CCXML is supported on VP 4.0 and later.

No. AVP 3.0 does not support the call recording functionality.

Yes, DD 4.0 does support outbound calling through CCXML. The FindADoctorCC CCXML sample application explains how to initiate an outbound call. This sample application comes bundled with Dialog Designer 4.0 installer. The DD 4.0 installer can be downloaded from the http://www.avaya.com/devconnect . The sample application is available in the directory < InstallPath >\Sample Applications\files\DialogDesigner_3.1_SampleApplications\ workspace\FindADoctorCC.

Outbound calling using outcall web service needs Avaya Special Application feature, SA8874, enabled on the Avaya Communication Manager. Use Avaya Communication Manager 3.1 build 369 or later with the Avaya Special Application SA8874 feature. Without this feature, supervised transfers do not have access to call progress information and will behave as blind transfers.

No. The full VoiceXML 2.1 conformance (such as handling all of the possible failure conditions properly) for < data > tag is not supported in Avaya Voice Portal versions prior to 4.0.

Voice Portal 4.x supports only VXML 2.0 compliant applications. Applications designed using Conversant-specific scripting languages (TAS, IRAPI) cannot be migrated to Voice Portal 4.x.

The current version of Voice Portal (version 4.x) only supports IBM and Nuance Speech Recognizers. Additional vendors may be supported in the future.

A Dialog Designer (and Voice Portal) license is required in order to run Dialog Designer applications deployed on a Voice Portal system.

Voice Portal 4.1 (Media Processing Platform) does not support mutual authentication with any application server (for Example: IBM WebSphere Application Server). The certificate provided by an application is used by Media Processing Platform (MPP) to trust an application server. Voice Portal does not provide an authentication certificate for an application server to trust MPP.

For both Avaya Interactive Response (IR) and Avaya Voice Portal, the browser has a default timeout set to 15 seconds.

The procedure to change the default timeout for the Avaya Voice Browser on the Avaya Voice portal (AVP) is shown by the following screenshots.

Save and apply the changes.

Note: Voice Portal does not indicate a separate service pack version number. On installing a service pack, the version number for the software is updated (For example: 4.1.0.1.2707).

Version numbers for Voice Portal components such as Voice Portal Management System (VPMS) and Media Processing Platform (MPP) are listed as shown below.

Log on to Voice Portal Management System (VPMS), as shown below:

The version number for Voice Portal Management System (VPMS) can be found on the VPMS home page or under System Maintenance -> System Monitor -> VPMS details as shown below.

The version number for Media Processing Platform (MPP) can be found under System Maintenance -> System Monitor -> MPP as shown below:

To fix the above issue, the following patches should be installed:

• Voice Portal Service Pack 4.1.0.1-2701.
• MPP hot-fixes wi00089856 and wi00075203

WebLM supports the following combinations of Tomcat and JRE:

• Java - JRE 1.4.2_03 Servlet container - Tomcat 5.0.28
• Java - JRE 1.5.0_02 Servlet container - Tomcat 5.5.9

The session variable available in VXML application has an attribute named channel which returns the port number on which the call resides.

The value in the channel attribute can be accessed by the following methods:

1. For a Dialog Designer VXML (speech) application, the channel field (attribute) can be accessed under session variable through the project. variables form as shown below.

   

2. This attribute value can be fetched at runtime by overriding the servlet implementation method as shown by the sample code snippet below:

   public void servletImplementation(SCESession mySession)
                           {
   	                        // TODO Auto-generated method stub
   	                        super.servletImplementation(mySession);
   	                        IVariableField field = mySession.getVariableField(IProjectVariables.SESSION, IProjectVariables.SESSION_FIELD_CHANNEL);
   	                        String channel = field.getStringValue();
   	                        System.out.println("channel is.."+ channel);
                           }
                       

The getReturnCode() method in LaunchVXML error object can be used to access the return codes in the event of a failure. Place the LaunchVXML code in the try block and in the catch block, catch the exception as LaunchVXMLFault and access the return code by using method getReturnCode() as shown in the code snippet below.

Refer to Voice Portal documentation for a list of return values and associated error messages.

Note: The catch block is executed only when LaunchVXML fails and hence these return codes are obtained only in case of a failure.

try
                {
	                AppIntfWS_ServiceLocator locator = new appIntfWS_ServiceLocator(config);
	                AppIntfWS_PortType appintfws = locator.getAppIntfWS(new URL( "http://" + vpmsHost + "/axis/services/AppIntfWS"));
	                LaunchVXMLResponse response = appintfws.launchVXML(request);
                }

                catch (LaunchVXMLFault e) 
                { 
	                int retcode = e.getReturnCode();
	                System.out.println("Return code is" + retcode);
                }
            

Voice Portal does not provide facility to route calls over selected trunks, since it does not have any association with the dial plan. Call routing is typically performed by a telephony switch. For Example: Avaya Communication Manager.

Voice Portal supports the following audio file formats:

1. Raw (headerless) 8kHz 8-bit mono mu-law [PCM] single channel. (G.711)
2. Raw (headerless) 8kHz 8 bit mono A-law [PCM] single channel. (G.711)
3. WAV (RIFF header) 8kHz 8-bit mono mu-law [PCM] single channel.
4. WAV (RIFF header) 8kHz 8-bit mono A-law [PCM] single channel.

