Designing multiple data center HA configuration

The multiple data centers HA configuration ensures availability across geographically-distributed and redundant Jive platforms as an active or passive configuration. Note that you cannot have Jive running in multiple data centers simultaneously.

Note: You may choose to configure redundant databases and replicate data between them. Jive does not provide database replication configuration or support. Your data center or DBA team, or both, must provide database support for you and your configuration. We have seen customers successfully deploy Oracle RAC and SQL server HA configurations with Jive.

As an example, here is how a multiple data center HA configuration might look (your configuration may vary).


Multiple HA Data Centers Diagram

In this configuration, the web application nodes are configured in a cluster and deployed behind a load balancer, preferably an enterprise-grade load balancer such as the F5 BIG-IP. For more information about how to set up a cluster, see Clustering in Jive.

In the passive standby data center system, you can leave the web application nodes booted up at the operating system level, but not the Jive application (while the active production data center is running). However, the cache nodes, the Document Conversion service nodes, the Activity Engine nodes, and the database nodes in the passive standby data center may be left on.

CAUTION:
The web app nodes in the passive standby data center cannot be up and available all the time. If you attempted to do this, the web application nodes in the passive data center would communicate with the active production cluster as if they were part of it, which would cause catastrophic issues in your production cluster.

For information on how to bring up Data Center B in the case of a failure, see Starting up after failover.