Spring Cloud Gateway 中基于用户计划的速率限制

Question

假设我的用户订阅了一个计划。那么是否可以使用 Spring Cloud Gateway 根据订阅计划对用户请求进行速率限制？鉴于有 Silver 和 Gold 计划，是否允许 Silver 订阅具有 5/10 和 Gold 50/100 的补充率/burstCapacity？

我天真地想过将 RedisRateLimiter 的新实例（见下文，我用 5/10 设置构造一个新实例）传递给过滤器，但我需要以某种方式从请求中获取有关用户的信息，以便能够了解它是否是白银计划和黄金计划。

@Bean
public RouteLocator myRoutes(RouteLocatorBuilder builder) {
    return builder.routes()
        .route(p -> p
            .path("/get")
            .filters(f ->
              f.requestRateLimiter(r -> {
                  r.setRateLimiter(new RedisRateLimiter(5, 10))
              })
            .uri("http://httpbin.org:80"))
            .build();
}

我是否正在尝试通过 Spring Cloud Gateway 实现甚至可能实现的目标？如果有的话，你会推荐哪些其他产品来检查这个目的？

谢谢！

Answer 1

好的，可以通过在RedisRateLimiter 类之上创建自定义速率限制器。不幸的是，该类尚未针对可扩展性进行架构设计，因此该解决方案有点“hacky”，我只能装饰普通的RedisRateLimiter 并在其中复制它的一些代码：

@Primary
@Component
public class ApiKeyRateLimiter implements RateLimiter {

    private Log log = LogFactory.getLog(getClass());

    // How many requests per second do you want a user to be allowed to do?
    private static final int REPLENISH_RATE = 1;
    // How much bursting do you want to allow?
    private static final int BURST_CAPACITY = 1;

    private final RedisRateLimiter rateLimiter;
    private final RedisScript<List<Long>> script;
    private final ReactiveRedisTemplate<String, String> redisTemplate;

    @Autowired
    public ApiKeyRateLimiter(
        RedisRateLimiter rateLimiter,
        @Qualifier(RedisRateLimiter.REDIS_SCRIPT_NAME) RedisScript<List<Long>> script,
        ReactiveRedisTemplate<String, String> redisTemplate) {

        this.rateLimiter = rateLimiter;
        this.script = script;
        this.redisTemplate = redisTemplate;
    }

    // These two methods are the core of the rate limiter
    // Their purpose is to come up with a rate limits for given API KEY (or user ID)
    // It is up to implementor to return limits based up on the api key passed
    private int getBurstCapacity(String routeId, String apiKey) {
        return BURST_CAPACITY;
    }
    private int getReplenishRate(String routeId, String apiKey) {
        return REPLENISH_RATE;
    }

    public Mono<Response> isAllowed(String routeId, String apiKey) {

        int replenishRate = getReplenishRate(routeId, apiKey);
        int burstCapacity = getBurstCapacity(routeId, apiKey);

        try {
            List<String> keys = getKeys(apiKey);

            // The arguments to the LUA script. time() returns unixtime in seconds.
            List<String> scriptArgs = Arrays.asList(replenishRate + "", burstCapacity + "",
                Instant.now().getEpochSecond() + "", "1");
            Flux<List<Long>> flux = this.redisTemplate.execute(this.script, keys, scriptArgs);

            return flux.onErrorResume(throwable -> Flux.just(Arrays.asList(1L, -1L)))
                .reduce(new ArrayList<Long>(), (longs, l) -> {
                    longs.addAll(l);
                    return longs;
                }) .map(results -> {
                    boolean allowed = results.get(0) == 1L;
                    Long tokensLeft = results.get(1);

                    Response response = new Response(allowed, getHeaders(tokensLeft, replenishRate, burstCapacity));

                    if (log.isDebugEnabled()) {
                        log.debug("response: " + response);
                    }
                    return response;
                });
        }
        catch (Exception e) {
            /*
             * We don't want a hard dependency on Redis to allow traffic. Make sure to set
             * an alert so you know if this is happening too much. Stripe's observed
             * failure rate is 0.01%.
             */
            log.error("Error determining if user allowed from redis", e);
        }
        return Mono.just(new Response(true, getHeaders(-1L, replenishRate, burstCapacity)));
    }

    private static List<String> getKeys(String id) {
        String prefix = "request_rate_limiter.{" + id;
        String tokenKey = prefix + "}.tokens";
        String timestampKey = prefix + "}.timestamp";
        return Arrays.asList(tokenKey, timestampKey);
    }

    private HashMap<String, String> getHeaders(Long tokensLeft, Long replenish, Long burst) {
        HashMap<String, String> headers = new HashMap<>();
        headers.put(RedisRateLimiter.REMAINING_HEADER, tokensLeft.toString());
        headers.put(RedisRateLimiter.REPLENISH_RATE_HEADER, replenish.toString());
        headers.put(RedisRateLimiter.BURST_CAPACITY_HEADER, burst.toString());
        return headers;
    }

    @Override
    public Map getConfig() {
        return rateLimiter.getConfig();
    }

    @Override
    public Class getConfigClass() {
        return rateLimiter.getConfigClass();
    }

    @Override
    public Object newConfig() {
        return rateLimiter.newConfig();
    }
}

所以，路线应该是这样的：

@Component
public class Routes {

    @Autowired
    ApiKeyRateLimiter rateLimiter;

    @Autowired
    ApiKeyResolver apiKeyResolver;

    @Bean
    public RouteLocator theRoutes(RouteLocatorBuilder b) {
        return b.routes()
            .route(p -> p
                    .path("/unlimited")
                    .uri("http://httpbin.org:80/anything?route=unlimited")
            )
            .route(p -> p
                    .path("/limited")
                    .filters(f ->
                            f.requestRateLimiter(r -> {
                                r.setKeyResolver(apiKeyResolver);
                                r.setRateLimiter(rateLimiter);
                            } )
                    )
                    .uri("http://httpbin.org:80/anything?route=limited")
            )
            .build();
    }

}

希望它能为某人节省一个工作日......