The above audio file formats are defined in the VXML specification available at http://www.w3.org/TR/voicexml20/#dmlAAudioFormats

A Blind transfer operation uses only one channel or telephony port to perform hold/dial/transfer steps. The channel is released at the end of a call.

A Dialog Designer Speech project has the below URL format:

http://<IP of App_server> :<Port>/<Applicaton_name>/Start.

An example URL for a CCXML project is:

http://<IP of App_server> :< Port>/<Applicaton_name>/<start page>

Note: Application developers can choose any appropriate start page for a CCXML application as shown in the screenshot below.

Voice Portal Application Summary report can be populated by using the Report item in a Dialog Designer application. A Report item is used to track and submit application data that is used to analyze and evaluate application performance in a run-time environment. Drag and drop a Report item in the call flow node and select a variable whose value is to be reported as shown in the screenshot below.

To view a variable and its value on Voice Portal generated reports, login to the Voice Portal web administration page. Browse to Reports -> Application. Select the appropriate Optional Filters for viewing Application Summary Reports and select Summarize by Variable Name under Report Type.

The variable name and its value are displayed in the Application Summary Report.

The AAI data field is supported only if the outbound call is placed by a Voice Portal system through a SIP trunk. Data cannot be passed to another application using AAI in an H.323 configuration.

The call recording functionality is supported in Voice Portal 4.0 and later versions by integrating Voice Portal with a third party call recording solution such as Proactive Contact. Voice Portal does not have any inbuilt functionality for call recording.

Call Classification feature enables a Voice Portal system to detect whether a call is answered by a human voice or answering machine. The 'green feature' SA8874 (Special Applications 8874) should be enabled on Communication Manager to enable the Call Classification feature for 7434ND (Voice Portal) station type. The SA8874 feature is enabled through licensing.

The following databases are supported on Voice Portal 5.0:

1. Oracle 9 and later versions.
2. PostgreSQL 8.2 and later versions.
3. MSSQL.

In order to generate Dialog Designer application reports:

1. Configure the web service authentication parameters on a Voice Portal Management System (VPMS).
2. Use the report item in the Dialog Designer application to submit application data to Voice Portal. Report items are used to add custom traces to the call flow.

These steps are elaborated below.

1. Configuring the Application Logging web service authentication parameters:

The Application Logging web service is used to log data to a Voice Portal system. The Application Logging web service conforms to all W3C standards and can be accessed through any web service client using the Avaya-provided Web Services Description Language (WSDL) file.

The WSDL file for the web service is located at: http://VPMS-server/axis/services/VPReport4?wsdl

Note: The application logging web service is used internally by Dialog Designer Report item to log information to Voice Portal. Third party applications could use this web service to log information to Voice Portal.

To configure the web service authentication parameters on Voice Portal, follow the steps given below:

1. Login to VPMS web interface and select System Configuration -> VPMS Servers. On the VPMS Servers page, click VPMS Settings.
2. On the VPMS Settings page, go to the Application Reporting section in the Web Service Authentication group.
3. Enter the user name and password for Digest Authentication. This is the same user name and password that must be used when accessing the web service though the WDSL file.

2. Using the Report item in Dialog Designer Applications:

Refer to FAQ - 'How to generate Voice Portal Application Summary reports from a Dialog Designer application' for information on using Dialog Designer report nodes to generate Voice Portal application reports.

In order to change the configurable application variable value on a Voice Portal 5.0 system, login to Voice Portal Management System (VPMS) web interface and browse to System Configuration -> Applications. On the Applications page, go to the desired application and click on the pencil icon in Configurable Application Variables column. Enter desired values for the variables retrieved from the application and save the changes made.

The Voice Portal Application Interface Web Service uses Digest Authentication to authenticate web service client requests. Error "401: Unauthorized" is received if the Application Interface Web Service fails to authenticate a web service client request.

To resolve this error, ensure that the user name and password are included in the web service client request as specified in the outcall section on the VPMS settings page of the Voice Portal Management System.

To configure and verify outcall parameters, follow the steps given below:

1. Log in to the Voice Portal Management System (VPMS) web interface with appropriate credentials and browse to System Configuration->VPMS Servers->VPMS Settings. On the VPMS settings page, configure outcall parameters under Web Service Authentication Group.
2. Try to access the Application Interface Web Service from web browser by specifying the URL http://VPMS-server/axis/services/AppIntfWS?wsdl where VPMS-server is the domain name or IP address of the system where VPMS software is installed. When prompted enter the username and password as specified in the outcall section on the VPMS settings page. This will open the WSDL file for the Application Interface Web Service.

Refer to the ClickToCall sample application available with Dialog Designer installation CD for information on how to use the Application Interface Web Service.

Yes. Starting from Voice Portal 5.0, reports can be scheduled and delivered as email attachments or via RSS feed.

To schedule report generation, follow the steps given below

1. Log in to the Voice Portal Management System web interface with the appropriate credentials and browse to System Configuration ->Reports->Scheduled Reports. Click on the Add button to schedule the generation of reports.
2. On clicking the Add button, add scheduled report page is displayed. On this page specify the source report, schedule date and time, report date and time and configure the notification methods and output options. For detailed description of scheduled report page fields refer to Voice Portal documentation library available on http://www.avaya.com/devconnect.

