Lab 1 - Bot advanced config ########################### 1. Enabling Malicious User Detection is done at the **HTTP Load Balancer** level. In our example we will enable only **Spam Sources** and **Anonymous Proxies** a) Web App & API Protection -> Load Balancers -> HTTP Load Balancer -> Click the 3 dots under the **arcadia-re-lb** row -> Manage Configuration -> Edit Configuration -> Fill the bellow data .. table:: :widths: auto ========================================== ======================================================================================== Object Value ========================================== ======================================================================================== **Bot Defense** Enable **Bot Defense Region** EU ========================================== ======================================================================================== b) On the same place click **Configure** under **Bot Defense Policy** -> Configure -> App Endpoint Type -> Configure -> Add Item -> Fill the bellow data -> Apply -> Apply -> Apply -> Save and Exit .. table:: :widths: auto ========================================== ======================================================================================== Object Value ========================================== ======================================================================================== **Name** login **HTTP Methods** POST **Endpoint Label** Undefined **Prefix** /v1/login **Select Bot Mitigation action** Block ========================================== ======================================================================================== .. raw:: html