转载

springboot之HTTP连接池

服务间基于HTTP通信相对于grpc、dubbo之类的通信效率要低得多，一方面是后者的传输数据结构紧凑，使用了序列化和压缩；另一方面，后者使用了TCP连接池，而前者默认情况下每一次服务间的通信会创建一个新的HTTP请求，会产生不小的性能消耗，对于需要额外非对称加密的HTTPS请求，性能消耗更加严重。

非连接池

默认情况下springboot的 RestTemplate 使用的 org.springframework.http.client.SimpleClientHttpRequestFactory ，即对每一次HTTP请求均新建一个新的TCP连接，请求结束后则关闭该连接。

RestTemplate restTemplate = new RestTemplate();
for(int i=0;i<100;i++) {
	ResponseEntity<String> responseEntity = restTemplate.getForEntity("https://wakzz.cn", String.class);
	System.out.println(responseEntity.getBody());
}
复制代码

通过抓包如下图，每次的HTTP请求都会断开上次的TCP连接，然后重新3次握手新建一个TCP连接。

连接池

HttpComponentsClientHttpRequestFactory 默认使用HTTP连接池 PoolingHttpClientConnectionManager ，其默认参数 maxTotal 为10即连接池最大HTTP连接数量为10， maxPerRoute 为2即连接池中相同目标IP和端口号的HTTP连接数量最多为2。

HttpComponentsClientHttpRequestFactory httpClientFactory = new HttpComponentsClientHttpRequestFactory();
httpClientFactory.setConnectTimeout(2000);
httpClientFactory.setReadTimeout(10000);

RestTemplate restTemplate = new RestTemplate(httpClientFactory);
for(int i=0;i<100;i++) {
	ResponseEntity<String> responseEntity = restTemplate.getForEntity("https://wakzz.cn", String.class);
	System.out.println(responseEntity.getBody());
}
复制代码

或者用户自定义 PoolingHttpClientConnectionManager 的参数：

PoolingHttpClientConnectionManager poolingConnectionManager = new PoolingHttpClientConnectionManager();
poolingConnectionManager.setMaxTotal(100);
poolingConnectionManager.setDefaultMaxPerRoute(10);

HttpClient httpClient = HttpClientBuilder.create()
		.setConnectionManager(poolingConnectionManager)
		.build();

HttpComponentsClientHttpRequestFactory httpClientFactory = new HttpComponentsClientHttpRequestFactory();
httpClientFactory.setConnectTimeout(2000);
httpClientFactory.setReadTimeout(10000);
httpClientFactory.setHttpClient(httpClient);

RestTemplate restTemplate = new RestTemplate(httpClientFactory);
for(int i=0;i<100;i++) {
	ResponseEntity<String> responseEntity = restTemplate.getForEntity("https://wakzz.cn", String.class);
	System.out.println(responseEntity.getBody());
}
复制代码

通过抓包如下图，每次的HTTP请求都复用的同一个TCP连接，无需在建立TCP连接上花费多余的性能消耗。

连接池连接状态：

源码分析

HttpComponentsClientHttpRequestFactory 缺省值情况下在 org.apache.http.impl.client.HttpClientBuilder#build 创建HTTP连接池：

final PoolingHttpClientConnectionManager poolingmgr = new PoolingHttpClientConnectionManager(
        RegistryBuilder.<ConnectionSocketFactory>create()
            .register("http", PlainConnectionSocketFactory.getSocketFactory())
            .register("https", sslSocketFactoryCopy)
            .build(),
        null,
        null,
        dnsResolver,
        connTimeToLive,
        connTimeToLiveTimeUnit != null ? connTimeToLiveTimeUnit : TimeUnit.MILLISECONDS);
if (defaultSocketConfig != null) {
    poolingmgr.setDefaultSocketConfig(defaultSocketConfig);
}
if (defaultConnectionConfig != null) {
    poolingmgr.setDefaultConnectionConfig(defaultConnectionConfig);
}
if (systemProperties) {
    String s = System.getProperty("http.keepAlive", "true");
    if ("true".equalsIgnoreCase(s)) {
        s = System.getProperty("http.maxConnections", "5");
        final int max = Integer.parseInt(s);
        poolingmgr.setDefaultMaxPerRoute(max);
        poolingmgr.setMaxTotal(2 * max);
    }
}
if (maxConnTotal > 0) {
    poolingmgr.setMaxTotal(maxConnTotal);
}
if (maxConnPerRoute > 0) {
    poolingmgr.setDefaultMaxPerRoute(maxConnPerRoute);
}
复制代码

org.apache.http.impl.execchain.MainClientExec#execute 从连接池获取HTTP连接：

final ConnectionRequest connRequest = connManager.requestConnection(route, userToken);
if (execAware != null) {
    if (execAware.isAborted()) {
        connRequest.cancel();
        throw new RequestAbortedException("Request aborted");
    } else {
        execAware.setCancellable(connRequest);
    }
}

final RequestConfig config = context.getRequestConfig();

final HttpClientConnection managedConn;
try {
    final int timeout = config.getConnectionRequestTimeout();
    managedConn = connRequest.get(timeout > 0 ? timeout : 0, TimeUnit.MILLISECONDS);
} catch(final InterruptedException interrupted) {
    Thread.currentThread().interrupt();
    throw new RequestAbortedException("Request aborted", interrupted);
} catch(final ExecutionException ex) {
    Throwable cause = ex.getCause();
    if (cause == null) {
        cause = ex;
    }
    throw new RequestAbortedException("Request execution failed", cause);
}
复制代码

每次发起HTTP请求时通过路由route（目标IP和端口号）从连接池获取链接，默认timeout为0即从连接池无限等待获取连接不超时。

原文 https://juejin.im/post/5d2c72cde51d4555fd20a3f6

正文到此结束