Segmentation fault when I try to access to a vertex

Eric · January 28, 2016, 10:44pm

Hi,

I have a segmentation fault create with a specific data set (other data set
works).
I have some difficulty to find the origin. I try to create a small example,
but for the moment I can only give you the output and backtrack of gdb:

*Output:*
Program received signal SIGSEGV, Segmentation fault.
0x00007ffff0473d62 in uw_init_context_1 () from
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib64/libgcc_s.so.1
Missing separate debuginfos, use: debuginfo-install atlas-3.8.4-2.el6.x86_64
boost-iostreams-1.41.0-27.el6.x86_64 boost-python-1.41.0-27.el6.x86_64
boost-regex-1.41.0-27.el6.x86_64 boost-system-1.41.0-27.el6.x86_64
boost-thread-1.41.0-27.el6.x86_64 bzip2-libs-1.0.5-7.el6_0.x86_64
fontconfig-2.8.0-5.el6.x86_64 freetype-2.3.11-15.el6_6.1.x86_64
glibc-2.12-1.166.el6.x86_64 keyutils-libs-1.4-5.el6.x86_64
krb5-libs-1.10.3-42.el6.x86_64 libX11-1.6.0-6.el6.x86_64
libXau-1.0.6-4.el6.x86_64 libXext-1.3.2-2.1.el6.x86_64
libXrender-0.9.8-2.1.el6.x86_64 libcom_err-1.41.12-22.el6.x86_64
libicu-4.2.1-12.el6.x86_64 libjpeg-turbo-1.2.1-3.el6_5.x86_64
libpng-1.2.49-2.el6_7.x86_64 libselinux-2.0.94-5.8.el6.x86_64
libtiff-3.9.4-10.el6_5.x86_64 libxcb-1.9.1-3.el6.x86_64
openssl-1.0.1e-42.el6_7.2.x86_64 pango-1.28.1-10.el6.x86_64
python-libs-2.6.6-64.el6.x86_64 sqlite-3.6.20-1.el6_7.2.x86_64
zlib-1.2.3-29.el6.x86_64

