Matter AI | Code Reviewer Documentation

Insecure Deserialization Overview

Unsafe Java Deserialization

// Anti-pattern: Unsafe Java deserialization
import java.io.ObjectInputStream;
import java.io.ByteArrayInputStream;
import java.util.Base64;

public class UserManager {
    public User loadUserFromToken(String token) {
        try {
            // Decode the base64 token
            byte[] serializedData = Base64.getDecoder().decode(token);
            
            // Deserialize without any validation
            ObjectInputStream ois = new ObjectInputStream(
                new ByteArrayInputStream(serializedData));
            
            // Unsafe deserialization of untrusted data
            return (User) ois.readObject();
        } catch (Exception e) {
            e.printStackTrace();
            return null;
        }
    }
}

// Better approach: Using a safer alternative
import com.fasterxml.jackson.databind.ObjectMapper;

public class UserManager {
    private final ObjectMapper objectMapper = new ObjectMapper();
    
    public User loadUserFromToken(String token) {
        try {
            // Decode the base64 token
            byte[] data = Base64.getDecoder().decode(token);
            String json = new String(data, "UTF-8");
            
            // Use JSON deserialization instead of native Java serialization
            return objectMapper.readValue(json, User.class);
        } catch (Exception e) {
            e.printStackTrace();
            return null;
        }
    }
}

// Even better: Using a whitelist approach with Jackson
import com.fasterxml.jackson.databind.ObjectMapper;
import com.fasterxml.jackson.databind.json.JsonMapper;
import com.fasterxml.jackson.databind.jsontype.BasicPolymorphicTypeValidator;

public class UserManager {
    private final ObjectMapper objectMapper;
    
    public UserManager() {
        // Configure Jackson with a whitelist of allowed classes
        BasicPolymorphicTypeValidator validator = BasicPolymorphicTypeValidator.builder()
            .allowIfBaseType(User.class)
            .allowIfSubType("com.example.model.")
            .build();
        
        objectMapper = JsonMapper.builder()
            .activateDefaultTyping(validator, ObjectMapper.DefaultTyping.NON_FINAL)
            .build();
    }
    
    public User loadUserFromToken(String token) {
        try {
            byte[] data = Base64.getDecoder().decode(token);
            String json = new String(data, "UTF-8");
            
            return objectMapper.readValue(json, User.class);
        } catch (Exception e) {
            e.printStackTrace();
            return null;
        }
    }
}

Java’s native serialization mechanism is particularly vulnerable to deserialization attacks, as it allows for arbitrary code execution during the deserialization process.

To prevent unsafe Java deserialization:

Avoid using Java’s native serialization (ObjectInputStream) for untrusted data
Use safer alternatives like JSON, XML, or YAML with appropriate libraries
If native serialization is necessary, implement a whitelist of allowed classes
Consider using serialization filtering in Java 9+ (ObjectInputFilter)
Keep all libraries and dependencies updated
Implement proper exception handling
Consider using the OWASP SerialKiller or other serialization security libraries

Unsafe PHP Deserialization

// Anti-pattern: Unsafe PHP deserialization
<?php
// User data stored in a cookie
$userData = $_COOKIE['user_data'];

// Unsafe deserialization of untrusted data
$user = unserialize($userData);

if ($user->isAdmin) {
    // Grant admin access
    showAdminPanel();
}
?>

// Better approach: Using safer alternatives
<?php
// Option 1: Using JSON instead of PHP serialization
$userData = $_COOKIE['user_data'];

// JSON decode with assoc=true to get an array instead of objects
$userData = json_decode($userData, true);

// Validate the data structure
if (!isset($userData['username']) || !isset($userData['role'])) {
    die('Invalid user data');
}

// Use the data safely
if ($userData['role'] === 'admin') {
    // Verify admin status from database, not just from the cookie
    if (verifyAdminStatus($userData['username'])) {
        showAdminPanel();
    }
}

// Option 2: If serialization is necessary, use a whitelist approach
function safeUnserialize($serialized) {
    // Define allowed classes
    $allowedClasses = ['UserData', 'UserPreferences'];
    
    // Set up a whitelist for allowed classes
    $options = ['allowed_classes' => $allowedClasses];
    
    // Unserialize with options (PHP 7+)
    return unserialize($serialized, $options);
}

$userData = $_COOKIE['user_data'];
$user = safeUnserialize($userData);
?>

PHP’s unserialize() function can lead to object injection vulnerabilities if used with untrusted data, potentially allowing attackers to instantiate arbitrary classes and trigger unexpected code execution.

