Partition Size

The following is the set of requirements for partitioning specific to OS capability:

CentOS Setup

The below table is an example:

Download the Latest Swarm Bundle Package

For OVF-based installation, see Swarm VMware Bundle Packages to download the latest Swarm VMWare bundle.

For RPM-based installations, see Swarm Bundles to download the latest Swarm bundle.

Install Elasticsearch

Refer to the following steps:

1. Update CentOS.

   yum -y update

2. Install EPEL.

   yum -y install epel-release

3. Install NTP Server.

   yum -y install ntp systemctl enable ntpd hwclock --systohc systemctl restart ntpd

4. Unzip the Swarm bundle package to /root/datacore.

5. Obtain the latest Elasticsearch RPM and Swarm Search RPM from the downloaded Swarm bundle package.

6. Install Swarm RPM public key included with the distribution bundle.

7. Install and configure the Elasticsearch components on each Elasticsearch node.

Configure a 3-Node Elasticsearch Cluster for Swarm

Verify the Data Folder for Swarm Elasticsearch

1. The default data folder for Swarm Elasticsearch on all nodes is /var/lib/elasticsearch. This folder must not be on the boot partition to verify Elasticsearch cannot fill the boot partition.

2. Verify ownership of that folder. Use ls -la /var/lib/elasticsearch. It should be elasticsearch:elasticsearch.

3. Configure the firewall to allow Elasticsearch vi /etc/firewalld/services/elasticsearch.xml.

4. Restart the firewall service:

5. Add Elasticsearch firewall rule permanently:

6. Reload the firewall service:

Configure Elasticsearch Cluster

The following information is about the example deployment used to configure a 3-node Elasticsearch cluster:

1. Run the Swarm Elasticsearch configuration script /usr/share/caringo-elasticsearch-search/bin/configure_elasticsearch_with_swarm_search.py on the first Elasticsearch node (e.g., es75-01).

2. Enter the Elasticsearch cluster name (e.g., swarm-es75).

3. Enter the list of all Elasticsearch server names or private network IP addresses.

4. Enter the name of the current Elasticsearch node.

   The configuration script generates a custom Elasticsearch configure file for each node in the Elasticsearch cluster:

   1. The other nodes’ files are /etc/elasticsearch/elasticsearch.yml.<node-ip>

   2. The current node’s file is /etc/elasticsearch/elasticsearch.yml

5. Verify the Elasticsearch configuration files of all Elasticsearch nodes.

6. Copy the generated Elasticsearch configure file to other nodes.

7. SSH to other nodes of the Elasticsearch cluster to run the configuration script.

   1. On ES75-02 node,

   2. On ES75-03 node,

See Configuring Elasticsearch to update the environment and log settings. Perform a rolling restart of Elasticsearch services to apply changes once all updates are made.
See Hardware Requirements for Elasticsearch to know the hardware requirements for Elasticsearch.