Creating a Cluster

Note: To create a cluster with pgEdge Distributed Postgres (Cloud Edition), you must have a pgEdge Enterprise Edition account (opens in a new tab) and a linked Cloud provider account.

To create a cluster, select the + New Cluster button in the left navigation pane. Then:

• Provide a unique name for your cluster in the Cluster Name field.

• Use the drop-down listbox in the Cloud Account field to select the cloud provider account that will be used to provision your cluster.

• Select up to 5 availability zones in any regions to which you have access. To use the map for location selection, use the Show map toggle to enable the map.

Once selected, the region name is added to the list of selected Regions. Use the +/- control or the drop-down list to the right of the Region name to select from the supported Availability Zones for node deployment. One node will be deployed in each zone listed; you can use the X in the zone icon to remove a zone from the deployment list.

Each Region of your cluster is represented by a pane in the Network Settings section; within each region, subnet details for each availability zone are displayed. Toggles at the top of the section control the display:

• Use the Configuration toggle to accept the Default address assignments, or move the toggle to Manual to specify custom address assignments.
• Use the Network Type toggle to specify if the connection values displayed or edited should be Public or Private addresses; note that private networking is only supported in AWS clusters.

The Network address specifies the VPC CIDR range in use by your cluster; the public/private subnet ranges ensure that each node in your cluster does not encounter conflicts with other nodes in your cluster. To modify the CIDR range to work with network address ranges currently in use on your private network, toggle to select Manual configuration and customize the addresses. Note that all addresses should meet Class C address specifications (opens in a new tab).

• Use the Network address for outside connections from client applications, SSH requests, and connections from web clients.
• Use the Public Subnet and Private Subnet addresses for VPC peering within your own network.

Next, specify the vendor-specific deployment details for your cluster.

• Use the Instance Type drop-down to select the size and configuration of your cluster instance type. Each node in the cluster will reside on the instance type selected. The instance type is one factor in determining the operating cost of your cluster.

• Use the Volume Size (GB) field to specify the size of the data volume for each node of your cluster. Volume size must be between 8 GB and 10 TB.

• Use the SSH Key drop-down to select a key pair from the list of your imported key pairs. A key pair is required to SSH to a cluster; without a key, you can connect to databases you provision on the cluster, but will not be able to connect with the host via SSH.

Select the backup stores (opens in a new tab) that will be available to use for database backups and log file backups for the new cluster.

As you select backup stores, remember that selecting backup stores near your cluster nodes will reduce network latency during the backup process. Use the drop-down field or select locations on the map to include a specific store.

Use the Firewall Rules field to open ports on each node of your cluster for connections. Click the + Add Firewall Rule icon to add a rule.

• Use the Type drop-down to select the connection type you are defining:

  • Select PostgreSQL to create a connection to port 5432 ; this is the port that connecting clients that wish to query the database should use.

  • Select HTTPS to open port 443 for secure communication between PostgreSQL and a web server.

  • Select SSH to open port 22 for SSH connections.

• Use the Sources drop-down to select the IP address(es) that will be allowed to connect to the port.

  • Choose My IP:x.x.x.x to open a port for the detected IP address from which you are connected.
  • Choose All IPV4 to allow connections to the selected port from any source. This setting is not recommended for connections to the database server.

After selecting a protocol type and an address, choose the + Add Rule button to add the list of firewall rules. To remove a rule from the list, select the X in the upper-right corner of the rule box.

To create the cluster without firewall rules, check the box to the left of Deploy cluster without firewall rules. Note that if you do not open ports for connection, client software will be unable to connect to the OS or databases that reside on the cluster.

In the Options section you'll find the option to create the cluster as a single-database cluster.

Single database clusters are beneficial because:

• they allow connections to the database with an IP address (rather than by domain name).
• they allow connections to the database via DNS names and records that you create outside of the pgEdge console.
• a single-purpose, dedicated cluster guarantees that all cluster resources are available for the single database.

When you've completed the dialog, select the Deploy Cluster button to deploy a provisioned cluster.

A progress bar tracks the progress as the deployment goes through each step; when the cluster is ready to use, it is added to the list of clusters in the navigation tree.

If you encounter any errors during cluster creation, please see the Troubleshooting section.