Diagnosing and Correcting Cisco CRA Problems

Table of Contents

Diagnosing and Correcting Cisco CRA Problems
General Troubleshooting Steps
Cisco Agent Desktop Problems
Cisco CallManager Automated Attendant Problems
Cisco CallManager Extension Mobility Problems
Cisco IP ICD Problems
RM-CM subsystem is out of service
Agent or CSQ does not appear in Cisco Desktop Administrator
Supervisors do not appear in the Cisco Desktop Administrator
Agents do not appear in the Resources area in the ICD Configuration web page
The ICD radio button is not available
You cannot select the order of agents
Changes are not saved when you modify a resource group, skill, resource, or CSQ
An agent does not go to Work state after handling a call
Error when trying to select skills in the Team View pane
IP Phone Agent statistics do not appear
An agent cannot see how many agents are logged in
A media step causes a Could not create PlayPromptDialog Object exception
Calls to ICD route points are disconnected
Calls are not routed to agents
Agents do not show in a CSQ
Caller gets dropped when an attempt is made to recall an ICD agent's extension after the agent previously parked the call
Agents cannot take calls in the correct order
Supervisor gets an error when trying to monitor an agent or a CSQ
Supervisor cannot monitor a CSQ
Supervisor unable to start Monitoring device <extension>
Supervisor can start monitoring but hears nothing
Supervisor cannot record an agent's call
Supervisor cannot play back recording of an agent's call
Cisco IP IVR Problems
CRA Administration Problems
The CRA Administration Authentication web page is not available
The Unauthorized web page appears when you try to log in
User cannot log in to the CRA web page
The CRA system logs you out automatically
Error message when selecting a profile
Engine status is unavailable
Changes to certain parameters do not take effect
The Subsystem Status does not show the subsystems that are running on the CRA server
The CTI Route Point Directory Number field does not appear as a drop-down list
The message "The Engine is not running" appears even though the CRA Engine is running.
The Cisco banner appears in English even when the system language is not English
You can configure more IVR ports than are licensed
The correct number of licensed IVR ports does not appear
Calls to the CRA system are aborted
The message "There has been an error while reading or saving to the database" appears
A JTAPI trigger is not available in the CTI Route Point Directory Number drop-down list after deleting an application
Refreshing subflow script does not update parent scripts
User deleted from CallManager cannot log in to Cisco CRA
CRA Supervisor web page cannot be viewed from CRA Server
Database table fields used by wallboard store data in milliseconds
Changing host name or IP address causes problems
All user records not displayed on CRA User Management web page
CRA Database Problems
CRA Editor Problems
CRA Engine Problems
The CRA Engine does not start
The CRA Engine does not start and an RMI port in use error appears
Attempting to start the Cisco CRA Engine service causes an
error 1067

Attempting to start the Cisco CRA Engine service causes an
error 1069

Application subsystem is in partial service
The CRA Engine is running but calls are not answered
An error message plays when calling a CTI route point
Changes to applications do not register
Call drops during transfer over gateway
H.323 client DTMF digits not detected
Subsystem status is not available
Subsystem status is incorrect
Redirected call is disconnected
The CRA server runs out of disk space
CRA Server runs at 100% capacity
Database Subsystem goes into partial service
JTAPI subsystem is in partial service
Unable to connect to JTAPI provider
The Simple Recognition step takes the unsuccessful branch
Calling party and CRA do not have common codec
Prompts with incorrect codec being played out
Prompt Exception in CRA Engine log file
CRA Engine does not start
Application subsystem in partial service and application running for an unexpectedly long time
CRA Real-Time Reporting Problems
CRA Historical Reporting Problems
Error when choosing an option from the Historical Reporting web page
The SQL Command Failed dialog box appears when you try to generate a historical report
Some information appears in English on a German system
The Print dialog box displays an incorrect printer
The Historical Reporting client computer cannot connect to the CRA server
A Database Connection Error 5051 error appears
Export file name does not appear in Export dialog box
User Configuration page has empty installed products packages
Cannot point to local applications from the Database Server Configuration page
Cannot reuse the previous Historical Reports Database Server after uninstalling Cisco CRA, reinstalling it, and dropping the CRA database
Attempt to log in to the CRA Server from the Historical Reporting client fails and an error message is returned
CRA Installation Problems
LDAP Directory Problems
Nuance ASR Problems
Nuance TTS Problems
Serviceability Problems
CRA Internationalization Problems

Diagnosing and Correcting Cisco CRA Problems


This manual describes common problems that you might encounter when using the Cisco Customer Response Applications (CRA) system. For each problem, this manual lists symptoms, possible causes, and corrective actions that you can take.

This manual assumes that you are familiar with the CRA web interface, CRA web pages, CRA trace and log files, and various Windows administrative tasks. For more information, refer to Cisco Customer Response Applications Administrator Guide, Cisco Customer Response Applications Serviceability Guide, and your Windows documentation.

The following sections include these topics:

General Troubleshooting Steps

The following troubleshooting steps can help you diagnose most problems with your Cisco CRA products:

1. Verify that Cisco CallManager is running.

2. Verify that the LDAP Directory service is running.

3. Verify that the CRA Engine service is registered.

4. Verify that you uploaded the application.aef files to the repository using the Repository Manager and that you refreshed the CRA Engine after making a change to an application.

5. Refer to the Release Notes for known problems.

6. Verify that the Cisco CRA Engine service is running under a user account with Administrator privileges.

7. Stop and start the Internet Information Server (IIS).

8. Save log files to prevent them from being overwritten.

9. Save the application (.aef) file.

10. Before debugging CRA Application Administration problems, turn on the Debugging trace level option for the ADM subfacility.

The detailed output will be in the following file:

c:\program files\wfavvid\tomcat_appadmin\logs\jvm.stdout.

The error output will be in the following file:

c:\program files\wfavvid\tomcat_appadmin\logs\jvm.stderr.

Cisco Agent Desktop Problems

The following section describes common problems with the Cisco Agent Desktop. For additional troubleshooting information about the Cisco Agent Desktop, refer to Service Information—Cisco Desktop Product Suite 4.5 (ICD).

Agent unable to log in to Cisco Agent Desktop

Symptom   An agent receives an error message when trying to log in to the Cisco Agent Desktop. See the following table for various messages, possible causes and recommended actions.

The following table provides the error messages, possible causes, and recommended actions for this symptom:

Message Possible Cause Recommended Action

Error Reading File \\server\DESKTOP_CFG\
config\ag_default\DataFields.ini
, where server is the IP address of the CRA server.

  • The DESKTOP_CFG configuration file is not read-write accessible to Cisco Agent Desktop users.
  • The IP connection to the CRA server is lost.
1. Make sure that the shared location and read-write file access is properly established for the DESKTOP_CFG configuration file.

2. Make sure that the IP connection to the CRA server is established.

Failed to login into CTI Manager Server! Please talk to your administrator.

  • Enable CTI Application Use is not configured for the agent user ID.
  • You are using an invalid password.
  • The Cisco CTIManager service is not running on the Cisco CallManager server.
1. From Cisco CallManager, choose User > Global Directory, and check the Enable CTI Application check box.

2. Verify that you are using the correct Cisco CallManager password.

3. Make sure that the Cisco CTIManager service is running on the Cisco CallManager server.

The ID you entered was not found.

The user ID you entered is invalid.

Make sure that you are typing the user ID correctly. User IDs are case sensitive. Verify that you are using the correct Cisco CallManager password.

Unable to log agent in.

The agent's phone is not associated with the RM JTAPI provider in the Cisco CallManager.

In the User ID Field in the Cisco CallManager ICD Configuration web page, associate the agent's phone with the RM JTAPI provider.

A critical error has been received. Either your phone or the CallManager is offline. If you are not already logged out, you may need to logout and try to log in again.

The Cisco CallManager server is offline or the Agent's IP phone has reset.

1. Make sure that the Cisco CallManager server is online.

2. Verify that the agent's phone is in service.

No data appears in the Enterprise Data fields

Symptom   When an Agent receives a call, the Enterprise Data window does not display the expected data.

Possible Cause   The CRA server is not correctly passing enterprise data from the enterprise data server to the Cisco Agent Desktop. This situation can be a result of incorrect step configuration in the script or in the Enterprise Data Configuration section of the Cisco Desktop Administrator. This situation can also be a result of an out of sync condition between the Enterprise Data subsystem and the Cisco Desktop Enterprise Server service.

Recommended Action   

1. Verify the step configuration in the script and in the Enterprise Data Configuration section in the Cisco Desktop Administrator.

2. Stop and restart the Cisco Desktop Enterprise Server service.

3. If the problem persists, stop and restart the CRA Engine.

Enterprise Server is down and is attempting to recover

Symptom   A user receives the message, Enterprise Server is down and is attempting to recover. The agent might not get Enterprise data during this time and any Enterprise data changes that the agent makes might not be propagated.

Possible Cause   The network connectivity between the client and server went down.

Recommended Action   Restart the Cisco Agent Desktop (CAD). Normal operation of the enterprise data starts when the message, Enterprise Server recovery has been completed. appears. This happens after calls queued prior to enterprise data recovery have been flushed out of the system. Sometimes after the engine restarts, an agent might find enterprise data missing for a few calls.

Chat Server is down and is attempting to recover

Symptom   User receives the message, Call Chat Server is down and is attempting to recover. The agent is unable to use the chat feature. The supervisor is unable to use the desktop.

Possible Cause   The network connectivity between the client and server went down.

Recommended Action   None. The client will recover automatically. Normal operation is observed after the user receives the message, Call Chat Server recovery has been completed.

Recording and Statistics Server is down and is attempting to recover

Symptom   User receives the message, Recording and Statistics Server is down and is attempting to recover... Recording and Agent Statistics will be unavailable. The agent is unable to see agent statistics, call logs, or agent logs during this time. If recording is enabled, the agent cannot start recording. The supervisor cannot see agent statistics and is unable to start recording.

Possible Cause   The network connectivity between the client and server went down.

Recommended Action   None. The client will recover automatically. Normal operation is observed after the message, Recording and Statistics Server recovery has been completed. Call may now be recorded.

VoIP Monitor Server is down and is attempting to recover

Symptom   User receives the message, At least one Voice over IP Server is down and is attempting to recover. You can't monitor or record some of the agents. Current monitoring and recording sessions may be affected. The supervisor cannot start recording or monitoring agents in the failed VoIP Server's domain during this time. Ongoing recordings and monitorings of agents in the failed domain are unaffected. Note that agents belonging to a failed VoIP server's domain will not get an indication that the VoIP server is down. If agent recording is enabled and if the agent tries to start a recording, an error is returned.

Possible Cause   The network connectivity between the client and server went down.

Recommended Action   None. The client will recover automatically. Normal operation is observed after the user receives the message, Voice over IP Server recovery has been completed. You can now monitor and record all agents.

Cisco CallManager Automated Attendant Problems

The following section describes common problems with the Cisco CallManager Automated Attendant.

Dial by name does not find the specified user

Symptom   The Cisco CallManager Automated Attendant cannot find a user that a caller specifies when dialing by name.

