The rolling of ports in these 2 clusters lead to a lot of timing issues and failed test cases; as witnessed in our patch process.
The way around is to let the OS pick the port for the NameNode/JobTracker and let the the DataNode/TaskTracker query them for the port to connect to and then use that port.
|Field||Original Value||New Value|
|Fix Version/s||0.12.1 [ 12312347 ]|
|Fix Version/s||0.13.0 [ 12312348 ]|
|Priority||Critical [ 2 ]||Major [ 3 ]|
|Status||Patch Available [ 10002 ]||Open [ 1 ]|
|Resolution||Fixed [ 1 ]|
|Status||Patch Available [ 10002 ]||Resolved [ 5 ]|
|Status||Resolved [ 5 ]||Closed [ 6 ]|