Cisco Cisco Unified Contact Center Enterprise 9.0(2) User Guide

network problems between the agent’s phone and its Cisco CallManager, the agent is logged
out automatically. The agent must manually log in again once the phone has connected to the
back up Cisco CallManager.

When the agent is logged out, the agent is removed from the real-time status reports until the
agent logs back in again. The historical information restarts when the agent logs back in again.
The agent Media Logout Status reports for the agent, agent skill group, agent team, and agent
peripheral show a logout reason code of 50003. The previous state that the agent was in before
the fail-over or recovery condition is not maintained.

If the agent is on a call when this failover or recovery condition occurs, the agent does not
failover or revert to the back up or primary Cisco CallManager until the call is disconnected.
The agent might be able to stay on the call, but does not get credit for that call within the historical
reports because signaling to the PG stops until the agent fails-over or reverts to the back up or
primary Cisco CallManager. Once the agent has connected to its back up CallManager, the
agent must log in again. The previous state that the agent was in before the fail-over or recovery
condition is not maintained.

About Application Instance/MR PG Fail-over

If the connection between the Application Instance and MR PG shuts down or either component
shuts down, the ICM/IPCC Central Controller discards all pending NEW_TASK requests
received from the application. The Application Instance waits for the connection to be restored
and continues to send messages regarding existing tasks and new tasks assigned by the
Application Instance to the Agent PG CTI server. When the connection, MR PIM, or Application
Instance is restored, the Application Instance resends any pending NEW_TASK requests for
which it has not received a response from the ICM/IPCC Central Controller. The tasks that are
assigned to the agent by the Application Instance while the connection is down and completed
before the connection is restored do not appear in WebView reports.

Note: If the Application Instance shuts down, this also affects Agent PG CTI server connections.

If the connection between the MR PIM and the ICM/IPCC Central Controller shuts down or
the ICM/IPCC Central Controller shuts down, the MR PIM sends a ROUTING_DISABLED
message to the Application Instance that causes the Application Instance to stop sending routing
requests to the ICM/IPCC Central Controller. Any request that is sent while the connection is
down is rejected with a NEW_TASK_FAILURE message. The Application Instance continues
to send messages regarding existing tasks and new tasks assigned by the Application Instance
to the Agent PG CTI server. When the connection or ICM/IPCC Central Controller is restored,
the MR PIM sends the Application Instance a ROUTING_ENABLED message that causes the
Application Instance to start sending routing requests to the ICM/IPCC Central Controller again.
The tasks that are assigned to the agent by the Application Instance while the connection is
down and completed before the connection is restored do not appear in reports. If the connection
between the ICM/IPCC Central Controller and the MR PG fails, the ICM/IPCC router deletes
all pending new tasks. When the connection is restored, the application connected to MR PG
will resubmit all the tasks.

Note: If the ICM/IPCC Central Controller shuts down, this also affects the application instance/
Agent PG CTI server interface.

Reporting Guide for Cisco IPCC Enterprise & Hosted Editions 7.0(0)