*bt*:
#0 0x00007ffff0473d62 in uw_init_context_1 () from
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib64/libgcc_s.so.1
#1 0x00007ffff047420e in _Unwind_RaiseException () from
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib64/libgcc_s.so.1
#2 0x00007ffff06d833b in __cxa_throw () at
../../../../libstdc++-v3/libsupc++/eh_throw.cc:82
#3 0x00007fffe4ae40f3 in
_ZNK10graph_tool6detail12graph_actionISt5_BindIFNS_14add_new_vertexESt12_PlaceholderILi1EESt17reference_wrapperINS_14GraphInterfaceEEmS6_IN5boost6python3api6objectEEEENS0_15all_graph_viewsEN4mpl_5bool_ILb0EEEJEEclIJEEEvDpOT_
() from
/*******/lib/python2.7/site-packages/graph_tool/libgraph_tool_core.so
#4 0x00007fffe4a51dee in
graph_tool::add_vertex(graph_tool::GraphInterface&, unsigned long) () from
/*******/lib/python2.7/site-packages/graph_tool/libgraph_tool_core.so
#5 0x00007fffe4a92bfe in
boost::python::objects::caller_py_function_impl<boost::python::detail::caller<boost::python::api::object
(*)(graph_tool::GraphInterface&, unsigned long),
boost::python::default_call_policies,
boost::mpl::vector3<boost::python::api::object,
graph_tool::GraphInterface&, unsigned long> > >::operator()(_object*,
_object*) () from
/***********/lib/python2.7/site-packages/graph_tool/libgraph_tool_core.so
#6 0x00007fffe303ca86 in boost::python::objects::function::call(_object*,
_object*) const () from /usr/lib64/libboost_python.so.5
#7 0x00007fffe303cd98 in ?? () from /usr/lib64/libboost_python.so.5
#8 0x00007fffe30452a3 in
boost::python::detail::exception_handler::operator()(boost::function0<void>
const&) const () from /usr/lib64/libboost_python.so.5
#9 0x00007fffe3ea37a3 in
boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool,
boost::python::detail::translate_exception<graph_tool::ValueException,
void (*)(graph_tool::ValueException const&)>,
boost::_bi::list3<boost::arg<1>, boost::arg<2>, boost::_bi::value<void
(*)(graph_tool::ValueException const&)> > >, bool,
boost::python::detail::exception_handler const&, boost::function0<void>
const&>::invoke(boost::detail::function::function_buffer&,
boost::python::detail::exception_handler const&, boost::function0<void>
const&) () from
/********/lib/python2.7/site-packages/graph_tool/libgraph_tool_core.so
#10 0x00007fffe304527d in
boost::python::detail::exception_handler::operator()(boost::function0<void>
const&) const () from /usr/lib64/libboost_python.so.5
#11 0x00007fffe3ea36d3 in
boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool,
boost::python::detail::translate_exception<graph_tool::IOException, void
(*)(graph_tool::IOException const&)>,
boost::_bi::list3<boost::arg<1>, boost::arg<2>, boost::_bi::value<void
(*)(graph_tool::IOException const&)> > >, bool,
boost::python::detail::exception_handler const&, boost::function0<void>
const&>::invoke(boost::detail::function::function_buffer&,
boost::python::detail::exception_handler const&, boost::function0<void>
const&) () from
/*******/lib/python2.7/site-packages/graph_tool/libgraph_tool_core.so
#12 0x00007fffe304527d in
boost::python::detail::exception_handler::operator()(boost::function0<void>
const&) const () from /usr/lib64/libboost_python.so.5
#13 0x00007fffe3ea3603 in
boost::detail::function::function_obj_invoker2<boost::_bi::bind_t<bool,
boost::python::detail::translate_exception<graph_tool::GraphException,
void (*)(graph_tool::GraphException const&)>,
boost::_bi::list3<boost::arg<1>, boost::arg<2>, boost::_bi::value<void
(*)(graph_tool::GraphException const&)> > >, bool,
boost::python::detail::exception_handler const&, boost::function0<void>
const&>::invoke(boost::detail::function::function_buffer&,
boost::python::detail::exception_handler const&, boost::function0<void>
const&) () from
/*******/lib/python2.7/site-packages/graph_tool/libgraph_tool_core.so
#14 0x00007fffe3045063 in
boost::python::handle_exception_impl(boost::function0<void>) () from
/usr/lib64/libboost_python.so.5
#15 0x00007fffe303a928 in ?? () from /usr/lib64/libboost_python.so.5
#16 0x00000000004197c7 in PyObject_Call (func=0x1c0f070, arg=<value
optimized out>, kw=<value optimized out>) at Objects/abstract.c:2529
#17 0x00000000004a80b0 in do_call (f=<value optimized out>, throwflag=<value
optimized out>) at Python/ceval.c:4239
#18 call_function (f=<value optimized out>, throwflag=<value optimized out>)
at Python/ceval.c:4044
#19 PyEval_EvalFrameEx (f=<value optimized out>, throwflag=<value optimized

) at Python/ceval.c:2666

#20 0x00000000004aa627 in PyEval_EvalCodeEx (co=0x10b66b0, globals=<value
optimized out>, locals=<value optimized out>, args=<value optimized out>,
argcount=1, kws=0x44c8a56c8, kwcount=0, defs=0x1bf7b28, defcount=1,
closure=0x0) at Python/ceval.c:3253
#21 0x00000000004a87d4 in fast_function (f=<value optimized out>,
throwflag=<value optimized out>) at Python/ceval.c:4117
#22 call_function (f=<value optimized out>, throwflag=<value optimized out>)
at Python/ceval.c:4042
#23 PyEval_EvalFrameEx (f=<value optimized out>, throwflag=<value optimized

) at Python/ceval.c:2666

.
. (loop)
.
#124095 0x00000000004a9ab5 in fast_function (f=<value optimized out>,
throwflag=<value optimized out>) at Python/ceval.c:4107
#124096 call_function (f=<value optimized out>, throwflag=<value optimized

