Skip to content

INSTALL.md

Latest commit

 

History

History
105 lines (72 loc) · 4.19 KB

INSTALL.md

File metadata and controls

105 lines (72 loc) · 4.19 KB

INSTALLATION INSTRUCTIONS

The INET Framework can be compiled on any platform supported by OMNeT++.

Recommended Installation

The recommended way to install INET is to use the opp_env tool to automatically install both INET and OMNeT++. It will handle the installation of all dependencies and configure the environment variables and guaranties a reproducible installation. With opp_env, you will be able to install various other models as well.

Manual Installation

If you opt to install INET manually, you will need to install OMNeT++ first.

General

  1. Make sure your OMNeT++ installation works OK (e.g. try running the samples) and it is in the path (to test, try the command "which omnetpp", it should print the path of the executable). On Windows, open a console with the mingwenv.cmd command. The PATH and other variables will be automatically adjusted for you. Use this console to compile and run INET.

  2. Extract the downloaded tarball into a directory of your choice (usually into your workspace directory, if you are using the IDE). NOTE: The built-in Windows archiver has bugs and cannot extract the file correctly. Use some other archiver or do it from command line (tar xvfz inet-x.y.z-src.tgz)

If you are building from command line:

  1. Change to the INET directory and source the setenv script.

    $ source setenv

  2. Make sure that any required Python modules are properly installed by executing pip install -r python/requirements.txt

  3. Type make to build the inet executable (release version). Use make MODE=debug to build debug version.

  4. You can run specific examples by changing into the example's directory and executing inet

If you are using the IDE:

  1. Open the OMNeT++ IDE and choose the workspace where you have extracted the inet directory. The extracted directory must be a subdirectory of the workspace dir.

  2. Import the project using: File | Import | General | Existing projects into Workspace. Then select the workspace dir as the root directory, and be sure NOT to check the "Copy projects into workspace" box. Click Finish.

  3. Open the project (if already not open). Now you can build the project by pressing CTRL-B (Project | Build all)

  4. To run an example from the IDE open the example's directory in the Project Explorer view, find the corresponding omnetpp.ini file. Right click on it and select Run As / Simulation. This should create a Launch Configuration for this example.

Note:

  • by default INET is creating a shared library (libINET.dll, libINET.so etc.) in the src directory. To use the shared library you can use the inet command to load it dynamically.

Note to GIT users:

If you want to check out INET directly from the repository, we recommend using the

$ git clone git@github.com:inet-framework/inet.git

To make the installation simple, the GIT repo contains all IDE configuration files. If you make local changes in the IDE you may need to disable the change tracking on those files, so GIT will not insist committing those changes back on your next commit. You can use the _scripts/track-config-files-[on/off] scripts to enable/disable the change tracking.

To further ease the merging/rebasing operation the .cproject .nedfolders .oppbuildspec .project files are configured to be resolved using the 'ours' merge strategy. Depending on your GIT version, you may need to enable the 'ours' merge driver for the project:

$ git config merge.ours.driver true

VoipStream feature

VoipStream has only been tested on Linux. This does not mean it won't work on other systems, but your mileage may vary on getting it up and running.

PREREQUISITES.

VoipStream requires a "devel" package of the avcodec library (part of FFmpeg) to be installed on your system. On Ubuntu, this package can be installed with the following command:

$ sudo apt-get install libavcodec-dev libavformat-dev libavresample-dev libavutil-dev

The package name and installation command may vary for other Linux systems.