Hello Sébastien,
Regarding the maximum read length, I've found 16kb to be the limit not 64kb
as you posted previously.
I've been trying to run Ray iteratively by including the cleaned up and
shredded contigs (to less then 50kb) from the previous assembly as one of
the inputs, and am running into a problem after the "Estimation of outer
distances for paired reads" step. I tried in on 2.2.0, 2.3.0 and the
latest version from github: v2.3.0-1-g2b0a6a5. Through trial and error, I
discovered that if I include shredded contigs greater than ~16kb, then it
segfaults, but if they are 16000 bases or less, then everything completes
successfully. (I did not find the actual limit)
It looks to me that MPI_Test is being called with a NULL pointer for the
request, but in looking at the code, I can't see where that would happen.
This is the command: mpirun -bynode -pernode -bycore -bind-to-core
-npernode 16 --mca oob_tcp_if_exclude ib0 memtime
/usr/common/jgi/assemblers/Ray/2.3.0/Ray -o R
ay-jgi_job_wrapper.sh-7460945-k51.d -k 51 -minimum-contig-length 250
-bloom-filter-bits 0 -read-write-checkpoints
/scratch/Ray-jgi_job_wrapper.sh-7460945-k51.d.ckpt -disable-network-test -i
1m.fastq -s 400.fastq -s unmap-min5-contigShreds.fastq.fasta
And here are some excerpts from the log:
Ray version 2.3.0
License for Ray: GNU General Public License version 3
RayPlatform version: 2.0.0
License for RayPlatform: GNU Lesser General Public License version 3
MAXKMERLENGTH: 160
Maximum coverage depth stored by CoverageDepth: 4294967295
MAXIMUM_MESSAGE_SIZE_IN_BYTES: 4000 bytes
FORCE_PACKING = n
ASSERT = n
HAVE_LIBZ = y
HAVE_LIBBZ2 = y
PROFILER_COLLECT = n
MINIRANKS = y
CLOCK_GETTIME = n
__linux__ = y
_MSC_VER = n
__GNUC__ = y
RAY_32_BITS = n
RAY_64_BITS = y
MPI standard version: MPI 2.1
MPI library: Open-MPI 1.6.4
Compiler: GNU gcc/g++ 4.8.1
MPI I/O: n
….
***
Step: Estimation of outer distances for paired reads
Date: Mon Nov 4 14:24:38 2013
Elapsed time: 33 seconds
Since beginning: 52 minutes, 29 seconds
***
….
Rank 6 starts on seed 2526, length is 454, flow 0 [2526/6125]
Rank 12 traversed 2436293 nucleotide symbols
Rank 12: assembler memory usage: 766920 KiB
Rank 12 starts on seed 2138, length is 520, flow 0 [2138/6278]
Rank 10: assembler memory usage: 753516 KiB
Rank 10 starts on seed 2486, length is 455, flow 0 [2486/6164]
Rank 13 traversed 2319776 nucleotide symbols
Rank 13: assembler memory usage: 758376 KiB
Rank 13 starts on seed 2089, length is 501, flow 0 [2089/6150]
Rank 2: assembler memory usage: 756088 KiB
Rank 2 starts on seed 2657, length is 437, flow 0 [2657/6068]
[mc0555:08747] *** Process received signal ***
[mc0555:08747] Signal: Bus error (7)
[mc0555:08747] Signal code: (128)
[mc0555:08747] Failing at address: (nil)
[mc0555:08747] [ 0] /lib/libpthread.so.0(+0xeff0) [0x2acd0fc8bff0]
[mc0555:08747] [ 1]
/global/common/genepool/usg/hpc/openmpi/gnu4.6/1.6.4/ib_ofed2/lib/libmpi.so.1(ompi_request_default_test+0x24)
[0x2acd0e7b2fb4]
[mc0555:08747] [ 2]
/global/common/genepool/usg/hpc/openmpi/gnu4.6/1.6.4/ib_ofed2/lib/libmpi.so.1(MPI_Test+0x6c)
[0x2acd0e7d48cc]
[mc0555:08747] [ 3]
/usr/common/jgi/assemblers/Ray/2.3.0/Ray(_ZN13RingAllocator17cleanDirtyBuffersEv+0x8c)
[0x59dbfc]
[mc0555:08747] [ 4]
/usr/common/jgi/assemblers/Ray/2.3.0/Ray(_ZN15MessagesHandler12sendMessagesEP12StaticVectorP13RingAllocator+0x39)
[0x5b73c9]
[mc0555:08747] [ 5]
/usr/common/jgi/assemblers/Ray/2.3.0/Ray(_ZN11ComputeCore12sendMessagesEv+0x156)
[0x5ba236]
[mc0555:08747] [ 6]
/usr/common/jgi/assemblers/Ray/2.3.0/Ray(_ZN11ComputeCore15runWithProfilerEv+0x534)
[0x5bc754]
[mc0555:08747] [ 7]
/usr/common/jgi/assemblers/Ray/2.3.0/Ray(_ZN11ComputeCore3runEv+0xaa)
[0x5bf16a]
[mc0555:08747] [ 8]
/usr/common/jgi/assemblers/Ray/2.3.0/Ray(_ZN7Machine5startEv+0x1416)
[0x477e96]
[mc0555:08747] [ 9] /usr/common/jgi/assemblers/Ray/2.3.0/Ray(main+0x297)
[0x4757c7]
[mc0555:08747] [10] /lib/libc.so.6(__libc_start_main+0xfd) [0x2acd0feb7c8d]
[mc0555:08747] [11] /usr/common/jgi/assemblers/Ray/2.3.0/Ray() [0x4727f1]
[mc0555:08747] *** End of error message ***
Killed [7]
On Wed, Aug 28, 2013 at 6:13 AM, Sébastien Boisvert <
sebastien.boisver...@ulaval.ca> wrote:
> On 28/08/13 04:15 AM, Hornung, Bastian wrote:
> > Hi Sébastien,
> >
> > the assembly is now running, so it seems to be working, thanks :).
