Disaster Recovery

You can use the disaster recovery feature to resume FTL communications after the main operations site becomes disabled. Application systems can continue after the interruption using replicated persistence data at a remote disaster recovery site.

The first diagram illustrates a set of realm servers and persistence servers configured to prepare for recovery from a potential disaster, along with the application processes that they serve. Components at the main site are on the left. Components at the disaster recovery site are on the right. Application processes run only at the main site, and interact only with realm servers and persistence servers at that site.

For simplicity, the diagram shows only one persistence cluster, but a realm could include more than one persistence cluster.

In the first diagram, the persistence cluster transport connects all the persistence servers of the primary set in a mesh, and all the standby persistence servers in another mesh. Data traverses the WAN link only once between the primary and standby sets.

In the second diagram, a disaster has disabled the main site. Administrators have manually cut over to the disaster recovery site:
  1. Administrators have restarted the disaster recovery realm server as the new primary realm server.
  2. Administrators have reconfigured the standby set of persistence servers to be the primary set.

The disaster recovery site now operates as the new main site. Persistence server 6 interacts with application processes.

Administrators can begin to set up another disaster recovery site.