The failure of the test with rmw_opensplice_cpp is something we are also having on our CI system (see http://ci.ros2.org/view/packaging/job/packaging_linux/902/). While the bridge itself work with OpenSplice the test is currently not working correctly. We haven’t had a chance to look at the test failure - it might be related to the warnings printed by Opensplice related to the metaconfig file as you mentioned.
The test with FastRTPS is expected to pass though. At least it does so on our CI as well as on the developer machines when we tested the recent release. Maybe you can post more information why the test failed in your case.