We are looking for the proper way to handle unique key for feature classes in ESRI Enterprise database.
In our scenario we are adding a second database instance to our application (GeoTrove). One of the requirements is to synchronize the two database instances, i.e. when a user edit the data (create, update, and delete) in one database instance, the change is propagated to the other instance automatically. After some research we decided to use Oracle Golden Gate.
For the most part we are able to synchronize the two database with Golden Gate. One tricky part is to ensure the database sequences are unique across the two database instances, we follow Oracle recommendation and use odd numbers in one instance and even numbers in the other. This works fine for regular tables, but when it comes to feature classes we are not sure if we can do the same. Normally feature class uses OBJECTID field which is populated by SDE automatically. We want to know what we need to do to synchronize feature classes in the above said scenario.
I found a link with info on how to manage Objectid field in ArcGIS: https://gis.stackexchange.com/questions/119604/where-is-the-sequence-in-oracle-associated-with-the-o....
Not 100% certain so would be happy if someone able to shed some light on this topic?
