I'm building exactly the same topology, but there is no explanation why do I need two physical interfaces in IRF stack?
So, can someone please explain me, why do I need two physical members in IRF stack?
Network Engineering Stack Exchange is a question and answer site for network engineers. It only takes a minute to sign up.Sign up to join this community
The main point of redundancy concepts such as IRF is to avoid a single-point-of-failure.
For example, if the switch (2 switches in an IRF behave as if they were one switch) on the left broke down, or was rebooted, the link to the switch on the right would go down. To avoid this, you use 2 switches which share a common interface, management, setup etc. using IRF. The probability of 2 independent hardware devices to die at the same time is far lower than that of a single device. But...
you still have only one interconnect between the left IRF switch stack and the right switch. So, in order to make that link redundant as well, you create an LACP link where both member links are active. If now one of the IRF switches does down, the other can still send data over the remaining LACP member link.
To complete this setup, the right switch should also consist of a switch stack.
What about the hosts that are connected to the left switch ports? Those connected to a failing switch would lose connectivity. So, you connect hosts using 2 links ("dual-homed"), possibly in an LACP link. As this would increase the cost a bit, only important hosts such as servers are cross-connected to each switch stack member.