Switching in Detail

Switch Basics
Switching Process
MAC Table
VLAN (Virtual Local Area Network)
Trunking
STP (Spanning Tree Protocol)
Exercises

1. Switch Basics

What is a Switch?

A Layer 2 device that connects multiple devices and forwards frames using MAC addresses.

A network switch is a Layer 2 (Data Link Layer) device that connects devices within the same network and forwards data based on MAC addresses.

+-----------------------------------------------------------+
|                       SWITCH                               |
|                                                          |
|  +------+  +------+  +------+  +------+  +------+       |
|  |Port 1|  |Port 2|  |Port 3|  |Port 4|  |Port 5|       |
|  +--+---+  +--+---+  +--+---+  +--+---+  +--+---+       |
|     |         |         |         |         |            |
|  +--+--+   +--+--+   +--+--+   +--+--+   +--+--+       |
|  | PC  |   | PC  |   | PC  |   | PC  |   | PC  |       |
|  |  A  |   |  B  |   |  C  |   |  D  |   |  E  |       |
|  +-----+   +-----+   +-----+   +-----+   +-----+       |
+-----------------------------------------------------------+

Switch vs. Hub vs. Router

Comparing intelligence, forwarding behavior, and collision domains across devices.

Feature	Hub	Switch	Router
Layer	Layer 1 (Physical)	Layer 2 (Data Link)	Layer 3 (Network)
Decision Basis	None	MAC Address	IP Address
Forwarding	Broadcast to all ports	Selective to target port	Based on routing table
Collision Domain	1 (shared)	Per port	Per port
Broadcast Domain	1	1 (per VLAN)	Per interface
Intelligence	None	MAC learning	Routing protocols

Types of Switches

Different switch categories based on features, management, and use cases.

Type	Description	Use Case
Unmanaged	Plug-and-play, no configuration	Home/small office
Managed	Configurable, VLAN/STP/SNMP support	Enterprise networks
PoE	Power over Ethernet (powers devices via cable)	IP cameras, VoIP phones
Layer 3	Can route between VLANs (has IP routing)	Campus networks
Stackable	Multiple switches operate as one	Data centers

Switch Port Types

Access ports carry one VLAN; trunk ports carry multiple VLANs.

Port Type	Function
Access Port	Carries traffic for ONE VLAN (connects to end devices)
Trunk Port	Carries traffic for MULTIPLE VLANs (connects to other switches)

2. Switching Process

The Core Switching Steps

+--------------------------------------------------------------+
|                     SWITCHING PROCESS                        |
|                                                              |
|                      Frame Arrives                           |
|                           |                                  |
|                           v                                  |
|    +----------------------------------+                     |
|    |  1. LEARN                        |                     |
|    |     Source MAC -> Port, add to    |                     |
|    |     MAC Table                    |                     |
|    +--------------+-------------------+                     |
|                   v                                          |
|    +----------------------------------+                     |
|    |  2. LOOKUP                       |                     |
|    |     Destination MAC in MAC Table |                     |
|    +------+---------------+-----------+                     |
|           |               |                                  |
|           v               v                                  |
|    +-----------+    +--------------+                        |
|    | FOUND     |    | NOT FOUND     |                        |
|    | Forward   |    | Flood to all  |                        |
|    | to port   |    | ports (except |                        |
|    |           |    | source port)  |                        |
|    +-----------+    +--------------+                        |
+--------------------------------------------------------------+

Step-by-Step Example

Initial State: MAC Table is EMPTY

+----------+          +----------+          +----------+
|   PC A   |          |  SWITCH  |          |   PC B   |
| MAC: AA  |          |          |          | MAC: BB  |
| Port 1   |--------->|          |<---------| Port 2   |
+----------+          +----------+          +----------+

1. PC A sends frame to PC B
   +-------------------------------------------------+
   | Source MAC: AA | Dest MAC: BB | DATA           |
   +-------------------------------------------------+

2. Switch receives frame on Port 1
   -> LEARN: MAC AA -> Port 1  (added to MAC table)
   -> LOOKUP: MAC BB -> ?? (not in table)
   -> FLOOD: sends frame out ALL ports except Port 1
     (Port 2 receives it)

3. PC B sends reply to PC A
   +-------------------------------------------------+
   | Source MAC: BB | Dest MAC: AA | DATA           |
   +-------------------------------------------------+

4. Switch receives frame on Port 2
   -> LEARN: MAC BB -> Port 2  (added to MAC table)
   -> LOOKUP: MAC AA -> Port 1 (FOUND!)
   -> FORWARD: sends frame ONLY to Port 1

5. MAC Table is now populated:
   +------------+--------+
   |   MAC      | Port   |
   +------------+--------+
   | AA:AA:AA   |   1    |
   | BB:BB:BB   |   2    |
   +------------+--------+

Next time PC A sends to PC B -> forwards directly to Port 2

Switching Methods (Forwarding Modes)

Different techniques a switch uses to forward frames, balancing speed vs error checking.

