Anti-DDoS Advanced health checks identify the running status of backend servers, where abnormal servers will be isolated to reduce the impact on overall business availability.
ping
command.http_1xx
, http_2xx
, http_3xx
, http_4xx
and http_5xx
are returned. You can select http_1xx
and http_2xx
to indicate that the service is normal, while the unselected codes represent that the service is not working properly.Note:If only one real server IP is configured in a single forwarding rule, health checks will not be enabled. This feature is used when multiple real server IPs are configured.
Note:
- By default, layer-4 health check is enabled. We recommend you use the default values when you configure this feature.
- Under TCP protocol, the layer-4 health check determines if the port is connected, while under UDP protocol, it determines whether the port is reachable with the
ping
command.
Note:By default, layer-7 health check is disabled.
Layer-4 health check
Item | Description |
---|---|
Response timeout | Maximum timeout for each health check response. If the backend server does not respond correctly within the specified period, then the health check fails. |
Check interval | The interval at which a health check is performed. |
Unhealthy threshold | If the health check fails n times (n is the entered number) in a row, the healthy backend server will become unhealthy, and the status displayed in the console will be Abnormal. |
Healthy threshold | If the health check succeeds n times (n is the entered number) in a row, the unhealthy backend server will become healthy, but the status will not be displayed in the console. |
Layer-7 health check
Item | Description |
---|---|
Check interval | The interval at which a health check is performed. Default: 15 seconds. |
Unhealthy threshold | If the health check fails n times (n is the entered number) in a row, the healthy backend server will become unhealthy, and the status displayed in the console will be Abnormal. |
Healthy threshold | If the health check succeeds n times (n is the entered number) in a row, the unhealthy backend server will become healthy, but the status will not be displayed in the console. |
HTTP request method and URL | By default, the HEAD method is used and the backend server returns the message header only in the response. If the GET method is used, the complete message will be returned. The backend server must support both HEAD and GET methods. |
HTTP status code detection | It determines whether the HTTP status code is healthy. When the default configuration is used or no selection is made, http_1xx , http_2xx , http_3xx and http_4xx are returned. If non-default ones are returned, the HTTP status code is considered unhealthy, and you can modify the configuration. |
Was this page helpful?