Seal is tested and developed on Ubuntu and CentOS. However, in principle Seal should work on any Linux platform.
If Seal doesn’t build or run on your Linux distribution of interest please submit a bug report and we can try to fix it.
We recommend you download the latest release from here: http://sourceforge.net/projects/biodoop-seal/files/.
On the other hand, if you want to try the latest improvements or contributions, you can checkout the latest code from our repository:
git clone https://github.com/crs4/seal.git
The typical installation procedure is:
Note that the software you installed (both Seal and its dependencies) need to be accessible to all nodes in the Hadoop cluster. See the deployment instructions/suggestions in the section Installation - Deploying.
There is no particular upgrade procedure (but see the notes below). Just build and deploy Seal following the instructions in the section above and take noticed of the notes below.
If you were setting configuration properties, either through the command line (-D argument) or through a .sealrc file, make sure you update the names. The news page contains a full list of the names that changed.
The names of the seal executables have changed. Now everything is invoked as:
seal <subcommand>
Make sure you update any scripts. A table with the name conversions is available on the news page.
There are several new options in the Seal programs. Have a look at them to make sure you make the most of the new features.
Make sure you copy any custom property settings from the old launcher scripts under bin/ into a Seal configuration file.