CLUSTER Connection Manager configuration parameter

The CLUSTER parameter specifies that a connection unit is composed of a high-availability cluster, and specifies a name for that connection unit. A high-availability cluster is a primary server plus one or more secondary servers, which can be high-availability data replication secondary servers, shared-disk secondary servers, or remote stand-alone secondary servers.

Syntax

1  CLUSTER connection_unit_name

Usage

Each CLUSTER parameter value must be unique within the Connection Manager configuration file.

CLUSTER parameter values cannot use multibyte characters.

CLUSTER connection units can use the following redirection policies:
  • Round-robin
  • Secondary apply backlog
  • Workload

Each CLUSTER connection-unit definition that references a specific high-availability cluster must have a unique PRIORITY attribute value. For example, the following high-availability cluster, composed of server_1 and server_2, must have unique PRIORITY values in each definition.

my_connection_manager_1's configuration file:
NAME my_connection_manager_1
LOG 1
LOGFILE $INFORMIXDIR/tmp/my_cm_1.log

CLUSTER my_cluster
{
   INFORMIXSERVER server_1,server_2
   FOC ORDER=ENABLED \
       PRIORITY=1
}
my_connection_manager_2's configuration file:
NAME my_connection_manager_2
LOG 1
LOGFILE $INFORMIXDIR/tmp/my_cm_2.log

CLUSTER my_cluster
{
   INFORMIXSERVER server_1,server_2
   FOC ORDER=ENABLED \
       PRIORITY=2
}

Example 1: Specifying a high-availability cluster as a CLUSTER connection unit

In the following example, a high-availability cluster composed of the servers in my_server_group is specified as the CLUSTER connection unit my_cluster:
CLUSTER my_cluster
{
   INFORMIXSERVER my_server_group
   SLA sla_1 DBSERVERS=ANY
   FOC ORDER=ENABLED
       PRIORITY=1
}