Mode	How It Works	Latency	Error Checking
Store-and-Forward	Receives entire frame, checks CRC, then forwards	High	Full
Cut-Through	Reads dest MAC, forwards immediately (before full frame arrives)	Low	None
Fragment-Free	Reads first 64 bytes (checks for collisions), then forwards	Medium	Partial

+------------------------------------------------------------+
|                    FRAME RECEPTION                          |
|                                                          |
|  +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+      |
|  |Pr|SF|  |  |  |  |  |  |  |  |  |  |  |  |  |CRC|     |
|  |e |D |  |DST MAC|SRC MAC|  DATA   |FCS|  |  |   |     |
|  +--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+      |
|     ^            ^         ^                               |
|     |            |         |                               |
|  Cut-Through  Fragment  Store-and-Forward                  |
|  (reads dest)  (64 bytes)  (entire frame + CRC)           |
+------------------------------------------------------------+

Frame Forwarding Decision Table

Rules a switch follows to flood, forward, or drop a frame.

Condition	Action
Destination MAC = Broadcast (FF:FF:FF:FF:FF:FF)	Flood to all ports (except source)
Destination MAC = Multicast	Flood to interested ports
Destination MAC found in table	Forward to specific port
Destination MAC NOT found	Flood to all ports (except source)
Source MAC = Dest MAC	Drop (invalid)

3. MAC Table

What is the MAC Table?

The MAC table (also called CAM table -- Content Addressable Memory) is a database on a switch that maps MAC addresses to physical ports.

Anatomy of a MAC Table Entry

+--------------+----------+--------------+--------------+
|   MAC Address |  Port   |   VLAN       |   Type       |
+--------------+----------+--------------+--------------+
| AA:BB:CC:DD  |  Gi0/1  |   10         |   Dynamic    |
| EE:FF:00:11  |  Gi0/2  |   10         |   Dynamic    |
| 22:33:44:55  |  Gi0/3  |   20         |   Dynamic    |
| 66:77:88:99  |  Gi0/24 |   --         |   Static     |
+--------------+----------+--------------+--------------+

Field	Description
MAC Address	The learned or configured MAC address
Port	Which switch port the device is connected to
VLAN	Which VLAN the device belongs to
Type	How the entry was learned (Dynamic/Static)

MAC Entry Types

How MAC addresses are learned — dynamically, statically, or via port security.

Type	How Added	Description
Dynamic	Learned automatically	Expires after aging time (default 300s)
Static	Manually configured	Permanent entry
Secure	Port Security feature	Locked to specific MAC

MAC Table Aging

Dynamic MAC entries expire after a timeout period (default 300 seconds) to keep the table current.

+--------------------------------------------------+
|  Time     |  Event                               |
+-----------+--------------------------------------+
|  0s       | PC A sends frame -> MAC AA added      |
|  50s      | PC A sends another -> timer resets     |
|  100s     | PC A idle -- timer counting...         |
|  350s     | Still idle...                         |
|  400s     | Entry EXPIRED -> MAC AA removed        |
|  401s     | PC A sends frame -> MAC AA re-learned  |
+--------------------------------------------------+

Default aging time: 300 seconds (5 minutes) on most switches. Can be configured between 10 and 1,000,000 seconds.

Viewing the MAC Table

Commands to inspect MAC address entries on Cisco switches and Linux.

Cisco Switch

show mac address-table
show mac address-table dynamic
show mac address-table static

Linux (as a switch)

bridge fdb show

Sample Output

Mac Address Table
Vlan    Mac Address       Type        Ports
----    -----------       ----        -----
  10    aabb.cc01.0101    DYNAMIC     Gi0/1
  10    aabb.cc02.0202    DYNAMIC     Gi0/2
  20    aabb.cc03.0303    DYNAMIC     Gi0/3
  20    aabb.cc04.0404    DYNAMIC     Gi0/4
Total Mac Addresses for this criterion: 4

Why MAC Table Size Matters

A full MAC table forces flooding of unknown frames, reducing performance and enabling attacks.

Switches have limited CAM space (varies by model: 8K to 1M entries)
When MAC table is full, new entries cannot be learned
Frames to unknown MACs get flooded (reduces performance)
MAC flooding attacks exploit this by filling the table intentionally

MAC Table vs. Routing Table

Side-by-side comparison of Layer 2 switching and Layer 3 routing databases.

Feature	MAC Table (Switch)	Routing Table (Router)
Layer	Layer 2	Layer 3
Key	MAC Address	IP Network
Learns from	Incoming frames	Routing protocols / config
Scope	Single broadcast domain	Entire internetwork
Timeout	Yes (aging timer)	No (unless dynamic route fails)

4. VLAN (Virtual Local Area Network)

What is a VLAN?

A VLAN (Virtual LAN) is a logical subdivision of a physical network. It allows you to group devices together as if they were on the same physical network, even if they aren't.