>
> Thank you for testing the patch.
>
> I merged it in the master branch.
>
>
> https://github.com/sebhtml/ray/commit/20d20c1cef4136281b46e14115631a2326920550
>
> >
> > Bastian
> >
> > ________________________________________
> > From: Sébastien Boisvert [sebastien.boisver...@ulaval.ca]
> > Sent: Tuesday, August 27, 2013 4:37 PM
> > To: Hornung, Bastian
> > Cc: denovoassembler-users@lists.sourceforge.net
> > Subject: Re: [Denovoassembler-users] maximum read length for ray
> >
> > (Please use the list)
> >
> > Hello,
> >
> > I think I found the problem.
> >
> >
> > Can you try with the enclosed patch ?
> >
> >
> > To build Ray v2.2.0 with this patch:
> >
> >
> > wget
> http://sourceforge.net/projects/denovoassembler/files/Ray-v2.2.0.tar.bz2
> > wget
> https://raw.github.com/sebhtml/patches/master/ray/Ray-v2.2.0-fix-for-amos-and-long-reads.patch
> >
> > tar -xjf Ray-v2.2.0.tar.bz2
> > cd Ray-v2.2.0
> > patch -p1 < ../Ray-v2.2.0-fix-for-amos-and-long-reads.patch
> >
> > make -j 20
> >
> > mkdir -p ~/software/ray/2.2.0-1-long-reads-amos
> > cp Ray~/software/ray/2.2.0-1-long-reads-amos
> >
> >
> >
> >
> > -Sébastien
> >
> > On 27/08/13 03:58 AM, Hornung, Bastian wrote:
> >> Okay, that's then obviously not the problem.
> >> I've for testing purposes installed a version locally, once with
> different parameters, afterwards with the standard parameters, and loaded
> the PacBio data into it. And it runs.
> >> If I now try it again on our server (both the local linux machine as
> well as the server are Ubuntu 12.04), no matter if I freshly compile it or
> use the already present installation (also without any different
> installation parameters) I get again a stack overflow warning, see the
> command line output:
> >>
> >>
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test_illumina_read_1/NG-5257_test_read_1.fastq with
> lazy loading (please wait...)
> >> Rank 3 is fetching file
> /home/bastian/data/RAW/test1_GTCCGC_L008_R2_001_AC0RCJACXX.filt.fastq with
> lazy loading (please wait...)
> >> Rank 2 is fetching file
> /home/bastian/data/RAW/test1_GTCCGC_L008_R1_001_AC0RCJACXX.filt.fastq with
> lazy loading (please wait...)
> >> Rank 1 is fetching file
> /home/bastian/data/RAW/test_illumina_read_2/NG-5257_test_read_2.fastq with
> lazy loading (please wait...)
> >> Rank 4 is fetching file /home/bastian/data/RAW/PacBio/all_ccs.fastq
> with lazy loading (please wait...)
> >> Rank 4: File /home/bastian/data/RAW/PacBio/all_ccs.fastq (Number 4) has
> 3699 sequences
> >> Rank 2: File
> /home/bastian/data/RAW/test1_GTCCGC_L008_R1_001_AC0RCJACXX.filt.fastq
> (Number 2) has 9886736 sequences
> >> Rank 3: File
> /home/bastian/data/RAW/test1_GTCCGC_L008_R2_001_AC0RCJACXX.filt.fastq
> (Number 3) has 9886736 sequences
> >> Rank 0: File
> /home/bastian/data/RAW/test_illumina_read_1/NG-5257_test_read_1.fastq
> (Number 0) has 34807392 sequences
> >> Rank 1: File
> /home/bastian/data/RAW/test_illumina_read_2/NG-5257_test_read_2.fastq
> (Number 1) has 34807392 sequences
> >> Rank 0 wrote
> /home/bastian/data/RAW/test_illumina_untrimmed_matepair_pacbio_ray_k31/NumberOfSequences.txt
> >> Rank 0 wrote
> /home/bastian/data/RAW/test_illumina_untrimmed_matepair_pacbio_ray_k31/SequencePartition.txt
> >>
> >> ***
> >> Step: Counting sequences to assemble
> >> Date: Tue Aug 27 09:42:39 2013
> >> Elapsed time: 18 seconds
> >> Since beginning: 18 seconds
> >> ***
> >>
> >>
> >> Rank 0 is adding sequences to
> /home/bastian/data/RAW/test_illumina_untrimmed_matepair_pacbio_ray_k31/AMOS.afg
> >>
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test_illumina_read_1/NG-5257_test_read_1.fastq with
> lazy loading (please wait...)
