SIP Dialer and MR PIM Resets with HB Timeout

Introduction

This document describes a problem encountered where the Media Routing (MR) Protocol Independent Multicast (PIM) resets due to Heartbeat (HB) timeout.

Prerequisites

Requirements

Cisco recommends that you have knowledge of the Unified Contact Center Enterprise (UCCE) Session Initiation Protocol (SIP) Dialer.

Components Used

This document is not restricted to specific 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.

Problem

You receive this error message:

08:29:18:003 dialer-baDialer Trace: CSipUATelephonyServer::
HandleKeepAliveTimeoutTimer: Missing HB ACK 1
08:29:18:050 dialer-baDialer Blended Agent connection to SIP Server
[xx.xx.xx.xx] on computer [XXX] is down, heart beat failure detected
08:29:18:128 dialer-baDialer Trace: Error receiving data from PIM.
Last API Error [10053]: An established connection was aborted by the
software in your host machine.

The MR PIM resets with this message:

08:29:18:128 PG2A-pim1 Trace: Transport::RecvBytes. SOCKET_ERROR=10054
08:29:18:144 PG2A-pim1 Error receiving data from MR application.
  Last API Error [10054]: An existing connection was forcibly closed by the
remote host.
08:29:18:175 PG2A-pim1 Trace: MR_Application::Receiver. Problem with receiving
application messages. Exit Receiver thread.
08:29:18:175 PG2A-pim1 Notifying Out Of Service to OPC (ThreadManager::
ClearActivated: MR Application Receiver thread failed.).
08:29:18:191 PG2A-pim1 Trace: Transport::SendMessage. SOCKET_ERROR=10054
08:29:18:206 PG2A-pim1 Error sending data to MR application.
  Last API Error [10054]: An existing connection was forcibly closed by the
remote host.
08:29:18:222 PG2A-pim1 TCP connection to MR application has been broken.
08:29:18:222 PG2A-pim1 ProcessPIMSetIdleReq: Peripheral 5003 going idle.

This scenario is caused when the SIP HB is enabled, and the HB timeout is set too aggressively. This scenario is common when the Cisco Unified SIP Proxy (CUSP) or SIP gateway crosses an unreliable WAN.

Solution

Modify this key, preferably the HBNumTries, in order to accomodate the WAN performance.

HKEY_LOCAL_MACHINE\SOFTWARE\Cisco Systems, Inc.\ICM\cicm\Dialer
HBNumTries and/or HBTimeout

In order to confirm that the network is reliable, take Wireshark captures at both ends. Ensure that the dialer on the Windows 2008 server is configured in order to disable the offload. Receive side scaling and chimney as described in the Contact Center Networking: Offload, Receive Side Scaling and Chimney article.