Configure Java 2 security

 

Protecting system resources and APIs (Java 2 security)

Java 2 security is a programming model that is very pervasive and has a huge impact on application development.

It is disabled by default, but is enabled automatically when global security is enabled. However, Java 2 security is orthogonal to Java 2 Platform, Enterprise Edition (J2EE) role-based security; you can disable or enable it independently of Global Security.

However, it does provide an extra level of access control protection on top of the J2EE role-based authorization. It particularly addresses the protection of system resources and application programming interfaces (API). Administrators need to consider the benefits against the risks of disabling Java 2 security. The following recommendations are provided to help enable Java 2 security in a test or production environment:

  1. Make sure the application is developed with the Java 2 security programming model. Developers have to know whether or not the APIs that are used in the
    applications are protected by Java 2 security. It is very important that the required permissions for the APIs used are declared in the policy file (was.policy), or the application fails to run when Java 2 security is enabled. Developers can reference the Web site for Development Kit APIs that are protected by Java 2 security. See the Programming model and decisions section of the Security: Resources for learning topic to visit this Web site.

  2. Make sure that migrated applications from previous releases are given the required permissions. Because Java 2 security is not supported or partially supported in previous WebSphere Application Server releases, applications developed prior to Version 5 most likely are not using the Java 2 security programming model. No easy way to find out all the required permissions for the application is available. The following are activities you can perform to determine the extra permissions that are required by an application:

    • Code review and code inspection

    • Application documentation review

    • Sandbox testing of migrated enterprise applications with Java 2 security enabled in a preproduction environment. Enable tracing in WebSphere Java 2 security manager to help determine the missing permissions in the application policy file. The trace specification is: com.ibm.ws.security.core.SecurityManager=all=enabled.

    • Use the com.ibm.websphere.java2secman.norethrow system property to aid debugging. Do not use this property in a production environment..

The default permission set for applications is the recommended permission set that is defined in the J2EE 1.3 Specification. The default is declared in the app_server_root/profiles/profile_name/config/cells/cell_name/nodes/node/app.policy policy file with permissions defined in the Development Kit (JAVA_HOME/jre/lib/security/java.policy) policy file that grant permissions to everyone. However, applications are denied permissions that are declared in the profiles/profile_name/config/cells/cell_name/filter.policy file. Permissions that are declared in the filter.policy file are filtered for applications during the permission check.

Define the required permissions for an application in a was.policy file and embed the was.policy file in the application enterprise archive (EAR) file as YOURAPP.ear/META-INF/was.policy, see Configuring Java 2 security policy files for details.

The following steps describe how to enforce Java 2 security on the cell level for WebSphere Application Server Network Deployment and the server level for WebSphere Application Server and WebSphere Application Server Express:

 

Procedure

  1. Click Security > Global security. The Global security panel is displayed.

  2. Select the Enforce Java 2 security option.

  3. Click OK or Apply.

  4. Click Save to save the changes.

  5. Restart the server for the changes to take effect.

 

Results

Java 2 security is enabled and enforced for the servers. Java 2 security permission is selected when a Java 2 security protected API is called. When to use Java 2 security

  1. Enable protection on system resources, for example when opening or listening to a socket connection, reading or writing to operating system file systems, reading or writing Java virtual machine system properties, and so on.

  2. Prevent application code from calling destructive APIs, for example, calling the System.exit method brings down the application server.

  3. Prevent application code from obtaining privileged information (passwords) or gaining extra privileges (obtaining server credentials).

 

What to do next

You can enforce Java 2 security on the server level for WebSphere Application Server Network Deployment by completing the following steps:

  1. Click Servers > Application servers > server_name.

  2. Under Security, click Server security.

  3. Under Additional properties, click Server-level security.

  4. Select the Enforce Java 2 security option.

  5. Click OK or Apply.

  6. Click Save to save the changes.

  7. Restart the server for the changes to take effect.
The Java 2 security manager is enhanced to dump the Java 2 security permissions that are granted to all classes on the call stack when an application is denied access to a resource. The java.security.AccessControlException exception is created. However, this tracing capability is disabled by default. You can enable this capability by specifying the server trace service with the com.ibm.ws.security.core.SecurityManager=all=enabled trace specification. When the exception is created, the trace dump provides hints to determine whether the application is missing permissions or the product runtime code or the third-party libraries that are used are not properly marked as privileged when accessing Java 2 security-protected resources. See the Security Problem Determination Guide for details.


 

Sub-topics


Using PolicyTool to edit policy files

Configuring Java 2 security policy files

Configuring static policy files

 

Related tasks


Migrating Java 2 security policy
Migrating, coexisting, and interoperating – Security considerations
Developing applications that use programmatic security