To prevent unsafe PHP deserialization:

Avoid using unserialize() with untrusted data
Use JSON or other safer formats instead
If serialization is necessary, use the allowed_classes option in PHP 7+
Implement proper input validation
Consider using cryptographic signatures to verify serialized data integrity
Keep PHP and all libraries updated
Implement proper exception handling

Unsafe Python Deserialization

# Anti-pattern: Unsafe Python deserialization
import pickle
import base64

def load_user_preferences(token):
    # Decode the base64 token
    serialized_data = base64.b64decode(token)
    
    # Unsafe deserialization of untrusted data
    user_prefs = pickle.loads(serialized_data)
    
    return user_prefs

# Better approach: Using safer alternatives
import json
import base64

def load_user_preferences(token):
    try:
        # Decode the base64 token
        data = base64.b64decode(token)
        
        # Use JSON instead of pickle
        user_prefs = json.loads(data.decode('utf-8'))
        
        # Validate the structure
        if not isinstance(user_prefs, dict):
            raise ValueError("Invalid user preferences format")
        
        # Additional validation of expected fields
        required_fields = ['theme', 'language', 'notifications']
        for field in required_fields:
            if field not in user_prefs:
                raise ValueError(f"Missing required field: {field}")
        
        return user_prefs
    except Exception as e:
        # Log the error
        print(f"Error loading user preferences: {e}")
        # Return default preferences
        return {'theme': 'default', 'language': 'en', 'notifications': True}

# If complex objects are needed, use a safer serialization library
import marshmallow

class UserPreferencesSchema(marshmallow.Schema):
    theme = marshmallow.fields.String(required=True)
    language = marshmallow.fields.String(required=True)
    notifications = marshmallow.fields.Boolean(required=True)

def load_user_preferences_with_schema(token):
    try:
        data = base64.b64decode(token)
        json_data = json.loads(data.decode('utf-8'))
        
        # Use schema to validate and deserialize
        schema = UserPreferencesSchema()
        user_prefs = schema.load(json_data)
        
        return user_prefs
    except Exception as e:
        print(f"Error loading user preferences: {e}")
        return {'theme': 'default', 'language': 'en', 'notifications': True}

Python’s pickle module is notoriously dangerous for deserializing untrusted data, as it can execute arbitrary code during deserialization.

To prevent unsafe Python deserialization:

Never use pickle, marshal, or shelve modules with untrusted data
Use safer alternatives like JSON, YAML, or MessagePack
Implement proper input validation
Consider using schema validation libraries like marshmallow or pydantic
If complex object serialization is needed, consider libraries like jsonpickle with safe_mode enabled
Implement proper exception handling
Keep all dependencies updated

Unsafe Node.js Deserialization

// Anti-pattern: Unsafe Node.js deserialization
const serialize = require('node-serialize');

function getUserData(userToken) {
  // Decode the base64 token
  const serializedData = Buffer.from(userToken, 'base64').toString();
  
  // Unsafe deserialization of untrusted data
  return serialize.unserialize(serializedData);
}

// Better approach: Using safer alternatives
function getUserData(userToken) {
  try {
    // Decode the base64 token
    const data = Buffer.from(userToken, 'base64').toString();
    
    // Use JSON.parse instead of node-serialize
    const userData = JSON.parse(data);
    
    // Validate the structure
    if (!userData || typeof userData !== 'object') {
      throw new Error('Invalid user data format');
    }
    
    // Additional validation of expected fields
    const requiredFields = ['id', 'username', 'role'];
    for (const field of requiredFields) {
      if (!(field in userData)) {
        throw new Error(`Missing required field: ${field}`);
      }
    }
    
    return userData;
  } catch (error) {
    console.error('Error parsing user data:', error);
    return null;
  }
}

// If complex objects are needed, use a schema validation library
const Joi = require('joi');

function getUserDataWithSchema(userToken) {
  try {
    const data = Buffer.from(userToken, 'base64').toString();
    const userData = JSON.parse(data);
    
    // Define a schema for validation
    const userSchema = Joi.object({
      id: Joi.string().required(),
      username: Joi.string().alphanum().min(3).max(30).required(),
      role: Joi.string().valid('user', 'admin', 'moderator').required(),
      preferences: Joi.object({
        theme: Joi.string().valid('light', 'dark', 'system').default('system'),
        notifications: Joi.boolean().default(true)
      }).default()
    });
    
    // Validate against the schema
    const { error, value } = userSchema.validate(userData);
    
    if (error) {
      throw error;
    }
    
    return value;
  } catch (error) {
    console.error('Error parsing user data:', error);
    return null;
  }
}

Node.js applications can be vulnerable to deserialization attacks, particularly when using libraries like node-serialize that support serializing functions.

