[aspect-devel] Boundary composition (Ying He)

Cox, Samuel P. spc29 at leicester.ac.uk
Thu Oct 6 03:26:03 PDT 2016


Hi Lev,

could you take a look at whether the fix in https://github.com/dealii/dealii/pull/3210 solves your issue? I tried with your parameter file and this alteration and it now seems to run fine for me, but I haven’t checked that it’s doing the ‘right thing’ as far as simulation goes, just that it runs!

Bests,
Sam Cox,
PhD Researcher,
University of Leicester, UK


On 5 Oct 2016, at 23:47, Lev Karatun <lev.karatun at gmail.com<mailto:lev.karatun at gmail.com>> wrote:

Thank you Ying. Average sized models (10m DOFs) with deal8.4.1 crash on the cluster that I'm using seemingly due to memory problems. I'm using the feature with deal8.5 without periodic b.c. right now without any problems. If you could keep me updated on the status of the fix, I would appreciate it =)

Best regards,
Lev Karatun.

2016-09-30 19:26 GMT-04:00 Ying He <yinghe at math.ucdavis.edu<mailto:yinghe at math.ucdavis.edu>>:
Hi Lev,

Currently I only have my Macbook available to use, on which I don't have the newest deal_II.g.so.8.5.0-pre installed. However, your prm files is running OK on my mac machine with deal.ii 4.1. Also based on your debug error messages, it seems something wrong related to dealii::DoFTools::make_flux_sparsity_pattern function call. This deal.ii function has some major changes just implemented recently in deal.ii 8.5pre when running with multiple processors for DG method. This change only affect the memory consumption but not the accuracy. So before we can fix the problem, you could still run your test problems in Aspect with the release deal.ii version. Hope this help.




On 2016-09-30 12:00, aspect-devel-request at geodynamics.org<mailto:aspect-devel-request at geodynamics.org> wrote:
Send Aspect-devel mailing list submissions to
        aspect-devel at geodynamics.org<mailto:aspect-devel at geodynamics.org>

To subscribe or unsubscribe via the World Wide Web, visit
        http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel
or, via email, send a message with subject or body 'help' to
        aspect-devel-request at geodynamics.org<mailto:aspect-devel-request at geodynamics.org>

You can reach the person managing the list at
        aspect-devel-owner at geodynamics.org<mailto:aspect-devel-owner at geodynamics.org>

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Aspect-devel digest..."


Today's Topics:

   1. Re: Boundary composition (John Naliboff)


----------------------------------------------------------------------

Message: 1
Date: Fri, 30 Sep 2016 11:41:46 -0700
From: John Naliboff <jbnaliboff at ucdavis.edu<mailto:jbnaliboff at ucdavis.edu>>
To: aspect-devel at geodynamics.org<mailto:aspect-devel at geodynamics.org>
Subject: Re: [aspect-devel] Boundary composition
Message-ID: <767c0289-21c3-8717-f0c1-0d2314960f73 at ucdavis.edu<mailto:767c0289-21c3-8717-f0c1-0d2314960f73 at ucdavis.edu>>
Content-Type: text/plain; charset="utf-8"; Format="flowed"

Hi Lev,

I just ran the file and looked at the error message with Ying He (one of
DG authors).  Neither of us has seen this error message before.  Ying is
going to take a closer look later, but in the meantime I would try
simplifying your model down further (one compositional field, basic
inflow/outflow) to see if you can point to the issue.

Cheers,
John

*************************************************
John Naliboff
Assistant Project Scientist, CIG
Earth & Planetary Sciences Dept., UC Davis

On 09/30/2016 10:36 AM, Lev Karatun wrote:
Sorry about multiple emails, forgot to attach the prm file that I used.

Best regards,
Lev Karatun.

2016-09-30 13:32 GMT-04:00 Lev Karatun <lev.karatun at gmail.com<mailto:lev.karatun at gmail.com>
<mailto:lev.karatun at gmail.com<mailto:lev.karatun at gmail.com>>>:

    Hi Timo,

    this is what I got from running debug mode through gdb:

    --------------------------------------------------------
    An error occurred in line <917> of file
    </home/lev/aspect/dealii_debug/include/deal.II/grid/tria_iterator.h>
    in function
        const Accessor& dealii::TriaRawIterator<Accessor>::operator*()
    const [with Accessor = dealii::CellAccessor<3, 3>]
    The violated condition was:
        Accessor::structure_dimension!=Accessor::dimension || state()
    == IteratorState::valid
    The name and call sequence of the exception was:
        ExcDereferenceInvalidCell(accessor)
    Additional Information:
    You tried to dereference a cell iterator for which this is not
    possible. More information on this iterator: level=-1, index=-1,
    state=past_the_end