Note: Scheduled reports can optionally be configured to generate notification only when the record count reaches specified minimum value. To configure the record threshold restriction, check the 'Enable Record Threshold restriction' field on the scheduled report page and specify the record count.

LaunchVXML and LaunchCCXML methods of Voice Portal Application Web Interface Service can be used for outbound calling.

LaunchVXML uses the platform's default CCXML page to initiate an outbound call, and start the specified VoiceXML application. There can be only one outbound call per invocation of the 'LaunchVXML' method.

The LaunchCCXML method of Voice Portal Application Web Interface Service can be used for outbound calling by placing the outbound call from the CCXML application. Using LaunchCCXML for outbound calling provides more control as we define what the CCXML application does. For example, a single CCXML application can be used to launch multiple outgoing calls simultaneously.

For more information on LaunchVXML and LaunchCCXML methods, refer to the Voice Portal Documentation Library available at http://www.avaya.com/devconnect.

Note: When launching multiple outgoing calls simultaneously, application should verify the number of available ports on the Media Processing Platform (MPP) so that it does not exceed the outbound resources. Refer to the topic "Best practices" in the Voice Portal Documentation Library while using the Application Interface Web Service methods to launch multiple outgoing calls.

To perform outbound calling with the Application Interface web service, Avaya special application feature SA8874 should be enabled on Avaya Communication Manager. Without the SA8874 feature, the web service has no access to call progress information and may initiate a VoiceXML application before the call is answered.

To check if the SA8874 feature is enabled on Avaya Communication Manager, refer to the FAQ 'How to check whether the SA8874 feature is enabled on the Avaya Communication Manager (CM)?' on http://www.avaya.com/devconnect.

Note: The SA8874 feature can be provisioned on Avaya Communication Manager 3.1 or later versions and requires a separate license before it can be enabled.

A Session Detail report created using the Voice Portal Management System can be used to determine the ASR server name that handles an application session.

To include the ASR server name in the Session Detail report, check the field "AES Server" in the optional filters group on the Session Detail (Filters) page. For information on how to create the Session Detail report, refer to the topic "Creating a Session Detail report" in the Voice Portal Documentation Library available at http://www.avaya.com/devconnect.

Note: Session detail information is stored in the 'sdr' table of the Voice Portal database.

Yes, when deploying the cticonnector.war file and expanding it to a directory a data/logs directory is created.

CTIConnector specific logging can be found at <ConnectorAppLocation>/data/logs/trace.log.

Voice Portal was renamed to Experience Portal for version 6. Therefore, the Voice Portal Management System (VPMS) is now called the Experience Portal Manager (EPM). The Media Processing Platform (MPP) name has not changed. However, a new media platform, the Avaya Media Server is now also supported with Experience Portal ONLY in the Microsoft Windows environment deployment.

Inside the VXML application, insert a prompt with the format: <prompt><audio src="builtin://senddigit/<dtmf_digits" >/></prompt> where <dtmf_digits> is a string of digits to be sent.

Documentation is not available for these messages. The following information is pertinent however:

The data that EPM sends to the syslog server comes from the database table named csadminauditlog. Within the EPM web application, the Audit Log Viewer displays the data from csadminauditlog, though it chooses to leave out some of the uninteresting bits.

For instance, here is sample output on a syslog server where the EPM sends login/logout events.

Dec 13 16:13:09 localhost @2011-12-13 16:13:09,245
                PST-PADTL00005:messages.vpmsMsgCode-INFO-Security-VoicePorta
                l-TP-Processor32-User:mvg-JNDI Login ,1323821589244, , , ,
                -String:1323821589244-String:-String:-String:-String:--malms
                teen-1234567890#### 
                Dec 13 16:13:25 localhost @2011-12-13 16:13:25,475
                PST-PADTL00006:messages.vpmsMsgCode-INFO-Security-VoicePorta
                l-TP-Processor37-User:mvg-Logoff ,1323821605474, , , ,
                -String:1323821605474-String:-String:-String:-String:--malms
                teen-1234567890####
            

Here is how this example breaks down...

Dec 13
                16:13:09 - Timestamp from syslog server
                localhost - Name of host that
                sent data to syslog server
                @2011-12-13 16:13:09,245 PST - Timestamp
                from Experience Portal (Timestamp column of Audit Log
                Viewer)
                PADTL00005 - Event code (Action column of Audit Log
                Viewer)
                messages.vpmsMsgCode - File that Experience Portal uses to
               find display string associated with event code (Ignore)
               INFO -
               Severity of event (Ignore-Always INFO)
               Security - Category of event
               (Category column of Audit Log Viewer)
               VoicePortal - Program that
               logged event (Ignore-Always VoicePortal)
               TP-Processor32 - Thread that
               logged event (Ignore)
               User:mvg - User that performed action (User
               column of Audit Log Viewer)
               JNDI Login - Action that was performed
               (Action column of Audit Log Viewer)
               1323821589244 - Timestamp in
               unreadable format
                , , , , - Values for other parameters logged with
               event (Columns Component, Property, From, To of Audit Log
               Viewer)
               String:1323821589244 - Type and value for parameter
               Timestamp
               String: -
               Type and value for parameter Component
               String: -
               Type and value for parameter Property
               String: - Type and value for
               parameter From
               String: - Type and value for parameter To
               malmsteen -
               Hostname
               1234567890 - Product ID
               #### - End of message
               indicator
            

Notice that the first couple of fields here actually come from the syslog server, rather than Experience Portal.

