Hadoop slave files configuration

Question

So I have a master and 3 slave machines and the slaves file on master machine has following entries:

master
slave1
slave2
slave3

Do I have to place this same slaves file in all of the slave machines or should i remove the first line having master and then place in slave machines?

Answer 1

The conf/master file corresponds to the Secondary Name Node and conf/slaves files corresponds to the TaskTracker/DataNode. Based on the cluster configurations pseudo distributed or fully distributed, the master/slave file configuration changes. When running in pseudo distributed mode, the master/slaves will have localhost and in fully distributed mode they will have the appropriate hostnames.

These files need to be present in the masters and not on the slaves. Check the Apache documentation on the purpose of the master/slaves configuration files, this will make it clear where to place the files.