This document describes one reason for the frequent failure of sessions between the Real-Time Distributor (RTD) process on the Distributor Admin Workstation (AW) and the Real-Time Server (RTS) process on the Cisco CallRouter. This document also provides a solution in a Cisco Intelligent Contact Management (ICM)/IP Contact Center (IPCC) Enterprise environment.
Cisco recommends that you have knowledge of these topics:
Cisco ICM Enterprise
Cisco IPCC Enterprise
The information in this document is based on these software and hardware versions:
The information in this document was created from the devices in a specific lab environment. All of the devices used in this document started with a cleared (default) configuration. If your network is live, make sure that you understand the potential impact of any command.
Refer to Cisco Technical Tips Conventions for more information on document conventions.
An RTD is an Admin Workstation (AW) that you configure in order to establish and maintain a connection to the Cisco CallRouter. Each administrative site has one or more AWs that you can configure as RTDs.
Real-time feed is the connection between a Cisco CallRouter and a Distributor AW. The rtdist.exe process is an RTD process on the Distributor AW that maintains the real-time feed to rtsrvr.exe. The rtsrvr.exe process is a Real-Time Server (RTS) process that runs on the Cisco CallRouter.
This document explains why users experience frequent session drops between the RTD process on the Distributor AW and the RTS process on the Cisco CallRouter.
A review of the RTS process log on the Cisco CallRouter reveals these important findings:
Figure 1 – Log of the RTS Process
Real-time connection fails (see arrow A in Figure 1).
Write of call type real time base records to client connection fails (see arrow B in Figure 1).
The value of current output queue bytes is equal to the value of the highest output queue bytes (see arrows C and D in Figure 1).
Item 3 represents the key finding. If the value of current output queue bytes approaches or is equal to the highest output queue bytes, you must increase the value of BufferMaxQueue.
On the Cisco CallRouter, double the value of the BufferMaxQueue registry key (see arrow A in Figure 2). You must change this value on both, CallRouterA and CallRouterB. Here is the navigation path:
Figure 2 – BufferMaxQueue for RealTimeServer on CallRouter
On the Distributor AW, double the value of the BufferMaxQueue registry key (see arrow A in Figure 3). Change this value on all running Distributor AWs. Here is the navigation path:
Figure 3 – BufferMaxQueue for RealTimeDistributor on Distributor Admin Workstation
The BufferMaxQueue registry key is not dynamic. Therefore, after you increase the number, cycle the RTS process on the CallRouter (see arrow A in Figure 4), and the RTD process on the Distributor AW (see arrow A in Figure 5).
Figure 4 – Cycle the RTS Process on the CallRouter
Figure 5 – Cycle the RTD Process on the Distributor AW
This problem no longer occurs after you increase the BufferMaxQueue number on the CallRouter and the Distributor Admin Workstation.