Possible Cause   The extension of the requested user is not valid because the user does not have a primary extension assigned in Cisco CallManager, or the ccndir.ini file is missing information.

Recommended Action   

1. In the Cisco CallManager User Information web page, verify that the user has an entry in the AutoAttendant Dialing field, that the User record has an associated phone, and that the Primary Extension radio button is selected

2. On the CRA server, verify that the ccndir.ini file contains the correct userbase and profilebase information. For example:
USERBASE "ou=Users, o=cisco.com"
PROFILEBASE "ou=profiles, ou=CCN, o=cisco.com"

Cisco CallManager Automated Attendant prompt is not played

Symptom   The Cisco CallManager Automated Attendant prompt is not played.

Possible Cause   An incorrect welcome prompt is specified in the welcomePrompt field in the Cisco Script Application web page.

Recommended Action   From the CRA Administration web page, choose System > System Parameters. Make sure that the following information appears in the User Prompt Directory field:

C:\program files\cisco\wfavvid\Prompts\User

Cisco CallManager Extension Mobility Problems

The following section describe common problems with Cisco CallManager Extension Mobility. For additional information about Cisco CallManager Extension Mobility, refer to this URL:

http://www.cisco.com/univercd/cc/td/doc/product/voice/c_callmg/3_2/sys_ad/
3_2_1/ccmsys/a07hotel.htm

The message HTTP Error (12) appears when logging into a phone using Cisco CallManager Extension Mobility

Symptom   An agent attempts to log in to a phone using Extension Mobility and receives the message HTTP Error (12).

Possible Cause   Port 8080 is assigned to an application other than the Tomcat web server. Port 8080 must be open for the Tomcat web server, otherwise the required HTTP subsystem will not run.

Recommended Action   Make sure that no application other than the Tomcat web server is using port 8080. If another application is using port 8080, change the port for that application and then stop and restart the CRA Engine.

You can verify that the HTTP subsystem works by accessing the URL http:// hostname:8080, where hostname is the host name or IP address of your CRA server. If the Cisco copyright notice appears, the HTTP subsystem is running.

Cisco IP ICD Problems

The following section describes common Cisco IP Integrated Contact Distribution (ICD) problems.

Refer to the Cisco IP ICD Agent Desktop online help for important information about supported and unsupported agent actions and phone configurations.

RM-CM subsystem is out of service

Symptom   The Resource Manager Contact Manager (RM-CM) subsystem is out of service.

Possible Cause   The RM JTAPI user in Cisco CallManager is not configured properly.

Recommended Action   

1. From the CRA Administration web page, choose Subsystems > ICD, click the RM JTAPI Provider hyperlink, and then make sure that the information in the RM JTAPI User ID and Password fields matches the information for the RM JTAPI user in Cisco CallManager.

2. In Cisco CallManager, verify that the Enable CTI Use option is enabled for the RM JTAPI user.

3. Make sure you have associated the agent's phone to the user.

Agent or CSQ does not appear in Cisco Desktop Administrator

Symptom   After adding an agent or a contact service queue (CSQ) in CRA Administration, the agent or the CSQ does not appear in the Cisco Desktop Administrator.

Possible Cause   The Cisco Desktop Administrator is automatically updated periodically to reflect configuration changes that are made in CRA Administration. This periodic update might not have occurred yet.

Recommended Action   To force an update, from the Cisco Desktop Administrator, choose Locations > Setup and then click Synchronize Directory Services.

Supervisors do not appear in the Cisco Desktop Administrator

Symptom   After adding supervisors in Cisco CallManager Administration and configuring them as resources in CRA Administration, the supervisors do not appear in the Cisco Desktop Administrator.

Possible Cause   The users have not been added as supervisors in the Cisco Desktop Administrator.

Recommended Action   Make sure that the Cisco Desktop Administrator has been updated to reflect the changes made in CRA Administration. To force an update, complete the following steps:

1. From the Cisco Desktop Administrator, choose Locations > Setup.

2. Click Synchronize Directory Services.

3. Choose Personnel Configuration, and then choose Supervisors.

4. Click Add, and then choose the desired userid from the list that appears.

Agents do not appear in the Resources area in the ICD Configuration web page

Symptom   No agents appear in the Resources area in the ICD Configuration web page.

Possible Cause   To appear as an agent in this area, a user must be configured as an ICD agent in the Cisco CallManager User Information web page.

Recommended Action   In Cisco CallManager, verify configuration information in the User Information web pages. For each user, under Associated Devices, verify that a phone is associated, and verify that the ICD extension radio button is selected.

The ICD radio button is not available

Symptom   The ICD radio button is not available in the Cisco CallManager Associated Devices web page.

Possible Cause   An error occurred during the installation process.

Recommended Action   Perform the following steps to modify the system profile object in the LDAP directory and set the iaq flag to true:

1. Log in to the LDAP server using DC Admin (or another login program based on your LDAP server type).

2. Navigate to the following location: CCN > systemProfile.

3. Right-click SystemProfile, and then set the IAQ Flag under the Application Install Status tab to true.

4. Wait 15 minutes or restart the IIS and its dependent services.

5. If you still do not see the agents, from the CRA Administration web page, choose System > Configuration and Repository and verify that the User Base field contains the location where users are stored.

You cannot select the order of agents

Symptom   When you configure a resource group, the system does not allow you to select the order of agents.

Possible Cause   You order agents at the CSQ level.

Recommended Action   When you configure the CSQ and select the desired Resource Group, click Show Resources and order the agents as desired.

Changes are not saved when you modify a resource group, skill, resource, or CSQ

Symptom   You click Update in CRA Administration after modifying a resource group, skill, resource, or CSQ, but your changes are not saved.

Possible Cause   The CRA database was not configured properly during the installation process.

Recommended Action   For Cisco CRA, Version 3.1, perform the following steps on the CRA server:

1. On the Cisco CRA Administration web page choose Tools > Historical Reporting > Database Configuration.

2. Be sure that the IP address shown points to the server containing the CRA databases, either local or remote (if you have configured a Historical Reports Database Server.)

If you are using an earlier version of Cisco CRA, such as 3.0(1), follow these steps:

1. On the CRA Server, verify that the following settings are correct in the wfengine.properties file, which is located by default in the C:\ProgramFiles\wfavvid directory:

HISTORICAL_REPORTING_SERVER should specify the IP address or host name of the CRA Server.

HISTORICAL_REPORTING_URL should be jdbc\:odbc\dsn_cra.

2. Choose Start > Settings > Control Panel > Administrative Tools > Data Sources (ODVC). Choose the System DSN tab and make sure that dsn_cra appears in the list of system data sources. If dsn_cra does not appear, re-run the CRA installation program.

An agent does not go to Work state after handling a call

Symptom   An agent does not go to Work State after handling a call, even though the CSQ is configured with Auto Work turned on.

Possible Cause   An agent will not go to Work State after handling a call if the agent presses the Ready button while in Talk state. In addition, if the agent services multiple CSQs, Auto Work may not be configured for each CSQ. The agent will only go to Work State if the call comes from a CSQ where Auto Work is enabled.

Recommended Action   None.

Error when trying to select skills in the Team View pane

Symptom   When you try to select skills in the Team View pane in the Cisco Supervisor Desktop, the following message appears: Cisco Agent Desktop must be active before call intervention, call recording, and queue stats are available!

Possible Cause   To view skill group statistics, you must log in to a Cisco Agent Desktop as a supervisor.

Recommended Action   Log in to a Cisco Agent Desktop as a supervisor.

IP Phone Agent statistics do not appear

Symptom   Statistics for an IP Phone Agent do not appear in the Team View pane on the Cisco Supervisor Desktop, but statistics do appear for a Desktop Agent.

Possible Cause   The Cisco Supervisor Desktop does not support viewing statistics for IP Phone Agents.

Recommended Action   None.

An agent cannot see how many agents are logged in

Symptom   An agent cannot see how many agents are logged in.

Possible Cause   Agents do not have access to this information. The Cisco Agent Desktop displays only the number of calls queued for the CSQ to which the agent belongs and the duration of the oldest contact in queue.

Recommended Action   Only supervisors have the ability to see how many agents are logged in. This information, and related information, must be viewed from the Cisco Supervisor Desktop.

A media step causes a Could not create PlayPromptDialog Object exception

Symptom   Any media step except SendDigitString causes the following exception in the CRA trace files: Could not create PlayPromptDialog Object: Exception=com.cisco.channel.ChannelUnsupportedException: com.cisco.dialog.PlayPromptDialog is not supported.

Possible Cause   A Primary Dialog Group was not specified when a trigger was defined.

Recommended Action   After you add an application in the CRA Configure Applications web page, you must define a trigger. When you define a trigger for the application, you must define both a Call Control Group and a Primary Dialog Group in the JTAPI Trigger Configuration window.

Calls to ICD route points are disconnected

Symptom   Callers are disconnected when calling ICD route points.

Possible Cause   The CSQ parameter is not correctly defined in the Cisco Script Application web page.

Recommended Action   From the CRA Administration web page, choose Applications > Configure Applications, click the name of the script that corresponds to ICD, and then enter the name of the configured CSQ in the CSQ field.

Calls are not routed to agents

Symptom   Calls are not routed to agents even though the agents are configured with the skills of the CSQ.

Possible Cause   The skill levels of the agents are not equal to or higher than the skill levels of the CSQ.

Recommended Action   Click Show Resources on the CSQ configuration page to determine that agents are part of the CSQ. If agents do not appear, verify that the skill levels of the agents are equal to or higher than the skill levels of the CSQ.

Agents do not show in a CSQ

Symptom   A CSQ is configured with a group of agents for Skill A and a group of agents for Skill B; however, the agents do not show up in the CSQ.

Possible Cause   Agents do not have all the skill levels of the CSQ or the skill level of the agents do not have equal or higher skill levels than that of the CSQ.

Recommended Action   Verify that agents have all the skill levels of the CSQ and that the agents have equal or higher skill levels than that of the CSQ.

Caller gets dropped when an attempt is made to recall an ICD agent's extension after the agent previously parked the call

Symptom   Agent A gets an ICD call and parks that call. After the parked call times out, a recall is attempted to Agent A's extension (if no other agent has picked up the call). If Agent A is busy handling another call on that ICD extension while the previously parked call is being routed, the caller gets dropped.

Possible Cause   A parked call gets dropped if an attempt is made to place the call again to a busy line that is not set with "forward busy."

Recommended Action   Configure Agent A's ICD extension with "Forward Busy" to a shared line on the same phone. Also, configure the shared line as "Forward Busy" to the ICD route point. When an attempt is made to recall Agent A's ICD extension, the call is forwarded to the shared line if the extension is busy. If the shared line is busy, the call is forwarded to the ICD route point and gets queued again instead of being dropped. You can set up the workflow of the ICD route point to increase the priority of the call.

Agents cannot take calls in the correct order

Symptom   After changing the configuration of a skill-based CSQ, agents cannot take calls in the order expected.

Possible Cause   The order of the agents is disturbed after a configuration change, and it takes awhile before calls are received in the correct order. Before that can happen, all the agents have to go through the Ready to Not Ready to Ready state transition. The maximum number of calls to the CSQ until all agents handle calls in the expected order is equal to the number of agents in the CSQ.

