Failure Detection – how does it work?

There are several user definable parameters that determine how fast HAProxy will be able to detect that a server is not available.

As default, the interval between health checks for a server that is "up", transitionally "up or down" or not yet checked is 2 seconds. The interval is 5 seconds when the server is 100% down or unreachable.

A server will be considered down/unavailable after 3 consecutive unsuccessful health checks, and will be operational again after 2 successful health checks.