) at Python/ceval.c:4042

#124097 PyEval_EvalFrameEx (f=<value optimized out>, throwflag=<value
optimized out>) at Python/ceval.c:2666
#124098 0x00000000004aa627 in PyEval_EvalCodeEx (co=0x7ffff7ece530,
globals=<value optimized out>, locals=<value optimized out>, args=<value
optimized out>, argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0,
closure=0x0)
    at Python/ceval.c:3253
#124099 0x00000000004aa732 in PyEval_EvalCode (co=<value optimized out>,
globals=<value optimized out>, locals=<value optimized out>) at
Python/ceval.c:667
---Type <return> to continue, or q <return> to quit---
#124100 0x00000000004cad5e in run_mod (fp=0x8f3600, filename=0x7fffffffdc71
"/*****/nektar_graph.py", start=<value optimized out>, globals=0x7fe690,
locals=0x7fe690, closeit=1, flags=0x7fffffffd760)
    at Python/pythonrun.c:1370
#124101 PyRun_FileExFlags (fp=0x8f3600, filename=0x7fffffffdc71
"/*******/nektar_graph.py", start=<value optimized out>, globals=0x7fe690,
locals=0x7fe690, closeit=1, flags=0x7fffffffd760)
    at Python/pythonrun.c:1356
#124102 0x00000000004caf87 in PyRun_SimpleFileExFlags (fp=0x8f3600,
filename=0x7fffffffdc71 "/*********/nektar_graph.py", closeit=1,
flags=0x7fffffffd760) at Python/pythonrun.c:948
#124103 0x0000000000414e72 in Py_Main (argc=<value optimized out>,
argv=<value optimized out>) at Modules/main.c:640
#124104 0x000000339ee1ed5d in __libc_start_main () from /lib64/libc.so.6

I use the last version of graph-tool(2.12), with python-2.7.6, gcc-4.9.3,
boost-1.59.0.
Should I try with python-3.X ?

It will be useful, if you have any advice or information to found the origin
of this segmentation fault. Without I comeback when I have more
information.

Regards,
Eric.

Joel_Moberg · January 28, 2016, 11:07pm

If this segfault is easy to reproduce I can try it here, if you send a
script. I have gt 2.12 python 2.7.11 gcc 4.9.3 boost 1.59.0.

attachment.html (10.6 KB)

Eric · January 29, 2016, 6:21pm

Thanks for your offer, but I can't reproduce this segmentation fault on small
script/example. I let you know if I found a way.

When I check libgraph_tool_core.so with ldd:
  linux-vdso.so.1 => (0x00007ffdd6d57000)
  libpython2.7.so.1.0 =>
/soft/bioinfo/linux_RH6/python-2.7.6/lib/libpython2.7.so.1.0
(0x00007ff7cbc49000)
  libboost_iostreams.so.5 => /usr/lib64/libboost_iostreams.so.5
(0x00007ff7cba10000)
  libboost_python.so.5 => /usr/lib64/libboost_python.so.5
(0x00007ff7cb7bf000)
  libboost_regex.so.5 => /usr/lib64/libboost_regex.so.5 (0x00007ff7cb4cb000)
  libboost_coroutine.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_coroutine.so.1.59.0
(0x00007ff7cb2be000)
  libCGAL.so.11 => /u/eaudemard/bin/lib/libCGAL.so.11 (0x00007ff7cb095000)
  libexpat.so.1 => /soft/bioinfo/linux_RH6/gcc-4.9.3/lib/libexpat.so.1
(0x00007ff7cae6c000)
  libbz2.so.1 => /lib64/libbz2.so.1 (0x00007ff7cac5a000)
  libstdc++.so.6 =>
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib/../lib64/libstdc++.so.6
(0x00007ff7ca948000)
  libm.so.6 => /lib64/libm.so.6 (0x00007ff7ca6c4000)
  libc.so.6 => /lib64/libc.so.6 (0x00007ff7ca32f000)
  libgcc_s.so.1 =>
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib/../lib64/libgcc_s.so.1
(0x00007ff7ca119000)
  libpthread.so.0 => /lib64/libpthread.so.0 (0x00007ff7c9efc000)
  libdl.so.2 => /lib64/libdl.so.2 (0x00007ff7c9cf7000)
  libutil.so.1 => /lib64/libutil.so.1 (0x00007ff7c9af4000)
  libz.so.1 => /lib64/libz.so.1 (0x00007ff7c98de000)
  libpython2.6.so.1.0 => /usr/lib64/libpython2.6.so.1.0 (0x00007ff7c9537000)
  libicuuc.so.42 => /usr/lib64/libicuuc.so.42 (0x00007ff7c91e5000)
  libicui18n.so.42 => /usr/lib64/libicui18n.so.42 (0x00007ff7c8e4f000)
  libboost_context.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_context.so.1.59.0
(0x00007ff7c8c43000)
  libboost_thread.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_thread.so.1.59.0
(0x00007ff7c8a21000)
  libboost_chrono.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_chrono.so.1.59.0
(0x00007ff7c881b000)
  libboost_system.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_system.so.1.59.0
(0x00007ff7c8617000)
  librt.so.1 => /lib64/librt.so.1 (0x00007ff7c840f000)
  libmpfr.so.4 => /u/eaudemard/bin/lib/libmpfr.so.4 (0x00007ff7c81b0000)
  libgmp.so.10 => /u/eaudemard/bin/lib/libgmp.so.10 (0x00007ff7c7f39000)
  libboost_thread-mt.so.5 => /usr/lib64/libboost_thread-mt.so.5
(0x00007ff7c7d24000)
  libboost_system-mt.so.5 => /usr/lib64/libboost_system-mt.so.5
(0x00007ff7c7b21000)
  /lib64/ld-linux-x86-64.so.2 (0x0000003756800000)
  libicudata.so.42 => /usr/lib64/libicudata.so.42 (0x00007ff7c69d5000)

