Skip to main content

Secure Random Number Generators should not output predictable values

Critical
securitycryptography

What is it?​

This practice addresses the proper use of secure random number generators (CSPRNG) to ensure their output remains unpredictable, which is critical for cryptographic security.

Why apply it?​

Using predictable seeds with secure random number generators compromises the security of cryptographic algorithms that depend on random numbers, potentially exposing sensitive data to attackers.

How to fix it?​

Do not set predictable seeds on CSPRNGs like SecureRandom. Use the default constructor or SecureRandom.getInstanceStrong() to ensure unpredictability.

Examples​

Example 1:​

Negative​

The negative example sets a predictable seed, which makes random number output predictable and insecure for cryptographic uses.

import java.security.SecureRandom;

public class InsecureExample {
    public static void main(String[] args) {
        SecureRandom sr = new SecureRandom();
        sr.setSeed(987654321L);  // Noncompliant
        byte[] bytes = new byte[20];
        sr.nextBytes(bytes);

        System.out.println("Generated insecure random bytes:");
        for (byte b : bytes) {
            System.out.format("%02X ", b);
        }
    }
}

Example 2:​

Positive​

The positive example uses SecureRandom properly by not setting a seed, allowing it to initialize with an unpredictable source of randomness.

import java.security.SecureRandom;

public class SecureExample {
    public static void main(String[] args) {
        SecureRandom sr = new SecureRandom(); 
        byte[] bytes = new byte[20];
        sr.nextBytes(bytes);

        System.out.println("Generated secure random bytes:");
        for (byte b : bytes) {
            System.out.format("%02X ", b);
        }
    }
}

Negative​

The negative example initializes SecureRandom with a byte array seed derived from a constant string, making the random output predictable.

import java.security.SecureRandom;
import java.nio.charset.StandardCharsets;

public class InsecureSeedExample {
    public static void main(String[] args) {
        byte[] seedBytes = "predictableseed".getBytes(StandardCharsets.UTF_8);
        SecureRandom sr = new SecureRandom(seedBytes);  // Noncompliant
        
        byte[] bytes = new byte[20];
        sr.nextBytes(bytes);

        System.out.println("Generated insecure seed random bytes:");
        for (byte b : bytes) {
            System.out.format("%02X ", b);
        }
    }
}

Example 3:​

Positive​

The positive example utilizes SecureRandom.getInstanceStrong() to ensure the use of a strong algorithm for random number generation.

import java.security.SecureRandom;

public class StrongSecureExample {
    public static void main(String[] args) throws Exception {
        SecureRandom sr = SecureRandom.getInstanceStrong();
        byte[] bytes = new byte[20];
        sr.nextBytes(bytes);

        System.out.println("Generated strong secure random bytes:");
        for (byte b : bytes) {
            System.out.format("%02X ", b);
        }
    }
}