Recommended Action   The agent should do one of the following actions:

  • Handle an ICD call.
  • Go from Ready to Not Ready to Ready manually.
  • Log off and log back in.

Supervisor gets an error when trying to monitor an agent or a CSQ

Symptom   When trying to monitor an agent or a CSQ, the supervisor gets an error.

Possible Cause   The VoIP Monitor subsystem is out of service and the VoIP monitor server is down.

Recommended Action   Restart the VoIP Monitor server.

Supervisor cannot monitor a CSQ

Symptom   The supervisor cannot remotely monitor a CSQ.

Possible Cause   The CSQ name might be invalid.

Recommended Action   To verify that the CSQ name is valid do the following:

1. Go to Applications > Configure Applications and select the application that is configured for remote monitoring and make sure the CSQx_Name exists in the system.

1. From the Administration web page, go to Subsystems > ICD.

2. Click the Contact Service Queues link.

A page appears showing a list of all the configured CSQs in the system.

Supervisor unable to start Monitoring device <extension>

Symptom   The supervisor cannot start monitoring and gets the error, Supervisor is unable to start Monitoring device extension. There are multiple monitoring domains and multiple VoIP monitoring servers installed.

Possible Cause   The agent's phone device is not associated with a VoIP monitor server and there is no default VoIP monitoring server selected.

Recommended Action   

1. Select Cisco Desktop Administrator > Locations > your CRA Server LCC > Enterprise Data Configuration > VoIP Monitor.

2. Ensure that either the agent's phone device is associated with the correct VoIP Monitor or the correct VoIP Monitor Server is selected as the "Default VoIP Monitor." Do the latter only if it makes sense for that server to be the default; that is, the majority of your agent phone devices need to be associated with that server.

Supervisor can start monitoring but hears nothing

Symptom   In a situation with multiple monitoring domains and multiple VoIP monitoring servers installed, the supervisor can start monitoring but hears nothing.

Possible Cause   One of the following conditions might exist:

  • The agent's phone device is associated with an incorrect VoIP Monitor Server.
  • The agent's phone device is not associated with a VoIP Monitor Server and an incorrect VoIP Monitor Server is selected as the default VoIP Monitor Server.
  • Spanning is not set up correctly on the switch.

Recommended Action   

1. Select Cisco Desktop Administrator > Locations > your CRA Server LCC > Enterprise Data Configuration > VoIP Monitor.

2. Ensure that either the agent phone device is associated with the correct VoIP Monitor Server or the correct VoIP Monitor Server is selected as the "Default VoIP Monitor Server." Do the latter only if it makes sense for that server to be the default; that is, the majority of your agent phone devices need to be associated with this server.

3. Make sure the span settings for your switch are correct.


Note    Incorrect span settings can be a problem whether or not you are using multiple monitoring domains.

Supervisor cannot record an agent's call

Symptom   In a situation with multiple monitoring domains and multiple VoIP monitoring servers installed, the supervisor cannot start recording an agent's call.

Possible Cause   The agent's phone device is not associated with a VoIP Monitor Server and there is no default VoIP monitoring server selected.

Recommended Action   

1. Select Cisco Desktop Administrator > Locations > your CRA Server LCC > Enterprise Data Configuration > VoIP Monitor.

2. Ensure that the agent phone device is associated with the correct VoIP Monitor Server or the correct VoIP Monitor Server is selected as the "Default VoIP Monitor Server." Do the latter only if it makes sense for this server to be the default; this is, the majority of your agent phone devices need to be associated with this server.

Supervisor cannot play back recording of an agent's call

Symptom   In a situation with multiple monitoring domains and multiple VoIP monitoring servers installed, the supervisor can record an agent's call, but when the recording is played back, it just starts and stops immediately.

Possible Cause   One of the following conditions might exist:

  • The agent's phone device is associated with an incorrect VoIP Monitor Server.
  • The agent's phone device is not associated with a VoIP Monitor Server and an incorrect VoIP Monitor Server is selected as the default VoIP monitoring server.
  • Spanning is not set up correctly on the switch.

Recommended Action   

1. Select Cisco Desktop Administrator > Locations > your CRA Server LCC > Enterprise Data Configuration > VoIP Monitor.

2. Ensure that either the agent phone device is associated with the correct VoIP Monitor Server or the correct VoIP Monitor Server is selected as the "Default VoIP Monitor Server. Do the latter only if it makes sense for this server to be the default; that is, the majority of your agent phone devices need to be associated with this server.

3. Be sure the span settings for your switch are correct.

Note: Incorrect span settings can be a problem whether or not you are using multiple monitoring domains

Cisco IP IVR Problems

The following section describes common problems with Cisco IP Interactive Voice Response (IVR).

Cisco IP IVR drops callers when transferring to an extension

Symptom   After Cisco IP IVR transfers a call to an extension, the called party hears a busy signal when taking the call and the caller is dropped.

Possible Cause   If a call gets dropped, one potential cause is a codec mismatch between the endpoint and the CRA Server. The CRA Server supports either the G.729 or the G.711 protocol, but not both simultaneously. To support these protocols, a transcoder is required.

Recommended Action   Install a transcoder for Cisco IP IVR.

Calls to CTI route points are disconnected

Symptom   Callers are disconnected when calling CTI route points.

Possible Cause   A script includes steps that require media channels but the JTAPI trigger does not have a valid dialog control group configured.

Recommended Action   

1. From the CRA Administration web page, choose Subsystems > Cisco Media and verify that a valid CMT Dialog Control Group exists.

2. From the CRA Administration web page, choose Subsystems > JTAPI, click the JTAPI Triggers hyperlink, click the JTAPI trigger corresponding to the route point, and then choose a valid CMT Dialog Control Group name in the Primary Dialog Group field.

3. Stop and start the engine after changes are made and try calling route point again.

A script is assigned to a route point and set to a language but callers do not hear prompts

Symptom   When calling a script that has been assigned to a route point and set to a language, callers do not hear any prompts.

Possible Cause   The script is invalid or the language to which the script has been set was not installed successfully.

Recommended Action   

1. Validate the script.

2. Set the language at the route point to en_US and verify that the script operates correctly. If it does not, follow these steps:

a. From the CRA Administration web page, choose System > Engine, click the Trace Configuration hyperlink, and then check the Debugging check boxes for the LIB_MEDIA and the SS_TEL subfacilities.

b. Run the script again and refer to the CRA trace files. If prompt exceptions appear in the CRA trace files, reinstall the desired language.

Prompts play in an incorrect language

Symptom   A script was assigned to a language at the route point but it plays prompts in another language.

Possible Cause   This problem can be caused by the following situations:

  • The system default language is set incorrectly.
  • The language specified in the Set Contact step is incorrect.
  • The language specified in the Play Prompt step is incorrect.

Recommended Action   Verify that system default language is set correctly. Verify that the correct language is set in the Set Contact step or the Play Prompt step if these steps are used.

Some prompts do not play

Symptom   A prompt in a script does not play. The script may or may not continue executing.

Possible Cause   A prompt is missing in the language directory for the language used by the script. By default, the Play Prompt step is set to continue if it encounters an error and the script will continue to play if it encounters a missing prompt. If you have changed the Play Prompt step to not continue if it encounters an error, the script will stop executing.

Recommended Action   Refer to the CRA trace files to find the missing prompt. Provide the missing prompt in the language folder shown in the CRA trace files.

Some prompts in a script play in the language specified and other prompts play in English

Symptom   A script is set to a language other than US English, but some prompts play in US English.

Possible Cause   A prompt is missing in the language directory for the language used by the script. If the default language for the script uses the same rules as US English, the system will automatically replace the missing prompt with a US English prompt.

Recommended Action   Refer to the CRA trace files to find the missing prompt and provide the missing prompt in the language folder shown in the CRA trace files.

A prompt plays phrases in the wrong order

Symptom   A prompt played by the Create Generated Prompt step plays the correct language but plays phrases in the wrong order. For example, a prompt that should play as "month, day, year" plays as "year, month, day."

Possible Cause   The Create Generated Prompt step is using incorrect rules for the language.

Recommended Action   If creating a new language or adapting an existing language for a new locale, check the PromptGenerator.properties file and make sure that it is using the correct rules for the language.

CRA Administration Problems

The following section describes common problems with administering CRA.

The CRA Administration Authentication web page is not available

Symptom   You cannot browse to the CRA Administration URL and a Page Cannot be Displayed error appears.

Possible Cause   The system cannot access the CRA Administration web page. A required service may not be running or required files may be missing.

Recommended Action   

1. Make sure that the following services are running:

  • CiscoCRAServletEngine service
  • IIS Admin service
  • World Wide Web Publishing service

2. If these services are running, verify that files exist in the install_directory\tomcat_appadmin\webapps\appadmin\ directory, where install_directory is the folder in which the CRA system is installed. (By default, the CRA system is installed in the c:\Program Files\wfavvid folder.) If no files exist in this directory, perform the following steps:

a. Stop the CiscoCRAServletEngine service.

b. Delete the appadmin folder from the tomcat_appadmin\webapps folder in the folder in which you installed the CRA system. (By default, the CRA system is installed in the c:\Program Files\wfavvid folder.)

c. Start the CiscoCRAServletEngine service.

d. Wait for a few minutes and try to browse to the URL again.

The Unauthorized web page appears when you try to log in

Symptom   After you enter a User Identification and Password and click Log On in the CRA Administration Authentication web page, the Unauthorized web page appears.

Possible Cause   You entered an incorrect ID or password, you have not been configured properly in CRA Administration, or the DC Directory Server service (or another LDAP service) is not running.

Recommended Action   

1. Make sure that you are using a User ID that is a configured as Administrator. Even though a user exists in Cisco CallManager, a user must be configured as Administrator in CRA Administration.

2. User ID and Password are case sensitive, so make sure that you enter these items correctly.

3. Do not use "administrator" as the user ID or "ciscocisco" as the password. These names work only when the system is first set up. After the system is set up and users are configured as administrators, the administrator/ciscocisco identification no longer works.

4. Make sure that the DC Directory Server service running. Or, if you have specified another LDAP as the active directory, make sure that service is running.

User cannot log in to the CRA web page

Symptom   A user cannot log in to the Cisco CRA web pages after the CRA Administration times out.

Possible Cause   Cisco CRA Administration has timed out, the CRA Authentication web page appears, but you cannot log back in.

Recommended Action   Exit your web browser, then start it again and reconnect to the Cisco CRA web pages.

The CRA system logs you out automatically

Symptom   You are logged in to CRA but have performed no activity for a while and the system requires you to log in again before continuing.

Possible Cause   If you perform no activity for 30 minutes, the CRA system automatically logs you out.

Recommended Action   Log in again to continue.

Error message when selecting a profile

Symptom   You enter a profile in the Profile Name field in the Configuration Setup area and the following message appears: The 2.x profile you selected must be converted to 3.0 using the conversion tool.

Possible Cause   You are using a profile from CRA version 2.x.

Recommended Action   To use a profile from CRA 2.x in CRA 3.0, you must convert the old profile using the LDAP conversion tool. For more information, refer to Appendix C, "Cisco CRA 2.x to 3.0 Profile Conversion," in Getting Started with Cisco Customer Response Applications.