+-----------------------------------------------------+
|  WITHOUT VLANS                                       |
|                                                       |
|  +-----------------+  +-----------------+           |
|  |  Sales Dept     |  |  Engineering    |           |
|  |  10 PCs         |  |  10 PCs         |           |
|  +--------+--------+  +--------+--------+           |
|           |                    |                      |
|           +----------+---------+                      |
|                      |                                |
|              +-------+--------+                       |
|              |    SWITCH      |   ALL IN ONE          |
|              |   (Flat LAN)   |   BROADCAST DOMAIN    |
|              +----------------+                       |
|                                                       |
|  X All traffic is visible to everyone                |
|  X No isolation between departments                  |
|  X Broadcasts reach ALL devices                      |
+-----------------------------------------------------+

+-----------------------------------------------------+
|  WITH VLANS                                          |
|                                                       |
|  +-----------------+  +-----------------+           |
|  |  VLAN 10        |  |  VLAN 20        |           |
|  |  Sales Dept     |  |  Engineering    |           |
|  |  10 PCs         |  |  10 PCs         |           |
|  +--------+--------+  +--------+--------+           |
|           |                    |                      |
|           +----------+---------+                      |
|                      |                                |
|              +-------+--------+                       |
|              |    SWITCH      |   TWO SEPARATE        |
|              |  VLAN 10 & 20  |   BROADCAST DOMAINS   |
|              +----------------+                       |
|                                                       |
|  Sales traffic stays in VLAN 10                      |
|  Engineering traffic stays in VLAN 20                |
|  Broadcasts isolated per VLAN                        |
|  ONE switch serving BOTH departments                 |
+-----------------------------------------------------+

Why Use VLANs?

VLANs improve security, reduce broadcast traffic, and add flexibility to network design.

Benefit	Explanation
Security	Isolate sensitive departments (HR, Finance)
Broadcast Control	Reduce broadcast traffic (smaller broadcast domains)
Performance	Less unnecessary traffic reaching devices
Flexibility	Group users by function, not physical location
Simplified Management	Move a device by changing its VLAN, not its cable
Cost Savings	Fewer switches needed (one switch = many VLANs)

VLAN Ranges

VLAN IDs are divided into ranges — normal (1-1005) and extended (1006-4094).

VLAN Range	Type	Purpose
0, 4095	Reserved	Cannot be used
1	Default	Default VLAN for all ports (cannot be deleted)
2-1001	Normal	Standard user VLANs
1002-1005	Reserved	Token Ring / FDDI (legacy)
1006-4094	Extended	Used in service provider networks

VLAN Configuration

Creating a VLAN on a Cisco Switch

! Create VLANs
vlan 10
 name SALES

vlan 20
 name ENGINEERING

! Assign access ports to VLANs
interface Gi0/1
 switchport mode access
 switchport access vlan 10

interface Gi0/2
 switchport mode access
 switchport access vlan 20

Verification Commands

show vlan brief
show vlan id 10
show interfaces status

Sample Output

VLAN Name                             Status    Ports
---- -------------------------------- -------- ---------------------
1    default                          active    Gi0/3, Gi0/4, Gi0/5
10   SALES                            active    Gi0/1
20   ENGINEERING                      active    Gi0/2
1002 fddi-default                     act/unsup
1003 token-ring-default               act/unsup
1004 fddinet-default                  act/unsup
1005 trnet-default                    act/unsup

VLAN Communication Rules

Devices in the same VLAN can communicate directly; different VLANs require a Layer 3 device.

+---------------------------------------------------------+
|                    VLAN 10 (SALES)                       |
|                                                          |
|   PC-A -------> PC-B    Same VLAN -> Direct             |
|   PC-A -------> PC-C    Same VLAN -> Direct             |
+---------------------------------------------------------+
+---------------------------------------------------------+
|                    VLAN 20 (ENGINEERING)                 |
|                                                          |
|   PC-D -------> PC-E    Same VLAN -> Direct             |
|   PC-D -------> PC-F    Same VLAN -> Direct             |
+---------------------------------------------------------+

VLAN 10 -------> VLAN 20    Different VLANs
    ^                            Need a Layer 3 device
    +----- Router / Layer 3 Switch ---- to route between them

VLANs are isolated by default. Devices in different VLANs cannot communicate unless traffic passes through a router or Layer 3 switch (VLAN routing / SVI).

Inter-VLAN Routing

Routing between different VLANs using a router or Layer 3 switch (router-on-a-stick or SVI).

+-----------------------------------------------------------+
|             INTER-VLAN ROUTING                             |
|                                                       |
|                    +----------+                       |
|                    |  Router  |  "Router-on-a-stick"  |
|                    |          |                       |
|                    |  .1      |                       |
|                    +----+-----+                       |
|                         | Trunk                        |
|                         |                              |
|  +----------------------+---------------------------+  |
|  |                   SWITCH                       |  |
|  |           VLAN 10 = Sales                      |  |
|  |           VLAN 20 = Engineering                |  |
|  +------+---------------------------+-------------+  |
|         |                           |                  |
|    +----+----+                 +----+----+           |
|    |  PC-A   |                 |  PC-B   |           |
|    | VLAN 10 |                 | VLAN 20 |           |
|    | 10.0.0.2|                 | 20.0.0.2|           |
|    +---------+                 +---------+           |
|                                                       |
|  PC-A -> PC-B (10.0.0.2 -> 20.0.0.2)                   |
|  1. PC-A sends to default gateway (10.0.0.1)          |
|  2. Router receives on sub-interface VLAN 10          |
|  3. Router checks routing table -> forwards via VLAN 20|
|  4. Switch delivers to PC-B                           |
+-----------------------------------------------------------+

VLAN Design Best Practices

Guidelines for planning and managing VLANs effectively in enterprise networks.

