<p><strong><span style="font-size:18px">Do the HTTP/HTTPS listeners support the WS/WSS protocol?</span></strong></p> <p>Support by default and no configuration is required. HTTP listening supports the unencrypted version of WebSocket protocol (WS protocol) by default, and HTTPS listening supports the encrypted version of WebSocket protocol (WSS protocol) by default. The premise is that the connection between ELB and back-end server needs to use HTTP/1.1, and the Web server deployed on the back-end server also needs to support HTTP/1.1.</p> <p><strong><span style="font-size:18px">Which SSL versions does an HTTPS listener support?</span></strong></p> <p>SSLv2, SSLv3, TLSv1, TLSv1.1 and TLSv1.2 are supported.</p> <p><strong><span style="font-size:18px">What is the HTTP protocol version of HTTP/HTTPS listening accessing back-end server?</span></strong></p> <p>When the protocol requested by the client is HTTP/1.1 or HTTP2/0 version, the HTTP protocol version of seven-layer listening accessing the back-end server is http/1.1.</p> <p>When the version of the protocol requested by the client is other than HTTP/1.1 and HTTP 2/0, the HTTP protocol version of the seven-layer listening accessing back-end server is HTTP/1.0.</p> <p><strong><span style="font-size:18px">How to select health check mode for TCP monitoring?</span></strong></p> <p>TCP listening supports two health check methods: HTTP and TCP.</p> <p>TCP protocol health check detects whether the server port is alive by sending SYN handshake message.</p> <p>HTTP protocol health check tests whether the server application is healthy by sending GET request and simulating the browser&#39;s access behavior.</p> <p>TCP health check mode consumes less performance resources of the server. If you are highly sensitive to the load of the back-end server, select TCP health check; if the load is not very high, select HTTP health check.</p> <p><strong><span style="font-size:18px">Why cannot each connection reach the bandwidth peak?</span></strong></p> <p>Because the ELB system provides services for ELB instances through cluster deployment, all external access requests will be evenly distributed to these ELB system servers for forwarding. Therefore, the set bandwidth peak will be set on multiple system servers evenly.</p> <p><strong><span style="font-size:18px">What is the default connection timeout for each listener of ELB?</span></strong></p> <p>TCP listening: 900 seconds</p> <p>UDP listening: 90 seconds</p> <p>HTTP listening: 60 seconds</p> <p>HTTPS listening: 60 seconds</p> <p>&nbsp;</p>