> My ideal solution would have : > - Clients access the servers using rr or any other algorithm. > - Clients stop accessing a given server if they can't use it anymore. > - Servers replicate their data directly between each other. > > Am I missing any better solution in 1.3.x? with the current codebase you will have to compromise on either 1. clients load balancing on servers OR 2. servers replicate between each other. These two are mutually exclusive at the moment. but the HA xlator (not yet released) used in combination with AFR will match your requirement set. the HA xlator does not really need the code infrastructure change of 1.4, but in terms of the project roadmap (priority) it sits in that timeframe. avati