I have done a second build with a little improvement in results (94% tests passed, 40 tests failed out of 695 vs the previous result of 94% tests passed, 42 tests failed out of 693) and I can now find the uploaded results here:
Hi @philip_rhoades nice work
If you have spare hardware you could run Nightly CI jobs for us (Debug and Release)
You will find these in the build folder for example: ctest -S CI_Nightly_Release.cmake
and of course come testnet 3 we will be looking for lots of community help
I have been busy with other stuff for a while but have I have done the odd build every now and again to see how things were going - I just did another one which got all the way through but ended with:
Submit files (using http)
Using HTTP submit method
Drop site:http://dash.maidsafe.net/submit.php?project=MaidSafe
Uploaded: /home/phil/src/c++/build_maidsafe/Testing/20150313-0155/Build.xml
Uploaded: /home/phil/src/c++/build_maidsafe/Testing/20150313-0155/Configure.xml
Uploaded: /home/phil/src/c++/build_maidsafe/Testing/20150313-0155/Test.xml
Submission successful
Errors while running CTest
CMakeFiles/Experimental.dir/build.make:49: recipe for target âCMakeFiles/Experimentalâ failed
make[3]: *** [CMakeFiles/Experimental] Error 8
CMakeFiles/Makefile2:370: recipe for target âCMakeFiles/Experimental.dir/allâ failed
make[2]: *** [CMakeFiles/Experimental.dir/all] Error 2
CMakeFiles/Makefile2:378: recipe for target âCMakeFiles/Experimental.dir/ruleâ failed
make[1]: *** [CMakeFiles/Experimental.dir/rule] Error 2
Makefile:291: recipe for target âExperimentalâ failed
make: *** [Experimental] Error 2
To get it to work I had to restore a line in my README recipe file to fix gcc problems again:
git clone git@github.com:maidsafe/MaidSafe
git -C MaidSafe submodule update --init
vim ./MaidSafe/cmake_modules/add_boost.cmake
Change:
find_library(Icui18nLib libicui18n.a)
find_library(IcuucLib libicuuc.a)
find_library(IcudataLib libicudata.a)
to:
.so
cmake -HMaidSafe -Bbuild_maidsafe -DCMAKE_BUILD_TYPE=Debug
cmake.sh # Removes âtoolset=gcc-4.9;â from cmake files
cd build_maidsafe
make
make Experimental
but the main thing that I have been meaning to ask is: How do I do an incremental build? - whenever I have tried to do a âgit pullâ (instead of a âgit clone . .â) in an existing MaidSafe dir and then proceed as normal through the rest of my README recipe, I always have a problem - so I have always had to start from a clean âgit cloneâ again. Is there some trick I am missing?
I finally got around to building a separate (beefier) workstation from my server and have just now tried the newer Rust-based builds. For building crust on Fedora 22 I get:
Compiling crust v0.2.1 (file:///home/phil/src/rust/maidsafe/crust)
src/beacon.rs:168:14: 168:58 error: no method named set_read_timeout found for type std::net::udp::UdpSocket in the current scope
src/beacon.rs:168 .set_read_timeout(Some(Duration::new(10, 0)));
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
error: aborting due to previous error
Could not compile crust.
Sorry, I thought you were talking about nightlies of crust not Rust! OK, so I got past that problem and after installing libsodium[-devel] RPMs everything built OK with only a couple of lint warnings.
Firstly - I donât know how much of the C++ code this has replaced but it is SO much faster to get through this process!
Secondly - is there the equivalent of âmake Experimentalâ to test the code?
Look in the examples sub directory to find out what examples each library had and insert the name to build it. Then run it and hope the help / UI is self explanatory!
!00% plus we are much farther forward now. Significant improvements all around especially security. Itâs a cool project now with a lot of easy to test conjectures, which makes the whole thing faster and much more efficient.
I do a âpullâ serially but run the builds and tests in parallel on my new machine (putting each dir test in a new xterm) - as of just now, âcrustâ is the only one with errors and âcargo testâ hangs indefinitely . .