To prevent unsafe Node.js deserialization:

Avoid using libraries that support serializing functions (like node-serialize)
Use JSON.parse for deserialization of untrusted data
Implement proper input validation
Consider using schema validation libraries like Joi or Yup
Implement proper error handling
Keep all dependencies updated
Consider using Object.freeze() to prevent modification of deserialized objects

Deserialization of Untrusted Data in APIs

// Anti-pattern: Unsafe deserialization in API
const express = require('express');
const app = express();
const serialize = require('node-serialize');

app.use(express.text());

app.post('/api/process-data', (req, res) => {
  try {
    // Deserialize data from request body
    const data = serialize.unserialize(req.body);
    
    // Process the data
    processData(data);
    
    res.json({ success: true });
  } catch (error) {
    res.status(500).json({ error: error.message });
  }
});

// Better approach: Safe API deserialization
const express = require('express');
const app = express();
const Joi = require('joi');

// Use JSON middleware instead of text
app.use(express.json());

// Define a schema for validation
const dataSchema = Joi.object({
  id: Joi.string().required(),
  name: Joi.string().required(),
  values: Joi.array().items(Joi.number()).required(),
  metadata: Joi.object().pattern(Joi.string(), Joi.string()).optional()
});

app.post('/api/process-data', (req, res) => {
  try {
    // Validate request body against schema
    const { error, value } = dataSchema.validate(req.body);
    
    if (error) {
      return res.status(400).json({ error: error.message });
    }
    
    // Process the validated data
    processData(value);
    
    res.json({ success: true });
  } catch (error) {
    console.error('Error processing data:', error);
    res.status(500).json({ error: 'Internal server error' });
  }
});

APIs that accept serialized data from clients are particularly vulnerable to deserialization attacks, as they often process untrusted data from various sources.

To prevent deserialization vulnerabilities in APIs:

Use safe formats like JSON for data exchange
Implement proper input validation and schema validation
Set strict content type requirements
Implement proper error handling without exposing sensitive details
Use rate limiting to prevent DoS attacks via deserialization
Implement proper logging and monitoring
Keep all dependencies updated
Consider implementing API gateways with additional security controls

Deserialization with Gadget Chains

// Anti-pattern: Vulnerable to gadget chain attacks
import java.io.ObjectInputStream;
import java.io.ByteArrayInputStream;
import java.util.Base64;

public class DataProcessor {
    public void processData(String serializedData) {
        try {
            byte[] data = Base64.getDecoder().decode(serializedData);
            
            // Vulnerable to gadget chain attacks
            ObjectInputStream ois = new ObjectInputStream(
                new ByteArrayInputStream(data));
            
            Object obj = ois.readObject();
            processObject(obj);
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    
    private void processObject(Object obj) {
        // Process the deserialized object
    }
}

// Better approach: Using serialization filters
import java.io.ObjectInputStream;
import java.io.ByteArrayInputStream;
import java.io.InvalidClassException;
import java.util.Base64;
import java.io.ObjectInputFilter;

public class DataProcessor {
    public void processData(String serializedData) {
        try {
            byte[] data = Base64.getDecoder().decode(serializedData);
            
            ByteArrayInputStream bais = new ByteArrayInputStream(data);
            ObjectInputStream ois = new ObjectInputStream(bais);
            
            // Set up serialization filter (Java 9+)
            ObjectInputFilter filter = ObjectInputFilter.Config.createFilter(
                "com.example.model.*;java.util.ArrayList;java.util.HashMap");
            
            ois.setObjectInputFilter(filter);
            
            Object obj = ois.readObject();
            processObject(obj);
        } catch (InvalidClassException ice) {
            System.err.println("Security violation: " + ice.getMessage());
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    
    private void processObject(Object obj) {
        // Process the deserialized object
    }
}

Gadget chains are a particularly dangerous aspect of deserialization vulnerabilities, where attackers chain together multiple classes to achieve malicious outcomes during deserialization.