Engine status is unavailable

Symptom   In the Engine Status web page, the Engine status appears as UNAVAILABLE.

Possible Cause   Cisco CRA Engine is not registered as a Windows service.

Recommended Action   Make sure that Cisco CRA Engine appears in the Windows Services window. If it does not, perform the following steps:

1. Open a command window.

2. In the command window, type cd progra~1\wfavvid. (If the CRA system is installed in another folder, replace progra~1 with that folder name.)

3. Type WFEngineService -service to register the Cisco CRA Engine service as a Windows service.

Changes to certain parameters do not take effect

Symptom   You make to any of the following fields in the CRA Administration web pages but the changes do not take effect:

  • The Profile Name field in the Directory Setup web page.
  • Any fields in the Engine Configuration web page.
  • The RTP Start Port field in the System Parameters Configuration web page.
  • All fields in the JTAPI Configuration web page.
  • The Speech Server Host Name(s) field in the Nuance ASR Configuration web page.
  • Any field in the Nuance TTS Server Configuration web page that you modify by clicking the parameter, changing its value on the Vocalizer TTS Server Configuration pane, and then clicking Update.
  • Any field in the RM JTAPI Provider web page.
  • Adding or removing a Nuance TTS server in the Nuance TTS Server Configuration web page.

Possible Cause   These changes require you to restart the CRA Engine.

Recommended Action   Stop and restart the CRA Engine.

The Subsystem Status does not show the subsystems that are running on the CRA server

Symptom   Subsystems that are running on the CRA server do not appear in the Engine Status area in the Engine web page.

Possible Cause   An incorrect IP address or host name for the CRA server is specified in CRA Administration.

Recommended Action   From the CRA Administration web page, choose System > Engine, click the Engine Configuration hyperlink, and then make sure that the correct IP address or host name for the CRA server appears in the Application Engine Hostname field.

The CTI Route Point Directory Number field does not appear as a drop-down list

Symptom   The CTI Route Point Directory Number field in the JTAPI Configuration web page appears does not have a drop-down list of choices.

Possible Cause   The CRA Engine is not running or the JTAPI User configured in Cisco CallManager has no more unassigned route points that are configured as associated devices.

Recommended Action   From the CRA Administration web page, choose System > Engine, click the Engine Status hyperlink, and then make sure that the CRA Engine is running. In Cisco CallManager, add more CTI route points as associated devices to the JTAPI user.

The message "The Engine is not running" appears even though the CRA Engine is running.

Symptom   You click Associate CTI Port in the JTAPI Call Control Group Configuration web page and the following message appears, even though the CRA Engine is running: The Engine is not running.

Possible Cause   Incorrect information is entered in the JTAPI Provider web page.

Recommended Action   From the CRA Administration web page, choose Tools > JTAPI Configuration and enter the correct information in the JTAPI Provider fields. Then stop and restart the CRA Engine.

The Cisco banner appears in English even when the system language is not English

Symptom   The CRA system language is not English, but the Cisco banner appears in English.

Possible Cause   The banner has not been translated in the CRA version that you are using.

Recommended Action   None.

You can configure more IVR ports than are licensed

Symptom   The CRA system allows you to configure more than the licensed number of IVR ports.

Possible Cause   There is no limit on the number of IVR ports that you can configure. The licensed number of IVR ports limits the number of simultaneous calls that the system will handle.

Recommended Action   None.

The correct number of licensed IVR ports does not appear

Symptom   After you upgrade the number of IVR ports on your CRA system, the correct number of licensed IVR ports does not appear on the JTAPI Call Control Group Configuration web page.

Possible Cause   After upgrading, you did not restart the CRA Engine or the CiscoCRAServletEngine service.

Recommended Action   Restart the CRA Engine and restart the CiscoCRAServletEngine service.

Calls to the CRA system are aborted

Symptom   Some calls to the Cisco CRA system are aborted in an unspecified manner.

Possible Cause   If the RTP start port is not 32256, calls to the Cisco CRA system can be aborted in an unspecified manner.

Recommended Action   Complete the following steps:

1. From the CRA Administration web page, choose System > System Parameters.

2. Enter 32256 in the RTP Start Port field.

3. Click Update.

4. Stop and then restart the CRA Engine.

The message "There has been an error while reading or saving to the database" appears

Symptom   When you update CSQ, skills, or resource information. the following message appears: There has been an error while reading or saving to the database. Please contact Admin.

Possible Cause   The CRA database was not properly installed or configured.

Recommended Action   

1. On the CRA server, verify that the CRA database is installed by opening a command window and typing osql -Usa -Ppassword -ddb_cra, where password is the password for the sa log in to the CRA database.

2. On the CRA server, choose Start > Settings > Control Panel > Administrative Tools > Data Sources (ODBC) and verify that the DSN is properly configured as the db_cra database.

3. If the database is missing, reinstall the CRA software.

A JTAPI trigger is not available in the CTI Route Point Directory Number drop-down list after deleting an application

Symptom   After you delete a Cisco Script Application, the JTAPI trigger that was associated with that application is no longer available in the CTI Route Point Directory Number field in the JTAPI Trigger Configuration web page.

Possible Cause   A JTAPI trigger remains configured after an associated Cisco Script Application is deleted, but the trigger is no longer associated with an application.

Recommended Action   To use the trigger for a different application, choose the desired application in the Application Name field in the JTAPI Trigger Configuration web page.

To make the JTPAI Trigger available in the CTI Route Point Directory Number drop-down list in the JTAPI Trigger Configuration web page, delete the JTAPI trigger from the JTAPI Trigger Configuration web page.

Refreshing subflow script does not update parent scripts

Symptom   Refreshing a subflow script does not update its parent scripts.

Possible Cause   If a script is referenced in other scripts, refreshing a subflow script does not update its parent scripts.

Recommended Action   Manually refresh all parent scripts.

User deleted from CallManager cannot log in to Cisco CRA

Symptom   A user who is deleted from Cisco CallManager cannot log in to Cisco CRA.

Possible Cause   If only one user is configured as Administrator in CallManager and that user is deleted from CallManager, that user will not be able to log in to Cisco CRA.

Recommended Action   You can work around this problem in one of the following ways:

  • Add the user in Cisco CallManager with the same user name and password that the user has in CRA. The user should then be able to log in to CRA.
  • Or follow these steps:

1. Log in to DC Directory and browse to Directory > cisco.com > CCN Apps > configurations > usergroup.

2. Click users?array in the right pane.

3. Right-click users?array and choose Properties.

4. In the Properties window, click the value in the Value area and then click Modify.

5. Click Remove and then click OK.

The user will then be able to log in to CRA using the "Administrator" ID.

CRA Supervisor web page cannot be viewed from CRA Server

Symptom   The Cisco CRA Supervisor web page cannot be viewed from the Cisco CRA Server.

Possible Cause   If the Cisco CRA Supervisor is running on a computer with the language set to Simplified Chinese, you cannot view the Cisco CRA Supervisor web page from the Cisco CRA Server.

Recommended Action   Use a client computer to view the Cisco CRA Supervisor web page.

Database table fields used by wallboard store data in milliseconds

Symptom   Some database table fields used by a wallboard store data in milliseconds instead of in HH:MM:SS.

Possible Cause   The avgTalkDuration, avgWaitDuration, longestTalkDuration, longestWaitDuration, and oldestContact database table fields in the RtCSQsSummary and the RtICDStatistics database tables store date in milliseconds.

Recommended Action   If you want to include information described by these fields on your wallboard, use the following fields, which store the same information but use the HH:MM:SS format:

  • Instead of avgTalkDuration, use convAvgTalkDuration.
  • Instead of avgWaitDuration, use convAvgWaitDuration.
  • Instead of longestTalkDuration, use convLongestTalkDuration.
  • Instead of longestWaitDuration, use convLongestWaitDuration.
  • Instead of oldestContact, use convOldestContact.

Changing host name or IP address causes problems

Symptom   Changing the host name or the IP address causes problems.

Possible Cause   If you change the host name or IP address in the Application Engine Hostname field in the Cisco CRA Administration Engine web page, and do not restart the CRA Engine after making the change, you might experience various problems.

Recommended Action   After changing the host name or IP address, stop and restart the CRA Engine.

Note: Only change the host name or IP address if the host name or IP address of the CRA Server changes.

All user records not displayed on CRA User Management web page

Symptom   All of the expected user records are not displayed on the Cisco CRA User Management web page.

Possible Cause   The default number of records that an Active Directory search returns is 1,000. If there are more than 1,000 records of users in your corporate directory, some of them will not display on the web page.

Recommended Action   You can increase the number of users displayed by increasing the default Active Directory search limit or by adding users to an LDAP entry. To increase the default Active Directory search limit, follow these steps:

1. Log into the Active Directory server using an administrator account and open a command prompt.

2. At the command prompt, type ntdsutil and press Enter.

The command prompt changes to nudsutil:.

3. Type LDAP Policies and press Enter.

The command prompt changes to ldap policy:.

4. Type connections and press Enter.

The command prompt changes to server connections:.

5. Type the following command, where name is the DNS name of the Active Directory server on which you are entering the command:

connect to server %name%

Press Enter to bind to the Active Directory server.

6. Press q and press Enter.

The command prompt changes to ldap policy:.

7. Type Show Values, press Enter, and determine the current setting for MaxPageSize.

8. Type the following command and press Enter to increase the MaxPageSize limit, where limit is the new limit to set:

Set MaxPageSize to limit

9. Type Commit Changes and press Enter.

10. Type Show Values, press Enter, and confirm the change to the MaxPageSize setting.

11. Type q and press Enter until the original Windows command prompt appears.

12. Stop and then restart the CiscoCRSServletEngine service.

To add users to the LDAP entry, complete the following steps:

1. Use a tool such as ADSI Edit to modify the following LDAP entry:

Cisco > CCN Apps > Configurations > usergroup.xxxxxxx/users?array

2. Add users to the property CiscoCCNatConfigInfoCESValue in this LDAP entry. Enter users in the following format, where (1) specifies administrator access and (2) specifies Supervisor access:

userID(1) or userID(2)

CRA Database Problems

The following section describes common problems with the CRA databases.

The CRA Databases are not purged as expected

Symptom   The CRA databases are not automatically purged as expected.

Possible Cause   See the following table for possible causes and recommended actions.

Possible Cause Recommended Action

Cisco Purging Scheduler service is not running

Start the Cisco Purging Scheduler service

Automatic purging is not configured properly

From the CRA Administration web page, choose Tools > Historical Reporting and configure automatic purging

You have changed the system clock on the CRA server

Stop and restart the Cisco Purging Scheduler service

You have altered the size of the CRA databases

If you change the size of the CRA databases, make sure that the CRA database size is equal to the maxsize

E-mail notification of database purging activities are not sent

Symptom   The CRA system does not send e-mail notification of database purging activities.

Possible Cause   The Email subsystem is not configured or e-mail notification is not set up properly in CRA Administration.

Recommended Action   

1. From the CRA Administration web page, choose Subsystems > eMail and make sure that correct information is entered in the Mail Server and eMail Address fields.

