In a perfect world we would never need redundancy on a network infrastructure, but as you know as well as I know we don’t live in a perfect world. Hardware will eventually fail, bottlenecks will appear, and the speed of our network will become slower when we max the bandwidth on links. So having redundancy in routers, connections, and having a hierarchy network is one of best choices to make when improving the efficiently of the network. But in the OSI model there is one layer that can’t have redundancy at least logically and that is Layer two of the OSI model.
If you follow this blog way back when I talked about the OSI model, layer two belongs to the switches and by design the ethernet frame does not support a TTL (Time-To-Live) like the IP packet. So bring forth STP, the Spanning Tree Protocol which works at layer two of the OSI model. If more than one path is available in the network the STP algorithm basically shutdowns the redundant links until active link is unavailable (shutdown, disabled, etc.) So although there are multiple links connected physically to offer redundancy, logically the switches in the network only see one path that ethernet frame can travel.
Wait…I have plugged into multiple Cisco switches before to create a redundant path and I did not notice anything wrong?
That’s because by default STP is enabled, to the average person they can plug away and create redundant paths with switches and cause no harm to the network. But if STP has been disabled on those switches or switches that don’t support STP (dumb switches), a Layer 2 loop can occur. A Layer 2 loop is a nightmare if this happens in a production network and can happen in seconds. Loops result in high CPU load on all switches caught in the loop. Because the same frames are constantly being forwarded back and forth between all switches in the loop, the CPU of the switch ends up having to process a lot of data. This slows down performance on the switch when legitimate traffic arrives. A host caught in a network loop is not accessible to other hosts on the network. Because the MAC address table is constantly changing with the updates from the broadcast frames, the switch does not know which port to send the unicast frames out to reach the destination. The unicast frames end up looping around the network as well. As more and more frames end up looping on the network, a broadcast storm occurs.
What makes this even more fun is if you try connecting to the switches to solve this problem, good luck :). Most of the time you have to physically turn off the switches to kill the broadcast storm and if it happened once expect it to happen again. To make things worse if any end-devices are connected to the switch ports that are experiencing a broadcast storm they too have a hard time keeping up with the mountain of broadcast frames being sent and sometimes malfunction. Usually you would have to disconnect the switch from the redundant links and apply STP to it and then re-connect the redundant links in order for your network to be broadcast storm free.
Although I’m just getting started I think by just introducing this topic will help you better understand it later, I will get into more of this STP magic and how this really helps us design a hierarchy network. Like always I hope this information was informative and if you can’t wait a simple search about STP will help. You can also go right to the source for more STP information at Cisco’s Support and Documentation. Have an idea of the next topic either ICND1 or ICND2 let me know about it.