0Pricing
Spring Security 6 & JWT Authentication · レッスン

JWTトークン失効戦略

ブラックリストや短期間のトークンなど、漏洩したJWTやログアウト済みのJWTを失効させる方法を学びます。

「JWTトークン失効戦略」はCoddyKit上の無料Spring Security 6 & JWT Authenticationレッスンです。 これはレッスン2/4です。 下記で完全なレッスンを無料で読むことができます。その後、ブラウザ内の組み込みコードエディタと24時間対応のAIチューターでハンズオン演習できます。 これはSpring Security 6 & JWT Authentication学習パスの一部であり、ウェブとCoddyKitアプリ全体で進捗が同期されます。 Spring Security 6 & JWT Authenticationコースには全4レッスンが含まれています。

このレッスンの一部はまだ翻訳されておらず、英語で表示されています。

Why Revoke JWTs?

JSON Web Tokens (JWTs) are designed to be stateless, meaning the server doesn't need to store session information. While this offers great scalability, it presents a challenge: how do you invalidate a token before its natural expiration?

We need revocation for scenarios like:

  • User logout
  • Token compromise (e.g., stolen token)
  • User role change or account disablement

The Statelessness Challenge

A core principle of JWTs is that once issued and signed, they can be validated without needing to query a database or external service. This means a server doesn't inherently 'know' if a token has been logically invalidated.

To revoke a JWT, you must introduce a mechanism that re-introduces a form of state, allowing the server to check if a token is still considered valid.

Strategy 1: Short-Lived Tokens

The simplest and most fundamental defense against compromised JWTs is to make them short-lived. If an access token expires quickly (e.g., 5-15 minutes), the window of opportunity for an attacker using a stolen token is minimized.

This strategy often pairs with Refresh Tokens (covered in another lesson) to provide a smooth user experience without requiring frequent re-logins.

Strategy 2: Blacklisting Tokens

To achieve immediate revocation, a common strategy is blacklisting. A blacklist is a storage (like a database table or a high-speed cache like Redis) that holds the unique identifiers (JTI claims) of tokens that have been explicitly invalidated.

When a server receives a JWT, it first checks if the token's JTI is present in the blacklist. If it is, the token is rejected, even if it hasn't expired.

Implementing a Blacklist

For an effective blacklist:

  • Unique ID: Ensure each JWT has a unique JTI (JWT ID) claim.
  • Storage: Use a fast, persistent store (e.g., Redis, database table) to hold blacklisted JTIs.
  • Check: Every time a JWT is presented, validate its signature, then check if its JTI is in the blacklist.
  • Expiration: Blacklisted tokens should still have their expiry respected. The blacklist entry itself can also have a TTL (Time To Live) matching the token's original expiry, to prevent the list from growing indefinitely.

Simulating a Blacklist

Let's look at a simple Java example to understand the blacklisting concept. We'll simulate a token's unique ID and an in-memory blacklist. In a real application, the blacklist would be a persistent, distributed store like Redis.

Blacklist in Action

This code demonstrates how a token ID can be added to a conceptual blacklist and then checked for revocation. Run it to see the output.

import java.util.HashSet;
import java.util.Set;
import java.util.UUID;

public class Main {
    private static Set<String> revokedTokens = new HashSet<>();

    public static void main(String[] args) {
        String userTokenId = UUID.randomUUID().toString();
        System.out.println("User token ID: " + userTokenId);

        // Simulate token validation
        if (!isTokenRevoked(userTokenId)) {
            System.out.println("Token is valid (not revoked).");
        } else {
            System.out.println("Token is revoked!");
        }

        // User logs out or token is compromised
        revokeToken(userTokenId);
        System.out.println("\n--- Token has been revoked ---");

        // Try to validate again
        if (!isTokenRevoked(userTokenId)) {
            System.out.println("Token is valid (not revoked).");
        } else {
            System.out.println("Token is revoked!");
        }
    }

    public static void revokeToken(String tokenId) {
        revokedTokens.add(tokenId);
    }

    public static boolean isTokenRevoked(String tokenId) {
        return revokedTokens.contains(tokenId);
    }
}

Strategy 3: Whitelisting

An alternative, though less common for raw JWTs, is whitelisting. Instead of listing invalid tokens, you maintain a list of all currently valid tokens or session IDs.

When a request comes in, you check if the token/session ID is on this 'allow list'. If it's not present, it's considered invalid. This approach is more typical for traditional session management but can be adapted.

  • Every active session gets a unique ID.
  • Store these valid IDs in a database or cache.
  • Upon logout, remove the ID from the whitelist.

Comparing Strategies

Each strategy has its place:

  • Short-Lived Tokens: Essential for minimizing risk. Always combine with other strategies.
  • Blacklisting: Best for specific, immediate invalidation (e.g., logout, compromise). Introduces a state check per request.
  • Whitelisting: Useful when you need to manage a finite set of active sessions and invalidate many at once (e.g., user disabled, revoke all sessions). Requires state for *all* tokens, which can be a performance consideration for very high-traffic APIs.

Revocation Strategies Check

Which of the following are valid strategies or important considerations for revoking JSON Web Tokens (JWTs) before their natural expiration?

Lesson Summary

In this lesson, we explored how to tackle the challenge of revoking stateless JWTs. We learned that while JWTs are stateless by design, scenarios like user logout or token compromise necessitate invalidation.

Key strategies include:

  • Using short-lived tokens to minimize risk.
  • Implementing a blacklist to mark specific tokens as invalid using their JTI.
  • Considering whitelisting for managing active sessions, though less common for raw JWTs.

The most robust solutions often combine short expiry times with a blacklisting mechanism for immediate revocation needs.

よくある質問

「JWTトークン失効戦略」レッスンは無料ですか?

はい。「JWTトークン失効戦略」の完全なテキストはこのウェブで無料で読めます。インタラクティブに演習し(組み込みコードエディタと24時間対応のAIチューター)、Spring Security 6 & JWT Authenticationコースの残りをアンロックするには、CoddyKit PROにアップグレードしてください。 Spring Security 6 & JWT Authenticationコースには全4レッスンが含まれています。

「JWTトークン失効戦略」で何を学びますか?

ブラックリストや短期間のトークンなど、漏洩したJWTやログアウト済みのJWTを失効させる方法を学びます。 ブラウザで直接実行するハンズオンコードでSpring Security 6 & JWT Authenticationを演習し、24時間対応のAIチューターがレッスンを進める中での質問に答えます。

Spring Security 6 & JWT Authenticationを始めるのに経験は必要ですか?

事前経験は必要ありません。CoddyKitのSpring Security 6 & JWT Authenticationは初級者から上級者向けに構成されているため、ここから始めるか最初から始めて、自分のペースで進むことができます。 これはレッスン2/4です。

「JWTトークン失効戦略」レッスンにはどのくらい時間がかかりますか?

ほとんどのCoddyKitレッスンは約5~10分かかります。各レッスンはコンパクトでインタラクティブなので、着実に進歩し、ウェブとアプリ全体で正確に前回の場所から再開できます。

このSpring Security 6 & JWT Authenticationレッスンでコードを書いて実行できますか?

はい。すべてのSpring Security 6 & JWT Authenticationレッスンに組み込みコードエディタが含まれているため、ブラウザでリアルコードを書いて実行し、即座のAIフィードバックを取得できます。ローカル設定は不要です。

このコースのすべてのレッスン

  1. リフレッシュトークンの実装
  2. JWTトークン失効戦略
  3. 安全なトークン保存の実践
  4. 署名キーのローテーションとキー管理
← Spring Security 6 & JWT Authenticationに戻る