Legend:
- Unmodified
- Added
- Removed
-
INSTALL
r10642 r10700 5 5 about Ariba and its application. 6 6 7 7 8 Prerequisites 8 9 ============= 9 10 ariba depends on libraries that may not be installed on your system: 10 11 11 * Boost (version >=1. 39)12 * Log4cxx (version >= 0.10.0)13 * GMP (no specific version)12 * Boost (version >=1.42) 13 * GMP 14 * CMake (version >=2.8) 14 15 15 Furthermore, you need default development tools - that are most likely already installed on your system - such as 16 gcc/g++, autoconf, automake, aclocal, libtool, liblt-dev ... 16 and optionally on 17 18 * Log4Cxx (version >= 0.10.0) for more sophisticated logging 19 * Avahi for more efficient bootstrapping in local networks 20 * LibBluetooth/Bluez for bluetooth support 21 * Doxygen to build the documentation 22 23 Furthermore, you need default development tools - that are most likely already 24 installed on your system - such as gcc/g++, libtool, liblt-dev ... 25 17 26 18 27 Quick Install (If all dependencies are satisfied) 19 28 ============= 20 29 21 ariba currently build on Linux systems. Our reference platform is 22 Ubuntu 8.04 with the g++ compiler version 4.1. However, latest 23 Ubuntu releases should work, too. Download the latest 24 ariba package from the the download site 25 http://ariba-underlay.org/downloads 26 or get the latest trunk code: 27 svn co https://svn.tm.kit.edu/SpoVNet-KA/entwicklung/ariba/trunk ariba-trunk 30 ariba currently builds on Linux systems. Our reference platform is Ubuntu 8.04 31 with the g++ compiler version 4.1. However, latest Ubuntu releases should work 32 too. 28 33 29 Extract the archive and change into the project directory: 34 Download the latest ariba package from the the download site 35 http://ariba-underlay.org/downloads or get the latest trunk code: 36 > svn co https://svn.tm.kit.edu/SpoVNet-KA/entwicklung/ariba/trunk ariba-trunk 30 37 38 If you downloaded ariba from the website extract the archive and change into the 39 project directory: 31 40 > tar xfz ./ariba-x.x.x.tar.gz 32 41 > cd ariba-x.x.x 33 42 34 For SVN checkout you have to 43 For SVN checkout you have to: 35 44 > cd ariba-trunk 36 45 37 Now , configure, compile, and install ariba (if not all libraries38 needed by Ariba are available on your system, read the section 39 'Prequisites'). 46 Now create a directory to build ariba in: 47 > mkdir build 48 > cd build 40 49 41 If no configure script is available (e.g. when you checked out an svn 42 version), run the ./bootstrap script first and then type: 43 44 > ./configure 50 Next the makefiles have to be generated and the source compiled: 51 > cmake .. 45 52 > make 46 > make install47 53 48 54 HINT: you may use … … 52 58 the compilation stops, try make without the -j option again. 53 59 54 55 In case you don't want to install Ariba into your system but to a 56 local place, do: 57 58 > mkdir build 59 > ./configure --prefix=$PWD/build 60 > make 60 And finally ariba will be installed into the system: 61 61 > make install 62 62 63 63 64 Local install (Download library dependencies and install ariba in a 65 ============= local subdirectory) 64 Custom Build Options 65 ==================== 66 The build may be customized in various ways by setting CMake options. This can 67 be done by giving them as arguments on the command line: 68 > cmake .. -DOPTION=value 66 69 70 by using the CMake GUI which lets you set the variables graphically: 71 > cmake-gui .. 67 72 68 If the required libraries are not available on your system or don't 69 want to install them to the system you can use the bootstrap_libs 70 script that comes with the ariba package. It will download all 71 dependencies, compile and install them locally, and compile and 72 install ariba: 73 or running cmake in interactive mode: 74 > cmake -i .. 73 75 74 > tar xfz ./ariba-x.x.x.tar.gz 75 > cd ariba-x.x.x 76 > ./bootstrap_libs 76 The last two ways also give an overview which options exist. 77 77 78 This script will automatically download all necessary libraries and 79 compile them inside a new ./libs subdirectory and install them into 80 ./build. In the end the ariba library itself is compiled to 81 ./build. You can safely delete the ./libs subdirectory after the 82 process. 