Also notice that the data values for Audit Log Viewer columns Component, Property, From, and To are included twice. In the first instance they are run together with the event description as one long string. In the second instance, they are split out as separate fields.

Two scenarios, Inbound Calls and Outbound Calls are relevant here.

Inbound Calls:

The only call classification done is looking for fax tone. If Experience Portal sees a fax tone and the phone number used is that of a fax machine, the incoming call will be transferred to the fax machine rather than application processing happening as normal.

Outbound Calls:

There are essentially three ways to generate an outbound call with Experience Portal:

1. CCXML tag {createcall}
2. VoiceXML tag {transfer}
3. Application Interface web service method LaunchVXML

In all cases, the only call classification done by Experience Portal is to discriminate between calls answered by humans and calls answered by machines. The algorithm used is rather convoluted, but it boils down to looking at the pattern of voice and silence when the call is answered. It is assumed that humans answer with something short (e.g. 'Hello', 'Hello...Hello?...Hello?') while machines answer with something long (e.g. 'Hi, this is Mike; I'm away from my desk; please leave a message; I'll get back to you.')

Experience Portal counts on the underlying telephony server to indicate when the call was answered and at that point, a decision can be made whether there is human or machine at the other end. In a SIP world, this answer notification actually originates with the remote SIP user agent. In an H.323 world, Communication Manager does the answer detection, but Communication Manager can only do the answer detection if special application 8874 (SA8874) is installed.

This error can occur if:

1. the MIME type is not set to CCXML when the CCXML application is saved, or
2. the Media Processing Platform (MPP) is not version 4.0 or later.

No, the name attribute in <transition> tags is not supported on Voice Portal 4.1. The name attribute has been replaced by the variable event$. The following code snippets show the usage of the <transition> tag and the syntactical differences between VP 4.0.x and 4.1.

Sample code for Voice Portal versions 4.0.x:

<transition event="connection.alerting" name="evt">
                <log expr="'Received connection.alerting for the following connection: '+ evt.connectionid"/>
                </transition>
            

Sample code for Voice Portal versions 4.1:

<transition event="connection.alerting">
                <log expr="'Received connection.alerting for the following connection: '+ event$.connectionid"/>
                </transition>
            

Voice Portal 5.0 supports the <createccxml> tag.

Note:

1. The <createccxml> tag is not supported in versions prior to Voice Portal 5.0.
2. Refer to the CCXML specification (http://www.w3.org/TR/2005/WD-ccxml-20050629/) for details on implementing the <createccxml> tag.
3. The <createccxml> tag creates a new CCXML session. The new session is independent of the previous session. Conversely, a connection cannot be moved from the older session to the newly created session.

Login into the Nuance Speech Server(s) machine and check if the OSSServer.cfg file is configured to contain all of the voices that are installed on the server.

This file can be located in the following location

Note: If RealSpeak Voices or OSR language packs are installed after having installed the Nuance SpeechWorks Media Server (SWMS), then the above file needs to be modified to include these new voices and languages.

Add the following parameters to the OSSServer.cfg file for enabling a supported Text-to-Speech voice.

Ensure that the Nuance services are restarted after modifying the OSSServer.cfg file. These services can be located under the services menu.

This error indicates that the grammar cache(s) on the IBM WVS server need to be cleared of existing (cached) grammar files.

Follow the steps below to complete the process:

• Shut down the IBM WVS and WebSphere Application Server (WAS) by clicking on 'Stop Server' in the 'First steps' utility. This utility can be accessed by clicking on the 'Start' menu as shown below.



Figure 1: Accessing ?First Steps? from the start menu

• Select 'Stop the Server' as shown below in Figure 2 to stop the WAS server. This would also stop the WVS service.



Figure 2: Stopping the WAS service using 'First Steps'

• Figure 3 displays the message box indicating that the services are stopped.



Figure 3: 'Stop the Server' confirmation screen

• Using Windows Explorer access the 'rrcache' folder under the 'VoiceServer' folder in the <Installation directory>\VoiceServer. The default location of this folder is: C:\Program Files\WebSphere\VoiceServer\rrcache

  Select all the cached versions of the grammar files and delete them from this directory.

• Using Windows Explorer access the 'gmcache' folder under the 'VoiceServer' folder in the <Installation directory>\VoiceServer. The default location of this folder is: C:\Program Files\WebSphere\VoiceServer\gmcache

  Select all the cached versions of the grammar files and delete them from this directory.

• Restart the IBM WAS service by using the 'First Steps' as shown in Figure 4 below.



Figure 4: 'Start the Server' using 'First steps'

• Verify that the WVS service is started before placing a call.

1. Log into the Voice Portal Management System (VPMS) machine using the web interface.
2. Click on 'Applications' on the left hand pane.



3. Open the page for the desired application. Insert vendor specific ASR parameters in the 'Vendor Parameters' field under 'ASR parameters'.



