Guest

Cisco Unified Web Interaction Manager

Cisco Unified Web and E-Mail Interaction Manager version 4.2(4) Listener Service Instance hangs while Starting

Document ID: 109596

Updated: Feb 09, 2009

   Print

Introduction

Cisco Unified Web and E-Mail Interaction Manager version 4.2(4) listener service instance hangs while starting. This document explains how to troubleshoot this issue.

Prerequisites

Requirements

Cisco recommends that you have knowledge of installing Cisco Unified Web and E-Mail Interaction Manager. Refer to Installing Cisco Unified Web and E-Mail Interaction Manager for more information.

Components Used

The information in this document is based on the Cisco Unified Web and E-Mail Interaction Manager version 4.2(4).

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.

Conventions

Refer to Cisco Technical Tips Conventions for information on document conventions.

Problem

With Cisco Unified Web and E-Mail Interaction Manager version 4.2(4), a process instance hangs in Starting mode in some circumstances. One common example is the Listener instance that hangs in Starting mode if it cannot successfully communicate with the designated CTI server(s).

Solution

Complete these steps in order to resolve this issue:

  1. On the CIM file server, open the file %install_dir%\CIM\eService\config\egpl_process_ids.properties.

  2. Take a note of the process ID that is associated with the process whose instance is stuck in Starting mode.

  3. On the CIM database server, open SQL Query Analyzer.

  4. Run this query against the CIM master database (by default called eGMasterDB):

    SELECT * FROM egpl_dsm_instance WHERE instance_name like '%listener%'
  5. In the results, take a note of the instance ID (column = instance_id) for the listener process that is in the Starting state.

    Note: For the column state, 3 = Running, 4 = Stopped, 6 = Starting.

  6. Run this query to update the state field in the table to the Stopped state:

    UPDATE egpl_dsm_instance SET state = 4 WHERE instance_id = XXXX
  7. Open the CIM system console, and navigate to the process whose instance was previously stuck in the Starting state. For example, navigate to System > Shared Resources > Services > Listener > Listener > Listener-process.

  8. Click the red X icon in order to stop the process, and verify the state is Stopped.

  9. On the CIM services server, open Windows Task Manager.

  10. Click the Processes tab, and verify that the process ID noted in step 1 no longer shows in the list of running processes. If it still appears in the process list, click the End Process button in order to manually end the process.

    Note: You may have to add the Process ID (PID) column to the display if it is not already. (Choose View > Select Columns > PID (Process Identifier).)

  11. On the CIM system console, start the process first and then the instance. For example:

    1. Navigate to System > Shared Resources > Services > Listener > Listener > Listener-process.

    2. Click the Start button (green play icon), and verify that the process successfully enters the Running state.

    3. Navigate to System > Partitions > <partition> > Services > Listener > Listener > Listener-instance.

    4. Click the Start button (green play icon), and verify that the instance successfully enters the Running state.

  12. If the instance continues to hang in the Starting state, collect the applicable logs and investigate the root cause for the process or instance being unable to start successfully.

For the Listener process, this issue is typically related to these situations:

  • Incorrect CTI server configuration (that is, <ip_address>: <port>)

  • Nonexistent secondary CTI server definition (for a simplex CTI deployment, the single server should be listed as both primary and secondary for the Listener instance)

  • Network connectivity failure between the CIM services server and the CTI server(s)

After you complete the above procedure, if you receive the INVALID_MRD rejection message from CTISVR (in the log) when Listener tries to log in, you must run the exit_opc command on the PG where CTISVR runs in order to clear the messages.

Note: The exit_opc command instructs the OPC process to exit on both sides of the PG, if duplexed. Node Manager forces the process to restart, which then forces a reload of the configuration for the Call Router. All internal peripheral and agent states are flushed (which means logged out), and then the OPC and Peripheral Interface Manager (PIM) relearn the PG and its configuration.

Related Information

Updated: Feb 09, 2009
Document ID: 109596