Install from Source (macOS)
These instructions outline how to install an all in one Opencast system on the macOS operating system. Tested on OS X 10.14.1 Mojave.
Open a Terminal and switch to the directory, in which the Opencast installation should be placed, e.g.
~/develop/ or whatever you prefer.
Get Opencast source
You can get the Opencast source code by either downloading a tarball of the source code or by cloning the Git repository. The latter option is more flexible, it is easier to upgrade and in general preferred for developers. The prior option, the tarball download, needs less tools and you don't have to download nearly as much as with Git.
Cloning the Git repository:
git clone https://github.com/opencast/opencast.git cd opencast git tag <- List all available versions git checkout TAG <- Switch to desired version
Using the tarball:
Select the tarball for the version you want to install from the GitHub releases section under the "Tags" tab and download it directly from there or with the curl command specified below.
# Download desired tarball, replace [...] with the desired version curl -OL https://github.com/opencast/opencast/archive/[...].tar.gz tar xf [...].tar.gz
Please make sure to install the following dependencies.
Xcode jdk 11 ffmpeg >= 3.2.4 maven >= 3.1 python >= 2.6
Required (not necessarily on the same machine):
ActiveMQ >= 5.10 (older versions untested) Elasticsearch 7.9.x
Required for text extraction:
tesseract >= 3
Required for hunspell based text filtering:
hunspell >= 1.2.8
Required for audio normalization:
sox >= 14.4 (with MP3, FLAC and OGG support)
Required for animate service:
You can download Xcode in the Mac App Store. JDK 8 for OS X is available from Oracle.
Homebrew is a package manager for OS X. For installation instruction see their website.
brew install maven brew install ffmpeg brew install apache-activemq brew install tesseract brew install hunspell brew install sox brew install synfig
Elasticsearch on macOS
If you want to install Elasticsearch in the same machine run Elasticsearch as a Docker container
docker run -d --name elasticsearch -p 9200:9200 -p 9300:9300 -e 'discovery.type=single-node' elasticsearch:7.9.3
Using pre-built binaries
Pre-built versions of most dependencies can be downloaded from the respective project website:
Switch to the opencast folder. If you downloaded the tarball, this is the folder you just unpacked (called something
opencast-community-opencast-[…]). If you chose to download via git, use
cd opencast. You can proceed by
building opencast (depending on the folder permissions, you might need to start the command with
mvn clean install -Pdev
Please be patient, as building Opencast for the first time will take quite long.
Please follow the steps of the Basic Configuration guide. It will help you to set your host
name, login information, etc. Be aware that the config files now reside in the build folders for the desired
distribution. For the all-in-one distribution, this would be
/your/path/to/opencast/build/opencast-dist-allinone-[…]/etc/, again with
[…] representing the selected version.
As specified in the guide, make sure you replace the default ActiveMQ configuration with the one provided in
docs/scripts/activemq/activemq.xml. If you installed ActiveMQ using homebrew, you can find the installation path with
brew info activemq. The configuration is probably located in
ffprobe is used to analyze new videos. It is installed with FFmpeg but usually not on the path to be automatically
executed. You have to link the ffprobe to
/usr/local/bin/. You can find the FFmpeg install directory with
brew info ffmpeg. Usually you would link the file with
ln -s /usr/local/Cellar/ffmpeg/<version>/bin/ffprobe /usr/local/bin/ffprobe.
Make sure you have ActiveMQ running (unless you're running it on a different machine). Then you can start Opencast using the start-opencast script:
activemq start cd /your/path/to/opencast/ cd build/opencast-dist-allinone-[…] ./bin/start-opencast
As soon as Opencast is completely started, browse to localhost:8080 to get to the administration interface.