The NameNode in a Hadoop cluster keeps track of the metadata for the whole HDFS filesystem. Unfortunately, as of this book's writing, the NameNode in the current stable version of Hadoop is a single point of failure. If the metadata of the NameNode is corrupted, for example, due to hard drive failure, the whole cluster will become unavailable. So, it is important to safeguard the NameNode from these disastrous failures.
There are multiple ways we can increase the resilience of a HDFS cluster. In this recipe, we will show you how to configure a SecondaryNameNode as a backup NameNode and how to recover from NameNode failures.
We assume that our Hadoop cluster has been properly configured, and we have one machine, master1
, as the NameNode and a second machine, master2
, to run the SecondaryNameNode.