hadoop2.5.2
Summarize the steps related to dynamically adding datanode nodes in a Hadoop cluster.
1. Install hadoop on the new node and copy the relevant configuration files of namenode to the node
2. Modify the masters and slaves files of the namenode node and add the node
3. Set each node to ssh into and out of the node without password, and set IP mapping
4. Start the Datanode and Nodemanager on the node separately
hadoop-daemon.sh start datanode
yarn-daemon.sh start nodemanager
5. Run start-balancer.sh for data load balancing. Function: When a node fails or a new node is added, the distribution of data blocks may be uneven. Load balancing can rebalance the distribution of data blocks on each datanode.