All Products
Search

This Product

    E-MapReduce:Manually recover a Standby NameNode

    Document Center

    E-MapReduce:Manually recover a Standby NameNode

    Last Updated:Mar 26, 2026

    Restore a Standby NameNode when its metadata is corrupted or out of sync with the Active NameNode — for example, when a data directory is accidentally deleted, a large number of edit logs are accumulated on a NameNode, or you have manually run a checkpoint on the Active NameNode.

    Prerequisites

    Before you begin, ensure that you have:

    • A healthy, running Active NameNode

    • SSH access to the node where the Standby NameNode resides

    Restore the Standby NameNode

    1. In the E-MapReduce (EMR) console, stop the Standby NameNode.

    2. Log in to the node where the Standby NameNode resides via SSH. For details, see Log on to a cluster.

    3. Switch to the hdfs user:

      su hdfs

    4. Resync the Standby NameNode from the Active NameNode:

      hdfs namenode -bootstrapStandby

    5. In the EMR console, start the Standby NameNode.

    6. Verify that the Standby NameNode is running correctly. Check the NameNode status on the web UI. For details, see Web UIs of HDFS components.