Some libboost are linked with the good version of boost but some others are
linked an oldest version (1.41).
I try to fixe this with the option "--with-boost-python=" of configure but
it doesn't work:
./configure --prefix=/u/eaudemard/bin/
--with-python-module-path=/u/eaudemard/bin/lib/python2.7/site-packages/
--with-boost=/soft/bioinfo/linux_RH6/boost-1.59.0/
--with-boost-libdir=/soft/bioinfo/linux_RH6/boost-1.59.0/lib/
--with-boost-python=/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0
...
checking for boostlib >= 1.54.0... yes
checking whether the Boost::Python library is available... yes
checking for exit in
-l/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0... no
checking for exit in
-lboost_python-/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0...
no
checking for exit in -lboost_python... yes

ls -l /soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0
-rwxr-xr-x. 1 adsoft users 404020 Oct 19 13:39
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0

After this i try apm -e of the old version and this time the configure don't
find "boost_python"
checking for boostlib >= 1.54.0... yes
checking whether the Boost::Python library is available... yes
checking for exit in -lboost_python... no
checking for exit in -lboost_python... (cached) no
checking for exit in -lboost_python... (cached) no
checking for exit in -lboost_python... (cached) no
checking for exit in -lboost_python... (cached) no
configure: error: No usable boost::python found

Have you an idea on that point ?

Regards,
Eric.

Joel_Moberg · January 29, 2016, 7:45pm

Could be this that cause the segfault.

Thanks for your offer, but I can't reproduce this segmentation fault on
small
script/example. I let you know if I found a way.

When I check libgraph_tool_core.so with ldd:
        linux-vdso.so.1 => (0x00007ffdd6d57000)
        libpython2.7.so.1.0 =>
/soft/bioinfo/linux_RH6/python-2.7.6/lib/libpython2.7.so.1.0
(0x00007ff7cbc49000)
        libboost_iostreams.so.5 => /usr/lib64/libboost_iostreams.so.5
(0x00007ff7cba10000)
        libboost_python.so.5 => /usr/lib64/libboost_python.so.5
(0x00007ff7cb7bf000)
        libboost_regex.so.5 => /usr/lib64/libboost_regex.so.5
(0x00007ff7cb4cb000)
        libboost_coroutine.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_coroutine.so.1.59.0
(0x00007ff7cb2be000)
        libCGAL.so.11 => /u/eaudemard/bin/lib/libCGAL.so.11
(0x00007ff7cb095000)
        libexpat.so.1 =>
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib/libexpat.so.1
(0x00007ff7cae6c000)
        libbz2.so.1 => /lib64/libbz2.so.1 (0x00007ff7cac5a000)
        libstdc++.so.6 =>
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib/../lib64/libstdc++.so.6
(0x00007ff7ca948000)
        libm.so.6 => /lib64/libm.so.6 (0x00007ff7ca6c4000)
        libc.so.6 => /lib64/libc.so.6 (0x00007ff7ca32f000)
        libgcc_s.so.1 =>
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib/../lib64/libgcc_s.so.1
(0x00007ff7ca119000)
        libpthread.so.0 => /lib64/libpthread.so.0 (0x00007ff7c9efc000)
        libdl.so.2 => /lib64/libdl.so.2 (0x00007ff7c9cf7000)
        libutil.so.1 => /lib64/libutil.so.1 (0x00007ff7c9af4000)
        libz.so.1 => /lib64/libz.so.1 (0x00007ff7c98de000)
        libpython2.6.so.1.0 => /usr/lib64/libpython2.6.so.1.0
(0x00007ff7c9537000)
        libicuuc.so.42 => /usr/lib64/libicuuc.so.42 (0x00007ff7c91e5000)
        libicui18n.so.42 => /usr/lib64/libicui18n.so.42
(0x00007ff7c8e4f000)
        libboost_context.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_context.so.1.59.0
(0x00007ff7c8c43000)
        libboost_thread.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_thread.so.1.59.0
(0x00007ff7c8a21000)
        libboost_chrono.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_chrono.so.1.59.0
(0x00007ff7c881b000)
        libboost_system.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_system.so.1.59.0
(0x00007ff7c8617000)
        librt.so.1 => /lib64/librt.so.1 (0x00007ff7c840f000)
        libmpfr.so.4 => /u/eaudemard/bin/lib/libmpfr.so.4
(0x00007ff7c81b0000)
        libgmp.so.10 => /u/eaudemard/bin/lib/libgmp.so.10
(0x00007ff7c7f39000)
        libboost_thread-mt.so.5 => /usr/lib64/libboost_thread-mt.so.5
(0x00007ff7c7d24000)
        libboost_system-mt.so.5 => /usr/lib64/libboost_system-mt.so.5
(0x00007ff7c7b21000)
        /lib64/ld-linux-x86-64.so.2 (0x0000003756800000)
        libicudata.so.42 => /usr/lib64/libicudata.so.42
(0x00007ff7c69d5000)

