Skip to main content

Failover and Data Recovery in Analytics and Core Databases

Here you can find how you can prepare for disaster recovery with the Jive Analytics and Core databases.

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.

Analytics Database Failover​

In Single Data Center HA Configuration ​

In a single data center HA configuration, if the Analytics database 1 fails over to Analytics database 2, this failure is transparent to the web application nodes due to the database driver layer controlling traffic. Community users would not notice the failure because the driver automatically redirects web app node requests over to Analytics database 2.

For more information, see Designing single data center HA configuration.

In Multiple Data Center HA Configuration ​

The Analytics database server connection string, user name, and password are stored in encrypted format in a table in the Core Application database. The setup of the connection string impacts how web application nodes resolve the Analytics database server. Be especially careful when setting this up during a failure in a multiple data center HA configuration.

Setting Up Connection String ​

You need to add a DNS name or IP address for the Analytics database server via the Admin Console at Reporting > Settings > Analytics. The connection string is stored in the Core Application database.

  • DNS Name Method (Preferred): The DNS name must resolve to an Analytics database server that is reachable by web application nodes in each respective data center (A or B).
  • IP Address Method: The IP address must be a virtual IP that points to the Analytics database server available from both data centers.

For more information, see Designing multiple data center HA configuration and Configuring Analytics database for high-availability.

Core Database Failover​

In Single Data Center HA Configuration ​

Refer to Configuring Core Application database for high-availability for details about a single data center HA configuration.

In Multiple Data Center HA Configuration​

Disaster Recovery (DR) architecture varies from data center to data center. If all systems are fully replicated to a DR facility, you can architect the Jive Platform similarly to a single data center HA configuration. However, many cases require a manual DR strategy, involving multiple data center HA configurations. For more information, see Designing multiple data center HA configuration and Designing single data center HA configuration.

You may follow one of these possible DR strategies:

Create Backup at a Remote Location (Simplest DR Strategy)​

If a recent full backup of the database is available at a remote location, it is possible to recover the system to the point in time of the available backup. Upon declaration of a disaster, perform the following steps:

  1. Set up a new cluster of application servers and point them to a new empty database at the recovery facility.
  2. After the initial setup, save specific property values from the new database's jiveProperty table or copy them to a backup table. For more information on which properties need to persist, see Restoring database with persistent properties.
  3. Restore the database backup over the newly created database.
  4. Apply the properties from the new database to the restored database jiveProperty table, as determined in Step 2. For more information, see Restoring database with persistent properties.
  5. Restart the application server.

Note: Minimizing data loss in the event of a disaster can be achieved by pushing incremental backups and transaction write-ahead logs to the remote facility. Point-in-time recovery may be performed up to the point of the disaster if backups are available.

Use Cold/Warm State Servers​

Streaming replication, or Write-Ahead Logging (WAL), maintains a recent copy of the database at the remote facility. Ensure a cluster of application servers is already set up and attached to an empty database. Upon declaration of disaster, the replicated database should replace the empty database, updating specific values in jiveProperty to reflect the DR environment. For more information, see Restoring database with persistent properties.

It is vital to consider which properties should be replaced with values from the original production site, and which should reflect the new facility's values. Review, validation, and live testing of a system failover eliminate potential issues during an actual disaster.

Related

Failover behavior of HA servers