Cisco Cisco Customer Voice Portal 8.0(1)

•

Ability to differentiate call recovery behavior by incoming DNIS and how long the call had
been in Unified CVP prior to failure.

•

Ability to differentiate call recovery behavior based on time of day and date.

•

Ability to hand off to an auto-attendant type application in the event of some downstream
failure (for example, WAN failure, Unified ICME failure, Unified CVP failure). This
auto-attendant functionality could be CME's BACD, a Unified CVP Standalone call flow
model, a VXML Server application, or a custom-written VXML application.

Installing the Call Survivability Script

To install the Call Survivability script, perform the following steps:

Using the Operations Console, copy all script/prompt files to the gateway.

On the gateway, do the following:

For a Unified CVP Comprehensive call flow model, first define two services:

application

service survive flash:survivability.tcl

paramspace callfeature med-inact-det enable

service handoff flash:handoff.tcl

Then add the following parameters:

 ip rtcp report interval 2000

gateway

timer receive-rtcp 4

• This will cause survivability to be invoked between 8 and 16 seconds ((2000 ms *4) * 2) for

an active call after a WAN failure. If IOS detects the absence of both RTP and RTCP packets
after 8 to 16 seconds, it will raise an error event and survivability will be invoked. (The factor
of 2 is a built-in IOS factor that cannot be configured. Do not adjust these values lower as
this could cause the survivability event to be prematurely invoked.)

• The timer receive-rtcp command configures a media activity timer that is common to both

H.323 and SIP. If set, it affects both H.323 and SIP calls.

For a Unified CVP Standalone call flow model, first define one service:

application

 service my-survivability-service flash:survivability.tcl

Configuration and Administration Guide for Cisco Unified Customer Voice Portal Release 8.0(1)