HAProxy SPOE Data

SPOE Data

To be able to use Darwin with HAProxy, you first need to use an HAProxy connector that will connect to Darwin via a Unix socket.

To do so, you can use the SPOA Data, which is a generic Stream Processing Offload Agent (documentation here) that takes any number of arguments, parses them, then sends them to Darwin in order to get a certitude score.

With this score obtained (named ip_score here), you can create ACLs to decide whether to block the request or not.

Here is an example of a configuration file:

[user_agent]

spoe-agent user_agent-agent
    messages check-user_agent

    option var-prefix user_agent

    timeout hello      100ms
    timeout idle       30s
    timeout processing 15ms

    use-backend user_agent-backend


spoe-message check-user_agent
    args var(proc.user_agent)
    event on-frontend-http-request