2. From the CRA Administration web page, choose Tools > Historical Reporting and click the Purge notification hyperlink.

a. If multiple e-mail addresses are specified in the Send Email Notifications To field, make sure that each address is separated with a semicolon (;), comma (,), or space.

b. Make sure that the Send Email Notifications To field contains no more than 255 characters.

Syslog or SNMP trap notification of database purging activities is not sent

Symptom   The CRA system does not send Syslog notification or SNMP trap notification of purging activities.

Possible Cause   The Cisco AVVID Alarm Service is not running, Syslog is not configured, or SNMP service is not configured.

Recommended Action   

1. Make sure that the Cisco AVVID Alarm Service is running.

2. Make sure that Syslog is properly configured.

3. Make sure that SNMP service is properly configured on the CRA server.

CRA Editor Problems

The following section describes common problems with the CRA Editor.

The file license.properties could not be found and the CRA Editor does not install

Symptom   When installing the CRA Editor, you receive the following message:
The file license.properties could not be found.

Possible Cause   Either you did not download the license.properties file or, when you did, your system renamed the file. Often, the file is renamed to license.properties.txt.

Recommended Action   

1. From the CRA Administration web page, choose Tools > Plug-ins > Cisco CRA Editor. Note the name of the default destination folder shown under Step 2 on the Plug-ins—Download CRA Editor web page.

2. Right-click Download the License.properties file and choose Save Target As from the pop-up menu.

3. In the Save As dialog box, browse to default folder that you noted in Step 1. In the File name field, enter license.properties. Choose All Files from the Save as type drop-down list.

4. In the Save As dialog box, click OK.

5. After the file downloads, verify that the file is named license.properties, not license.properties.txt.

Parameters from user-defined steps made using the CRA SDK are not configurable from CRA Administration

Symptom   You create steps using the CRA SDK and set variables to be configurable as parameters but you cannot access them from the CRA Administration web pages.

Possible Cause   The .jar files and classes are not included in the system classpath.

Recommended Action   Make sure the .jar file or classes are included in the system classpath.

Remote functionality is not available in the CRA Editor

Symptom   The Remote Functionality options, such as Debug, Test DB, and Refresh DB Scheme, are not available in the CRA Editor.

Possible Cause   Some parameters in the ccndir.ini file are incorrect.

Recommended Action   

1. From the CRA Administration web page, choose Tools > Plug-ins and click the Cisco CRA Editor hyperlink.

2. Choose Download the directory access configuration file and follow the on-screen instructions to save the file as specified.

3. Restart the CRA Editor.

Change a string variable to an integer

Symptom   You want to change a string variable to an integer.

Recommended Action   Use the Set step, which supports the conversion of a string to any numerical type.

Accept step error during debug

Symptom   While debugging an application, the following message appears, where n is the task ID: Task: n Accept Step: Trigger is not a Contact Application trigger.

Possible Cause   The debugger encountered the Accept step in the application but there was no call to answer.

Recommended Action   Debug the application as a Reactive Application and make the call before the Reactive Application times out.

The prompt "Please try again" plays

Symptom   A caller hears the prompt "please try again" after speaking a series of digits and then pressing a Dual Tone Multifrequency (DTMF) key other than the configured cancel key.

Possible Cause   For calls that have been allocated a channel from a Nuance ASR Dialog Control Group, the CRA system does not allow DTMF input other than the configured Cancel key to be mixed with voice input.

Recommended Action   Instead of using a terminating key to end voice input of a digit string, a caller should simply speak the string of digits and then wait to indicate the end of the string.

Search option throws an exception

Symptom   The Search option throws an exception.

Possible Cause   When using the Cisco CRA Editor on a CRA server, you click Search and receive a security warning regarding the jvm7.tmp file, and you then click No, the system throws an exception.

Recommended Action   Click Yes in response to the security warning.

Script variable loses value when renamed

Symptom   A script variable loses its value when it is renamed.

Possible Cause   If you change the name of a variable in a script, upload the modified script to the repository, and then refresh the Cisco CRA Engine, the variable loses its value.

Recommended Action   Set the modified variable to its previous value and click Update in the Application Configuration web page.

Error occurs with Reactive Debugging Tool

Symptom   An error occurs when using the Reactive Debugging tool.

Possible Cause   Using the CRA Editor Reactive Debugging tool on a translation routed call can cause an error.

Recommended Action   From the ICM Configuration Manager, choose Tools > List Tools > Network VRU Script List and temporarily increase the value in the Timeout field for the script.

CRA Engine Problems

The following section describes common problems with the CRA Engine.

The CRA Engine does not start

Symptom   The CRA Engine does not start when you click Start Engine in the Engine Status area in the Engine web page.

Possible Cause   The login settings in the Windows Service Manager are incorrect.

Recommended Action   Set the properties window of the CRA Engine service:

1. Open the Service Manager by selecting Start > Programs > Administrative Tools > Services.

2. Right-click the Cisco CRA Engine service.

3. Select Properties.

4. Select the Log On tab.

5. Select This Account and set it to .\Administrator.

6. Enter the correct password in the Password and Confirm Password fields and click OK.

The CRA Engine does not start and an RMI port in use error appears

Symptom   The CRA Engine does not start and an RMI port in use error appears in the CRA trace files.

Possible Cause   Another process is using the port that the CRA Engine is attempting to use.

Recommended Action   Assign another port for the CRA Engine to use. To assign another port, modify the C:\program files\wfavvid\wfengine.properties file and change the entry for com.cisco.wfframework.engine.RMI_PORT from 1099 to another unused port.

Attempting to start the Cisco CRA Engine service causes an
error 1067

Symptom   You attempt to start the Cisco CRA Engine service in the Windows Services window and the following message appears: Could not start the Cisco CRA Engine service on local computer. Error 1067: The process terminated unexpectedly.

Possible Cause   You have not yet run the initial setup for the CRA Server.

Recommended Action   Refer to Cisco Customer Response Applications Administrator Guide or the Application Administration online help for information about properly setting up the CRA Server.

Attempting to start the Cisco CRA Engine service causes an
error 1069

Symptom   You attempt to start the Cisco CRA Engine service in the Windows Services window and the following message appears: Could not start the Cisco CRA Engine service on local computer. Error 1069: The service did not start due to a logon failure.

Possible Cause   When you install Cisco CallManager or CRA, the Windows 2000 administrator password that you enter overwrites the existing Windows 2000 administrator password. Also, if you enter a password that includes spaces, it may not be recorded properly.

Recommended Action   Perform the following steps to change the password for the CRA Engine service:

1. On the CRA Server, choose Start > Settings > Control Panel > Administrative Tools > Services.

2. Double-click Cisco CRA Engine.

3. Choose the Log On tab.

4. Enter and confirm the Windows 2000 administrator password and click Apply. Do not include spaces in the password.

Application subsystem is in partial service

Symptom   The Engine Status area in the Engine web page shows that the Application subsystem is in partial service.

Possible Cause   Some applications are invalid.

Recommended Action   Performs these actions:

1. Refer to the CRA trace files to identify the invalid application.

2. Validate the corresponding script using the CRA Editor.

The CRA Engine is running but calls are not answered

Symptom   The CRA Engine is running but the CRA system does not answer calls.

Possible Cause   The JTAPI subsystem is out of service, the trigger is disabled, the application is disabled, the maximum number of sessions or maximum number of tasks were exceeded, or no CTI ports or media channels are available for the trigger.

Recommended Action   

1. From the CRA Administration web page, choose System > Engine and verify that the JTAPI subsystem is in service.

  • If the JTAPI subsystem is in partial service, see the "Database Subsystem goes into partial service" section.
  • If the JTAPI subsystem out of service, refer to the "CRA Provisioning for Cisco CallManager" section in Chapter 4 in Getting Started with Cisco Customer Response Applications for information about configuration.

2. From the CRA Administration web page, choose Subsystems > JTAPI and click the JTAPI Triggers hyperlink. If False appears in the Enabled column for the trigger, double-click the trigger, click the Enabled Yes radio button, and then click Update.

3. From the CRA Administration web page, choose Applications > Configure Applications. If No appears in the Enabled column for the application, double-click the application, click the Enabled Yes radio button, and then click Update.

4. In the CRA trace files, verify that the calls do not exceed the maximum number of allowed sessions.

5. In the CRA trace files, verify that the calls do not exceed the maximum number of allowed tasks.

6. In the CRA trace files, make sure that there are no messages regarding insufficient free CTI ports or media channels.

An error message plays when calling a CTI route point

Symptom   Callers hear the following message when calling a CTI route point: I'm sorry, we are currently experiencing system problems. The JTAPI subsystem might also be in partial service because the CTI route point cannot load the associated application script.

Possible Cause   The application script associated with the CTI route point did not load correctly.

Recommended Action   

1. Validate the application script in the CRA Editor as follows:

  • From the CRA Administration web page, choose Scripts > Manage Scripts.
  • Click the script and download it from the Repository.
  • Open the script in the CRA Editor.
  • Validate the script and save it.
  • Choose Scripts > Manage Scripts and upload the script to the Repository.

2. When prompted, click Yes to refresh both script and applications.

3. Refer to the CRA trace files to verify that the application script was loaded successfully.

4. If a script has been validated, saved, and uploaded to the repository, and still won't load, verify that any other dependencies are met. For example, if the script references a custom classes, make sure that the class is available to the CRA Engine.

Changes to applications do not register

Symptom   You make changes to an application script but the changes are not apparent to callers.

Possible Cause   The application script was not uploaded to the repository and refreshed.

Recommended Action   After making a change to an application script, perform the following steps:

1. Save the application script.

2. From the CRA Administration web page, choose Scripts > Manage Scripts and upload the application script to the repository.

3. When prompted, click Yes to refresh both script and applications.

Call drops during transfer over gateway

Symptom   When the CRA system receives a call made over a gateway, the CRA system drops the call if the call is transferred.

Possible Cause   The H.323 client does not support the Empty Capability Service and the H.323 port on the Cisco CallManager is not configured to use a Media Termination Point (MTP).

Recommended Action   Update the configuration of the Cisco CallManager H.323 port to require an MTP and reset the H.323 port.

H.323 client DTMF digits not detected

Symptom   When a call originates from an H.323 client, DTMF digits are not collected.

Possible Cause   The H.323 client only produces in-band DTMF signals. Cisco CallManager cannot detect in-band DTMF signals.

Recommended Action   None.

Subsystem status is not available

Symptom   The Engine Status web page does not display the status of the subsystems.

Possible Cause   The host name is not resolving to an IP address.

Recommended Action   Go to System > Engine > Engine configuration. Enter the IP address of the CRA Server instead of its host name in the Application Engine Hostname field.

Subsystem status is incorrect

Symptom   A subsystem is running but the Engine Status web page reports the status of the subsystem incorrectly.

Possible Cause   The CRA Engine is configured incorrectly.

Recommended Action   From the CRA Administration web page, choose System > Engine, click the Engine Configuration hyperlink, and makes sure that the information in the Application Engine Hostname field and the RMI Port Number field is correct.