4. Similarly, insert vendor specific TTS parameters in the 'Vendor Parameters' field under 'TTS parameters'. Click 'Save' to apply the setting changes.

NOTE: The Voice Portal does NOT support inserting vendor specific parameters in any location other than the ones mentioned above.

There is no 'safe' way to simulate failover strategy. Avaya recommends any one of the methods given below to test Media Processing Platform's (MPP) failover strategy:

1. 'reboot' command can be used to restart the machine.
2. Disconnect the power supply to simulate the Real failure scenario.

Application failover capability is not handled by the Voice Portal management System. Most Web Servers hosting applications have built in capabilities to provide redundancy in the event of a failure. Refer to the respective Web Server Vendor's documentation regarding failover scenarios.

There should not be any conflicts for sharing license ports between Natural Language Speech Recognizer (NLSR) and whole word recognizers. Ensure that the total number of ports allocated to each is less than (or equal to) the number if ports licensed.

The Media Processing Platform (MPP)(s) in the cluster will still handle calls even if the VPMS fails (as long as the MPP(s) are not restarted).. However, if there is a need to re-assign an MPP to another VPMS system then the 'authorize_vpms.php' script on that MPP will have to be executed as follows: authorize_vpms.php (VPMS IP address: 8080)

For additional details on running this script refer to Installing, upgrading and configuring voice portal ->Installation Troubleshooting->Reinstalling the VPMS software in the voice portal online documentation.

The following configuration needs to be done:

1. The application (VXML page) must be hosted on a web server (Tomcat / WebSphere) also ensure that the number to which the call is getting transferred is valid.
2. If the call is getting transferred to an external phone number, then follow the steps below:
   1. Set up a dial plan.
   2. Configure the Automatic Route selection routing table and set up a feature access code.
   3. Use the above Feature Access Code (FAC) as prefix to the dial string. For e.g.: 9, 1212456980 where 9 is the FAC.
3. Refer to the 'Administrator Guide for Avaya Communication Manager' Document ID: 03-300509 for details on configuring the communication Manager.

Note: If only blind transfer feature in the VXML application is tested using the test page then this can be directly deployed on the Media Processing Platform (MPP).The same page is also available as part of a sample application. Access the Voice Portal online documentation and click on Installing, upgrading and configuring voice portal -> Voice portal system configuration and installation -> Voice portal basic system configuration overview -> Adding the Voice Portal test application for details on accessing the sample application.

The Avaya Voice XML browser does not parse grammars > 1896 bytes in size. Therefore in case of size > 1896 bytes it is recommended to pass URL of the grammar to the speech Recognizer. The speech server fetches the grammar (hosted on a web server) and directly loads the grammar (instead of being passed through the IR).

Avaya Voice Browser (AVB) in Avaya Voice Portal 3.x does not support ECMA script processing.

The correct URL would be: http://(IP address of MPP machine)/(directory path to application). Use MPP IP address instead of localhost. When referencing the application on the VPMS application page, localhost implies that the application exists in the Voice Portal Management System (VPMS) machine which is not the case.

'AVAYA Voice Portal 3.0.1.3-0002 MPP Patch' has to be applied to resolve this issue along with following Nuance patch:

For Speech Works Media server (SWMS) 3.x, a Nuance patch is required to fix this issue. Nuance recommends an upgrade to SWMS 3.1.14 to resolve this issue.

Login to the Nuance OSR Speech server and check if the OSSServer.cfg file on the machine is configured to contain all of the voices that are installed on the server. This file can be found at the location specified below:

Note: If RealSpeak Voices or OSR language packs are installed after installing the Nuance SpeechWorks Media Server (SWMS), then the above file will have to be modified to include these new voices and languages.

Add the following parameters to the OSSServer.cfg file for enabling a supported Text-to-Speech voice.

Ensure that the Nuance services are restarted after modifying the OSSServer.cfg file. These services can be located under the Services menu.

Yes, Voice Portal 4.1 is compatible with Nuance Recognizer 9.0.1.

No, a GUI is not available while installing from the Avaya Enterprise Linux CD. This CD is available as part of a Voice Portal bundled offer and is a version of Red Hat Enterprise Linux 3.0 Portal with essential features. Additional components like GUI are not included to improve security and system performance.

No, at present there is a no upgrade CD available from Avaya. The recommended procedure is to register with Red Hat Linux and use the Linux utility 'up2date' to download the RPM packages for the Update 6 release. Note that 'up2date' utility will download all the latest RPM packages, so the installed version may be newer than update 6.

A minimum memory of 2 GB is required for installing a voice portal system (both VPMS and MPP).

Starting with GA release 3.0.1.0.2303, both the VPMS and MPP can be installed on the same server. However, this configuration does not support adding multiple MPP?s to the configuration and only a maximum of 24 ports (Telephony, Text-to-Speech, and Advanced Speech recognition ports) can be licensed for this system.

The MIB is included in the Voice Portal installation CD under 'Support/SNMP-MIBs' folder.

VPMS and MPP share an IP address when installed on the same server. In general, VPMS and MPP components are assigned the IP address of the server on which they are installed.

