Cisco Cisco IP Contact Center Release 4.6.2 Maintenance Manual
•
Single CallRouter. When a CallRouter on one side of the Central Controller fails, that entire
side of the Central Controller is removed from service. This is because the CallRouter plays
an integral part in forwarding historical data to the historical Logger on its side of the system.
The on-line side of the Central Controller runs as a simplexed system. Call processing
continues uninterrupted and reporting functions are still available. When the failed CallRouter
returns to service, both CallRouters and both Loggers return to synchronized execution.
side of the Central Controller is removed from service. This is because the CallRouter plays
an integral part in forwarding historical data to the historical Logger on its side of the system.
The on-line side of the Central Controller runs as a simplexed system. Call processing
continues uninterrupted and reporting functions are still available. When the failed CallRouter
returns to service, both CallRouters and both Loggers return to synchronized execution.
•
Logger and CallRouter (opposite sides) . In this failure scenario, side B of the Central
Controller is removed from service due to the CallRouter failure. Call routing continues
uninterrupted with the remaining Side A CallRouter; however, data in both databases slowly
becomes out of date. Some reporting functions are not available until the nodes are returned
to service and synchronized execution is restored.
Controller is removed from service due to the CallRouter failure. Call routing continues
uninterrupted with the remaining Side A CallRouter; however, data in both databases slowly
becomes out of date. Some reporting functions are not available until the nodes are returned
to service and synchronized execution is restored.
•
Both Loggers. In a double Logger failure, call routing continues uninterrupted. If it is the
historical Loggers that failed, all reporting functions are lost until at least one of the historical
Loggers returns. If it is the configuration Loggers that failed, you cannot make any
configuration changes until at least one configuration Logger is operational. Such a double
failure is extremely rare. The CallRouter continues to route calls because it has a copy of the
call center enterprise configuration data in its program memory. (The CallRouter loads the
configuration data into memory when it is started and keeps it up-to-date as configuration
changes occur.)
historical Loggers that failed, all reporting functions are lost until at least one of the historical
Loggers returns. If it is the configuration Loggers that failed, you cannot make any
configuration changes until at least one configuration Logger is operational. Such a double
failure is extremely rare. The CallRouter continues to route calls because it has a copy of the
call center enterprise configuration data in its program memory. (The CallRouter loads the
configuration data into memory when it is started and keeps it up-to-date as configuration
changes occur.)
•
One Side. If one side of the Central Controller goes off-line, processing continues
uninterrupted. The system software continues to function as a simplexed system until the
failed side of the Central Controller returns to service. All functions remain, but the system
is running simplexed (without protection against an additional failure). When the off-line
side of the Central Controller returns, normal duplexed operation is restored.
uninterrupted. The system software continues to function as a simplexed system until the
failed side of the Central Controller returns to service. All functions remain, but the system
is running simplexed (without protection against an additional failure). When the off-line
side of the Central Controller returns, normal duplexed operation is restored.
A double CallRouter failure would temporarily disrupt call routing and reporting functions.
This type of failure is extremely rare (especially in geographically distributed Central Controller
configurations).
This type of failure is extremely rare (especially in geographically distributed Central Controller
configurations).
Geographic Distribution
To provide maximum protection against disasters such as fires, floods, and earthquakes, the
two sides of the Central Controller can be in separate locations—even in separate cities. The
two Synchronizers communicate with each other via a private wide area network (WAN) to
ensure that they remain synchronized. This WAN, called the private WAN, is used for no other
purpose than to ensure synchronization between the sides of the Central Controller.
two sides of the Central Controller can be in separate locations—even in separate cities. The
two Synchronizers communicate with each other via a private wide area network (WAN) to
ensure that they remain synchronized. This WAN, called the private WAN, is used for no other
purpose than to ensure synchronization between the sides of the Central Controller.
See Also
For details on collocated and distributed Central Controller configurations, refer to the Installation
Guide for Cisco Unified ICM/Contact Center Enterprise & Hosted.
Guide for Cisco Unified ICM/Contact Center Enterprise & Hosted.
Administration Guide for Cisco Unified ICM/Contact Center Enterprise & Hosted Release 8.x
21
Chapter 2: Fault Tolerance
Central Controller