RIP uses a hold-down timer to make sure a route advertised with a\nhigher metric is indeed valid. RIP will put a higher metric advertised\nroute into hold-down for default 180 seconds, meaning it won't put this\nhigher advertised metric into it's routing table for at least 180\nseconds as long as it keeps getting updates for the same metric.\n\nSo what exactly happens with RIP split horizon poisoned reverse? If\nthe router receives information that a network is unreachable through\nsplit horizon, and the route is already in it's table with a lower\nmetric, wouldn't this route be eventually be replaced with an\nunreachable value?\n\nWait...I just realized that this same router would obtain updates from\nanother router with the lower metric for this route, so in essence,\nthis route would be in hold-down for only as long as the next update is\nreceived advertising the lower metric.\n\nSo, one router could advertise unreachable and another router could\nadvertise 1, and the router inbetween would only be in hold-down for\nthat route until it received the lower metric advertisement. Every\ntime it received a split horizon unreachable on one interface it would\ngo into hold-down. Then when the new update with the lower metric\ncomes through the other interface the table is not updated, but the\nhold-down timer is reset.