STP and Rapid PVST+ Basics

An IP packet carries a TTL that each router decrements until it reaches zero, which kills packets caught in a routing loop. An Ethernet frame has no such field. If a Layer 2 loop exists, a frame can circulate indefinitely — there is no mechanism in the frame itself to stop it.

Now combine that with the flooding behaviour from Lesson 1. A broadcast floods out both links of a redundant pair. Each copy reaches the other switch, which floods it again, back across both links. Copies multiply on every pass. Within seconds the links saturate, switch CPUs spike, and the MAC table flaps as the same source MAC keeps appearing on different ports on each lap. That is a broadcast storm, and it takes the entire Layer 2 network down.

Convergence follows five moves. Switches exchange Bridge Protocol Data Units and agree on one root bridge as the reference point. Every non-root switch then picks its single root port, its best path toward the root. Each link segment gets one designated port, the forwarding port onto that segment. Any remaining redundant port is put into blocking — the classic name — or discarding, the rapid name. If the topology changes, the tree recomputes and a standby port takes over. Classic 802.1D can take around 50 seconds to do this: max-age 20 seconds plus two 15-second forward-delay stages. Modern switches run the rapid version and converge in seconds.

The Bridge ID is the bridge priority followed by the switch's MAC address. The switch with the lowest Bridge ID becomes root. Priority is compared first — the range is 0 to 65535, the default is 32768, and it is configured in steps of 4096. Lower wins. If priorities tie, the MAC address breaks it, and again lower wins.

Cisco runs a separate spanning tree per VLAN, called PVST+ or Rapid PVST+. To keep Bridge IDs distinct per VLAN, the switch adds the VLAN ID to the base priority — the extended system ID — so the effective priority is normally the base plus the VLAN number. The practical consequence is useful: VLAN 10 and VLAN 20 can elect different roots and therefore block different physical ports, which lets you load-balance traffic across redundant links instead of leaving one idle.

1. Every port on the root bridge is designated and forwarding. The root never blocks — this alone answers many exam questions.
2. On each non-root switch, the root port is the port with the lowest cumulative path cost to the root.
3. On each segment, whichever end has the lower cost to the root owns the designated port.
4. The remaining port — the higher-cost side of the redundant link — blocks or discards.

Path cost is speed-based and cumulative: each link adds a cost according to its speed, and the switch sums them along the path to the root. Lower total wins.

VLAN0010
  Spanning tree enabled protocol rstp
  Root ID    Priority    24586
             Address     0011.1111.1111
             This bridge is the root
             Hello Time  2 sec  Max Age 20 sec  Forward Delay 15 sec

  Bridge ID  Priority    24586  (priority 24576 sys-id-ext 10)
             Address     0011.1111.1111

Interface        Role Sts Cost      Prio.Nbr Type
---------------- ---- --- --------- -------- --------------------
Gi0/1            Desg FWD 4         128.1    P2p
Gi0/2            Desg FWD 4         128.2    P2p

Read it top-down. The protocol line confirms rstp, meaning Rapid PVST+. Compare Root ID against Bridge ID: when they match and the output says this bridge is the root, this switch is the root for VLAN 10. On a non-root switch you would instead see the neighbour's address as Root ID, one interface with Role Root and status FWD, and the redundant interface with Role Altn and status BLK. The priority 24576 sys-id-ext 10 line shows the per-VLAN arithmetic: 24576 plus VLAN 10 gives the 24586 above it.

! Make THIS switch the root for VLAN 10 (lowers priority to 24576)
Switch(config)# spanning-tree vlan 10 root primary

! Equivalent explicit form
Switch(config)# spanning-tree vlan 10 priority 24576

root primary lowers the priority so this switch wins the election instead of leaving it to the lowest MAC. Do it per VLAN — that is exactly how you split VLAN 10 and VLAN 20 across different roots for load balancing. There is a matching root secondary for the backup core.

• Ethernet has no TTL, so a Layer 2 loop becomes a broadcast storm; STP blocks redundant paths to leave one active path.
• The root is the lowest Bridge ID — priority first, then MAC — and every port on the root bridge is designated and forwarding.
• Each non-root switch has exactly one root port, the lowest cumulative cost to the root; the leftover port blocks or discards.
• Cost is speed-based and cumulative: 1 Gbps is 4, 100 Mbps is 19.
• Ties break on sender Bridge ID, then sender port priority, then sender port ID.
• Rapid PVST+ is a fast per-VLAN tree with states discarding, learning, and forwarding.