High availability

NOTE: The high availability approach for Instana's Self-hosted Docker-based installer is deprecated, and will be entirely removed from Instana's documentation when the documented high availability setup for Self-hosted Instana backend on Kubernetes/OpenShift (on-premises) is available. All users that depend on high availability are encouraged to move to a Self-hosted Instana backend on Kubernetes/OpenShift (on-premises).

The backend can be installed on a single server or two servers (multi-host). By default, all the running backend components are supervised by Docker. Therefore, in case of a possible outage, the state automatically converges and repairs itself.

Instana supports an active-active high availability setup. By provisioning two backends, you reduce the impact of an unplanned service outage on your premises. The Instana agent is set up to report into both backends, and this covers use cases where data redundancy and high available service continuity are required.

For information about how to configure the agent to send to both backends, see the docs on configuring multiple backends.