Some libboost are linked with the good version of boost but some others are
linked an oldest version (1.41).
I try to fixe this with the option "--with-boost-python=" of configure but
it doesn't work:
./configure --prefix=/u/eaudemard/bin/
--with-python-module-path=/u/eaudemard/bin/lib/python2.7/site-packages/
--with-boost=/soft/bioinfo/linux_RH6/boost-1.59.0/
--with-boost-libdir=/soft/bioinfo/linux_RH6/boost-1.59.0/lib/

--with-boost-python=/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0
...
checking for boostlib >= 1.54.0... yes
checking whether the Boost::Python library is available... yes
checking for exit in
-l/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0... no

You need this library installed (v 1.59.0)? Probably you it's called
libboost-python, search your package repo.

attachment.html (6.45 KB)

Eric · February 1, 2016, 3:45pm

For boost, the answer was here :
http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/Problem-with-configure-and-Boost-Python-td4024944.html

Boost was installed with another version of python, than the version used
for graph-tool.
It's fixed now, but the segmentation fault is still there.

I will come back soon :),
Eric.

Eric · February 2, 2016, 6:44pm

Hi,

I check and fixe all lib: ldd
~/bin/lib/python2.7/site-packages/graph_tool/libgraph_tool_core.so
  linux-vdso.so.1 => (0x00007ffc8f5ea000)
  libpython2.7.so.1.0 =>
/soft/bioinfo/linux_RH6/python-2.7.6/lib/libpython2.7.so.1.0
(0x00007f0ff8088000)
  libboost_iostreams.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_iostreams.so.1.59.0
(0x00007f0ff7e70000)
  libboost_python.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_python.so.1.59.0
(0x00007f0ff7c22000)
  libboost_regex.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_regex.so.1.59.0
(0x00007f0ff7911000)
  libboost_coroutine.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_coroutine.so.1.59.0
(0x00007f0ff7704000)
  libCGAL.so.11 => /u/eaudemard/bin/lib/libCGAL.so.11 (0x00007f0ff74db000)
  libexpat.so.1 => /soft/bioinfo/linux_RH6/gcc-4.9.3/lib/libexpat.so.1
(0x00007f0ff72b2000)
  libbz2.so.1 => /lib64/libbz2.so.1 (0x00007f0ff7077000)
  libstdc++.so.6 =>
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib/../lib64/libstdc++.so.6
(0x00007f0ff6d65000)
  libm.so.6 => /lib64/libm.so.6 (0x00007f0ff6ae1000)
  libc.so.6 => /lib64/libc.so.6 (0x00007f0ff674c000)
  libgcc_s.so.1 =>
/soft/bioinfo/linux_RH6/gcc-4.9.3/lib/../lib64/libgcc_s.so.1
(0x00007f0ff6536000)
  libpthread.so.0 => /lib64/libpthread.so.0 (0x00007f0ff6319000)
  libdl.so.2 => /lib64/libdl.so.2 (0x00007f0ff6114000)
  libutil.so.1 => /lib64/libutil.so.1 (0x00007f0ff5f11000)
  libz.so.1 => /lib64/libz.so.1 (0x00007f0ff5cfb000)
  librt.so.1 => /lib64/librt.so.1 (0x00007f0ff5af2000)
  libicudata.so.42 => /usr/lib64/libicudata.so.42 (0x00007f0ff49a7000)
  libicui18n.so.42 => /usr/lib64/libicui18n.so.42 (0x00007f0ff4611000)
  libicuuc.so.42 => /usr/lib64/libicuuc.so.42 (0x00007f0ff42be000)
  libboost_context.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_context.so.1.59.0
(0x00007f0ff40b3000)
  libboost_thread.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_thread.so.1.59.0
(0x00007f0ff3e91000)
  libboost_chrono.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_chrono.so.1.59.0
(0x00007f0ff3c8a000)
  libboost_system.so.1.59.0 =>
/soft/bioinfo/linux_RH6/boost-1.59.0/lib/libboost_system.so.1.59.0
(0x00007f0ff3a87000)
  libmpfr.so.4 => /u/eaudemard/bin/lib/libmpfr.so.4 (0x00007f0ff3827000)
  libgmp.so.10 => /u/eaudemard/bin/lib/libgmp.so.10 (0x00007f0ff35b1000)