Practice	Why
Use descriptive VLAN names	`SALES` not `VLAN 10` -- easier to manage
Don't use VLAN 1 for user traffic	It's the default -- security risk
Create a separate management VLAN	Keep switch management isolated
Prune unused VLANs on trunks	Reduce unnecessary broadcast traffic
Use consistent VLAN numbering	Plan ahead (e.g., VLAN 10 = Sales everywhere)
Document VLAN to subnet mapping	Essential for troubleshooting

5. Trunking

What is Trunking?

Trunking allows a single link between switches to carry traffic for multiple VLANs. Without trunking, you'd need a separate physical link for each VLAN.

Trunking vs. Access Ports

Trunk ports carry multiple VLANs on a single link; access ports carry only one VLAN.

|  WITHOUT TRUNKING                                         |
|                                                            |
|  Switch A                  Switch B                       |
|  +--------+               +--------+                      |
|  | VLAN 10|---------------| VLAN 10| 1 cable per VLAN     |
|  | VLAN 20|---------------| VLAN 20| Expensive / messy    |
|  +--------+               +--------+                      |
|                                                            |
|  WITH TRUNKING                                             |
|                                                            |
|  Switch A                  Switch B                       |
|  +--------+               +--------+                      |
|  | VLAN 10|               | VLAN 10|                       |
|  | VLAN 20|----TRUNK----->| VLAN 20| 1 cable carries ALL  |
|  | VLAN 30|   802.1Q      | VLAN 30| Economical           |
|  +--------+               +--------+                      |

How Trunking Works -- 802.1Q Tagging

A 4-byte VLAN tag is inserted into the Ethernet frame to identify which VLAN the traffic belongs to.

Each frame traveling on a trunk gets a VLAN tag inserted in the Ethernet header:

+---------------------------------------------------------------+
|                    STANDARD ETHERNET FRAME                     |
|                                                              |
|  +--------+--------+--------+----------+--------------+       |
|  |  Dest  |  Src   |  Type  |  DATA    |     FCS      |       |
|  |  MAC   |  MAC   | /Len   |          |   (CRC)      |       |
|  +--------+--------+--------+----------+--------------+       |
|                                                              |
|                   802.1Q TAGGED FRAME                        |
|                                                              |
|  +--------+--------+-----+--+--------+----------+------+     |
|  |  Dest  |  Src   |Tag|TP|  Type   |  DATA    |  FCS |     |
|  |  MAC   |  MAC   |    |ID| /Len    |          |      |     |
|  +--------+--------+-----+--+--------+----------+------+     |
|                        ^    ^                                |
|                   Tag Proto   VLAN ID (12 bits)              |
|                  0x8100       0-4095                         |
+---------------------------------------------------------------+

802.1Q Tag Fields:

TPID (Tag Protocol Identifier): 0x8100 -- marks frame as tagged
Priority: 3 bits (802.1p QoS)
CFI (Canonical Format Indicator): 1 bit
VID (VLAN ID): 12 bits -- identifies the VLAN (0-4095)

Native VLAN

The one VLAN on a trunk that travels untagged — both sides must agree on its ID.

The Native VLAN is the one VLAN that traverses a trunk without a tag. By default, it's VLAN 1.

+-----------------------------------------------------------+
|                     TRUNK LINK                             |
|                                                          |
|  Switch A <========================> Switch B          |
|                                                          |
|  Tagged:  VLAN 10, 20, 30 ---- tagged frames            |
|  Untagged: VLAN 1 (Native)  ---- untagged frames         |
|                                                          |
|  Both switches must agree on Native VLAN!               |
|  Mismatch = native VLAN mismatch error                   |
+-----------------------------------------------------------+

Security best practice: Change the native VLAN to an unused VLAN (never VLAN 1) to prevent "VLAN hopping" attacks.

Trunking Configuration

Cisco Switch

! Configure trunk on an interface
interface Gi0/24
 switchport mode trunk
 switchport trunk native vlan 999
 switchport trunk allowed vlan 10,20,30

Allowed VLANs on Trunk

! Allow only specific VLANs
switchport trunk allowed vlan 10,20,30

! Add a VLAN to existing list
switchport trunk allowed vlan add 40

! Remove a VLAN
switchport trunk allowed vlan remove 20

! Allow all VLANs
switchport trunk allowed vlan all

Verification

show interfaces trunk
show interfaces Gi0/24 switchport

Sample Output

Port        Mode         Encapsulation  Status        Native vlan
Gi0/24      on           802.1q         trunking      999

Port        Vlans allowed on trunk
Gi0/24      10,20,30

Port        Vlans active and in the spanning tree
Gi0/24      10,20,30

DTP (Dynamic Trunking Protocol)

Cisco proprietary protocol that automatically negotiates trunk links between switches.

