To configure your database for continuous operation, you need to use database network connections, replicate your database, use replication sets, and ensure the ABL application that connects to the OpenEdge database use automatic database reconnection.

Database network connections

Any ABL application with the goal of continuous operations must use OpenEdge database network connections. Network connections permit a database to continue running on a separate set of shared OS images while PAS for OpenEdge instances and ABL sessions are added and removed from service. Using network connections also protects a database in the event any ABL session client disconnects while still holding a latch or resource in the database.

Part of configuring your database is ensuring it has enough user connections to service all of its ABL session clients. This configuration must be large enough to handle the maximum number of ABL Sessions for each instance in the load balancer pool. If you choose to have a hot-standby PAS for OpenEdge instance, your database configuration must also account for those additional connections. This strategy allows for quickly scaling up PAS for OpenEdge instances without having to create a larger database to handle the extra processing.

In most deployments, the OpenEdge database is running a different OS image than the PAS for OpenEdge instances. Best practice is to remove other applications and processes from the database machine, so they don’t adversely affect the performance of the database if a sudden spike in usage occurs.

Replicate your database

Using database replication is the best way to make your application continuously available and fault tolerant. If for any reason your production database, database machine, or network becomes unavailable you can quickly transition to the replicated database.

Data replication has two main real-time functions: to provide a disaster recovery or a failover site to ensure availability of an application and to distribute copies of data to one or more sites for horizontal read scaling. If you run multiple copies of replicated target databases, and if the source goes down, a target will transition to source and there will still be a target database. OpenEdge Replication provides near, real-time replication of an OpenEdge database from one location to another. The second location can be local to the production database, at a remote data center, or in the cloud.

OpenEdge Replication requires a minimum of two databases: a source database and a target database. OpenEdge can also replicate to a tertiary database. The source database is where the users do their work and make database updates. This database is the production database and has both read and write access for all processes.

The following are the criteria for an OpenEdge Replication source database:
  • The source database must be enabled as an OpenEdge Replication source database.
  • OpenEdge Replication server replicates data to the target database from the enabled source database.
  • After-imaging is required to be enabled on the database.
  • After-imaging is used to capture all database activity.

The target database originates from the source database, and contains the same data and schema as the source. The target database does not allow database updates by anything other than OpenEdge Replication. OpenEdge Replication Plus permits queries and reports by users, and any other non-database write activity. The OpenEdge Replication agent updates this database. For the most fault-tolerant deployment, you set two target databases.

The following are criteria for an OpenEdge Replication target database:
  • The target database must be enabled as an OpenEdge Replication target database.
  • OpenEdge Replication agent replicates data from the source database.
  • The OpenEdge Replication agent is the only process that can update the database.

Replication sets

A Replication Set is a two-target Replication configuration where the two targets are aware of each other, and are configured to continue replication in the event of a server failure.

Adding a Replication Set, to an OpenEdge Replication configuration, changes the failover activity. The biggest benefit of a Replication Set is the ability to transition all available databases together. The goal of the Replication Set is to minimize the risk of a single point of failure and the need to rebase.

With a Replication Set, if a Replication agent loses communication contact with the Replication server (for example, if the primary machine were to shut down unexpectedly), you can move all database update activity from the source database one of the target databases, and the other target becomes an agent to the new source.

For more information about OpenEdge Replication, see Introduction to OpenEdge Replication.