Peer Server
Last updated
Last updated
The Peer Server functionality strengthens system reliability and fault resilience for mission-critical applications by preventing two servers from being hosted on the same hypervisor.
A peer server is usually a secondary server that shares the same responsibilities as the primary server. For example, two web servers running behind a load balancer are considered peer servers since they both equally manage web traffic.
However, servers do not need to serve the same purpose to utilise this feature. It can be used to ensure that two servers are placed on separate hypervisors, improving service availability and minimising downtime in the event of a hypervisor failure.
When a peer server is assigned, the platform checks whether both servers are currently located on the same hypervisor. If they are, a live migration will be scheduled to separate them. During the migration, your server will remain fully operational, but management actions through the control panel will be temporarily unavailable.
Each server can have only one designated peer server. It is not possible to configure three or more servers to ensure they are all placed on different hypervisors.
The Peer Server feature is available for plans that utilise distributed storage. To enable it, follow these steps:
Navigate to the VM management page.
In the Overview section, click Set Peer Server:
Choose the server you want to designate as a peer, then click Update
Once configured, the platform ensures that peer servers are not placed on the same hypervisor and will initiate a live migration if necessary to meet this requirement.