Role Based Access Control (RBAC) Filter¶
The RBAC filter is used to authorize actions (permissions) by identified downstream clients (principals). This is useful to explicitly manage callers to an application and protect it from unexpected or forbidden agents. The filter supports configuration with either a safe-list (ALLOW) or block-list (DENY) set of policies based off properties of the connection (IPs, ports, SSL subject) as well as the incoming request’s HTTP headers. This filter also supports policy in both enforcement and shadow mode, shadow mode won’t effect real users, it is used to test that a new set of policies work before rolling out to production.
The RBAC filter configuration can be overridden or disabled on a per-route basis by providing a RBACPerRoute configuration on the virtual host, route, or weighted cluster.
The RBAC filter outputs statistics in the http.<stat_prefix>.rbac. namespace. The stat prefix comes from the owning HTTP connection manager.
|allowed||Counter||Total requests that were allowed access|
|denied||Counter||Total requests that were denied access|
|shadow_allowed||Counter||Total requests that would be allowed access by the filter’s shadow rules|
|shadow_denied||Counter||Total requests that would be denied access by the filter’s shadow rules|
The RBAC filter emits the following dynamic metadata.
|shadow_effective_policy_id||string||The effective shadow policy ID matching the action (if any).|
|shadow_engine_result||string||The engine result for the shadow rules (i.e. either allowed or denied).|