How datasource contents must be synchronized

Who: Sebastien Bahloul

When: first release 13/04/2008, never updated

Summary: How datasource contents must be synchronized

Full synchronizations are tasks initiated to create, update and delete entries in the target directory. Even if the entries may be completed by other tasks or engines, this kind of tasks are dedicated to manage to main synchronization process :

• create a entry : if data are found in the data source and the given entry does not exists in the target directory - identify by its distinguish name - the entry will be created
• updates entry attributes : if data are found in the data source and the given entry exists,
• delete the entry from the directory.

So the creation process is done by completing the LDAP related Plain Old Java Object (POJO) and sending it to the directory. But in the middle of that process, you way have included synchronization options. If so, attributes updates may be altered according to the synchronization options configuration (see synchronization options ).

Some requirements are mandatory in order to use the full synchronization tasks:

• The engine is not responsible for data consistency: if you miss to give the mandatory attributes, synchronization task items will failed to execute.
• You must implement the getDistinguishName() method: because the task is going to create new entries, the naming plan must be clearly described in this method.