Redundant Peering for Internet Access with BGP - Best Practices

Two BGP Deployment Options for Redundant Peering

Question

Customers connecting to a Service Provider for Internet access are intending to implement redundant peering.

The design requirements call for preferring a primary link for both ingress and egress traffic.

Secondary links should be used only during primary outages.

What two BGP deployment options will accomplish this design goal? (Choose two.)

Answers

Explanations

Click on the arrows to vote for the correct answer

A. B. C. D. E.

CE.

The design goal is to have a primary link for both ingress and egress traffic, and to use the secondary links only during primary link outages. This can be achieved using BGP (Border Gateway Protocol) deployment options.

Option A: On the router handling the secondary link, advertise routes with a MED value of 0. This option involves using the Multi-Exit Discriminator (MED) attribute in BGP. By setting the MED value to 0 for all routes advertised via the secondary link, the routes will be less preferred than the routes advertised via the primary link, which will have a higher MED value. This will ensure that traffic will be routed through the primary link as long as it is available. In the event of a primary link outage, the secondary link will become the preferred path due to the lower MED value.

Option B: On the router handling the primary link, set the weight for all incoming routes to be a value of 0. This option involves using the weight attribute in BGP. By setting the weight to 0 for all incoming routes on the router handling the primary link, the routes will not be used for forwarding traffic. This means that all traffic will be forwarded via the secondary link as long as the primary link is available. In the event of a primary link outage, the weight for the incoming routes on the primary link router will be changed to a non-zero value, and traffic will be forwarded via the primary link.

Option C: On the router handling the secondary link, advertise all routes with a longer AS-PATH value. This option involves manipulating the AS-PATH attribute in BGP. By advertising all routes with a longer AS-PATH value via the secondary link, the routes will be less preferred than the routes advertised via the primary link, which will have a shorter AS-PATH value. This will ensure that traffic will be routed through the primary link as long as it is available. In the event of a primary link outage, the secondary link will become the preferred path due to the longer AS-PATH value.

Option D: On the router handling the primary link, advertise all routes with a longer AS-PATH value. This option is similar to option C, but involves advertising all routes with a longer AS-PATH value via the primary link. This will ensure that traffic will be routed through the primary link as long as it is available. In the event of a primary link outage, the secondary link will become the preferred path due to the shorter AS-PATH value advertised via the secondary link.

Option E: On the router handling the secondary link, set the local preference for all incoming routes to be a value of 0. This option involves using the local preference attribute in BGP. By setting the local preference to 0 for all incoming routes on the router handling the secondary link, the routes will be less preferred than the routes advertised via the primary link, which will have a higher local preference value. This will ensure that traffic will be routed through the primary link as long as it is available. In the event of a primary link outage, the local preference for the incoming routes on the secondary link router will be changed to a non-zero value, and traffic will be forwarded via the secondary link.

In summary, options A and B are the correct options to achieve the design goal of preferring a primary link for both ingress and egress traffic, and using secondary links only during primary outages. Option A involves using the MED attribute, while option B involves using the weight attribute.