Class IPFailoverProxyProvider<T>

java.lang.Object
org.apache.hadoop.hdfs.server.namenode.ha.AbstractNNFailoverProxyProvider<T>
org.apache.hadoop.hdfs.server.namenode.ha.IPFailoverProxyProvider<T>
All Implemented Interfaces:
Closeable, AutoCloseable, org.apache.hadoop.io.retry.FailoverProxyProvider<T>

public class IPFailoverProxyProvider<T> extends AbstractNNFailoverProxyProvider<T>
A NNFailoverProxyProvider implementation which works on IP failover setup. Only one proxy is used to connect to both servers and switching between the servers is done by the environment/infrastructure, which guarantees clients can consistently reach only one node at a time. Clients with a live connection will likely get connection reset after an IP failover. This case will be handled by the FailoverOnNetworkExceptionRetry retry policy. I.e. if the call is not idempotent, it won't get retried. A connection reset while setting up a connection (i.e. before sending a request) will be handled in ipc client. The namenode URI must contain a resolvable host name.