Untrusted Java Deserialization in Apache OpenNLP

Executive Summary

This vulnerability is an untrusted Java deserialization issue in the Apache OpenNLP SvmDoccatModel component before version 3.0.0-M4. The method SvmDoccatModel.deserialize(InputStream) reads data from an attacker-controlled stream using java.io.ObjectInputStream without applying any ObjectInputFilter. This means that the entire object graph referenced in the stream is deserialized before any type casting occurs.

If a malicious Java deserialization gadget chain exists on the classpath of the application using this module, an attacker can craft a payload that executes arbitrary code when deserialized. Apache OpenNLP itself does not include such gadget chains, so the risk mainly applies to downstream applications that include vulnerable dependencies alongside the libsvm module.

Because the deserialize method is public and static, any caller can pass an untrusted stream directly to it, increasing the risk of exploitation.

Useful 0 Not Useful 0

Impact Analysis

The primary impact of this vulnerability is remote code execution (RCE) on processes that load SvmDoccatModel instances from untrusted or semi-trusted sources. An attacker who can supply a crafted serialized stream to the deserialize method may execute arbitrary code within the Java Virtual Machine hosting the application.

This can lead to full compromise of the affected system, including unauthorized access, data theft, or disruption of services.

Useful 0 Not Useful 0

Detection Guidance

Detection of this vulnerability involves identifying usage of the vulnerable Apache OpenNLP versions before 3.0.0-M4, specifically the libsvm document categorization module that includes the SvmDoccatModel.deserialize(InputStream) method.

You can check your system or application dependencies for Apache OpenNLP versions prior to 3.0.0-M4.

• Use dependency management tools to list versions, for example, in Maven: mvn dependency:tree | grep opennlp
• In Gradle: ./gradlew dependencies | grep opennlp
• Search your codebase for calls to SvmDoccatModel.deserialize() to identify if untrusted streams might be passed.

Network detection is difficult because the vulnerability is triggered by deserialization of crafted Java objects within the application. Monitoring for suspicious or unexpected serialized Java object streams being loaded by the application could help, but no specific commands are provided.

Useful 0 Not Useful 0

Mitigation Strategies

The primary mitigation is to upgrade Apache OpenNLP to version 3.0.0-M4 or later, where this vulnerability is fixed.

If upgrading immediately is not possible, treat all serialized SvmDoccatModel streams as untrusted input unless their provenance is verified.

Avoid invoking SvmDoccatModel.deserialize() on streams supplied by end users or fetched from third-party sources without integrity checks.

Useful 0 Not Useful 0

Compliance Impact

The provided context does not include any information about the impact of this vulnerability on compliance with common standards and regulations such as GDPR or HIPAA.

Useful 0 Not Useful 0