Details
-
Bug
-
Resolution: Won't Fix
-
Major
-
1.7.0
-
Security Level: Public
Description
steps to reproduce :
install 1.6.5.4 on ip(a)
install 1.6.5.4 on ip(b)
join ip(a) and ip(b) and rebalance
then change ip(a) to ip(X)
perform an in-place upgrade ( shut down each node and upgrade )
after upgrade process ip(b) will come back up but ip(X) says that the cluster in not initialized as if the data is lost and bucket did not exist in this machine