Using iHub

This page describes how to set up Integration Processes to integrate the customer's backend resources, such as web services and databases, with Genesys Intelligent Automation.

Integration Hub is a simple and powerful way to do this, and brings many other benefits such as support for multiple environments’ endpoints (dev/test/production, for example) and the ability to create automated test scripts. Using Integration Hub is a good way to keep your Intelligent Automation apps and their assorted presentation logic separate from the details of how to call onto your enterprise’s backend systems.

Script blocks are great for adding ‘presentation’ logic to your app - comparing values and branching out to different blocks, performing date calculations, combining prompts. It’s possible to invoke RESTful web services directly from a Script block, and this may be sufficient if the services are simple. It is advised not to the Script block for tasks like handling security certificates, or calling onto databases or SOAP-based web services

Tip

For information on working with Groovy, refer Apache Groovy documentation.

iHub Interface

The iHub interface is where you set up the Processes that integrate Intelligent Automation with the customer's resources, such as databases and web services.

iHub has four tabs, as follows:

Processes—In this tab, you define the Integration Processes that will handle HTTP(S) and JDBC requests that are received from the Intelligent Automation VUI. Use the scripting commands described in Scripting Commands to write any Process and Library scripts.
Deploy to Production—After you have created and tested your Processes, you put them into production using this tab. Deployments in the current calendar year are displayed in tabular and calendar (weekly) format. From the table of deployments, you can also rollback to previous productions if needed.
Import—You can import Process scripts and Shared scripts from other Processes.
Export—You can export Process scripts, including the latest production versions, and library entries into a compressed (.zip) file; compatible for use with the Import tab.

Processes

Defining Common Properties

To simplify your work, Intelligent Automation allows you to specify common properties, such as variables, functions, and environment settings, that can be used for all of your processes as required. In addition, if any Company-specific configuration information applies to a specific callflow engine and resides on a web site or in a database, you can specify the location and access credentials of the site or database to which the VUI must connect, even securing the connection if necessary. You can do all this on the right-hand panel of the Processes tab.

Tip

These settings can be modified or deleted at any time by opening the Processes tab and making the necessary changes.

Library tab

On the Library tab, create a list of additional variables and functions that can be shared between Processes. The Library is inserted at the top of a Process script, so you can reference the variables and functions throughout the script, as needed.

HTTP Settings tab

On the HTTP Settings tab, provide the Company-specific configuration information that applies to a given Intelligent Automation callflow engine installation on a website.

You can set up multiple secure connections between Intelligent Automation and external websites.

Enter only a Hostname, Port (defaults to the port used by the installation in which you are working), and the Connection Timeout. If you want to set up a secure connection, you will have to provide additional information. For more information about the security concepts, refer to the Genesys Security Deployment Guide.

When calling an external web site using Groovy, the list of trusted SSL certificates must be provided. Intelligent Automation supports both client-side authentication and server-side authentication.

The Send Username and Password option allows access to external servers.

When using Client-Side Authentication, Intelligent Automation allows uploading the client certificate including the public key certificate and a private key along with the password. IA stores this certificate information and uses it for verification.

You can upload a server certificate also. The Verify Hostname on Server's Certificate option verifies if the host name matches with the certificate.

IA supports both client-side and server-side authentication individually and also together.

JDBC Data Sources tab

If any of this Company-specific information resides in a database or in SQL connection pools, define the JDBC parameters on this tab. The contents of this tab can differ depending on what you specify in the Database Type field, and by what you enter in other fields. Enter the required information in this tab, noting the following:

The entry in the Name field is unique, but is for internal use only.
You can enter the JDBC URL manually by selecting Set Connection String Manually or have Intelligent Automation construct it from the information you enter in the fields that open when you click JDBC URL Generation Method.
The are four supported Database Types—SQL Server, PostgreSQL, MySQL, and Oracle (using either a System ID or a Service Name).
Warning
Genesys does not ship with a driver for MySQL and PostgreSQL, so if you are using the MySQL or PostgreSQL DBMS, you will have to obtain that separately.
Enter a driver name in Override Driver Class Name only if you need to override the driver class name, for example if you are using an older or custom version of a driver.
If you do not want to use the standard validation query specific to the database type (for example, SELECT 1 for SQL Server) that is used by the connection pool to determine if the connection is working, you can specify a custom one, perhaps based on the Company name, in Override Validation Query.
The Initial Connections field indicates the number of connections that are made initially when connecting to the database. The allowed permissible range is 0 to the value configured in Indy.ConnectionPooling.JDBC.MaxSize.
The Maximum Connections field indicates the maximum number of connections to the database. The default value is 10.

Important

To increase the maximum number of connections, modify the Indy.ConnectionPooling.JDBC.MaxSize parameter (Administration > Default Server Settings). The allowed permissible range is 0 to the value configured in Indy.ConnectionPooling.JDBC.MaxSize.

The Maximum Wait Time field (or maximum time to wait for a pool connection) is set by default as 1 and maximum of 300000 millisseconds.

Environment-specific Settings tab

In this tab, define any settings that are specific to the environment in which you are working, such as Lab, Test, or Production.

