Ignoring HAServiceMBeanSupport/HAServiceMBean for the moment, here's what we're now agreeing to implement, pending other objections. 1) Remove the following methods from relevant classes. ClusterPartitionMBean getClusterPartition() | void setClusterPartition(ClusterPartitionMBean) | void setPartitionName(String) 2) Add the following methods as replacements. HAPartition getHAPartition() | void setHAPartition(HAPartition) 3) Undeprecate the following "convenience" method. String getPartitionName() I'm not familiar with the use of HAServiceMBeanSupport/HAServiceMBean. The javadoc and class names suggest that this class is provided to allow clients to extend the implementation and integrate with their own clustered mbeans. If we modify the mbean/implementation so that clients access the cluster's interface rather than its mbean, does that affect clients who might want access to the mbean itself? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4062603#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...