Cisco switches support automatic trunk negotiation via DTP:

Mode	Behavior
switchport mode access	Forces access mode (never trunks)
switchport mode dynamic desirable	Actively tries to form trunk with neighbor
switchport mode dynamic auto	Passive -- becomes trunk if neighbor asks
switchport mode trunk	Forces trunk mode regardless of neighbor

Security best practice: Set unused ports to switchport mode access to prevent trunk negotiation with unauthorized devices.

6. STP (Spanning Tree Protocol)

What is STP?

STP (IEEE 802.1D) prevents bridging loops in networks with redundant paths. It creates a loop-free logical topology by blocking certain ports.

The Problem: Bridging Loops

Redundant links without STP cause broadcast storms, MAC table instability, and duplicate frames.

Without STP, redundant links cause catastrophic problems:

+-----------------------------------------------------------+
|                  BRIDGING LOOP                             |
|                                                       |
|                   +------+                            |
|         +---------| SW-A |----------+                |
|         |         +------+          |                |
|         |           |               |                |
|         |           |               |                |
|         |     Redundant Links       |                |
|         |           |               |                |
|         |           v               |                |
|         |         +------+          |                |
|         +---------| SW-B |----------+                |
|                   +------+                            |
|                                                       |
|  Frame from PC-A -> Broadcast -> Loop!                |
|  SW-A sends to SW-B -> SW-B sends to SW-A -> infinity|
|  Broadcast storm -> Network collapse!                 |
+-----------------------------------------------------------+

Three Deadly Problems from Loops

Problem	Description
Broadcast Storm	Broadcasts circulate infinitely, consuming all bandwidth
MAC Table Instability	Switch sees the same MAC on different ports repeatedly
Multiple Frame Copies	Destination receives duplicate frames (protocols break)

How STP Solves This

STP places redundant links in a blocking state so there's only one active path:

+-----------------------------------------------------------+
|                   STP SOLUTION                             |
|                                                       |
|                     +------+                          |
|   ROOT BRIDGE ---->| SW-A |<-- Designated Port      |
|                     +---+--+                          |
|                         |  Forwarding                  |
|                         |                              |
|                    +----+----+                         |
|                    |         |                         |
|                    v         v                         |
|                  +------+   +------+                   |
|                  | SW-B |   | SW-C |                   |
|                  +--+---+   +---+---+                   |
|                     |          |                        |
|                 BLOCKED    Forwarding                   |
|                 (Alternate)                             |
|                                                       |
|  SW-B's port to SW-C is BLOCKED -> No loop             |
|  If primary link fails -> STP unblocks backup          |
|  Convergence: ~30-50 seconds (traditional STP)        |
+-----------------------------------------------------------+

STP Port Roles

Each switch port in STP takes a role — Root, Designated, Alternate, or Backup — to maintain a loop-free topology.

Role	Description
Root Port (RP)	The best path to the Root Bridge (one per non-root switch)
Designated Port (DP)	The best port on a segment (one per segment)
Alternate Port (AP)	Backup to the Root Port (blocked)
Backup Port (BP)	Backup to a Designated Port (blocked, rare)

STP Port States

Ports transition through Blocking, Listening, Learning, and Forwarding — taking ~50 seconds total.

DISABLED --> BLOCKING --> LISTENING --> LEARNING --> FORWARDING
   |                                                 |
   v                                                 v
Admin Down                                     Sending/Receiving
(No BPDUs)                                          Data

+-----------------------------------------------------------+
|  State      |  Forward Traffic?  |  Learn MACs?           |
+------------+--------------------+------------------------+
|  Disabled   |         No         |          No           |
|  Blocking   |         No         |          No           |
|  Listening  |         No         |          No           |
|  Learning   |         No         |          Yes          |
|  Forwarding |         Yes        |          Yes          |
+-----------------------------------------------------------+

Time in each state:

Blocking -> Listening: 20s (Max Age timer)
Listening -> Learning: 15s (Forward Delay)
Learning -> Forwarding: 15s (Forward Delay)
Total convergence: ~50 seconds

STP Election Process

Three-step election: pick a Root Bridge, select Root Ports, then designate one port per segment.

STP uses a three-step election:

Step 1: Root Bridge Election
  "Who has the lowest Bridge ID?"
  Bridge ID = Priority (16 bits) + MAC Address (48 bits)

  Default Priority = 32768
  SW-A Priority 32768, MAC 0000.0A01.0101
  SW-B Priority 32768, MAC 0000.0A02.0202  <- Tie goes to lower MAC
  SW-C Priority 4096,  MAC 0000.0A03.0303  <- Lowest priority = ROOT!

