This documentation is for WSO2 Stream Processor 4.2.0. View documentation for the latest release.

All docs This doc
Skip to end of metadata
Go to start of metadata

WSO2 SP uses several artifacts for its functionality including deployment configurations for tuning its operation as well as deployable artifacts for extending its functionality. In each of these scenarios, there can be situations where the data specified is of a sensitive nature e.g., access tokens, passwords, etc.

The following topics cover how sensitive data can be obfuscated in both scenarios to avoid scrutiny,

Securing sensitive data in deployment configurations

WSO2 SP offers the Cipher tool to encrypt sensitive data in deployment configurations. This tool works in conjunction with WSO2 Secure Vault to replace sensitive data that is in plain text with an alias. The actual value is then encrypted and securely stored in the SecureVault. At runtime, the actual value is retrieved from the alias and used. For more information, see WSO2 Secure Vault.

To encrypt sensitive data to avoid scrutiny, follow the procedure below:

  1. Open the <SP_HOME>/conf/<RUNTIME>/ file and enter the following information.
    1. Enter the required sensitive data element with the value in plain text as shown in the example below.
      wso2.sample.password1=plainText ABC@123 
    2. Enter the alias to be used in the required configuration file instead of the actual value of sensitive data you specified in the previous step as shown in the example below.
      password: ${sec:wso2.sample.password1}
  2. To encrypt the sensitive data element and store it in the secure vault, run the Cipher tool by issuing a command similar to the following example for the required profile (i.e., worker, manager, dashboard, or editor).
    sh <SP_HOME>/bin/ -runtime worker

Protecting sensitive data in Siddhi applications

A parameter named ref is used to secure sensitive information in Siddhi applications that are deployed in WSO2 SP.  For Siddhi applications that use storage technologies supported by Carbon Datasourcesit is also possible to use Carbon datasources instead of specifying the connection parameters directly on the Siddhi file.

Using the ref parameter

Siddhi 4.0 supports the ref parameter that enables the user to specify parts of their definition outside the Siddhi App. Extensions that support this functionality include:

  • Stores
  • Sources
  • Sinks

This method of securing sensitive data involves defining the store parameters required by a connection instance in the <SP_HOME>/conf/<RUNTIME>/deployment.yaml file, and referring to those from Siddhi applications via the ref parameter.


In the <SP_HOME>/conf/<RUNTIME>/deployment.yaml file, some connection parameters are defined for a store named store1 as follows:

	        name: 'store1'
            type: 'rdbms'
      			jdbc.url: 'jdbc:h2:./repository/database/ANALYTICS_EVENT_STORE'
	      		username: 'root'
    	  		password: ${sec:store1.password}
    	  		field.length='currentTime:100' 'org.h2.Driver'

The Siddhi application includes the following configuration:

define table SmartHomeTable (id string, value float, property bool, plugId int, householdId int, houseId int, currentTime string);

Here @Store(ref='store1') refers to store1 defined in the deployment.yaml file, and as a result, the properties defined for this store is applicable to the Siddhi application when it is connected to the store.

Using carbon datasources

Currently, Carbon Datasources only support relational datasource definitions. You can also define RDBMS Store artifacts using Carbon Datasources or JNDI instead of directly specifying the connection parameters. Then the datasource definitions defined in the <SP_HOME>/conf/<RUNTIME>/deployment.yaml file can be secured via the process described under Securing sensitive data in deployment configurations.

The following is an example of an RDBMS store definition using Carbon datasources for looking up connection parameters.

@Store(type='rdbms', datasource='MY_DS')
define table SmartHomeTable (id string, value float, property bool, plugId int, householdId int, houseId int, currentTime string);

This can also be done via JNDI lookup as shown below:

@Store(type='rdbms', jndi.resource='jdbc/MyDS')
define table SmartHomeTable (id string, value float, property bool, plugId int, householdId int, houseId int, currentTime string);
  • No labels