Local Network Controller error

@BenMS thanks for your help.

When I do:

It builds to 100%.
But I am still getting the same error message when I run the Local Network Controller.

EDIT: I did this in a new build directory. I also used option one for new local network and the other 5 default options.

so, I started from scratch with a fresh VM. Here are the exact steps I performed:

sudo apt-get update
sudo apt-get install clang-3.5 cmake build-essential python-psutil libfuse-dev git-all libicu-dev valgrind binutils-gold

git clone git@github.com:maidsafe/MaidSafe
git -C MaidSafe submodule update --init
cd Maidsafe
git submodule foreach "git checkout next; git pull"
git checkout next; git pull
cd build_maidsafe

cmake -DCMAKE_BUILD_TYPE=Release -H/home/hillbicks/MaidSafe/ -Bbuild_maidsafe

cd build_maidsafe
make

build step for ā€˜boost_pythonā€™ still fails. If I can provides any additional information or test something, let me know :slight_smile:

Could you please run

python --version

and post the results.

Thanks

Python 2.7.6

1 Like

Thanks
I have created a new Digital Ocean - Ubuntu droplet and I am currently following the steps you posted to attempt to recreate the issue - I shall post my findings soon - just kicked off the build :wink:

1 Like

Ok I can recreate your issue :slight_smile: & :frowning:

    Scanning dependencies of target boost_python
[ 40%] Creating directories for 'boost_python'
[ 40%] No download step for 'boost_python'
[ 40%] No patch step for 'boost_python'
[ 40%] No update step for 'boost_python'
[ 40%] No configure step for 'boost_python'
[ 40%] Performing build step for 'boost_python'
CMake Error at /home/maidsafe/community_python_issue/MaidSafe/build_maidsafe_clang/boost_1_55_0/src/boost_python-stamp/boost_python-build.cmake:16 (message):
  Command failed: 1

   '/home/maidsafe/community_python_issue/MaidSafe/build_maidsafe_clang/boost_1_55_0_Clang_3_5_0/b2' 'link=static' 'threading=multi' 'runtime-link=shared' '--build-dir=Build' 'stage' '-d+2' '--hash' 'variant=release' 'cxxflags=-fPIC' 'cxxflags=-std=c++11' '-sNO_BZIP2=1' '--layout=tagged' 'toolset=clang-3.5' '--with-python'

  See also

    /home/maidsafe/community_python_issue/MaidSafe/build_maidsafe_clang/boost_1_55_0/src/boost_python-stamp/boost_python-build-*.log



src/third_party_libs/CMakeFiles/boost_python.dir/build.make:103: recipe for target 'boost_1_55_0/src/boost_python-stamp/boost_python-build' failed
make[2]: *** [boost_1_55_0/src/boost_python-stamp/boost_python-build] Error 1
CMakeFiles/Makefile2:1818: recipe for target 'src/third_party_libs/CMakeFiles/boost_python.dir/all' failed
make[1]: *** [src/third_party_libs/CMakeFiles/boost_python.dir/all] Error 2
Makefile:123: recipe for target 'all' failed
make: *** [all] Error 2

Thank you for reporting this - I shall now create a bug in our JIRA system to get it sorted and I am also now trying to recreate this on a real Ubuntu machine.

1 Like

Good to know that it wasnā€™t our incompetence :smiley:

In the meantime Iā€™tm trying to get it running on an arch VM.

1 Like

@hillbicks can you try on the ubuntu VM to

make local_network_controller

instead of just make that builds all targets. Maybe we have a lingering deprecated target.

2 Likes

So, the original thread was: the local network controller not running properly :). Back to that

2 Likes

Well, build succeeded but starting the local_network_controller now fails with a different error

Throw location unknown (consider using BOOST_THROW_EXCEPTION)
Dynamic exception type: N5boost16exception_detail10clone_implINS0_19error_info_injectorIN8maidsafe14maidsafe_errorEEEEE
std::exception::what: Defaulted

Build on arch also fails, Iā€™m still looking into that one and will open a new thread for that one.

