TCP-based health check

Overview

Global Traffic Manager (GTM) provides the TCP-based health check feature that can be used to check the health status of IP addresses. TCP-based health checks are performed to monitor multiple metrics of IP addresses. These metrics include network accessibility, port availability, and latency. If a monitored IP address is abnormal, the system automatically blocks this IP address. If the IP address becomes normal, the system automatically unblocks the IP address.

Parameter description

1. Health Check Protocol

   The protocol that is used to monitor the metrics of IP addresses. These metrics include the network accessibility, packet loss rate, and latency. In this example, TCP is selected.

2. Health Check Interval

   The interval between two consecutive TCP-based health checks. Default value: 1 Minute.

3. Health Check Port

   The port on which health checks are performed. When health checks are performed on IP addresses over TCP, GTM checks whether the port of the IP addresses allows access over Telnet. If access over Telnet is allowed, the application service that is running on these IP addresses is deemed normal. If access over Telnet is not allowed, the application service is deemed abnormal.

4. Timeout Period

   The timeout period of a TCP-based health check. During a TCP-based health check, the system calculates the response time of each TCP packet that is sent. If the response time of a packet exceeds the specified timeout period, the health check times out. Valid values: 2 Seconds, 3 Seconds, 5 Seconds, and 10 Seconds.

5. Consecutive Failures

   The number of consecutive health checks that an IP address of an application service fails before the IP address is deemed abnormal. An application service is deemed abnormal only if an IP address of the application service consecutively fails health checks for the specified times. This prevents the accuracy of health checks from being affected by issues such as instantaneous network jitters. Valid values: 1 Request, 2 Requests, and 3 Requests.

• 1 Request: An application service is deemed abnormal when one health check determines that an IP address of the application service is abnormal. In this case, one alert is reported.

• 2 Requests: An application service is deemed abnormal when two health checks determine that an IP address of the application service is abnormal. In this case, two alerts are reported.

• 3 Requests: An application service is deemed abnormal when three health checks determine that an IP address of the application service is abnormal. In this case, three alerts are reported.

6. Failure Rate

   The ratio of health check nodes that detect exceptions to all health check nodes during a TCP-based health check. If the actual ratio exceeds the specified threshold, the application service is deemed abnormal. Valid values: 20%, 50%, 80%, and 100%.

7. Monitored Nodes

   The geographical location in which the nodes that perform TCP-based health checks reside. The following table describes the default nodes that the system provides.

   Node type	Geographical location

   Node type	Geographical location
   Border Gateway Protocol (BGP) nodes	Zhangjiakou, Qingdao, Hangzhou, Shanghai, Hohhot, Shenzhen, and Beijing
   Nodes outside the Chinese mainland	Hong Kong, Germany, Singapore, Silicon Valley, Australia, Malaysia, and Japan
   Internet service provider (ISP) nodes	China Unicom in Wuhan, China Unicom in Dalian, China Unicom in Nanjing, China Unicom in Tianjin, China Telecom in Qingdao, China Telecom in Changsha, China Telecom in Xi'an, China Telecom in Zhengzhou, China Mobile in Shenzhen, China Mobile in Dalian, and China Mobile in Nanjing

Procedure

1. Log on to the Alibaba Cloud DNS console.

2. In the left-side navigation pane, click Global Traffic Manager. On the Global Traffic Manager page, find the GTM instance that you want to manage, click its ID, and then perform the following operations:

• If you have not configured the GTM instance, configure the instance by referring to the Tutorial overview and Enable health check topics.

• If the instance is configured, click the Address Pools tab. On this tab, turn on Health Check and click Modify. In the Modify Health Check Setting dialog box, modify the parameters based on the description of the parameters.