Transferring Elasticsearch and Hadoop Data to Another Disk

Introduction

In this article, we will explain how to transfer Elasticsearch and Hadoop data to a different disk and create a symbolic link in Cluster systems.

Elasticsearch

First, stop the Elasticsearch service with the following command:

#systemctl stop elasticsearch

Next, navigate to the /var/lib directory using the command below:

#mc /var/lib

Then, depending on where the new LVM or new disk is mounted, navigate to that location in the right pane, and on the left pane, select the "elasticsearch" directory. Use the "F6" key to move the"elasticsearch"  folder to the new location.

After the transfer process is completed, while you are on the newly moved elasticsearch folder, press the CTRL+X and then S key combination on your keyboard to create a symbolic link.

Next, navigate to the /var/lib directory and create the symbolic link to the new location by pointing to the moved elasticsearch folder. The system will treat the newly moved data as if it is still in the original location.

By using this method, the link will ensure that all the references to the original /var/lib/elasticsearch folder now point to the new disk where the data was moved.

After the process is completed, the symbolic link on the left side should appear as shown in the screenshot below.

Finally, the following command is executed to start the Elasticsearch service:

#systemctl start elasticsearch

Hadoop Data Transfer to Another Disk

To move the Hadoop data to a different disk, follow the steps below:

1. Stop All Hadoop Services: First, stop all Hadoop services to ensure data consistency during the move. Run the following command to stop Hadoop HDFS services:

# systemctl stop hadoop-hdfs*