    Stacktrace:
    -----------
    #0  /home/lev/aspect/aspect_debug/aspect:
    dealii::TriaRawIterator<dealii::CellAccessor<3, 3> >::operator*()
    const
    #1  /home/lev/aspect/aspect_debug/aspect:
    dealii::TriaRawIterator<dealii::CellAccessor<3, 3> >::operator->()
    const
    #2  /home/lev/aspect/dealii_debug/lib/libdeal_II.g.so.8.5.0-pre:
    dealii::CellAccessor<3, 3>::neighbor_of_neighbor_internal(unsigned
    int) const
    #3  /home/lev/aspect/aspect_debug/aspect: dealii::CellAccessor<3,
    3>::neighbor_face_no(unsigned int) const
    #4  /home/lev/aspect/dealii_debug/lib/libdeal_II.g.so.8.5.0-pre:
    #5  /home/lev/aspect/dealii_debug/lib/libdeal_II.g.so.8.5.0-pre:
    void
    dealii::DoFTools::make_flux_sparsity_pattern<dealii::DoFHandler<3,
    3>,
    dealii::TrilinosWrappers::BlockSparsityPattern>(dealii::DoFHandler<3,
    3> const&, dealii::TrilinosWrappers::BlockSparsityPattern&,
    dealii::ConstraintMatrix const&, bool, dealii::Table<2,
    dealii::DoFTools::Coupling> const&, dealii::Table<2,
    dealii::DoFTools::Coupling> const&, unsigned int)
    #6  /home/lev/aspect/aspect_debug/aspect:
    aspect::Simulator<3>::setup_system_matrix(std::vector<dealii::IndexSet,
    std::allocator<dealii::IndexSet> > const&)
    #7  /home/lev/aspect/aspect_debug/aspect:
    aspect::Simulator<3>::setup_dofs()
    #8  /home/lev/aspect/aspect_debug/aspect: aspect::Simulator<3>::run()
    #9  /home/lev/aspect/aspect_debug/aspect: main
    --------------------------------------------------------