To prevent gadget chain attacks:

Avoid using native serialization mechanisms with untrusted data
Implement class whitelisting for deserialization
Use serialization filters (e.g., ObjectInputFilter in Java 9+)
Keep all libraries and dependencies updated
Consider using deserialization security libraries
Implement proper exception handling
Consider using runtime application self-protection (RASP) solutions

Deserialization Denial of Service

// Anti-pattern: Vulnerable to DoS via deserialization
import java.io.ObjectInputStream;
import java.io.ByteArrayInputStream;

public class MessageProcessor {
    public void processMessage(byte[] serializedData) {
        try {
            // No size limits or timeouts
            ObjectInputStream ois = new ObjectInputStream(
                new ByteArrayInputStream(serializedData));
            
            Object message = ois.readObject();
            handleMessage(message);
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
    
    private void handleMessage(Object message) {
        // Process the message
    }
}

// Better approach: Implementing protections against DoS
import java.io.ObjectInputStream;
import java.io.ByteArrayInputStream;
import java.util.concurrent.*;

public class MessageProcessor {
    private static final int MAX_DATA_SIZE = 1024 * 1024; // 1MB
    private static final int DESERIALIZATION_TIMEOUT = 5; // 5 seconds
    
    public void processMessage(byte[] serializedData) {
        // Check data size
        if (serializedData.length > MAX_DATA_SIZE) {
            System.err.println("Data too large: " + serializedData.length + " bytes");
            return;
        }
        
        ExecutorService executor = Executors.newSingleThreadExecutor();
        Future<Object> future = executor.submit(() -> {
            try {
                ObjectInputStream ois = new ObjectInputStream(
                    new ByteArrayInputStream(serializedData));
                
                // Set up serialization filter
                ois.setObjectInputFilter(createSerializationFilter());
                
                return ois.readObject();
            } catch (Exception e) {
                throw new RuntimeException(e);
            }
        });
        
        try {
            // Apply timeout to deserialization
            Object message = future.get(DESERIALIZATION_TIMEOUT, TimeUnit.SECONDS);
            handleMessage(message);
        } catch (TimeoutException e) {
            System.err.println("Deserialization timed out");
            future.cancel(true);
        } catch (Exception e) {
            System.err.println("Error during deserialization: " + e.getMessage());
        } finally {
            executor.shutdownNow();
        }
    }
    
    private java.io.ObjectInputFilter createSerializationFilter() {
        return java.io.ObjectInputFilter.Config.createFilter(
            "com.example.model.*;java.util.ArrayList;!*");
    }
    
    private void handleMessage(Object message) {
        // Process the message
    }
}

Deserialization can be exploited for denial of service attacks by crafting serialized data that consumes excessive resources during deserialization.

To prevent deserialization denial of service:

Implement size limits for serialized data
Use timeouts for deserialization operations
Implement resource limits (memory, CPU)
Consider processing deserialization in a separate process or container
Implement proper monitoring and alerting
Use rate limiting for endpoints that accept serialized data
Consider using more efficient serialization formats
Implement circuit breakers for critical systems

Secure Deserialization Patterns

// Pattern 1: Data Transfer Objects with manual deserialization
public class UserDTO {
    private String username;
    private String email;
    private String role;
    
    // Getters and setters
    
    // Factory method for safe deserialization
    public static UserDTO fromJson(String json) {
        try {
            // Use a JSON library
            ObjectMapper mapper = new ObjectMapper();
            JsonNode node = mapper.readTree(json);
            
            UserDTO dto = new UserDTO();
            
            // Manual extraction and validation
            if (node.has("username") && node.get("username").isTextual()) {
                dto.setUsername(node.get("username").asText());
            } else {
                throw new IllegalArgumentException("Invalid or missing username");
            }
            
            if (node.has("email") && node.get("email").isTextual()) {
                String email = node.get("email").asText();
                if (!isValidEmail(email)) {
                    throw new IllegalArgumentException("Invalid email format");
                }
                dto.setEmail(email);
            } else {
                throw new IllegalArgumentException("Invalid or missing email");
            }
            
            if (node.has("role") && node.get("role").isTextual()) {
                String role = node.get("role").asText();
                if (!isValidRole(role)) {
                    throw new IllegalArgumentException("Invalid role");
                }
                dto.setRole(role);
            } else {
                // Default role if not specified
                dto.setRole("user");
            }
            
            return dto;
        } catch (Exception e) {
            throw new IllegalArgumentException("Error deserializing user data", e);
        }
    }
    
    private static boolean isValidEmail(String email) {
        // Email validation logic
        return email.matches("^[A-Za-z0-9+_.-]+@(.+)$");
    }
    
    private static boolean isValidRole(String role) {
        // Role validation logic
        return Arrays.asList("user", "admin", "moderator").contains(role);
    }
}

