How are groups handled during the merge procedure?
Groups are identified in the following way during a merge run:
First, the GUID is taken into account to identify the group. If there is a group in target database with the same GUID, it is matched with the source. Depending on the setting this happens with or without merging the attributes.
Second, the path to the root is matched. If a group with the same path to root exists in the target database, it is matched with the source. If a group in the path in the target database has no name in the language in which the user is logged in, the GUID of this group is used as its name.
If no group can be found in the target with both criteria and you have Write access privileges for the superior group of the target database, a new group is created with the same GUID and the same name as in the source database. The new group receives all attributes which were specified in the source database. The current user will receive the Read, Write and Delete access privileges for the new group.