    Program received signal SIGABRT, Aborted.
    0x0000003edf232625 in raise () from /lib64/libc.so.6
    Missing separate debuginfos, use: debuginfo-install
    librdmacm-1.0.19.1-1.1.el6_7.x86_64
    (gdb) bt
    #0  0x0000003edf232625 in raise () from /lib64/libc.so.6
    #1  0x0000003edf233e05 in abort () from /lib64/libc.so.6
    #2  0x00007ffff4959156 in
    dealii::deal_II_exceptions::internals::abort (exc=..., nothrow=false)
        at /home/lev/aspect/dealii_debug/source/base/exceptions.cc<http://exceptions.cc>:322
    #3  0x0000000001467e2b in
    dealii::deal_II_exceptions::internals::issue_error<dealii::TriaRawIterator<dealii::CellAccessor<3,
    3> >::ExcDereferenceInvalidCell>
    (handling=dealii::deal_II_exceptions::internals::abort_on_exception,
    file=0x1dc0cc8
    "/home/lev/aspect/dealii_debug/include/deal.II/grid/tria_iterator.h",
        line=917, function=0x1dc5240 "const Accessor&
    dealii::TriaRawIterator<Accessor>::operator*() const [with
    Accessor = dealii::CellAccessor<3, 3>]",
        cond=0x1dc1648
    "Accessor::structure_dimension!=Accessor::dimension || state() ==
    IteratorState::valid",
        exc_name=0x1dc1620 "ExcDereferenceInvalidCell(accessor)",
    e=...) at
    /home/lev/aspect/dealii_debug/include/deal.II/base/exceptions.h:266
    #4  0x00000000019fb095 in
    dealii::TriaRawIterator<dealii::CellAccessor<3, 3> >::operator*
    (this=0x7fffffff7a90)
        at
    /home/lev/aspect/dealii_debug/include/deal.II/grid/tria_iterator.h:915
    #5  0x00000000019f4f8e in
    dealii::TriaRawIterator<dealii::CellAccessor<3, 3> >::operator->
    (this=0x7fffffff7a90)
        at
    /home/lev/aspect/dealii_debug/include/deal.II/grid/tria_iterator.h:959
    #6  0x00007ffff5052ccc in dealii::CellAccessor<3,
    3>::neighbor_of_neighbor_internal (this=0x7fffffff7e90, neighbor=3)
        at /home/lev/aspect/dealii_debug/source/grid/tria_accessor.cc:1675
    #7  0x000000000132bcf8 in dealii::CellAccessor<3,
    3>::neighbor_face_no (this=0x7fffffff7e90, neighbor=3)
        at
    /home/lev/aspect/dealii_debug/include/deal.II/grid/tria_accessor.templates.h:3469
    #8  0x00007ffff41a44a5 in dealii::DoFTools::internal::(anonymous
    namespace)::make_flux_sparsity_pattern<dealii::DoFHandler<3, 3>,
    dealii::TrilinosWrappers::BlockSparsityPattern> (dof=...,
    sparsity=..., constraints=..., keep_constrained_dofs=false,
    int_mask=..., flux_mask=..., subdomain_id=0)
        at
    /home/lev/aspect/dealii_debug/source/dofs/dof_tools_sparsity.cc:806
    #9  0x00007ffff42aaca3 in
    dealii::DoFTools::make_flux_sparsity_pattern<dealii::DoFHandler<3,
    3>, dealii::TrilinosWrappers::BlockSparsityPattern> (dof=...,
        sparsity=..., constraints=..., keep_constrained_dofs=false,
    int_mask=..., flux_mask=..., subdomain_id=0)
        at
    /home/lev/aspect/dealii_debug/source/dofs/dof_tools_sparsity.cc:1255
    #10 0x000000000144d1c6 in
    aspect::Simulator<3>::setup_system_matrix (this=0x7fffffff9380,
    system_partitioning=std::vector of length 8, capacity 8 = {...})
        at /home/lev/aspect/aspect_debug/source/simulator/core.cc<http://core.cc>:1106
    #11 0x00000000014463ce in aspect::Simulator<3>::setup_dofs
    (this=0x7fffffff9380) at
    /home/lev/aspect/aspect_debug/source/simulator/core.cc<http://core.cc>:1444
    #12 0x0000000001445363 in aspect::Simulator<3>::run
    (this=0x7fffffff9380) at
    /home/lev/aspect/aspect_debug/source/simulator/core.cc<http://core.cc>:2360
    #13 0x0000000001727506 in main (argc=2, argv=0x7fffffffe0f8) at
    /home/lev/aspect/aspect_debug/source/main.cc<http://main.cc>:487

    Best regards,
    Lev Karatun.

    2016-09-30 10:12 GMT-04:00 Timo Heister <heister at clemson.edu<mailto:heister at clemson.edu>
    <mailto:heister at clemson.edu<mailto:heister at clemson.edu>>>:

        Lev,

        are you running in debug mode? If not, please do so. It would
        also be
        great to run this in gdb and get a full backtrace for us (let
        me know
        if you need help on how to do this).



        --
        Timo Heister
        http://www.math.clemson.edu/~heister/
        <http://www.math.clemson.edu/%7Eheister/>
        _______________________________________________
        Aspect-devel mailing list
        Aspect-devel at geodynamics.org<mailto:Aspect-devel at geodynamics.org> <mailto:Aspect-devel at geodynamics.org<mailto:Aspect-devel at geodynamics.org>>
        http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel
        <http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel>





_______________________________________________
Aspect-devel mailing list
Aspect-devel at geodynamics.org<mailto:Aspect-devel at geodynamics.org>
http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.geodynamics.org/pipermail/aspect-devel/attachments/20160930/041b00ea/attachment-0001.html>

------------------------------

Subject: Digest Footer

_______________________________________________
Aspect-devel mailing list
Aspect-devel at geodynamics.org<mailto:Aspect-devel at geodynamics.org>
http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel

------------------------------

End of Aspect-devel Digest, Vol 58, Issue 21
********************************************

--
Best,
Ying He
_______________________________________________
Aspect-devel mailing list
Aspect-devel at geodynamics.org<mailto:Aspect-devel at geodynamics.org>
http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel

_______________________________________________
Aspect-devel mailing list
Aspect-devel at geodynamics.org<mailto:Aspect-devel at geodynamics.org>
http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/aspect-devel/attachments/20161006/fad5fe4c/attachment-0001.html>


More information about the Aspect-devel mailing list