as in version 8.0.

  • Brian Walenz
    Brian Walenz

    Unfortunately, these errors are impossible to fix without access to the data (which is why the report was ignored, my bad). There are some workarounds, namely runCA options cnsReuseUnitigs (enable it) and cnsMaxCoverage (set to 25x or so).

  • Brian Walenz
    Brian Walenz

    • status: open --> closed-out-of-date
    • assigned_to: Brian Walenz