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test_illumina_read_1/NG-5257_test_read_1.fastq with
> lazy loading (please wait...)
> >> adRank 0 is fetching file
> /home/bastian/data/RAW/test_illumina_read_2/NG-5257_test_read_2.fastq with
> lazy loading (please wait...)
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test_illumina_read_2/NG-5257_test_read_2.fastq with
> lazy loading (please wait...)
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test1_GTCCGC_L008_R1_001_AC0RCJACXX.filt.fastq with
> lazy loading (please wait...)
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test1_GTCCGC_L008_R1_001_AC0RCJACXX.filt.fastq with
> lazy loading (please wait...)
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test1_GTCCGC_L008_R2_001_AC0RCJACXX.filt.fastq with
> lazy loading (please wait...)
> >> Rank 0 is fetching file
> /home/bastian/data/RAW/test1_GTCCGC_L008_R2_001_AC0RCJACXX.filt.fastq with
> lazy loading (please wait...)
> >> Rank 0 is fetching file /home/bastian/data/RAW/PacBio/all_ccs.fastq
> with lazy loading (please wait...)
> >> Rank 0 is fetching file /home/bastian/data/RAW/PacBio/all_ccs.fastq
> with lazy loading (please wait...)
> >> *** stack smashing detected ***: Ray terminated
> >> [ssb3:23288] *** Process received signal ***
> >> [ssb3:23288] Signal: Segmentation fault (11)
> >> [ssb3:23288] Signal code: (128)
> >> [ssb3:23288] Failing at address: (nil)
> >>
> >> I have no idea what's going on.
> >> I doubt that I can run the assembly on my local machine, don't think I
> have enough RAM here.
> >> There shouldn't be any difference between the 2 machines, that's what
> puzzles me.
> >>
> >> Any ideas?
> >>
> >> Bastian
> >>
> >>
> >> ________________________________________
> >> From: Sébastien Boisvert [sebastien.boisver...@ulaval.ca]
> >> Sent: Monday, August 26, 2013 3:12 PM
> >> To: Hornung, Bastian
> >> Cc: denovoassembler-users@lists.sourceforge.net
> >> Subject: Re: [Denovoassembler-users] maximum read length for ray
> >>
> >> On 26/08/13 04:56 AM, Hornung, Bastian wrote:
> >>> Hi @all,
> >>>
> >>> I'd have a technical question about ray: What is the maximum read
> length, which can be used?
> >>> I ask because I get a stack overflow (or rather a message that the
> process has been aborted before it could happen) when I try to run ray with
> PacBio data (up to 2kb in that set, IIRC), and wonder what the maximum is
> what ray can take.
> >>>
> >>
> >> It is 65536 nucleotides.
> >>
> >>> Best regards,
> >>>
> >>> Bastian
> >>>
> >>>
> >>>
> ------------------------------------------------------------------------------
> >>> Introducing Performance Central, a new site from SourceForge and
> >>> AppDynamics. Performance Central is your source for news, insights,
> >>> analysis and resources for efficient Application Performance
> Management.
> >>> Visit us today!
> >>>
> http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk
> >>> _______________________________________________
> >>> Denovoassembler-users mailing list
> >>> Denovoassembler-users@lists.sourceforge.net
> >>> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
> >>>
> >>
> >>
> >>
> >
> >
> >
> >
> ------------------------------------------------------------------------------
> > Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
> > Discover the easy way to master current and previous Microsoft
> technologies
> > and advance your career. Get an incredible 1,500+ hours of step-by-step
> > tutorial videos with LearnDevNow. Subscribe today and save!
> >
> http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk
> > _______________________________________________
> > Denovoassembler-users mailing list
> > Denovoassembler-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
> >
>
>
>
> ------------------------------------------------------------------------------
> Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
> Discover the easy way to master current and previous Microsoft technologies
> and advance your career. Get an incredible 1,500+ hours of step-by-step
> tutorial videos with LearnDevNow. Subscribe today and save!
> http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk
> _______________________________________________
> Denovoassembler-users mailing list
> Denovoassembler-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
>
------------------------------------------------------------------------------
November Webinars for C, C++, Fortran Developers
Accelerate application performance with scalable programming models. Explore
techniques for threading, error checking, porting, and tuning. Get the most
from the latest Intel processors and coprocessors. See abstracts and register
http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk
_______________________________________________
Denovoassembler-users mailing list
Denovoassembler-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/denovoassembler-users
