Hi,
I would like to understand how inter process communication between hadoop nodes happen? I know that it uses Remote procedure call but would like to know if it needs passwordless ssh ? If it doesn't need passwordless ssh as well set up, then how does RPC work, especially when data is written on one of the datanodes and in turn that data node writes the same data(replication) to another datanode? How exactly this works when passwordless ssh is set up?
Thanks for your answers.