Integrate API Manager Analytics

The API Manager Analytics provides reports, statistics and graphs on the APIs deployed in WSO2 Open Banking API Manager. You can then configure alerts to monitor these APIs and detect unusual activity, manage locations via geo-location statistics and carry out a detailed analysis of the logs. 

This section explains how to integrate API Manager Analytics with the WSO2 Open Banking solution.

Configuring the API Management Analytics 

Follow the instructions below to configure the API Management Analytics to publish statistics.

1. Open the <WSO2_OB_APIM_HOME>/repository/conf/api-manager.xml file. 

2. Configure the properties under the <Analytics> sub element.

   <Analytics>
           <!-- Enable Analytics for API Manager -->
           <Enabled>true</Enabled>
   
           <!-- Server URL of the remote StreamProcessor server used to collect statistics. Must
               be specified in protocol://hostname:port/ format.
   
               An event can also be published to multiple Receiver Groups each having 1 or more receivers. Receiver
               Groups are delimited by curly braces whereas receivers are delimited by commas.
               Ex - Multiple Receivers within a single group
               tcp://localhost:7612/,tcp://localhost:7613/,tcp://localhost:7614/
   
               Ex - Multiple Receiver Groups with two receivers each
               {tcp://localhost:7612/,tcp://localhost:7613},{tcp://localhost:7712/,tcp://localhost:7713/} -->
           <StreamProcessorServerURL>{<PROTOCOL>://<ANALYTICS_SERVER_HOSTNAME>:<PORT>}</StreamProcessorServerURL>
   </Analytics>

3. Save the changes and restart the WSO2 Open Banking API Manager server.

Configuring databases

You can configure databases to persist, process and analyze API-related data. Follow the steps below to configure the databases.

1. Create the ANALYTICS_DB database in your database server.

2. Configure the worker profile.   

   Open the <WSO2_OB_API-M_ANALYTICS_HOME>/conf/worker/deployment.yaml file. Update the APIM_ANALYTICS_DB datasource with your database configurations.

   deployment.yaml

   - name: APIM_ANALYTICS_DB
   description: The Datasource used for APIM Analytics
   jndiConfig:
     name: jdbc/APIM_ANALYTICS_DB
   definition:
     type: RDBMS
     configuration:
       jdbcUrl: 'jdbc:<DB_TYPE>://<DB_HOST>:<DB_PORT>/ANALYTICS_DB?autoReconnect=true&useSSL=false'
       username: <DB_USERNAME>
       password: <DB_PASSWORD>
       driverClassName: <DB_DRIVER>
       maxPoolSize: 50
       idleTimeout: 60000
       connectionTestQuery: SELECT 1
       validationTimeout: 30000
       isAutoCommit: false

3. Configure the dashboard profile.   

   Open the <WSO2_OB_API-M_ANALYTICS_HOME>/conf/dashboard/deployment.yaml file. Update the APIM_ANALYTICS_DB datasource with your database configurations.

   deployment.yaml

   - name: APIM_ANALYTICS_DB
   description: The Datasource used for APIM Analytics
   jndiConfig:
     name: jdbc/APIM_ANALYTICS_DB
   definition:
     type: RDBMS
     configuration:
       jdbcUrl: 'jdbc:<DB_TYPE>://<DB_HOST>:<DB_PORT>/ANALYTICS_DB?autoReconnect=true&useSSL=false'
       username: <DB_USERNAME>
       password: <DB_PASSWORD>
       driverClassName: <DB_DRIVER>
       maxPoolSize: 50
       idleTimeout: 60000
       connectionTestQuery: SELECT 1
       validationTimeout: 30000
       isAutoCommit: false

   By default, the following datasource entries are configured to H2 databases. If required, you can create external databases and point these entries to them.

   Datasource Name	File
   BUSINESS_RULES_DB	<WSO2_OB_API-M_ANALYTICS_HOME>/conf/dashboard/deployment.yaml
   WSO2_PERMISSIONS_DB	<WSO2_OB_API-M_ANALYTICS_HOME>/conf/worker/deployment.yaml <WSO2_OB_API-M_ANALYTICS_HOME>/conf/dashboard/deployment.yaml

   The GEO_LOCATION_DATA datasource should be configured when you are using Geo-location based statistics

4. Integrate third-party products when configuring databases.

   WSO2 Stream Processor is an OSGi-based product. Therefore, when you integrate third party products such as Oracle and MS SQL with WSO2 API-M Analytics, you need to check whether the libraries you need to add are OSGi based. If they are not, you need to convert them to OSGi bundles before adding them to the <WSO2_OB_API-M_ANALYTICS_HOME>/lib directory. 

   To convert the jar files to OSGi bundles, follow the steps below.

   1. Download the non-OSGi jar for the required third party product, and save it in a preferred directory in your machine.

   2. Go to the <API-M_ANALYTICS_HOME>/bin directory. Run the command given below, to generate the converted file in the <API-M_ANALYTICS_HOME>/lib directory.

   ./jartobundle.sh <PATH_TO_NON-OSGi_JAR> ../lib

Configuring the Keystores

In the SSL handshake between the API Manager and API Manager Analytics servers, the client (i.e. API Manager) needs to verify the certificate presented by the server (i.e. API Manager Analytics). For this purpose, the client stores the trusted certificate of the server in the client-truststore.jks keystore.

If you use a custom keystore in API Manager and/or API Manager Analytics, import the public key certificate of API Manager Analytics into the client-truststore.jks file of the API Manager. To export the public key from the server and import it into the client's trust store, follow the steps given in Adding CA-signed certificates to keystores in the Administration Guide.

Starting the servers

1. Navigate to the <WSO2_OB_API-M_ANALYTICS_HOME>/bin directory, and give execution permissions to the worker.sh file.

2. Run the API-M Analytics worker.sh as follows:

   ./worker.sh

3. Navigate to the <WSO2_OB_API-M_ANALYTICS_HOME>/bin directory, and give execution permissions to the dashboard.sh file.

4. Run the API-M Analytics dashboard.sh as follows:

   ./dashboard.sh 

5. You can view analytics data as described below:
   1. Log in to the API Store using the credentials provided during signup.
   2. Click the Analytics tab.
   3. You can view API Usage, Top Users and Faulty Invocations details through the subtopics.
6. To observe data analysis based on geographical locations, see Configuring Geolocation Based Statistics.