Redirected call is disconnected

Symptom   A redirected call disconnects or a redirected call does not ring the IP phone to which it was directed.

Possible Cause   Some gateways do not support ringback.

Recommended Action   Use the following table to reconfigure the gateway and protocols so that they will support ringback.

Gateway Protocol

26XX FXO

Media Gateway Control

36XX FXO

Media Gateway Control

VG200 FXO

Media Gateway Control

DT-24+

Skinny

WS-6608-T1—(Cat6K 8-port T1 PRI)

Skinny

WS-6608-E1 (Cat6K 8-port E1 PRI)

Skinny

DE-30+

Skinny

AT-2, 4, 8



The CRA server runs out of disk space

Symptom   Possible symptoms are:

  • An out of memory error occurs on the CRA server.
  • Accessing purging or synchronization pages on the Administration UI returns an error.
  • Running historical reports returns SQL error 5048.

Possible Cause   The CRA database log files, the tempdb database, or the tempdb log files have grown large.

Recommended Action   There are two possible actions you can take:

1. To manually shrink a CRA database log file, open a command window on the CRA server and type the following commands:

a. osql -Usa -Ppassword -ddb_cra, where password is the password for the sa log in to the CRA database.

b. USE database_name, where database_name is db_cra or db_cra_ccdr, depending on the CRA database that you want to shrink.

c. GO

d. DBCC SHRINKFILE (database_name_log.mdf), where database_name db_cra or db_cra_ccdr, depending on the CRA database that you want to shrink.

e. GO

2. Alternatively, you can shrink the log files by running the batch file runTruncateHistDBLogs.bat, installed under the wfavvid directory. Depending on the arguments, it shrinks the log files of db_cra, db_cra_ccdr, or tempdb.

Examples:

  • Truncate the log files for both db_cra and db_cra_ccdr to 10MB:
runTruncatedHistDBLogs "sa" "sa_password" "db_cra_all" 10
  • Truncate the tempdb transaction log:

runTruncateHistDBLogs "sa" "sa_password" "tempdb"

Additional information can be found about truncating logs in the Managing Historical Reporting Databases section of the Cisco CRA Administrator Guide.


Note    This troubleshooting tip also applies to the CRA Database Problems section and the CRA Historical Reporting Problems section.

CRA Server runs at 100% capacity

Symptom   The CRA server CPU works at or close to 100 percent capacity. DTMF digits are delayed.

Possible Cause   One of the following configurations might be causing this problem:

  • Trace settings include debugging.
  • Cisco CallManager polling is enabled. (Polling is enabled by default, but it can consume server resources.)
  • You are running many applications on a smaller system simultaneously. For example, you are running Cisco CallManager, Cisco CRA Server, and the LDAP database all on a low-end MCS.

Recommended Action   

1. Turn off debugging as a trace level option. Debugging consumes substantial server resources and should only be used when you are actively debugging Cisco CRA.

2. Turn off Cisco CallManager polling. Polling enables JTAPI (and therefore the telephony applications that use JTAPI, such as CRA) to detect the addition of devices to an application or user's controlled list. For example, polling can detect when an agent is added to a call center or a CTI port is added to the CRA Engine. If you do turn off polling, Cisco CallManager does not update new devices automatically. For example, you must restart the CRA Server after adding a new CTI port or route point to Cisco CallManager.

3. If you are using a smaller system with many applications running at the same time, install the different telephony applications on separate servers or use an MCS-7835.

Database Subsystem goes into partial service

Symptom   The Database subsystem is in partial service when the Cisco CRA system is configured to use a Sybase database.

Possible Cause   If the Sybase datasource name that you enter in the Cisco CRA Administration Enterprise Database Subsystem Configuration web page does not match exactly the datasource name in the Windows ODBC DSN configuration window, the database connection will fail and the database will go into partial service.

Recommended Action   Be sure the Sybase datasource name on the Administration Enterprise Database Subsystem Configuration web page matches the Windows ODBC datasource name.

JTAPI subsystem is in partial service

Symptom   The Engine Status area in the Engine web page shows that the JTAPI subsystem is in partial service.

Possible Cause   The JTAPI client was not set up properly. At least one, but not all, of the CTI ports, route points, or dialog channels (CMT or Nuance) could not initialize.

Recommended Action   

1. Refer to the CRA trace files to determine what did not initialize.

2. Verify that all CTI ports and CTI route points are associated with the JTAPI user in Cisco CallManager.

3. Verify that the Cisco CallManager and JTAPI configuration IP addresses match.

4. Verify that the Cisco CallManager JTAPI user has control of all the CTI ports and CTI route points.

5. Verify that the LDAP directory is running on the computer specified in the Directory Host Name field in the Directory Setup web page Configuration Setup area.

6. Verify that the application file was uploaded to the repository using the Repository Manager.

Unable to connect to JTAPI provider

Symptom   The JTAPI provider is unavailable.

Possible Cause   See the following table for possible causes and recommended actions.

Possible Cause Recommended Action

Cisco CallManager is not running.

Troubleshoot the Cisco CallManager (refer to Cisco CallManager Administration Guide).

LDAP Directory Service is not running.

Troubleshoot the LDAP directory (refer to Cisco CallManager Administration Guide).

Incorrect JTAPI client version is installed on the CRA server.

Check the JTAPI version on the CRA server by selecting Start > Programs > Cisco JTAPI > Readme.

JTAPI user is not configured correctly.

From the CRA Administration web page, choose Subsystems > JTAPI, click the JTAPI Provider hyperlink, and then:

  • Verify that information in the User ID field matches the name of a valid user in Cisco CallManager.
  • Verify that information in the Password field is correct.

JTAPI client cannot communicate with the Cisco CallManager.

Verify that Cisco CallManager is running.

Configure Cisco CallManager using the IP address instead of the DNS1 name.

1DNS = Domain Name System

The Simple Recognition step takes the unsuccessful branch

Symptom   The Simple Recognition step in a script takes the unsuccessful branch even when the word spoken or DTMF key pressed is defined in the grammar.

Possible Cause   The Simple Recognition step is configured with a set of tags and output points. Most likely, the tag names defined in the step do not exactly match the tag names defined in the grammar used by the step. Such a mismatch can occur when a tag name is defined in the grammar but not configured in the step or when tag names use inconsistent spelling or case.

Recommended Action   

1. Verify that all tag names defined in the grammar are configured in the Simple Recognition step with matching spelling and case.

2. If the grammar contains tag names that are not configured in the Simple Recognition step, either configure the same tag name in the Simple Recognition step or remove the tag names from the grammar.

Calling party and CRA do not have common codec

Symptom   The calling party hears a fast busy signal when calling into a CRA application. The CRS log shows CTIERR_REDIRECT_CALL_PROTOCOL_ERROR.

Possible Cause   The calling device's codec is possibly incompatible with CRA.

Recommended Action   Use the transcoding service on Cisco CallManager or ensure that the calling device is using G.711 or G729, depending on what is configured on the CRA server.

Prompts with incorrect codec being played out

Symptom   The calling party does not hear prompts.

Possible Cause   The prompt being played does not match the system's codec.

Recommended Action   Use the correct version of the prompt.

Prompt Exception in CRA Engine log file

Symptom   A prompt exception containing the text open port failed appears in the Cisco CRA Engine log file.

Possible Cause   This error is generally caused when the Cisco CRA Engine is incorrectly shut down; for example, from the Windows Task Manager while there are RTP ports in use.

Recommended Action   If this prompt exception appears, reboot your Cisco CRA Server. To prevent this problem, stop the Cisco CRA Engine. Choose System > Engine from the CRA Administration menu bar and then click Stop Engine. Alternatively you can use the Windows services console to stop the Cisco CRA Engine.

CRA Engine does not start

Symptom   The Cisco CRA Engine does not start and the trace file reads Port already in use.

Possible Cause   If another process is using the Cisco CRA Engine default port 1099, the CRA Engine will not start.

Recommended Action   From the CRA Administration web page, complete the following steps:

1. Choose System > Engine > Engine Configuration.

2. Enter a different port in the RMI Port Number field.

3. Stop and then restart the CRA Engine.

Application subsystem in partial service and application running for an unexpectedly long time

Symptom   The Application subsystem is in partial service and the Application Tasks real-time report shows an application running for an unexpectedly long time.

Possible Cause   If an application does not receive a disconnect signal after a call, and the application does not have an error handling mechanism to detect that the call has ended, the Application subsystem might go into partial service. In addition, Application Tasks real-time report might show an application running for an unexpectedly long time.

Recommended Action   Make sure that the application script includes error handling that prevents infinite retries if a call is no longer present.

CRA Real-Time Reporting Problems

The following sections describe common problems with CRA Real-Time Reporting.

Attempting to run a real-time report causes an error

Symptom   The following message appears when you try to run any real-time report from the CRA Administration web page: Unable to connect to the server.

Possible Cause   The proxy server setting on the Browser impedes underlying RMI communication, or the RTR subsystem or the DB subsystem is not running.

Recommended Action   

1. From Internet Explorer, choose Tools > Internet Options > Connections > LAN settings and then uncheck the Use a proxy server check box.

2. Make sure that the RTR and DB subsystems are running.

Invalid page fault message while Real-time Reporting is running

Symptom   Intermittently, while using Microsoft Internet Explorer, a message box stating DDHelp caused an invalid page fault in module <unknown> at ... appears after RTR has been running for awhile. Subsequently, the machine pauses indefinitely.

Possible Cause   The DirectDraw function in the Java Runtime Environment (JRE) causes interruption to the video driver, as related in Bug4713003 reported by Sun Microsystems.

Recommended Action   Upgrade Internet Explorer to version 5.5SP2. If the problem persists, go to Start > Settings > Control Panel open the Java Plug-in panel and add -Dsun.java2d.noddraw=true flag as a parameter.

CRA Administration pauses indefinitely after Real-time Reporting closes

Symptom   After the real-time reporting applet is closed, the CRA Applications Administration stops taking input and pauses indefinitely. The machine must be rebooted.

Possible Cause   The JRE used by real-time reporting causes interruption of the video driver.

Recommended Action   The solution is to close Application Administration before exiting real-time reporting. Do not use JRE 1.4.1 on this particular machine because video driver-related issues have been reported by Sun Microsystems.

After installing a version of JRE, the user receives a message from real-time reporting saying to install a version of JRE

Symptom   Upon opening a Real-time Reporting applet, a message box with information about the version of JRE that is running and the required version of JRE appears. The user installs the required JRE version yet still gets this message when opening a real-time report applet.

Possible Cause   The message appears when another version of JRE is installed on the machine as default and is invoked at run time.

Recommended Action   Uninstall that version of JRE from the Control Panel and reinstall the required JRE version.

Co-locating Real-time Reporting and CallManager on a server and running them at the same time causes JRE incompatibility

Symptom   The user is notified of JRE incompatibility after opening a Real-time Reporting applet while CallManager is running.

Possible Cause   Currently, RTR must use JRE 1.3.1 while CallManager must use JRE 1.4.0, and the default JRE is 1.4.0.

Recommended Action   Real-time Reporting should not run on a co-located server with CallManager.

CRA Historical Reporting Problems

The following sections describe common problems with CRA Historical Reporting.

