Deploying Your Replication-Enabled Project

We discussed how to enable your project for replication. The distributions that you created there can be deployed in the usual way. Follow the installation instruction manual for this. As was mentioned during the discussing of how replication works, even when you don't run a cluster on the source, the repository must be configured to write its events to a journal. You will find instructions on how to set that up in the installation manual mentioned above.

After you have successfully deployed both the source and target distributions of your project to the respective environments you must first of all initiate a full sync. Note that this is only necessary during the first deployment. After subsequent upgrade deployments there should be no need for full syncs anymore, as long as the upgrade did not introduce new, to be replicated content.

The sync will cause the uuids of the JCR nodes on the source and the target to correspond. This is a prerequisite for replication to work properly. Another precondition is that the version of the replication addon is the same on the source and on the target. In fact, in general the version of your project must correspond due to the fact that the node type definitions (CNDs) must be compatible.

Disabling the Replication Source On Startup 

Per Bloomreach Experience Manager versions 16.3.0 and 15.7.2, it is possible to start source nodes with replication disabled, rather than enabled. This is convenient when a database has been copied from production to a lower environment, and restarting the lower reads production target configurations to try to replicate its data to, which is unwanted. 

To start a replication source cluster node with replication disabled, set this system property:

replication.disabled.on.source.at.startup=true