  Root Bridge = SW-C (priority 4096)

Step 2: Root Port Selection (on non-root switches)
  Each non-root switch picks its best path to the Root:
  Criteria (in order):
    1. Lowest Root Path Cost
    2. Lowest Neighbor Bridge ID
    3. Lowest Neighbor Port ID

Step 3: Designated Port Selection (on each segment)
  On each link, one port is designated:
  Criteria (in order):
    1. Switch with lower Root Path Cost
    2. Switch with lower Bridge ID

STP Path Costs (Old vs. New Standard)

Path cost values changed from the original 802.1D standard to the revised 802.1t to support higher link speeds.

Speed	Original (802.1D)	Revised (802.1t)
10 Mbps	100	2,000,000
100 Mbps	19	200,000
1 Gbps	4	20,000
10 Gbps	2	2,000
100 Gbps	1	200
1 Tbps	--	20

BPDU (Bridge Protocol Data Unit)

Special frames exchanged every 2 seconds between switches to communicate STP information.

STP uses BPDUs to communicate between switches:

+-----------------------------------------------------------+
|                   BPDU FRAME                               |
|                                                          |
|  +--------------+----------------------------------+    |
|  | Protocol ID  | 0x0000                           |    |
|  | Version      | 0x00 (802.1D)                    |    |
|  | Message Type | 0x00 (Config BPDU)               |    |
|  | Flags        | TC, TCA                           |    |
|  | Root ID      | Priority + MAC of Root Bridge    |    |
|  | Root Path Cost| Cost from sender to Root         |    |
|  | Bridge ID    | Sender's own Priority + MAC       |    |
|  | Port ID      | Sender's Port number              |    |
|  | Max Age      | 20s (default)                     |    |
|  | Hello Time   | 2s (default)                      |    |
|  | Forward Delay| 15s (default)                     |    |
|  +--------------+----------------------------------+    |
|                                                          |
|  BPDUs are sent every 2 seconds (Hello Timer)            |
|  Multicast address: 01:80:C2:00:00:00                   |
+-----------------------------------------------------------+

STP Variants

Different versions of STP improve convergence time and add per-VLAN capabilities.

Protocol	Standard	Convergence	Key Feature
STP	802.1D	30-50s	Original
RSTP	802.1w	< 10s	Fast convergence
MSTP	802.1s	< 10s	Multiple instances per VLAN group
PVST+	Cisco	Varies	Per-VLAN STP (Cisco proprietary)
Rapid-PVST+	Cisco	< 10s	Per-VLAN RSTP

RSTP (Rapid Spanning Tree Protocol -- 802.1w)

An improved STP that converges in under 10 seconds using a handshake mechanism instead of timers.

Key improvements over classical STP:

STP (802.1D)	RSTP (802.1w)
5 port states	3 port states (Discarding, Learning, Forwarding)
Blocking / Disabled / Listening = all Discarding in RSTP
Timer-based convergence	Proposal/Agreement handshake
30-50 seconds	< 10 seconds
Port roles: Root, Designated, Alternate, Backup	Same + Backup refined

RSTP Port Roles

                   +----------------+
                   |   EDGE PORT    |  <- Connected to end devices
                   | (Immediately   |     (no BPDUs expected)
                   |  Forwarding)   |
                   +----------------+

                   +----------------+
                   |  ROOT PORT     |  <- Best path to root
                   +----------------+

                   +----------------+
                   | DESIGNATED PORT|  <- Best port on segment
                   +----------------+

                   +----------------+
                   | ALTERNATE PORT |  <- Backup to root port
                   | (Discarding)   |
                   +----------------+

                   +----------------+
                   |  BACKUP PORT   |  <- Backup to designated
                   | (Discarding)   |
                   +----------------+

STP Configuration (Cisco)

! Configure Bridge Priority (to influence root election)
spanning-tree vlan 10 priority 4096

! Enable PortFast (edge ports)
interface Gi0/1
 spanning-tree portfast

! Enable BPDU Guard (protect against rogue switches)
interface Gi0/1
 spanning-tree bpduguard enable

! Configure UplinkFast (for fast convergence on access switches)
spanning-tree uplinkfast

! Verify STP status
show spanning-tree
show spanning-tree vlan 10

Sample `show spanning-tree` Output

VLAN0010
  Spanning tree enabled protocol rstp
  Root ID    Priority    4096
             Address     aabb.cc03.0303
             Cost        20000
             Port        24 (GigabitEthernet0/24)
             Hello Time   2 sec  Max Age 20 sec  Forward Delay 15 sec

  Bridge ID  Priority    32768  (priority 32768 sys-id-ext 10)
             Address     aabb.cc01.0101
             Hello Time   2 sec  Max Age 20 sec  Forward Delay 15 sec
             Aging Time  300 sec

Interface       Role    Sts      Cost      Prio.Nbr     Type
--------        ----    ---      ----      --------     ----
Gi0/1           Desg    FWD      20000     128.1         P2p Edge
Gi0/2           Desg    FWD      20000     128.2         P2p Edge
Gi0/24          Root    FWD      20000     128.24        P2p
Gi0/23          Altn    BLK      20000     128.23        P2p

STP Protection Features

Additional mechanisms to enhance STP stability and security against misconfiguration or attacks.