Cheers Ross

@hillbicks If you install python-dev

sudo apt-get install python-dev

then it installs the needed header files in /usr/include/python2.7 for the build target boost_python and

make all

should then run successfully.

2 Likes

Yeah, build is already running with python-dev installed (which was really obvious now that I think about it, but as always, hindsight is 20/20) Will comment on the bug in JIRA as soon as the build finished.

Iā€™m still trying to get LNC running, this time with ubuntu 14.10, error message is still the same, but Iā€™ll try again with clang and Debug.

Can you tell me which OS should work as of now? So far I didnā€™t have luck with 14.04, 14.10 or arch :smile:

I use the LNC routinely on Ubuntu 14.04, most often with clang release. I have some subtle changes different from the next branch, but these address potential errors seemingly different from the ones reported here. Iā€™ll have to do some more tests across different machines and different configurations to get to the bottom of this.

Hope to get back to you soon!

1 Like

Boost_python build successfully after installing python-dev.

Will report back later again with results on clang an 14.10.

1 Like

sigh

Still no luck with 14.04.

cmake with CMAKE_BUILD_TYPE=Debug results in

 /home/hillbicks/build_maidsafe/boost_1_55_0_Clang_3_5_0/boost/asio/basic_io_object.hpp:124:5: runtime error: member call on address 0x000007ba90d0 which does not point to an object of type 'boost::asio::waitable_timer_service<std::chrono::_V2::steady_clock, boost::asio::wait_traits<std::chrono::_V2::steady_clock> >'
0x000007ba90d0: note: object has invalid vptr
 00 00 00 00  00 e3 b9 07 00 00 00 00  10 b8 b1 03 00 00 00 00  00 00 00 00 00 00 00 00  58 bb ba 07
              ^~~~~~~~~~~~~~~~~~~~~~~
              invalid vptr
/home/hillbicks/build_maidsafe/boost_1_55_0_Clang_3_5_0/boost/asio/waitable_timer_service.hpp:90:5: runtime error: member access within address 0x000007ba90d0 which does not point to an object of type 'boost::asio::waitable_timer_service<std::chrono::_V2::steady_clock, boost::asio::wait_traits<std::chrono::_V2::steady_clock> >'
0x000007ba90d0: note: object has invalid vptr
 00 00 00 00  00 e3 b9 07 00 00 00 00  10 b8 b1 03 00 00 00 00  00 00 00 00 00 00 00 00  58 bb ba 07
              ^~~~~~~~~~~~~~~~~~~~~~~
              invalid vptr
Throw location unknown (consider using BOOST_THROW_EXCEPTION)
Dynamic exception type: N5boost16exception_detail10clone_implINS0_19error_info_injectorIN8maidsafe14maidsafe_errorEEEEE
std::exception::what: Defaulted

cmake with CMAKE_BUILD_TYPE=Release results in

Throw location unknown (consider using BOOST_THROW_EXCEPTION)
Dynamic exception type: N5boost16exception_detail10clone_implINS0_19error_info_injectorIN8maidsafe14maidsafe_errorEEEEE
std::exception::what: Defaulted

14.10 apparently doesnā€™t like clang, or in other words I canā€™t get it working. I guess I have to wait for the installer to be finished because as of now, I canā€™t get the local_network_controller to start no matter what I try.

EDIT: Damn it, the error is the same, just with more information when cmake is run with debug. Who wouldā€™ve thought :stuck_out_tongue:

So, is there any news to this?

I tried again a couple of days ago and no matter which plattform I use, I canā€™t get the local_network_controller to start, which is a bit odd, considering that testnet3 supposedly starts in march.

The routing_v2 means we will not need a local network controller, vaults themselves will be self starting. For this reason there is no work on LNC right now, not that it wont reappear though, but now we donā€™t need it to start networks :slight_smile:

1 Like

Thanks, thatā€™s good to know. So, there is no way to join the current test network? Iā€™m just very curious to see something in action, get a feeling for the performance impacts on the VMs, connection setups, etc.

I am not a patient man and this project is really yanking my chain :wink:

1 Like