Developing a grouper configuration
The installation, configuration and commissioning of a complex software package such as Grouper is often presented as a linear process, with one step following another until everything is finished. In reality the process is very different:
Our understanding of what can be done grows with our experience of using Grouper
Grouper is not stand-alone. It interacts with other parts of the infrastructure and the best way of doing this may only become clear after trying different approaches
Other parts of the infrastructure might require modification, upgrade or replacement to benefit the overall solution
I would therefore suggest that a linear process only occurs when you are ready to do the final installation and commissioning of Grouper. By this point in time you should have a detailed plan which you can follow step-by-step. Up until this point I would suggest that you follow a simple process:
The core Grouper binary is easy to install, and so installation will be relatively easy (not much more difficult than installation of the quick start). The interesting stuff really begins with the installation of the database, and configuration of data connections. It is at these stages that the process can be used. At first you will encounter errors and other problems, but as you continue your knowledge of Grouper will increase. It is important at each iteration of the process that you have realistic expectations of what represents success at that specific stage. For example, at first a reasonable expectation would be to get a working Grouper database
1 Comment
Gail Lift (umich.edu)
Is this good place to discuss config files vs grouper_config? Depending on your team's build and deploy processes, you might
If you design/test new provisioners, loaders, data providers, etc in a test environment, important to have an efficient way to deploy the configs into production, including environment-specific details (eg, connection strings)