Feature	Purpose	Configuration
PortFast	Bypass listening/learning for end devices	`spanning-tree portfast`
BPDU Guard	Shut down port if BPDU received (prevents rogue switches)	`spanning-tree bpduguard enable`
BPDU Filter	Suppress sending/receiving BPDUs on a port	`spanning-tree bpdufilter enable`
Root Guard	Prevent a port from becoming Root Port	`spanning-tree guard root`
Loop Guard	Prevent alternate ports from transitioning to forwarding if BPDUs stop	`spanning-tree guard loop`
UDLD	Detect unidirectional links (one-way traffic)	`udld enable`

STP Troubleshooting

Common STP issues, their symptoms, and how to resolve them.

Problem	Symptom	Solution
No root bridge elected	All ports blocking	Check BPDU transmission, link status
Wrong root bridge	Unexpected switch is root	Configure priority on intended root
Slow convergence	~50 second delay	Use RSTP / Rapid-PVST+
Duplicate frames	Intermittent connectivity issues	Check for loops, verify STP topology
BPDU Guard err-disable	Port is down (err-disabled)	`shutdown` then `no shutdown` on port

7. Exercises

Exercise 1: MAC Table Population & Frame Forwarding

Scenario

You have a switch with 8 ports. The network is initially powered on, and the MAC table is empty. Devices are connected as follows:

+-----------------------------------------------------------+
|                         SWITCH                             |
|                                                          |
|  Port 1    Port 2    Port 3    Port 4    Port 5         |
|    |          |         |         |         |            |
|  PC-A      PC-B      PC-C      PC-D      PC-E           |
| MAC: AA   MAC: BB   MAC: CC   MAC: DD   MAC: EE        |
| IP: .10   IP: .20   IP: .30   IP: .40   IP: .50        |
|                                                          |
|     Port 6     Port 7      Port 8                       |
|       |          |           |                           |
|    Server-F   Printer-G   Laptop-H                      |
|    MAC: FF    MAC: GG     MAC: HH                        |
|    IP: .60    IP: .70     IP: .80                        |
+-----------------------------------------------------------+

Trace the following sequence of events

Step	Action
1	PC-A (Port 1) sends frame to PC-C (MAC: CC)
2	PC-C (Port 3) sends reply to PC-A (MAC: AA)
3	PC-B (Port 2) sends frame to Server-F (MAC: FF)
4	PC-D (Port 4) broadcasts ARP request "Who has 192.168.1.60?"
5	Laptop-H (Port 8) sends frame to PC-A (MAC: AA)
6	Server-F (Port 6) sends reply to PC-B (MAC: BB)

Questions

After each step, write the updated MAC table (MAC -> Port)
After step 1: Is the frame flooded or selectively forwarded? Why?
After step 4: Which ports receive the broadcast frame?
After step 5: Which port does the switch forward the frame to?
Challenge: If PC-E (Port 5) has been idle for 10 minutes and the aging time is 300 seconds, is its MAC still in the table?

Click for Answers

After Step 1 -- PC-A -> PC-C (frame from AA to CC)

MAC	Port	Notes
AA	1	Learned from source

Dest MAC CC is NOT in table -> Flooded to all ports except Port 1 (Ports 2-8)
PC-C receives on Port 3

After Step 2 -- PC-C -> PC-A (reply from CC to AA)

MAC	Port
AA	1
CC	3

Dest MAC AA is IN the table -> Selectively forwarded to Port 1 only

After Step 3 -- PC-B -> Server-F (frame from BB to FF)

MAC	Port
AA	1
CC	3
BB	2

Dest MAC FF NOT in table -> Flooded to all ports except Port 2

After Step 4 -- PC-D broadcasts ARP (dest FF:FF:FF:FF:FF:FF)

MAC	Port
AA	1
CC	3
BB	2
DD	4

Broadcast -> Flooded to ALL ports except Port 4 (Ports 1, 2, 3, 5, 6, 7, 8)
Server-F (MAC FF) on Port 6 receives the ARP and will reply

After Step 5 -- Laptop-H -> PC-A (frame from HH to AA)

MAC	Port
AA	1
CC	3
BB	2
DD	4
HH	8

Dest MAC AA is IN the table on Port 1 -> Forwarded only to Port 1

After Step 6 -- Server-F -> PC-B (reply from FF to BB)

MAC	Port
AA	1
CC	3
BB	2
DD	4
HH	8
FF	6

Dest MAC BB is IN the table on Port 2 -> Forwarded only to Port 2

Final MAC Table:

+------------+--------+
| MAC        | Port   |
+------------+--------+
| AA         |   1    |
| BB         |   2    |
| CC         |   3    |
| DD         |   4    |
| FF         |   6    |
| HH         |   8    |
| EE         |   (?)  | <- EE was never seen -- NOT in table
| GG         |   (?)  | <- GG was never seen -- NOT in table
+------------+--------+

Answer 5: PC-E (MAC EE) on Port 5 -- EE was never learned because PC-E never sent any frame during this sequence. The aging timer only applies to entries that exist. Since EE was never in the table, it doesn't matter that 10 minutes passed.