The procedure to install a license file on VPMS is elaborated upon below:

1. Login to VPMS. Browse to System Configuration -> Licensing. Enter the license server's URL and click Verify. This sequence is shown in the screenshot below.



2. Login to WebLM with the default credentials, as shown below.



3. After logging into WebLM, the license installation page is displayed. Browse to the license path and click on Install to install the license file, as shown below.



Verify the successful installation of license file as shown in the screenshot below.

No, a new system must be installed fresh (both Operating System and Experience Portal). However, there is a migration path of data from an existing Voice Portal 5.0 or 5.1 systems to the new Experience Portal installation. Only Voice Portal 5.0 and 5.1 systems running on Avaya Enterprise Linux may be upgraded without doing a fresh install.

See the Experience Portal Installation Guide for more details.

Documentation for Intelligent Customer Routing functionality in Avaya Experience Portal topic is covered in the Administering Avaya Experience Portal guide.

The Voice Portal H.323 implementation supports Vectoring Converse-on step. Converse step can be implemented using either of the two methods below:

• On incoming call Converse-on data is passed as part of the application URL query string.

  The Application page in the VPMS should be configured so that the Converse-On field is set to 'Yes' as shown below in Figure 1.



Figure 1: Selecting 'Converse-On' for a voice portal application

The converse-on at the beginning of the call is passed by VP to the application as part of the initial query string in the form of:
&session___vpconverseondata=<Converse-On Digits>
• Voice Portal Applications can return data to vector by using blind transfer.

  For example, here's part of a vxml code using converse-on.

  <transfer name="mycall" dest="tel:160204" bridge="false">
  <prompt> Transferring to vector 160. </prompt>
  </transfer>

  The number used in the transfer, 160204, is actually composed of the converse-on vector #160 and some data ?204.?

The following means are available to access information about a particular MPP system.

The scripts are located in the '/opt/Avaya/VoicePortal/MPP/bin' directory but can be run directly from the command line as the directory location is entered as (by default) a 'PATH' environment variable in the MPP system.

Logon to the MPP server with the correct credentials and run the getmpplogs.sh --logs command on the command prompt.

The LaunchVXML method has a default timeout value of 120 seconds. The default timeout value is automatically applied when the user does not provide a timeout value or the timeout value is set to 0.

On the VPMS page, select MPP Manager -> System Maintenance. Select the required MPP and click Restart (as shown on the following page).

Whenever an MPP fails, calls in progress on that MPP system are terminated immediately. Subsequent calls will be redirected to the other MPPs systems as per the failover mechanism in place.

The CCXML, VXML and session related logs can be found under the CXI, VB and SessMgr folders respectively. These folders are located in the /opt/Avaya/Voice Portal/MPP/logs/process directory on the MPP server.

No, the MPP is only supported in the Linux environment. For the Windows environment, the Avaya Media Server (AMS) is the only media server type supported.

Use the SIP UUI field to transport the call data to a transferred destination. In the VXML <transfer> tag OR a transfer node in an Orchestration Designer flow, set the data in the "AAI" attribute of the transfer tag or node. This data will be populated in the SIP UUI field when a transfer occurs.

There is a setting in Communication Manager that will affect the UUI behavior. The UUI Treatment setting for the SIP trunk to the Experience Portal determines the UUI format that Communication Manager expects. In 'shared' mode, the UUI data is expected to be an Avaya specific formatted field (i.e. UCID, ASAI Data). In 'service-provider' mode, the data can be any format necessary for the end device (i.e. screen-pop application, device display). In 'service-provider' mode, Communication Manager expects a user-defined format (i.e. string, text, numbers) and will not try to interpret the UUI data. To affect this setting in Communication Manager, login into Communication Manager via the SAT terminal, type the command 'display trunk-group ##' where ## is the trunk group number for the SIP trunk routing to Experience Portal. Go to page 3 and in the 'UUI Treatment' field, make sure it's set to 'shared' if you are using Avaya formatted UUI data or "service-provider" if passing along customer specific UUI data (i.e. string data). In addition, for UUI to be delivered to the final destination, all the trunks in the call path to the transfer-to destination must have UUI setup on them.

In addition, there is a setting in the Experience Portal that affects UUI. In AAEP web admin, go to Applications -> APP_NAME, then scroll to the bottom. Expand the "Advanced Parameters" section and make sure the "Operation Mode" is set to the appropriate value to match what was set on Communication Manager.

No, the user must have the role of "Privacy Manager" assigned to view the transcription data. This features helps guard against the loss of sensitive data that may be collected during a call session (i.e. credit card, social security, bank account numbers, etc).

The key points are to have a role of manager and assign that role to a user 'tomcat' and set the password for the user 'tomcat'. Once Tomcat is restarted, point a browser to http://<tomcathost>:8080/ and select 'Tomcat Manager' in the Administration menu as shown in Figure 1 below.

Figure 1: Accessing Tomcat Manager

Operating two instances of Tomcat or possibly different Tomcat versions on the same server depends upon the usage scenario. Two common situations are described below:

1. If both the instances of Tomcat are not intended to be run at the same time, unzip the second Tomcat release to a different directory. It is likely that both Tomcat servers will contain the default settings (port numbers etc.) at startup.

   Note:Operating both instances in this case will throw exceptions during startup.

