Exam 300-620 topic 1 question 226 discussion

C : https://community.cisco.com/t5/data-center-and-cloud-blogs/aci-multi-pods-caveats-and-considerations/ba-p/3837927

To ensure minimal traffic disruption during a planned or unplanned spine reload in a multi-pod ACI deployment, the correct action is: Configure all spines as Route Reflectors in the BGP Route Reflector Policy. This configuration ensures that all spines have a complete view of the network topology and can forward traffic to the appropriate destination, even if one or more spines are down or undergoing a reload. By distributing the routing information across multiple spines, the network becomes more resilient and can tolerate failures without significant impact.

I think that would be D, because the question says "traffic loss in case of pllaned or umpllaned spine reload", It say spine not spines. So both spines should be route reflectos.

Prior to 5.0(1): Fabric > Fabric Policies > Policies > Pod > ISIS Policy Default > ISIS metric for redistributed routes From 5.0(1): System > System Settings > ISIS Policy > ISIS metric for redistributed routes

This ensures that when new spine switch is introduced (booting up), or during upgrades/reboot of spine, the spine is not in the forwarding path to external destinations until the full configuration of the spine is completed and the default metric is changed to the lower metric (recommended to 32, default 63). (Fabric>Fabric Policies>Policies>POD>ISIS Policy default) *Traffic loss for Multi-Pod traffic after a Spine reboot (BugID CSCvd75131)

Very difficult question but option A would probably be the more viable answer for me, as the COOP database will help optimize the traffic. The BGP Route Reflector (answer D) will simply reflect the routes learned from boarder leaf's the rest of the fabric to the rest of the fabric. COOP database on the other hand will use the IPN's to optimise traffic internally. Anyway im not 100 % sure!