78 Important options: 83 79 84 Here is the manual way to go: If you install Ariba locally and have 85 the required libraries also installed locally, you can use a 86 config.site script to make it easier. The config.site file must reside 87 in a folder called share. If your install path is /home/foo/local and 88 you do a ./configure --prefix=/home/foo/local, ariba headers will be 89 installed in /home/foo/local/include, and the ariba library in 90 /home/foo/local/lib. To use a config.site script, create a folder 91 /home/foo/local/share and create a file config.site. Such a file has 92 paths towards required header files and libraries. 80 CMAKE_INSTALL_PREFIX - Where to install the compiled files. The default on Unix 81 platforms is /usr/local/. If you for example don't want 82 or can't install system wide, you can specify a directory 83 you have control over. The files will be installed to 84 "${prefix}/include/", "${prefix}/lib/" and so on. 93 85 94 with_boost=/home/foo/Libraries/include 95 test -z "$CPPFLAGS" && CPPFLAGS='-I/home/foo/Libraries/include' 96 test -z "$LDFLAGS" && LDFLAGS='-L/home/foo/Libraries/lib' 86 CMAKE_BUILD_TYPE - One of "", "Release", "Debug", "RelWithDebInfo" or 87 "MinSizeRel". This influences the build in various ways 88 (e.g. which compiler optimizations are turned on, whether 89 debug symbols are included, what warnings to show). 97 90 98 If you now do a ./configure --prefix=/home/foo/local, the config.site 99 will be found and the paths therein used for finding libraries. If you 100 e.g. have multiple libraries distributed in their own include folders, 101 you can also have multiple includes: 91 ENABLE_{AVAHI,BLUETOOTH,LOG4CXX} 92 - If set to OFF or 0 (ON is the default) it disables the 93 support of the feature even if the corresponding library 94 (see above) was detected to be present. 102 95 103 test -z "$CPPFLAGS" && CPPFLAGS='-I/home/foo/Libraries/include -I/home/foo/otherlibrary/include' 96 <library>_INCLUDE_DIR - Where the directory containing the header files for 97 <library> is located. If the library is installed in the 98 usual system paths CMake should be able to automatically 99 find the right location. If the library is located 100 elsewhere (e.g. because you compiled it yourself in your 101 home directory) then you may need to set this variable 102 manually. 103 104 <library>_LIBRARY - Where the library file (aka the .so, .a or .dll file) for 105 <library> is located. If the library is installed in the 106 usual system paths CMake should be able to automatically 107 find the right location. If the library is located 108 elsewhere (e.g. because you compiled it yourself in your 109 home directory) then you may need to set this variable 110 manually. 111 112 DOCUMENTATION_GENERATE_GRAPHICS 113 - Whether the documentation should include graphics such as 114 inheritance and include graphs (OFF by default). This 115 might take a long time and consume a lot of space. 116 117 CMAKE_{C,CXX}_COMPILER - Which C/C++ compiler to use 118 CMAKE_{C,CXX}_FLAGS - Which additional flags to give to the compiler (e.g. -pg 119 for profiling support) 120 104 121 105 122 Running the PingPong Sample 106 123 =========================== 107 The PingPong binary pingpong is installed in build/bin. It has one 108 parameter, a configuration file. You can find sample configuration 109 files in the etc/pingpongconfig folder. If no configuration file is 110 given, the node will randomly select its NodeID but will not find 111 other nodes. This is because bootstrap modules are selected in the 112 configuration file. 124 The PingPong binary pingpong is installed in "${prefix}/lib/ariba/" or found 125 directly in the build tree at "sample/pingpong/pingpong". It has one parameter, 126 a configuration file. You can find sample configuration files in the 127 "etc/pingpong" folder. If no configuration file is given, the node will randomly 128 select its NodeID but will not find other nodes. This is because bootstrap 129 modules are selected in the configuration file. 130 > ./sample/pingpong/pingpong ../etc/pingpong/settings_node1.cnf 113 131 114 > ./pingpong ../../etc/pingpong/settings_node1.cnf 132 If this fails to find the libariba you may have to set the LD_LIBRARY_PATH 133 correctly in your current terminal, or better add it to your .bashrc 134 > export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/user/ariba/build/source/ariba 115 135 116 If this fails to find the libariba you may have to set 117 the LD_LIBRARY_PATH correctly in your current terminal, 118 or better add it to your .bashrc 136 When running the pingpong application it will output a large number of log 137 messages and the initiator will wait for other nodes to join. You can start them 138 using the configuration files settings_node1.cnf and settings_node2.cnf. You may 139 need to adjust the configurations files: currently both node1 and node2 try to 140 join the initiator on the local machine. This will only work if you start all 141 instances on a local machine. 