2. If both instances need to be run simultaneously, modify each Tomcat's settings so that there are no port conflicts. (Note: The Windows service installer cannot be used here.) The server.xml file needs to be edited and the ports changed as shown in the screenshot below. The server.xml file is located in the <TOMCAT installation directory>/conf folder.

This error means that the application server is not responding to a page fetch request. Some possible reasons are:

1. Tomcat has not been started, or,
2. The application is not deployed correctly.

Follow the steps below to build and re-deploy the application. The steps outlined assume that the application is run from a Dialog Designer development environment with the application context path pointing to Tomcat deployed on the same machine.

Note: If the application is deployed on a remote Tomcat machine then Steps 1 through 3 would involve un-deploying the application and stopping Tomcat.

1. Enter the URL http://localhost:<portnumber> (default port = 8080), and logon to the Tomcat home page via any Internet browser. Click on Tomcat Manager as shown in the screenshot below.



2. After clicking Tomcat Manager, enter the username and password as shown below.

   Note: To create a username and password, locate the Tomcat-user.xml file under<Tomcat Installation directory>\conf, add a manager role and a user for the Tomcat manager as shown below:

   <role rolename="manager"/>
   <user username="<username>" password="<password>"
   roles="manager"/>



3. After logging to Tomcat Manager, click on the Undeploy link (shown below) to uninstall the application.



4. Right-click on the application in the Navigator view, select Tomcat Project and click on Remove Context Definition to remove the application's context definition from Tomcat.



5. Stop Tomcat from the Dialog Designer environment by clicking on the icon (circled) shown in the screenshot below.



6. Right click on the application in the Navigator view and select Dialog Designer -> Generate from the drop down list as shown in the screenshot below.



7. Export the application by right-clicking on the application and selecting the Export option from the dropdown list and create a new .war file.



8. Start Tomcat and deploy the application. For deploying an application in Dialog Designer, please refer to Tutorial for Deploying Speech and Call Control Applications on Avaya Voice Portal and Avaya Interactive Response.

The user's account is locked by the system. Login into the system as administrator and click on the 'Users' link under 'User Management' in the menu on the left hand side of the screen.

Figure 1: Unlocking a 'Locked' User account

The value in the 'Account Lockout Duration' edit box can be changed to allow the user to log in again within the specified time period

User accounts can be created by clicking on the 'Add' button shown in Figure 1 above. The form shown in Figure 2 displays the following types of user accounts.

Figure 2: Choosing a user type

Refer to the Voice portal help documentation for additional details on the types of user accounts.

General recommendation is to use the audio format supported by the switch. Set the Audio format in the Voice Portal as follows:

1. Log into the Voice Portal system.
2. Under system configuration: Click on VoIP Settings -> VoIP Audio Formats -> MPP Native Format and select a format from the drop-down box.
3. Click 'Save' to apply the setting changes.

Other tested formats: Ulaw 8bit/8k/mono (raw) - .ulaw Alaw 8bit/8k/mono (raw) - .ulaw Pcm 16bit/8k/mono - .wav Pcm 8bit/8k/mono - .wav vis (Avaya proprietary format) ) - .vis

NOTE: When using vis & raw files the mime types for the Application server may need to be edited to add these entries. Otherwise, the Voice Browser may report errors while trying to fetch/play the prompt.

Reason for this error message is that the Media Processing Platform (MPP) could not accept incoming calls because all the stations assigned to the MPP are out of service.

Follow the steps below to correct this problem:

• Login to the VPMS web interface.
• On the Port Distribution web page, identify the ports that have been assigned to the MPP.
• Ensure that the private branch exchange (PBX) and the MPP are communicating properly.
• On the H.323 Connections web page, verify that the configurations for the IP Address, Phone Numbers, Passwords, and Ports for the PBX are correct.
• On the VoIP web page, verify that the VoIP Network Address configured for the MPP is correct.
• Verify that the PBX is functioning correctly.

The VPMS does not directly support this but it can be configured to send Simple Network Management Protocol (SNMP) traps in the "SNMP" page on Voice Portal web interface. Third party SNMP Trap receivers can be used to detect the incoming alerts and send out SMS/Email alerts.

Voice Portal Management System (VPMS) handles cache management in Voice Portal 3.0.1.1 and later versions.

Access the VPMS logon page by opening a web browser and entering the URL of the VPMS as http://< IP address of VPMS >. The log on screen shown below will be displayed. Enter the username and password in the respective fields.

On successful authentication, the VPMS Home page is displayed as shown below:

Under the System Configuration tab select the MPP Servers option. Now select the appropriate MPP server as shown below:

The Change MPP Server screen as shown below appears. Here, select the Categories and Trace Levels option at the bottom of the screen.

On selecting the option, the following screen appears which contains the Voice Browser INET option.

Set the Voice Browser INET option to Fine, Finer or Finest. The Voice Browser INET option manages the downloading of content such as VoiceXML and prompts from the application server .

For more information, refer the Online Help section under the VPMS web pages.

To change the authentication credentials of outcall web service, login to Voice Portal Management System using root username and password. Under System Configuration, select VPMS Servers > VPMS settings. Under Web Service Authentication tab enter the username and password settings in the section highlighted in the screenshot below.

