Configuring Connection Failover

Connection failover allows an application to connect to an alternate, or backup, SequeLink Server if the primary SequeLink Server is unavailable, for example, because of a hardware failure or traffic overload. Connection failover ensures that the data on which your critical JDBC applications depend is always available.

You can customize the JDBC Client for connection failover by configuring a list of alternate SequeLink servers that are tried if the primary server is not accepting connections. Connection attempts continue until a connection is successfully established or until all the alternate SequeLink servers have been tried the specified number of times.

For example, suppose you have the environment shown in Figure 5-2 with multiple SequeLink servers: SequeLink server A, B, and C. SequeLink server A is designated as the primary SequeLink server, SequeLink server B is the first alternate server, and SequeLink Server C is the second alternate server.

Figure 5-2. Connection Failover Using the JDBC Client

First, the application attempts to connect to the primary SequeLink server, SequeLink server A (1). If connection failover is enabled and SequeLink server A fails to accept the connection, the application attempts to connect to SequeLink Server B (2). If that connection attempt also fails, the application attempts to connect to SequeLink Server C (3).

In this scenario, it is probable that at least one connection attempt would succeed, but if no connection attempt succeeds, the SequeLink Client can retry each alternate SequeLink server (primary and alternate) for a specified number of attempts.

To configure connection failover, you must specify a list of alternate SequeLink servers that are tried at connection time if the primary server is not accepting connections. To do this, use the AlternateServers connection property. Connection attempts continue until a connection is successfully established or until all the SequeLink servers in the list have been tried once (the default).

Optionally, you can specify the following additional connection failover features:

The number of times the JDBC Client attempts to connect to the primary and alternate SequeLink servers after the initial unsuccessful connection attempt. By default, the JDBC Client does not retry. To set this feature, use the ConnectionRetryCount connection property. See "Using Connection Retry" for more information.
The wait interval, in seconds, between attempts to connect to the primary and alternate SequeLink servers. The default interval is 3 seconds. To set this feature, use the ConnectionRetryDelay connection property.
Whether the JDBC Client will use client load balancing in its attempts to connect to primary and alternate SequeLink servers. If load balancing is enabled, the JDBC Client uses a random pattern instead of a sequential pattern in its attempts to connect. The default value is not to use load balancing. To set this feature, use the LoadBalancing connection property. See "Using Client Load Balancing" for more information.

You can use a connection URL to direct the JDBC to use connection failover, or use a JNDI LDAP provider. Refer to the SequeLink Administrator's Guide for detailed information.