119 142 120 > export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/user/ariba/build/lib 143 Once the PingPong sample is running and the nodes have connected, each node will 144 send out ping messages to every node he knows in the overlay structure every 145 5 seconds. You can now e.g. test mobility of Ariba and change the IP address of 146 a node, or swith from LAN connection to WLAN. The links established by the 147 PingPong sample through Ariba are mobility invariant and automatically repaired. 121 148 122 When running the pingpong application it will output a large number of123 log messages and the initiator will wait for other nodes to join. You124 can start them using the configuration files settings_node1.cnf and125 settings_node2.cnf. You may need to adjust the configurations files:126 currently both node1 and node2 try to join the initiator on the local127 machine. This will only work if you start all instances on a local128 machine.129 149 130 Once the PingPong sample is running and the nodes have connected, each 131 node will send out ping messages to every node he knows in the overlay 132 structure every 5 seconds. You can now e.g. test mobility of Ariba and 133 change the IP address of a node, or swith from LAN connection to 134 WLAN. The links established by the PingPong sample through Ariba are 135 mobility invariant and automatically repaired. Selecting a compiler 150 Cross-Compiling for Maemo 151 ========================= 152 Ariba runs on Nokia Maemo 4 (tested) and probably Maemo 5. We have tested Ariba 153 on an N810 device. Cross-Compiling is done using Scratchbox. Use the 154 preassembled Scratchbox version provided by Nokia which will install and 155 configure the complete Scratchbox system automatically. 136 156 137 As the g++-4.3 compiler is very restrictive when compiling C++ and you 138 will have some trouble with Boost and Log4cxx, we suggest to use 139 e.g. g++-4.1. You then have to compile the libraries and Ariba with 140 this compiler. You can tell Log4cxx and Ariba to use a different 141 compiler using: 157 If you compile for Maemo you have to set the HAVE_MAEMO option in CMake. 142 158 143 ./configure --prefix=... CXX=g++-4.1 144 145 This will not work in Boost as the configure script is just a wrapper around the Boost Build.System bjam. You can edit the jamfile in the Boost root directory: 146 147 using gcc : 4.1 ; 148 149 and then build using bjam as described in http://www.boost.org/doc/libs/1_38_0/more/getting_started/unix-variants.html. 150 151 Cross-Comiling for Maemo 152 ======================== 153 Ariba runs on Nokia Maemo 4 (tested) and probably Maemo 5. We have 154 tested Ariba on an N810 device. Cross-Compiling is done using 155 Scratchbox. Use the preassembled Scratchbox version provided by Nokia 156 which will install and configure the complete Scratchbox system 157 automatically. 158 159 The Ariba configure will test for Maemo systems. Internally there are 160 a number of special cases where handling on Maemo is different from 161 normal Linux. If you require special handling, do the following in 162 your code: 159 Internally there are a number of special cases where handling on Maemo is 160 different from normal Linux. If you require special handling, do the following 161 in your code: 163 162 164 163 #include "ariba/config.h" … … 167 166 // special Maemo handling 168 167 #endif 169 170 Overview of special configure options171 =====================================172 There are several options to configure that are specific to Ariba:173 174 * --enable-debug=yes - for building a debug build175 * --enable-profiling=yes - for profiling with gprof176 * --enable-logcolors=yes - for colorful logging output177 * --enable-doxygen=yes - for generating doxygen documentation (do a make html-local in ariba/docu/doxygen)178
Note:
See TracChangeset
for help on using the changeset viewer.
