Chapter 3. Administering a Hadoop-YARN Cluster
In the previous chapter, we covered the installation steps to configure single and multi-node Hadoop-YARN clusters. As an administrator or a user of a Hadoop-YARN cluster, it is important to know how services are configured or managed. For example, an administrator must monitor the health of all the nodes across the cluster and a user should be able to view the logs of the applications submitted.
Hadoop-YARN has a predefined set of user, as well as administrative commands. It exposes monitoring data as service metrics and provides an easy integration of monitoring data with tools such as Ganglia, Nagios, and so on. It also defines a mechanism for High Availability and recovery.
In this chapter, we will cover:
- The YARN user and administration commands
- Configuring, managing, and monitoring YARN services
- ResourceManager's High Availability
- Monitoring NodeManager's health
