I am running Velocity CTP3. I use a cluster with 2 nodes. When I take down a node I get one of the following exceptions (depending on how I take it down): // ErrorCode<ERRTransport0002>:Unknown failure in transport layer. // ErrorCode<ERRCA0019>:Initialization: No alive servers could be detected. // ErrorCode<ERRCA0022>:Cache::Get: There is a temporary failure, please retry after some time.
My understanding is that this is happening because veocity is temporarily unavailable whilst repartitioning the cache (and maybe establishing a new primary?) and that the solution to the problem is to retry the read or write operation until it goes through. I assume this behaviour is part of Velocity CTP4 but it isn't in CTP3 and I have to implement it if I want to see it working with CTP3. That being said, can you please confirm that my assumptions are correct, and if so can you please recommend a number of retries and the delay between retries?
