A MAC-VRF with a VLAN-aware service type uses a single bridging instance for the routing instance, so all VLANs configured under that MAC-VRF share the same bridge table while still maintaining per-VLAN separation for forwarding.

Design a five-stage Clos IP fabric: A five-stage Clos fabric architecture provides greater scalability and can support larger numbers of spine and leaf switches compared to a traditional three-stage Clos fabric, thus better addressing growth.
Use QFX5700 Series devices as the super spines: The QFX5700 series is a high-performance switch suitable for use as super-spines in large-scale data center IP fabrics. These devices offer high throughput and low latency, making them ideal for managing the large amount of traffic typically seen in high-growth environments. Using such devices for the super-spine layer will support scalability and performance as the data center grows.

In all-active multihoming, multiple PEs can forward known unicast traffic simultaneously, but only one PE, the Designated Forwarder, is responsible for forwarding broadcast, unknown unicast, and multicast (BUM) traffic to the multihomed segment. This prevents duplicate BUM frames from being delivered to the CE device over multiple links, since all attached PEs receive the same BUM traffic from the fabric but only the DF forwards it downstream.

Edge-routed bridging allows integrating devices that are not EVPN-VXLAN capable by extending Layer 2 at the edge while using routing toward the fabric, enabling those leaves to connect without participating in the EVPN control plane. A collapsed fabric design can place overlay functions on a subset of capable devices while keeping other leaf nodes operating as simpler routed/bridged nodes, allowing coexistence within the same fabric.

* Issue Overview:
* The leaf devices in an IP fabric using eBGP are advertising and receiving all routes, but the routes are not being installed in the routing table and are marked as hidden. Thistypically indicates an issue with the BGP configuration, particularly with next-hop handling or AS path concerns.
* Corrective Actions:
* B. You need to configure a next-hop self policy:This action ensures that the leaf devices modify the next-hop attribute to their own IP address before advertising routes to their peers. This is particularly important in eBGP setups where the next-hop may not be directly reachable by other peers.
* D. You need to configure multipath multiple-as:This setting allows the router to accept multiple paths from different autonomous systems (ASes) and use them for load balancing.
Without this, the BGP process might consider only one path and mark others as hidden.
* Incorrect Statements:
* A. You need to configure as-override:AS-override is used to replace the AS number in the AS- path attribute to prevent loop detection issues in MPLS VPNs, not in a typical eBGP IP fabric setup.
* C. You need to configure loops 2:There is no specific BGP command loops 2 relevant to resolving hidden routes in this context. It might be confused with allowas-in, which is used to allow AS path loops under certain conditions.
Data Center References:
* Proper BGP configuration is crucial in IP fabrics to ensure route propagation and to prevent routes from being marked as hidden. Configuration parameters like next-hop self and multipath multiple-as are common solutions to ensure optimal route installation and load balancing in a multi-vendor environment.