Spanning Tree Protocol (STP) Explained | 2026

So what is Spanning Tree Protocol?

Spanning Tree Protocol, usually shortened to STP, is the system that keeps the network of switches inside a building from forming loops. A loop is simply more than one path between the same two points, and on a local network a loop is surprisingly dangerous.

It was invented by Radia Perlman back in 1985 and later written into an international standard, and the basic job hasn’t changed since: let a network have backup links for resilience, while making sure those backups never quietly form a loop that brings everything down.

You can think of it as a referee for your cabling. It looks at every possible path through the network, picks a sensible set that reaches everywhere without doubling back, and benches the rest until they’re needed.

Why a loop is so dangerous

Inside an office, switches move traffic around by flooding certain messages to every port until they learn where everything lives. If there are two paths between two switches, those messages can go round and round the loop, forever, multiplying as they go.

Within seconds you get what’s called a broadcast storm: the network fills with the same traffic echoing endlessly, the switches are overwhelmed, and real traffic can’t get through. To everyone in the building it looks like the whole network has simply died, all at once.

The cruel irony is that loops are usually created with the best of intentions. You add a second cable between two switches so there’s a backup if one fails. Without something watching over it, that act of caution is exactly what creates the loop.

How STP keeps the peace

STP’s solution is rather elegant. All the switches talk to each other and elect one of them as the reference point, the root. Each switch then works out its best path back to that root and quietly switches off any other path that would create a loop. The blocked links aren’t gone; they’re held in reserve.

The clever part is what happens when something fails. If the main link a switch was using goes down, STP notices, recalculates, and brings one of those reserved backup links into service. You get the resilience of having spare cabling without the danger of a loop. The wider internet runs on the same idea one level up, healing around failures out on the backbone. You plug in two paths for safety, and STP makes sure only one of them is ever live at a time.

Why this matters if you run a business

Like a lot of networking, STP is something you’ll never touch directly. Your switches handle it, and a well-built network has it working sensibly from day one. But it’s worth understanding for two opposite reasons.

The first is resilience. STP is the mechanism that lets a network have backup links at all. A properly designed network can lose a cable or a switch and keep running, because a spare path quietly takes over. That only works if the network was built with redundancy in mind, rather than as an afterthought.

The second is fragility. When STP is missing, misconfigured, or running on cheap unmanaged switches that don’t support it, a single careless loop can take everything down.

The hospital that went back to paper

The classic cautionary tale is Beth Israel Deaconess Medical Center in Boston. In November 2002 a single application began looping data across a network built almost entirely on simple switches that leaned heavily on spanning tree. The result was a network meltdown that came and went for the best part of four days. Staff fell back to paper records, lab results that normally took 45 minutes took five hours, and the emergency department had to divert patients. The network ended up being rebuilt from the ground up.

The everyday version

Most loops are far less dramatic. The common one is mundane: someone tidies a cupboard and plugs both ends of a spare cable into the same switch, or daisy-chains a little desk switch back into the wall. On a network with STP doing its job, that mistake is spotted and the offending path is blocked before anyone notices. On a network without it, the office goes dark until someone physically finds the cable.

You add a second cable for safety, and on a network without spanning tree, that very act of caution is what takes the whole office down.

The practical takeaway

You don’t need to learn spanning tree. You need a network that was set up by someone who has. In practice that comes down to a few things:

• Use proper managed switches. The cheapest unmanaged kind often don’t run spanning tree at all, which is precisely how an accidental loop turns into an outage.
• Design in redundancy on purpose. Backup links only help if the network was built to use them, with spanning tree configured to match, not bolted on later and hoped for.
• Switch on the modern safeguards. Newer versions and features, with names like RSTP, BPDU Guard and loop protection, make the network recover faster and stop a stray cable causing chaos. A good setup turns them on, and turns off the insecure ones, like automatic trunking.
• Keep an eye on it. Knowing that a backup link has taken over, or that something has started looping, before it spirals is the difference between a non-event and a very bad afternoon.

The short version

Spanning Tree Protocol is the referee that lets your network have backup cabling without those backups forming a loop that takes everything down. You’ll never configure it, but whether it’s set up properly is the difference between a spare link that quietly saves you and a single loose cable that floors the whole office.

The bottom line

Spanning Tree is one of those quietly brilliant pieces of plumbing that has been holding networks together since the 1980s. When it’s set up well you never think about it: links fail, backups take over, and nobody notices. When it’s missing or done badly, an ordinary mistake can take down everything at once.

The lesson for a business isn’t to understand the protocol. It’s to make sure the network underneath you was built properly, with real redundancy, sensible switches and the modern safeguards switched on, so that a single cable never gets to decide how your team’s afternoon goes.

If you’d like someone to take an honest look at how resilient your network actually is, that’s exactly the kind of thing we enjoy getting stuck into.