|Private Cloud Load Balancer and High Availability (HA) Best Practices|
|1/11/2022 - Cloudmersive Support|
Cloudmersive Private Cloud natively supports High Availability deployment. Because each node is stateless and does not need to communicate with other nodes, nodes can be deployed in different data centers, or brought online and offline at will. In addition, load balancers can be added in front of Cloudmersive Private Cloud servers to enhance availability and also to distribute traffic across the nodes.
The first step is to decide if you wish to use HTTP or HTTPS between your load balancer and the Cloudmersive Private Cloud nodes, and if you want to use an Application Load Balancer or a Network Load Balancer. Cloudmersive Private Cloud supports all combinations of these. If you wish to use HTTPS, be sure to configure TLS on the Cloudmersive Private Cloud Servers.
Next, install all of the Cloudmersive Private Cloud servers first, and test them to ensure they are available and working as expected. Once deployed, add each server endpoint to your load balancing target pool.
We recommend configuring a health check. A health check is a continuous health probe from the load balancer to each server to check if it is available and operating as expected. To configure a health check, add the following endpoint as a health check to your load balancer for each server:
You should check for the 200 response and the content "Online". We recommend that your health check run every 10 seconds for each target server. If two sequential health checks fail, then the load balancer should be configured to temporarily remove that target server from the load balancing pool until it becomes healthy again.
We recommend configuring Windows Update on each server to run in a staggered timing so that at any given time, at least one or more servers is available. In addition, we recommend performing maintenance in the same way.