Setting Up a Cluster During the Elasticsearch Installation

When you perform a fresh install of Elasticsearch using the PeopleSoft Deployment Packages (DPK), the installation process enables you to set up an Elasticsearch cluster with a single node or with multiple nodes.

For the step-by-step description of the installation process for your operating system, see PeopleSoft Deployment Packages for Elasticsearch Installation (Doc ID 2205540.2) on My Oracle Support.

During the installation process, you are prompted to specify various configuration parameters. If you are manually installing Elasticsearch, you are required to edit the elasticsearch.yml file where you can specify values for the parameters pertaining to a cluster. The configuration parameters that are specific to a cluster are described here:

  • A name for the cluster.

  • Minimum number of master nodes.

  • List of host names where Elasticsearch will be installed or is installed.

The setup script prompts you to enter a name for the Elasticsearch cluster.

If you are performing a manual installation of Elasticsearch, you need to enter a value for the cluster.name parameter in the elasticsearch.yml file.

A cluster is identified by a unique name. This name is important because a node can only be part of a cluster if the node is set up to join the cluster by its name. Make sure that you don’t reuse the same cluster names in different environments, otherwise you might end up with nodes joining the wrong cluster. For instance you could use logging-dev, logging-stage, and logging-prod for the development, staging, and production clusters.

The setup script prompts you to enter the minimum number of master nodes.

If you are performing a manual installation of Elasticsearch, you need to enter a value for the discovery.zen.minimum_master_nodes parameter in the elasticsearch.yml file.

Oracle PeopleSoft recommends a cluster with three nodes for high availability and to prevent a split-brain scenario. Then, depending on the hardware, memory availability, and search performance, you may add nodes to the existing cluster.

This setting must be set to a quorum of your master-eligible nodes. It is recommended to avoid having only two master-eligible nodes, since a quorum of two is two. Therefore, a loss of either master-eligible node will result in an inoperable cluster. To prevent the split-brain scenario, PeopleSoft recommends an odd number of nodes in the cluster, so configure the majority of nodes (total number of master-eligible nodes/ 2 + 1) .

Note: This feature of specifying multiple nodes during the installation of Elasticsearch are contained in the DPKs (ESK-DPK-LNX-6.1.2_01 for Linux and ESK-DPK-WIN-6.1.2_01 for Windows).

For more information on split-brain scenario, refer to the Elasticsearch online documentation.

The setup script prompts you to enter host names of nodes present in the cluster.

If you are performing a manual installation of Elasticsearch, you need to enter a value for the discovery.zen.ping.unicast.hosts parameter in the elasticsearch.yml file.

A node name is a symbolic name for identifying the node. Host name is the IP or DNS of the machine where Elasticsearch is installed. Host names(or IP/DNS) are required for letting each Elasticsearch server where it can ping and find other Elasticsearch servers during booting up.

Enter the host name for any nodes that are already members of a cluster. Enclose one or more host names in square brackets, with the host name or IP address in double quotes.

For example:

  • For one host, you would enter: ["host1.example.com"]

  • To list two or more hosts, use commas: ["host1.example.com", "192.0.2.1"]

This section provides the high-level steps to set up a cluster when you are installing Elasticsearch. For steps 1 and 2, you may use the PeopleSoft Deployment Packages for Elasticsearch Installation (Doc ID 2205540.2) on My Oracle Support for detailed instructions.

  1. Obtain the Elasticsearch DPK for your operating system.

  2. Run the script to install Elasticsearch. The script creates a cluster with one or multiple nodes, which you need to specify.

    While installing Elasticsearch ensure that you provide values for the three configuration parameters that are essential to set up a cluster, which are discussed in an earlier section.

    • Cluster name.

    • Minimum number of master nodes.

    • List of host names where Elasticsearch is installed.

  3. Verify the cluster setup by executing the following command in a browser:

    http(s)://host:port/_cluster/stats?

    Where, host refers to the Elasticsearch host.

    Enter the Elasticsearch user credentials (esadmin) when prompted for login information.

    The cluster and node information can be found in the nodes/count/total section of the response.

    Other methods to verify the cluster and node setup, which involves using the PeopleSoft Search Framework and PeopleSoft Health Center, are discussed in a later section of this topic. See Verifying the Cluster and its Nodes.

To complete the search functionality setup, you will need to follow the post-installation instructions in the PeopleSoft Deployment Packages for Elasticsearch Installation (Doc ID 2205540.2) on My Oracle Support.

Also, refer to Understanding PeopleSoft Search Framework Administration.