Hi, I hope this question is a bit less obvious than my last one. When i read the wiki article on clustering, i understand that some parts of the repository.xml must be configured to point to the same location and others should not. I begin to be dubitative when it explains that worspace level have to be private and FileSystem (there is also a FileSystem configuration under Workspace configuration bean) have to be shared. When i see read the sample section of ConnectionPooling, everything is shared even Workspace and Versionning. Does the sample in ConnectionPooling is OK with the Clustering doc or not ? Thanks in advance for your lights http://wiki.apache.org/jackrabbit/ConnectionPooling http://wiki.apache.org/jackrabbit/Clustering