The Voice Portal Application Interface Web Service is automatically installed on the server running Voice Portal Management System (VPMS) 4.1.x software. In order to configure the Application Interface Web Service, download the Avaya-provided WSDL file and build a custom web service client (that uses the WSDL file) as described in the steps below.

1. 1. Open the following page in a web browser to access the WSDL file:

   http://< VPMS-server >/axis/services/AppIntfWS?wsdl where < VPMS-server > is the domain name or IP address of the system where the VPMS software is installed.



2. When prompted, enter the user name and password. This user name and password should be same as the one specified in the Outcall section of the System Settings page. Refer to the FAQ "How to change the outcall web service authentication credentials in Voice Portal" for changing the authentication credentials.
3. Save WSDL file and use it to build the web service client for accessing Application Interface web service.

The Application Interface Web Service provides the following functionality:

1. Use Application Interface Web Service to start a CCXML or VXML application that has been added to Voice Portal using the 'Add Application' page. An Application can use the LaunchVXML and LaunchCCXML functions provided by Application Interface Web Service to start another VXML or CCXML application respectively. The Application Interface web service automatically examines each MPP in the Voice Portal system and starts the session on the first available MPP that has the required outbound resources available.
2. To send an event to a specific application session running on an MPP, the Applications can use SendCCXMLEvent method provided by the Application Interface web service. This method instructs the MPP to dispatch a user-named event with an accompanying parameter string to the specified CCXML session
3. To query the system for the total number of:
   1. Used and unused outbound resources available
   2. Unused SIP outbound resources
   3. Unused H.323 outbound resources
4. Applications can use the QueryResources method provided by the Application Interface web service to determine the approximate availability of outbound resources, before using the LaunchCCXML method or the LaunchVXML method to start a new outbound session.

Voice Portal requires the following components to support SIP connections:

1. SIP Enablement Services version 4.0 or later versions
2. Communication Manager (CM) Version 3.0 or later versions

Yes, IPCoDE supports SIP connections to a Voice Portal system. For more information on this topic, refer to
DevConnect portal > Products & SDKs > Application Enablement Services > Avaya IP Communications Development Environment.

To configure a Voice Portal SIP connection, follow the steps below:

1. Open a web browser and type http://< VP IP address >/VoicePortal in its address field to login to the VPMS administration site. The following screenshot shows the Voice Portal login page.



2. Browse to System Configuration > VoIP Connections > SIP tab and add the new SIP connection by clicking on the Add button. The screenshot on the following page highlights the relevant tabs and buttons.



3. Clicking on the Add button brings up the Add SIP Connection page (shown in the screenshot below) wherein the parameters for the new SIP connection may be configured through the fields provided. For more information on these fields, click on the Help link shown below.

The following steps need to be followed to achieve this:

1. The first step is to add the converse-on vector command in the associated vector to pass data between the switch and an IVR application. Use the variable wait in the converse-on vector command as:

   converse-on split X pri l passing wait and none

   For a complete description of passing information from the switch to an IVR application refer to section "Passing EWT to the VRU"from document Avaya Call center, Call Vectoring and Expert Agent Selection (EAS) Guide 07-600780, Release 5.0, January 2008.
2. In DD Speech application, use the session variable vpconverseondata to extract the EWT information

   Note: 'vpconverseondata' session variable is available only with Voice Portal platform. A 'prompt and collect' node is used to gather digits if performing a converse-on step with an application deployed on an Interactive Response platform.

Modify the file /opt/Tomcat/apache-tomcat-5.5.20/webapps/VoicePortal/WEB-INF/classes/messages/languages.properties on the Voice Portal Management System (VPMS) server to include the desired language ID. Remember to take a backup of this file before modifying it.

Login to the VPMS Admin page, browse to System configuration > Speech Servers and click on the Text-to-Speech (TTS) tab. Select the TTS server.

Select the appropriate language ID from the Voices dropdown box.

This can be done in two ways:

1. 1. While deploying (or adding) the application on Avaya Voice Portal, set the Generate UCID field from the Advanced Parameter section on the Add Application page. This setting will enable Media Processing Platform (MPP) to provide a UCID for each call.



The Call Control XML (CCXML) application will receive a UCID generated by an MPP in call events such as connection.connected in the event.avaya.ucid field which can be monitored in an MPP log file. Log onto the MPP system and access VB log file under the /opt/Avaya/VoicePortal/MPP/logs/process directory. Locate the UCID generated as shown below.



This UCID can then be accessed in the CCXML 'transition' element as show in the following code snippet

< transition event="connection.connected" state="init">
                        < var name="ucid" expr="event$.avaya.ucid"/>
                        < /transition>
                    

2. The second method is to use the CTI connector in Avaya Dialog Designer applications. The cticallinfo variable (shown below) will then contain all the call related information along with UCID which is provided by Avaya Communication Manager (CM).

For H.323 stations, the IP stations assigned on Communication Manager to Voice Portal must be of type 7434ND. Voice Portal does not support any other H.323 phone type.

The 7434ND phone type should be enabled on Avaya Communication Manager. The steps to enable this feature on Avaya Communication Manager are:

1. Login to Avaya Communication Manager with the appropriate credentials.
2. Type the command change system-parameter features.
3. Set the 7434ND field to 'y'