Once a cluster is launched, administrators should start monitoring the Hadoop cluster. Apache Hadoop provides number of software to manage the cluster; in addition to that there are dedicated open sources as well as third party application tools to do the management of Hadoop cluster.
By default, Hadoop provides two web-based interfaces to monitor its activities. A JobTracker web interface and NameNode web interface. A JobTracker web interface by default runs on a master server (http://localhost:50070
) and it provides information such as heap size, cluster usage, and completed jobs. It also provides administrators to drill down further into completed as well as failed jobs. The following screenshot describes the actual instance running in a pseudo distributed mode:
Tip
Downloading the example code
You can download the example code files for all Packt books you have purchased from your account at http://www.packtpub.com. If you purchased this book elsewhere, you can visit http://www.packtpub.com/support and register to have the files e-mailed directly to you.
Similarly, the NameNode interface runs on a master server (http://localhost:50030
), and it provides you with information about HDFS. With it, you can browse the current file system in HDFS through the Web; you can see disk usage, its availability, and live data node related information.