Error when choosing an option from the Historical Reporting web page

Symptom   An error message appears when you choose an option from the Historical Reporting web page.

Possible Cause   The CiscoCRAServletEngine service, the DC Directory Server service (or another LDAP service), or the MSSQLServer service may not be running, or there may be a problem connecting to the CRA database.

Recommended Action   

1. Make sure that the CiscoCRAServletEngine service is running on the CRA server.

2. Make sure that the DC Directory Server service is running on the CRA server. Or, if you have specified another LDAP as the active directory, make sure that that service is running.

3. Make sure that the MSSQLServer service is running on the CRA server.

4. On the CRA server, choose Start > Settings > Control Panel > Administrative Tools > Data Sources (ODBC) and make sure that ODBC System DSNs dsn_cra and DSN_SCH_DB are properly configured.

The SQL Command Failed dialog box appears when you try to generate a historical report

Symptom   The CRA Historical reports client computer displays the SQL Command Failed dialog box when you try to generate a historical report. This dialog box specifies an error number, n.

Possible Cause   This error can occur in a variety of situations.

Recommended Action   On the computer on which you received the SQL Command Failed dialog box, open the most recent Historical Reports log file. Search for the error number. The cause of the error will appear near the error number. You can use this information to resolve the problem.

Some information appears in English on a German system

Symptom   On a CRA Historical reports client computer with an English version of Windows for which the system locale has been set to German, the Search dialog box and the Preview tab on the Report Viewer still appear in English.

Possible Cause   Limitation of the Report Viewer.

Recommended Action   Install a German version of the operating system.

The Print dialog box displays an incorrect printer

Symptom   When printing a historical report on a CRA Historical reports client computer running Windows98, the Print dialog box displays an incorrect printer.

Possible Cause   You have changed the default printer while CRA Historical Reports is running and have not exited and restarted CRA Historical Reports.

Recommended Action   Exit and restart CRA Historical Reports.

The Historical Reporting client computer cannot connect to the CRA server

Symptom   The CRA Historical Reporting client computer is unable to connect to the CRA server. The Historical Reports Client log file shows the following message: Not associated with a trusted connection.

Possible Cause   SQL server is not being accessed with the proper authentication.

Recommended Action   Perform the following steps on the CRA Historical Reporting client computer:

1. Choose Start > Programs > Microsoft SQL Server n > Enterprise Manager, where n is a version number.

2. Double-click Microsoft SQL Servers.

3. Double-click SQL Server Group.

4. From the SQL Server group, right-click the name of the server on which the CRA databases reside.

5. Choose Properties.

6. Choose the Security tab.

7. Click the SQL Server and Windows NT radio button.

8. Click OK.

A Database Connection Error 5051 error appears

Symptom   When you try to log into the CRA Historical Reporting client software on the client computer, a Database Connection Error 5051 is displayed.

Possible Cause   Network connectivity is down or the client connection setting is incorrect.

Recommended Action   

1. From the Windows Control Panel on the CRA Historical reports client computer, choose Data Sources (ODBC).

2. Create a DSN to the db_cra database on the CRA server that the client computer is attempting to connect to. Choose SQL authentication (you will need your database login information) while creating the DSN.

3. If the DSN cannot be created, verify that network connectivity exists between the CRA Historical Reports client computer and the CRA server.

If you are able to connect successfully using the DSN, update the hrcConfig.ini file on the client computer with the appropriate network library.

Export file name does not appear in Export dialog box

Symptom   A default export file name does not appear in the Export dialog box.

Possible Cause   If you click the Export Report tool in the Report Viewer on a Cisco CRA Historical Reports client computer on which the language is set to Simplified Chinese, the Export dialog box will not contain a default export file name.

Recommended Action   The name of the report is shown on the header of the report in the report viewer. In the Export dialog box, specify a name for the exported report using the report name; for example, you can name your PDF report <ReportName>_<startdatetime>_<enddatetime>.pdf.

User Configuration page has empty installed products packages

Symptom   With a Historical Reports Database Server installation, the User Configuration page has empty installed products packages.

Possible Cause   Synchronization between CRA and the Historical Reports Database Server was not completed.

Recommended Action   After configuring CRA with the Historical Reports Database Server, choose Tools > Historical Reporting > Synch now page and perform a synchronization.

Cannot point to local applications from the Database Server Configuration page

Symptom   With a Historical Reports Database Server installed, the user cannot point to the CRA local database from the Database Server Configuration page.

Possible Cause   Once a Historical Reports Database Server (remote database) is configured from Cisco CRA, the user cannot point back to CRA.

Recommended Action   None. This is working as designed. To go back to a single-box solution (and lose the Historical Reports Database Server configuration in CRA), uninstall Cisco CRA and reinstall it (and accept the dropping of the CRA databases when prompted).

Cannot reuse the previous Historical Reports Database Server after uninstalling Cisco CRA, reinstalling it, and dropping the CRA database

Symptom   With a Historical Reports Database Server installed, the user cannot reuse the previous relational database (RDB) after uninstalling Cisco CRA, reinstalling it, and dropping the CRA database.

Possible Cause   Cisco CRA and the RDB are tightly coupled in a one-to-one relationship. When you uninstall CRA, reinstall it, and accept the dropping of the Historical Reports Database Server when prompted during installation, you lose the Historical Reports Database Server configuration in CRA. CRA becomes a single-box installation using the local database and therefore the Historical Reports Database Server loses its CRA.

Recommended Action   There are two alternatives:

  • If you do not need the previous data, uninstall CRA, reinstall it, and drop the CRA database. This is done on the Historical Reports Database Server.
  • If you want to maintain the previous data and configuration, restore CRA and RDB using the backup and restore utility with a backup of the servers that was taken when both machines were configured together.

Attempt to log in to the CRA Server from the Historical Reporting client fails and an error message is returned

Symptom   The CRA servlet service or web server is not reachable from the client machine.

Possible Cause   An authentication request timeout has occurred. The client log indicates Request timed out error.

Recommended Action   Complete the following steps:

1. On the CRA server, check to be sure the web server and servlet service are running. On the client, check your browser Internet options for the connection setting.

2. Make sure you are able to connect to the CRA Administration web page from the client machine. Refresh the page to make sure it is not cached.

3. If the error persists after doing the above, modify your client hrcConfig.ini file by updating AuthReqTimeOut to a larger value than the current one (default is 15 seconds).

4. Restart your client and attempt to log in again.

CRA Installation Problems

The following section describes common problems that can occur during CRA installation.

The installation program reports that there is not enough memory

Symptom   When installing CRA, you see a message indicating that there is not enough memory to continue.

Possible Cause   The server on which you are installing CRA does not have enough disk space for the installation.

Recommended Action   Make sure that your server has the required amount of free disk space. Disk space requirements for the various CRA application packages are shown in the following table.

Application Disk space requirement

Extended Services

667 MB

Customer Response Applications

1,971 MB + 358 MB (if SQL Server is not present) + 20 MB per language supported maximum + 4 GB for a standalone system or 512 MB for a coresident system

ASR/TTS

2,004 MB for ASR + 977 MB for TTS install

+ if [en_US] language is selected for ASR, add 284 MB

+ if [en_CA] language is selected for ASR, add 3 MB

+ if [en_GB] language is selected for ASR, add 90 MB

+ if [fr_FR] language is selected for ASR, add 86 MB

+ if [fr_CA] language is selected for ASR, add 67 MB

+ if [de_DE] language is selected for ASR, add 127 MB

+ if [es_CO] language is selected for ASR, add 21 MB

+ if [es_MX] language is selected for ASR, add 21 MB

+ if [es_ES] language is selected for ASR, add 11 MB

+ if [it_IT] language is selected for ASR, add 58 MB

+ if [ja_JA] language is selected for ASR, add 230 MB

+ if [en_US] language is selected for TTS, add 258 MB

+ if [en_GB] language is selected for TTS, add 7 MB

+ if [fr_FR] language is selected for TTS, add 3 MB

+ if [de_DE] language is selected for TTS, add 11 MB

+ if [es_ES] language is selected for TTS, add 3 MB

+ if [it_IT] language is selected for TTS, add 14 MB

LDAP Directory Problems

The following section describes common problems with LDAP directory configuration for CRA.

LDAP configuration error message appears

Symptom   During the CRA directory setup process, the Directory Setup web page displays the error LDAP Configuration - Error.

Possible Cause   The IP address or host name of the LDAP directory is not configured properly.

Recommended Action   

1. From the CRA Administration web page, choose System > Configuration and Repository. Verify that the information in the fields in the Directory Setup web page are correct. If you are using a directory other than the DC Directory (for example, Netscape or Active Directory), verify that directory is specified correctly.

2. Make sure that the DC Directory Server service running. Or, if you have specified another LDAP as the active directory, make sure that service is running.

Nuance ASR Problems

The following section describes common problems with Nuance ASR (automatic speech recognition).

Names are not recognized

Symptom   For calls that have been allocated a channel from a Nuance ASR Dialog Control Group, names are not consistently recognized by the Name to User step.

Possible Cause   The Refresh Task was not run after a new CRA installation or after you added or changed an existing name. If names are not recognized after you run the Refresh Task, you have supplied one or more merge dictionaries in the C:\Cisco Dictionaries folder, which requires you to stop and restart the CRA Engine after adding or changing an existing name.

Recommended Action   Run the Nuance ASR Refresh Task. (It is recommended that you run the Nuance ASR Refresh Task at off-peak times.) If names are still not recognized, stop and restart the CRA Engine.

Alternate pronunciations and nicknames are not recognized

Symptom   For calls that have been allocated a channel from a Nuance ASR Dialog Control Group, some alternate pronunciations and nicknames are consistently not recognized by the Name to User step.

Possible Cause   The Refresh Task was not run after the specific names were added or modified.

Recommended Action   Run the Nuance ASR Refresh Task.

Speech recognition consistently fails

Symptom   Automatic speech recognition fails consistently and the following message appears in the CRA trace file: Refresh Task failed to copy packages over to nuance recserver.

Possible Cause   Network connectivity has been lost or the \wfnuance network share is not properly established.

Recommended Action   

1. Verify that there is network connectivity between the CRA server and the speech server.

2. Verify that the windows login credentials for the network share have not changed.

3. If the speech server is on DHCP and a host name is used in the Nuance ASR speech server configuration, verify the IP address of the speech server.

4. From the CRA Administration web page, choose Subsystems > Nuance ASR.

a. In the Nuance ASR Configuration area, verify that the correct IP address or host name is entered for speech server.

b. In the Nuance ASR Configuration area, look at the Network share status displayed in the window. If it shows Network share is not available, click Update to reestablish the connection. In the Network Share Log-on Configuration web page that appears, enter the login credentials, and then click Go.

A call using the Nuance ASR Dialog Channel goes to the default script instead of to the configured script

Symptom   A call to a CTI route point that uses ASR dialog channels goes to the default script instead of to the configured script. Initial prompts may be played correctly if they are played by the Play Prompt step. The CRA trace files will include the following exception: NUANCE_SERVER_NOT_ACCESSIBLE.