// Pattern 2: Builder pattern with validation
public class User {
    private final String username;
    private final String email;
    private final String role;
    
    private User(Builder builder) {
        this.username = builder.username;
        this.email = builder.email;
        this.role = builder.role;
    }
    
    // Getters only, no setters for immutability
    
    public static class Builder {
        private String username;
        private String email;
        private String role = "user"; // Default value
        
        public Builder username(String username) {
            this.username = username;
            return this;
        }
        
        public Builder email(String email) {
            this.email = email;
            return this;
        }
        
        public Builder role(String role) {
            this.role = role;
            return this;
        }
        
        public User build() {
            // Validate before creating object
            if (username == null || username.isEmpty()) {
                throw new IllegalArgumentException("Username cannot be empty");
            }
            
            if (email == null || !email.matches("^[A-Za-z0-9+_.-]+@(.+)$")) {
                throw new IllegalArgumentException("Invalid email format");
            }
            
            if (!Arrays.asList("user", "admin", "moderator").contains(role)) {
                throw new IllegalArgumentException("Invalid role");
            }
            
            return new User(this);
        }
    }
    
    // Factory method for JSON deserialization
    public static User fromJson(String json) {
        try {
            ObjectMapper mapper = new ObjectMapper();
            JsonNode node = mapper.readTree(json);
            
            Builder builder = new Builder();
            
            if (node.has("username")) {
                builder.username(node.get("username").asText());
            }
            
            if (node.has("email")) {
                builder.email(node.get("email").asText());
            }
            
            if (node.has("role")) {
                builder.role(node.get("role").asText());
            }
            
            // Validation happens in build()
            return builder.build();
        } catch (Exception e) {
            throw new IllegalArgumentException("Error deserializing user data", e);
        }
    }
}

Implementing secure deserialization patterns can help mitigate the risks associated with deserializing untrusted data.

Key secure deserialization patterns:

Use Data Transfer Objects (DTOs) with manual deserialization
Implement the Builder pattern with validation
Use immutable objects to prevent post-deserialization manipulation
Implement factory methods for controlled object creation
Use schema validation libraries
Implement proper input validation
Consider using safer serialization formats
Implement proper exception handling

Insecure Deserialization Prevention Checklist

Insecure Deserialization Prevention Checklist:

1. Avoid Native Serialization
   - Avoid using native serialization mechanisms (Java ObjectInputStream, PHP unserialize(), Python pickle, etc.)
   - Use safer alternatives like JSON, XML, or YAML with appropriate libraries
   - If native serialization is necessary, implement strict controls

2. Input Validation
   - Validate all serialized data before deserialization
   - Implement schema validation
   - Use whitelisting instead of blacklisting
   - Validate data structure, types, and values

3. Implement Class Restrictions
   - Use whitelists of allowed classes for deserialization
   - Implement serialization filters
   - Avoid deserializing arbitrary classes

4. Integrity Verification
   - Implement cryptographic signatures for serialized data
   - Verify data integrity before deserialization
   - Use HMAC or digital signatures

5. Resource Protection
   - Implement size limits for serialized data
   - Use timeouts for deserialization operations
   - Implement resource limits (memory, CPU)
   - Consider processing in isolated environments

6. Secure Libraries and Dependencies
   - Keep all libraries and dependencies updated
   - Use security-focused serialization libraries
   - Regularly check for vulnerabilities in dependencies

7. Monitoring and Logging
   - Implement proper logging for deserialization operations
   - Monitor for suspicious deserialization attempts
   - Implement alerts for potential attacks

Preventing insecure deserialization requires a comprehensive approach that addresses multiple aspects of the deserialization process.

Key prevention strategies:

Use safer alternatives to native serialization
Implement proper input validation and integrity verification
Restrict which classes can be deserialized
Protect against resource exhaustion
Keep all dependencies updated
Implement proper monitoring and logging
Follow the principle of defense in depth

Introduction

Getting Started

Product

Patterns

Integrations

Enterprise

Insecure Deserialization Vulnerabilities