In a high availability configuration, a fully redundant secondary GitHub Enterprise Server appliance is kept in sync with the primary appliance through replication of all major datastores.
In an active/passive configuration, the replica appliance is a redundant copy of the primary appliance. If the primary appliance fails, high availability mode allows the replica to act as the primary appliance, allowing minimal service disruption.
You can failover to a GitHub Enterprise Server replica appliance using the command line for maintenance and testing, or if the primary appliance fails.
After failing over to a GitHub Enterprise Server appliance, you should regain redundancy as soon as possible rather than rely on a single appliance.
You can stop replication to a GitHub Enterprise Server replica temporarily, or permanently remove replication.
Geo-replication on GitHub Enterprise Server uses multiple active replicas to fulfill requests from geographically distributed data centers.