Automatic Failover and Standby Servers

Rodrigo

I would like to suggest a feature that could improve the availability of websites hosted with Enhance: support for standby servers with automatic failover.

The idea would be to keep two servers synchronized. If the primary server became unavailable, Enhance could detect the failure through health checks and automatically redirect traffic to the secondary server by updating DNS records or using a similar mechanism.

This would not necessarily need to be a traditional load balancer. It could be a high-availability solution that includes:

File synchronization between servers;
Database replication or synchronization;
Automatic health checks;
Fast DNS updates when a failure is detected;
Controlled failback after the primary server recovers;
Alerts and a history of failover events.

Some hosting companies already build similar solutions using file synchronization, database replication, and external automation to update DNS records. Having this functionality integrated into Enhance would make the process much simpler and reduce the risk of configuration errors.

This feature could help hosting providers deliver higher availability and meet more demanding SLAs, especially during hardware failures, data center outages, or emergency maintenance.

Is anything like this currently planned for Enhance? If not, it would be great to consider an official failover solution—or an integration framework that would allow hosting providers to implement it safely.

CamielK

I have customers that would definitely pay more for a high-availability hosting solution.

Adam

My background for anyone who isn't aware is that I ran one of the largest hosting companies in the UK from 2002 until 2017. We offered high availability hosting both to enterprise clients and on our own mass shared hosting platform.

The problem with high availability hosting is that it often leads to worse overall uptime - the engineering often breaks more frequently than an individual server would.

You can already come fairly close on Enhance (without the complexity, exotic networking, shared storage or replication) using the built-in disaster recovery features. Set a low default TTL, set your backup frequently to hourly then, in the very unlikely event of a total server loss, just decommission the server in the UI and perform a new migration to restore the backup data to another server in your cluster, updating the DNS automatically.

It's not impossible that we will support high availability and replication in the future, this is still a common feature request, I just don't think it's necessary in 99.9% of cases.