tmsnan

It appears that the functionality is incomplete and unsuitable for production use. Should we enable it? https://www.envoyproxy.io/docs/envoy/latest/api-v3/extensions/resource_monitors/downstream_connections/v3/downstream_connections.proto#extensions-resource-monitors-downstream-connections-v3-downstreamconnectionsconfig

@arkodg If possible, I could first add other overload manager features that are already fully supported.

@arkodg Maybe we can refer to Google VRP edge server configuration. ``` overload_manager: refresh_interval: 0.25s resource_monitors: - name: "envoy.resource_monitors.fixed_heap" typed_config: "@type": type.googleapis.com/envoy.extensions.resource_monitors.fixed_heap.v3.FixedHeapConfig # TODO: Tune for your system. max_heap_size_bytes: 2147483648...

@arkodg, Its comments may be ambiguous, requiring `clear_route_cache` to be set. https://github.com/envoyproxy/envoy/blob/1f1c32083150dced241336a6e2734e05bcead4b8/source/extensions/filters/http/jwt_authn/authenticator.cc#L380C1-L382C4

No changes needed @arkodg. Recalculating routes may cause performance issues. In some scenarios, only header or metadata information needs to be added without requiring rerouting, such as ratelimit-based JWT claims....

Vote for `SecurityPolicy`, access control is reasonable as a security policy.

Is it possible to add a disclaimer stating that the use of routing override is not recommended?

How to set the approximate fixed heap value of the system through a script, i need help.

> +1 apart from upstream keep alive, are there other knobs we should be setting for the `xds_cluster` ? I think keep alive is enough to ensure that the network...