Get to grips with the intricacies of Hadoop monitoring using the power of Ganglia and Nagios
About This Book
- Track Hadoop operations, errors, and bottlenecks efficiently
- Employ Hadoop logging features to help manage Hadoop clusters better
- Visualize the data collected and present it in a systematic manner
Who This Book Is For
This book is useful for Hadoop administrators who need to learn how to monitor and diagnose their clusters. Also, the book will prove useful for new users of the technology, as the language used is simple and easy to grasp.
What You Will Learn
- Install Nagios and Ganglia and understand logging at the operating system level
- Create and configure Nagios nodes for monitoring with custom checks
- Monitor Hadoop daemons such as NameNode, DataNode, JobTracker, and so on
- Configure logs for various daemons and set up audits for the options done on the cluster
- Track important parameters for the File System, MapReduce, and other counters
- Set up Nagios master and client nodes with checks for the system and applications running on it
- Configure the Hadoop metrics collection and visualize it for nontechnical users
- Understand the communication between different daemons and protocols and the ports they use
In Detail
With the exponential growth of data and many enterprises crunching more and more data, Hadoop as a data platform has gained a lot of popularity. The Hadoop platform needs to be monitored with respect to how it works and functions. There is an ever-increasing need to keep the Hadoop platform clean and healthy.
This book will help you to integrate Hadoop and Nagios in a seamless and easy way. At the start, the book covers the basics of operating system logging and monitoring. Getting to grips with the characteristics of Hadoop monitoring, metrics, and log collection will help Hadoop users, especially Hadoop administrators, diagnose and troubleshoot clusters better. In essence, the book teaches you how to set up an all-inclusive and robust monitoring system for the Hadoop platform. The book also serves as a quick reference to the various metrics available in Hadoop.
Concluding with the visualization of Hadoop metrics, you will get acquainted with the workings of Hadoop in a short span of time with the help of step-by-step instructions in each chapter.
Gurmukh Singh
Gurmukh Singh has been an infrastructure engineer for over 10 years and has worked on big data platforms in the past 5 years. He started his career as a field engineer, setting up lease lines and radio links. He has vast experience in enterprise servers and network design and in scaling infrastructures and tuning them for performance. He is the founder of a small start-up called Netxillon Technologies, which is into big data training and consultancy. He talks at various technical meetings and is an active participant in the open source community's activities. He writes at http://linuxaddict.org and maintains his Github account at https://github.com/gdhillon.