And the segmentation fault is deferent (with gdb):
(gdb) bt
#0 0x000000339ee4406a in vfprintf () from /lib64/libc.so.6
#1 0x000000339ee69609 in vsprintf () from /lib64/libc.so.6
#2 0x000000339ee4f2b8 in sprintf () from /lib64/libc.so.6
#3 0x0000000000464451 in PyString_FromFormatV (format=0x530a57 "",
vargs=0x7fffff3ff8e0) at Objects/stringobject.c:370
#4 0x000000000046491b in PyString_FromFormat (format=<value optimized out>)
at Objects/stringobject.c:410
#5 0x000000000046f0ae in object_repr (self=0x6f51b10) at
Objects/typeobject.c:2996
#6 0x00000000004522a5 in PyObject_Repr (v=0x6f51b10) at
Objects/object.c:381
#7 0x0000000000452668 in internal_print (op=0x6f51b10, fp=0x339f18e7a0,
flags=<value optimized out>, nesting=0) at Objects/object.c:305
#8 0x000000000044ccbd in dict_print (mp=0x44c993830, fp=0x339f18e7a0,
flags=<value optimized out>) at Objects/dictobject.c:1063
#9 0x00000000004525a3 in internal_print (op=0x44c993830, fp=0x339f18e7a0,
flags=1, nesting=0) at Objects/object.c:315
#10 0x000000000042f9d9 in file_PyObject_Print (v=<value optimized out>,
f=0x7ffff7fae150, flags=1) at Objects/fileobject.c:110
#11 PyFile_WriteObject (v=<value optimized out>, f=0x7ffff7fae150, flags=1)
at Objects/fileobject.c:2574
#12 0x00000000004a70d9 in PyEval_EvalFrameEx (f=<value optimized out>,
throwflag=<value optimized out>) at Python/ceval.c:1770
#13 0x00000000004a9ab5 in fast_function (f=<value optimized out>,
throwflag=<value optimized out>) at Python/ceval.c:4107
#14 call_function (f=<value optimized out>, throwflag=<value optimized out>)
at Python/ceval.c:4042
#15 PyEval_EvalFrameEx (f=<value optimized out>, throwflag=<value optimized

) at Python/ceval.c:2666

#16 0x00000000004a9ab5 in fast_function (f=<value optimized out>,
throwflag=<value optimized out>) at Python/ceval.c:4107
#17 call_function (f=<value optimized out>, throwflag=<value optimized out>)
at Python/ceval.c:4042
#18 PyEval_EvalFrameEx (f=<value optimized out>, throwflag=<value optimized

) at Python/ceval.c:2666

The version of /lib64/libc.so come from CentOS 6:
glibc-2.17-106.el7_2.1.x86_64 (compiled with gcc 4.4.7)

We update your OS in 2 weeks, but do you know if this version can create an
segmentation fault ? Or have you an other idea ?

Regards,
Eric.

tiago · February 3, 2016, 9:49am

It is impossible to say anything without the proper context. Since you
haven't told us what usage pattern causes the segfault, by giving us a
minimal script that reproduces it, there is nothing we can say.

Note that your gdb backtrace points to functions deep inside python,
that have nothing to do with graph-tool...

Best,
Tiago

Eric · March 22, 2016, 6:50pm

Hi,

if someone have the same error.
In my case, this segmentation fault was created by a stack overflow.

You can fix it temporary with : "ulimit -s unlimited", and modify your code
if the segmentation fault disappear.

Regards,
Eric.