Variables used during processing (like hostname and port) can be declared here and accessed by Groovy/Java scripts later on.

Creating a Process

Create a new Process in the Processes tab, as indicated in the following diagram:

After you click New Process in the tab's header bar or click the Create a New Process button:

Enter a unique Process Name. Intelligent Automation automatically assigns a Process UUID for a new process. This UUID stays with this Process until the Process is deleted.
Enter any parameters that will be required in the request from VUI. These parameters will be used by the process as it is executing. If the request does not contain these parameters, it will be rejected.
Enter any confidential parameters. These parameters might contain customer-sensitive data, such as passwords or credit card numbers, and will not be written to logs or reporting databases.
Using the scripting commands provided with Intelligent Automation, enter the code for the script that will process the request.
Create the Response Template that will be returned by this Process.You must create at least one Response Templates - one for successful Process execution (Success). A Response Template for unsuccessful execution (Error) is optional.

Modifying a Saved Process

To modify a saved (existing) process, even its name, click Processes on the Processes tab to view the list of Processes. Select it by name on the Processes tab, and make the appropriate changes.

Deleting an Existing Process

To delete a process, click Processes on the Processes tab to view the list of Processes. Select the process you want to delete in the list, and click the garbage can icon to delete it.

Testing a Process

iHub allows you to test your Processes by defining test cases and running them against any Processes you have created, before the Processes are put into production. You can also test a production version by copying the currently deployed version of a Process to the test side of the environment.

Warning

If a Process by that name already exists, the test version will be overwritten by the copied production version, and any undeployed changes to the original test version will be lost.

Refer to the following diagram to create, run, and examine the results of a test case. The numbers in the diagram correspond to the tasks that follow.

Creating a Test Case

After clicking New Test Case:

Define the test case by giving it a specific Test Name, an optional Test Description, and values of any Request Parameters required by the process.
Specify the expected HTTP Response Code (the default, 200, indicates success); and optionally, the XPath Response Assertions, in which known elements or variables should match or not match specified values. The variety of assertions is based on whatever data is returned by the test stubs. The Author can write as many XPath assertions as they wish to inspect the generated XML, as long as he or she knows what data is returned by a test web service for a particular input. For example:
- XPath is /response/status/text()
- Match Type is matches
- Expected Values is success

Running and Evaluating a Test Case

Click Run this Test.
View the Execution Results, including the information from any Response Templates associated with this process.

Deploy to Production

After you have created the Process and tested it, it is ready to be put into the production side of the environment in which you are working. Note that when you deploy a process to production, it is not changed nor removed from the testing environment.

To deploy a process to production:

Select the Processes to Deploy. If there is more more than one listed, you can deploy one, some, or all of those that are listed.
Deployment Options give you the option to include the Library and/or Environment information that you entered on the Processes tab.
Tip
If you included any of the library variables or environment information in any of the processes to be deployed, you must include the corresponding deployment option here.
You must provide a Reason for Deployment, if for no other reason than to identify this particular deployment.

The Production Version section (see the diagram below) displays a history of your deployment activities, including which deployment is your Current Production Version.

You can also copy a deployment to a test version, effectively overwriting the test version with a copy of the deployed version.

Tip

If your current deployment is not operating as expected, you can copy the previously deployed version to the test side of the environment, test it to make sure that it is running properly, then redeploy this version. This version then becomes the Current Production Version, taking over this position from the more updated, but faulty version. Then, you can work to debug the latest (updated but faulty) version, and deploy it when tested.

The twelve-month calendar for the current year illustrates the distribution of all deployments throughout the current calendar year. Hovering over a red box displays the date and number of deployments on that date; clicking on the red box highlights the deployment in the table above the calendar.

Warning

The Copy-to-test feature in iHub replaces all test processes with the production processes that was in the specific deployment.

Adding Integration Processes to a Callflow

After you have created and tested an Integration Process, you can incorporate it into your callflow by adding it to an Interceptor block, as shown in the following diagram:

Open the Integration tab of the Interceptor block, and do the following:

Replace the values of the Web Service URL for Test Calls and Web Service URL for Production Calls fields with the following:
```
ihub://<Process_ID>
```
Where Process_ID is the Process UUID shown at the top of the Processes tab. You can also add additional parameters, if required.
Change the value of the Web Service Timeout field, if needed.

Import

On the Import tab, you can import iHub Scripts from other processes in the same or a different environment. The input file must be created by using the Export tab.

After you have selected the .zip files for importing, click Choose Components for Import. You can then choose to use the file contents to create a new process, or to overwrite the corresponding parts of an existing process.

Export

On the Export tab, you can export processes and shared scripts to a .zip file. This file can be used for such things as backup or storage, or to import back into iHub, perhaps into another environment.

You must select one or more processes and/or one or more shared component to export. If any process you selected is also in production, you can also choose to export the latest production version of it instead of the test version.

When you click Export, the export operation creates a file called Integration Export <datestamp> <timestamp>.zip. This file can then be imported using the Import tab.

Scripting Commands