Scenario 3.4 (BONUS) - Hard-down active node w/ GRE KA's & BFD enabled | No ISP reth's
Physical removal of power from active cluster node (control-plane). Failure Scenario 3 - Pull power on active cluster member
BONUS scenario
Someone asked me if having the ISP links assigned to reths made a difference with failover times. The answer should be "no", but I realized I hadn't tried it out.
It turns out no, when you are pulling the power, it doesn't make much of a difference. So why do it? Well, in a scenario where you lose the control link, you want your interfaces pinned to a reth so that they go hard-down immediately. In a control-link failure scenario, the cluster that is active for redundancy-group 0 (control-plane) becomes active for ALL redundancy groups. This ensures that you do not have a split-brain scenario.
I'll show what control link failover looks like in a later video. If you're reading this and I haven't done this yet, remind me :)