Possible Cause   When a call reaches a trigger that uses a Nuance ASR dialog channel and the trigger uses a language that was not installed for Nuance ASR, subsequent steps with speech recognition will fail. If a step fails, the call goes to the default script.

Recommended Action   Make sure that the language that the trigger uses is installed for ASR. (For a trigger that does not use a Nuance ASR dialog Channel, you can use any language that was installed for CRA.)

The ASR subsystem is out of service

Symptom   The Engine Status area in the Engine web page shows that the Nuance ASR subsystem is out of service.

Possible Cause   The Speech Server is not configured from the Nuance ASR Configuration web page.

Recommended Action   

1. Verify that the Nuance Speech Server is installed on the CRA Server and on the Speech Server.

2. Verify that Speech Servers are configured through the Nuance ASR Configuration web page. If no Speech Server is configured, configure the Speech Server(s) and then stop and restart the CRA Engine.

Nuance Subsystem takes a long time to go into service

Symptom   The Nuance subsystem takes a long time to go into service.

Possible Cause   The system is working as expected. The following factors can cause the Nuance subsystem to take a long time to go into service.

  • Multiple languages installed for ASR.
  • A large Cisco CallManager user directory.

Recommended Action   You can configure the JTAPI triggers with a secondary Cisco Media Termination (CMT) media group (non-ASR) to allow calls to be received as CMT only until the ASR channels go into service.

Nuance TTS Problems

The following section describes common problems with Nuance TTS (text to speech).

Text playback using TTS causes unexpected results

Symptom   Text playback using TTS causes unexpected results.

Possible Cause   The current TTS technology requires that you be aware of how a prompt text is processed. If you do not implement TTS correctly, you might experience unexpected results.

Recommended Action   When implementing TTS, make sure to follow the guidelines in the Nuance Vocalizer Developer's Guide, which you can find in the following directory on your speech server:

C:\Program Files\Nuance\Vocalizer\doc

A TTS Prompt will not play

Symptom   Callers do not hear a TTS Prompt when a TTS prompt is expected.

Possible Cause   See the following table for possible causes and recommended actions.

Possible Cause Recommended Action

The language specified in the Override Language field in the TTS Prompt step is not available.

From the CRA Administration web page, choose Subsystems > Nuance TTS and check whether the TTS server for the Override Language is configured. If it is not, configure a new TTS server with the desired language and then stop and restart the CRA Engine.

The text referenced in the Text Input expression is larger than 20 KB.

Text larger than 20 KB is not allowed. Make sure that you are using less than 20 KB of text.

There are not enough TTS port licenses. This situation can occur if the number of IVR ports exceeds the number of TTS port licenses and the number of active IVR calls making TTS requests exceeds the available TTS ports.

Obtain a license for additional TTS ports.



Response time can be slow under heavy loads

Symptom   Occasionally TTS prompts respond slowly.

Possible Cause   The problem generally occurs when the Nuance speech server, the Cisco CRA server, the Gateway, or the network is operating under heavy load.

Recommended Action   Identify and correct any network problems. In addition, if you are using VoiceXML, you can work around this problem as follows:

  • Reduce the complexity of the grammars used in the VoiceXML application.
  • Remove the reference to Document Type Definition (DTD) in the VoiceXML application.
  • Use local .wav, grammar, and VoiceXML files instead of files on another server.

A TTS prompt is not recognizable

Symptom   A TTS Prompt cannot be recognized by callers.

Possible Cause   The language of the text does not match the Override Language in the TTS Prompt step. For example, the text is in English, but the Override Language is Spanish.

Recommended Action   In the appropriate TTS Prompt step, make sure that the Text Input matches the Override Language Selection.

The TTS subsystem is out of service

Symptom   The Engine Status area in the Engine web page shows that the Nuance TTS Subsystem is out of service.

Possible Cause   No Nuance TTS server is configured or the Nuance TTS server component is not installed.

Recommended Action   

1. From the CRA Administration web page, choose Subsystems > Nuance TTS and verify that TTS servers are configured properly. If no TTS server is configured, configure the TTS server(s) and then stop and restart the CRA Engine.

2. Verify that the Nuance TTS server component is installed on the CRA server and on the TTS server.

Long TTS prompts consume significant memory on CRA Server

Symptom   Long TTS prompts consume significant memory on the Cisco CRA Server.

Possible Cause   Long text files will increase memory usage on the CRA server. For example, a 5 KB text file requires a 2.5 MB audio file. On a system with 40 TTS ports, the scenario can consume approximately 120 MB of memory

Recommended Action   If you are using long text files, make sure to provision your system appropriately.

Language specified for TTS request differs from language used by TTS

Symptom   The language specified for a TTS request differs from the language used by TTS.

Possible Cause   TTS uses a fallback mechanism if a specified language is not available. This mechanism can cause the language specified to differ from the language used. Here are the possible fallbacks:

  • If the specified language is US English or Canadian English and if US English is available, US English will be used as the fallback language. If US English is not available, and only UK English is available, UK English will be used as the fallback language.
  • If the specified language is UK English or Australian English and if UK English is available, UK English will be used as the fallback language. If UK English is not available, and only US English is available, US English will be used as the fallback language.
  • If the specified language is any other version of English and if any one English is available, it will be used as the fallback language. If both US English and UK English are available, US English will be used as the fallback language.
  • If the desired language is any other language, the European continental version will be used as the fallback language, if it is available. If it is not available, an error will occur.

Recommended Action   None. TTS is working as designed.

Serviceability Problems

The following sections describe common problems with Serviceability.

SNMP-based network management tools cannot monitor CRA components

Symptom   You are unable to monitor CRA components with SNMP-based network management tools, such as CiscoWorks 2000.

Possible Cause   The SNMP subagents that monitor CRA components are not loaded or configured properly, or the SNMP service is not running.

Recommended Action   

1. Make sure that each subagent has a key under the following SNMP service registry:

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\SNMP\Parameters\ExtensionAgents

2. Make sure that the subagent DLL exists under the directory specified in the registry. For example, the subagent SnmpSysAppAgent should have a registry under HKEY_LOCAL_MACHINE\SOFTWARE\
Cisco Systems, Inc.\SnmpSysAppAgent\CurrentVersion and a path name that points to the location of the SnmpSysApp subagent SnmpSysAppImpl.dll.

3. Make sure that the SNMP service is running on the CRA server. If it is not, start the SNMP service.

SNMP traps do not arrive at the trap receiver

Symptom   The network management system (NMS) does not receive SNMP trap messages.

Possible Cause   There is a misconfiguration in the SNMP service properties.

Recommended Action   Perform the following actions:

1. On the CRA server, choose Start > Settings > Control Panel > Administrative Tools > Services.

2. Double-click SNMP Services in the Name field.

3. In the SNMP Service Properties window, click the Traps tab and

a. Make sure that you use the correct case and name for the community name.

b. Make sure that the IP address or host name of the trap destination is correct.

4. In the SNMP Service Properties window, click the Security tab and:

a. Make sure that at least one community name is defined and that its rights are READ ONLY or READ WRITE. Note that "public" community name is not recommended as it might lead to security holes in your system. For more information about SNMP security, refer to your Microsoft Windows documentation.

b. Make sure that you use the correct case for the community name.

5. On the CRA server, choose Start > Settings > Control Panel > Administrative Tools > Services and verify that the SNMP service and Cisco AVVID Alarm Service are running.

6. On the CRA server, verify that the Cisco AVVID Alarm Service is receiving messages:

a. Open the AlarmService.ini file and verify that the correct port number is listed. The default port is 1444.

b. Open the Alarm Service error logs, AlarmTracen.log, where n indicates the log number. (For example, AlarmTrace11.log is the eleventh log.) By default, the error logs are in the following folder:
C:\Program Files\Cisco\AlarmService\ AlarmServiceLog.

7. On the trap receiver (the NMS system):

a. Make sure that the same community names are defined in the SNMP Service properties as you used in step 4.

b. Make sure traps are enabled by verifying that the cvaNotificationEnable table is set to true in the ciscoVoiceAppsMIB.

Syslog messages not received by receiver

Symptom   Your Network Management System (NMS) does not receive Syslog messages.

Possible Cause   There is a misconfiguration in the service properties. For more information, refer to Cisco Customer Response Applications Serviceability Guide.

Recommended Action   Perform the following actions on the CRA server:

1. Choose Start > Settings > Control Panel > Administrative Tools > Services and verify that the Cisco AVVID alarm Service is running.

2. Verify that the Cisco AVVID Alarm Service is receiving messages:

a. Open the AlarmService.ini file and verify that the correct port number is listed. The default port is 1444.

b. Open the Alarm Service error logs, AlarmTracen.log, where n indicates the log number. (For example, AlarmTrace11.log is the eleventh log.)

3. Verify the settings on the Syslog receiver (the NMS system).

The Alarm Service does not start

Symptom   The Cisco AVVID Alarm service does not start.

Possible Cause   AlarmNTService.exe does not exist, or the alarm service is not registered.

Possible Cause   

1. Verify that AlarmNTService.exe exists in the C:\program files\cisco\AlarmService directory. If it does not exist, reinstall the CRA system.

2. Make sure that the Cisco AVVID Alarm Service is running. If this service is not running, register this service by typing the following command in a command window:

C:\Program Files\Cisco\AlarmService\AlarmNTService -Service

Serviceability does not uninstall completely

Symptom   Uninstalling Serviceability does not completely remove all serviceability components.

Possible Cause   On a server on which CRA and Cisco CallManager are both installed, some Serviceability files are shared by each of these Cisco applications. If you uninstall Serviceability, you will be prompted for permission to delete the shared files. If you do not delete these files, they will remain on the system.

Recommended Action   None.

CRA Internationalization Problems

The following sections describe common problems related to internationalization.

Results not as expected for first name and last name in Chinese, Japanese, and Korean

Symptom   First name and last name information does not produce the expected results for Chinese, Japanese, and Korean.

Possible Cause   In the Cisco CRA User Options Alternate Pronunciations web page, entering information in the First Name and Last Name fields does not produce the expected results for Chinese, Japanese, and Korean.

Recommended Action   To work around this problem, enter the given name in the First Name field, and enter the family name in the Last Name field.

Language specified is not accepted or played

Symptom   A user specifies a language but the software does not accept it or play the language entered by the user.

Possible Cause   If you are trying to use a prompt or recognition grammars, the problem might be that the language was not successfully installed. Check the language and associated region in the ISO 639 and ISO 3166 standards. If the problem occurs with a Voice Browser, note that the VXML standard specifies that languages be entered in the form xml:lang="<ISO 639>-<ISO 3166>" (example: en-US, fr-CA). Note that the VXML delimiter is a hyphen (-); for the CRS Editor, Application Administration, locale customization, installation, and prompts, the separator is an underscore (_).

Recommended Action   Check the language installation. Check the following ISO references for the correct language and region representation:

  • ISO 639 can be viewed at

http://www.ics.uci.edu/pub/ietf/http/related/iso639.txt

  • ISO 3166 can be viewed at

http://www.chemie.fu-berlin.de/diverse/doc/ISO_3166.html

Check to be sure you have used the hyphen (-) delimiter for VXML and have used the underscore (_) delimiter everywhere else.