Exercise 2: VLAN & STP Topology Planning

Scenario

You are designing a network for a three-floor office building. Each floor has one switch, and all switches connect to a core switch in the server room. Redundant links exist for fault tolerance.

                              +----------------------+
                              |     CORE SWITCH       |
                              |      (SW-CORE)        |
                              +------+--------+-------+
                                     |        |
                +--------------------+        +--------------------+
                |                                                   |
           +----+----+                                        +----+----+
           | SW-F1   |                                        | SW-F2   |
           | Floor 1 |                                        | Floor 2 |
           +----+----+                                        +----+----+
                |                                                  |
                +----------------------+---------------------------+
                                       |
                                  +----+----+
                                  | SW-F3   |
                                  | Floor 3 |
                                  +---------+

Requirements

Department	Floor	VLAN ID	Subnet	PCs
Management	1	10	10.0.10.0/24	10
Engineering	2	20	10.0.20.0/24	30
Sales	3	30	10.0.30.0/24	25
IT Servers	Server Room	99	10.0.99.0/24	5

Tasks

Task A -- VLAN Configuration: Write the configuration commands for each switch to create the necessary VLANs. Which VLANs should exist on each switch?

Task B -- Trunking: Configure trunk ports between all switches (including SW-CORE). Which VLANs should be allowed on each trunk?

All switches need to reach the IT Servers (VLAN 99)
Each floor switch only needs its own floor's VLAN + management VLAN + server VLAN

Task C -- STP Root Bridge:

SW-CORE has MAC address: aaaa.bbbb.cc01
SW-F1 has MAC: aaaa.bbbb.cc02 (priority set to 4096)
SW-F2 has MAC: aaaa.bbbb.cc03
SW-F3 has MAC: aaaa.bbbb.cc04
All switches have default priority (32768)

Who becomes the root bridge? What command would you add to make SW-CORE the root bridge?

Task D -- Port Roles: Based on your root bridge selection in Task C:

Which ports on SW-F1/SW-F2/SW-F3 become Root Ports?
Which ports become Designated Ports?
Which ports become Alternate (Blocked) ports?

Assume path costs: 1 Gbps = 4 (original 802.1D cost).

Task E -- Inter-VLAN Routing:

Configure a router-on-a-stick or Layer 3 switch to route between VLAN 10, 20, 30, and 99
Which device should handle the routing?
What IP configuration does each PC need?

Click for Answers

Task A -- VLAN Configuration

Switch	Required VLANs
SW-CORE	10, 20, 30, 99
SW-F1	10, 99
SW-F2	20, 99
SW-F3	30, 99

SW-CORE Configuration:

vlan 10
 name MANAGEMENT
vlan 20
 name ENGINEERING
vlan 30
 name SALES
vlan 99
 name IT_SERVERS

SW-F1 Configuration:

vlan 10
 name MANAGEMENT
vlan 99
 name IT_SERVERS

interface Gi0/1
 switchport mode access
 switchport access vlan 10
!
interface Gi0/10
 switchport mode access
 switchport access vlan 99          ! Link to server room

Task B -- Trunking Configuration

SW-CORE trunk ports:

! Trunk to SW-F1
interface Gi0/1
 switchport mode trunk
 switchport trunk allowed vlan 10,99

! Trunk to SW-F2
interface Gi0/2
 switchport mode trunk
 switchport trunk allowed vlan 20,99

! Trunk to SW-F3
interface Gi0/3
 switchport mode trunk
 switchport trunk allowed vlan 30,99

SW-F1 trunk (to SW-CORE):

interface Gi0/24
 switchport mode trunk
 switchport trunk allowed vlan 10,99

SW-F2 to SW-F3 trunk (redundant link):

Should only allow VLANs that need to cross between these floors
Both need VLAN 99 -> allow only VLAN 99

switchport trunk allowed vlan 99

Task C -- STP Root Bridge Election

Switch	Priority	MAC	Lower MAC on tie?
SW-CORE	32768	aaaa.bbbb.cc01	Lowest MAC of all -> would win with default settings
SW-F1	4096	aaaa.bbbb.cc02	Lowest priority -> SW-F1 WINS!
SW-F2	32768	aaaa.bbbb.cc03
SW-F3	32768	aaaa.bbbb.cc04

Root Bridge: SW-F1 (priority 4096 < 32768 of others)

This is likely not what we want -- the core switch should be root!

Fix: Set SW-CORE priority lower than 4096:

spanning-tree vlan 10 priority 0
spanning-tree vlan 20 priority 0
spanning-tree vlan 30 priority 0
spanning-tree vlan 99 priority 0

Or simpler: spanning-tree vlan 10,20,30,99 root primary

After fix, SW-CORE becomes root (priority 0 < 4096).

Task D -- Port Roles (after fixing root = SW-CORE)

Using the fixed topology (root = SW-CORE). Path cost: 1 Gbps = 4.

Port	Path to Root	Cost	Role	State
SW-CORE -> SW-F1	Direct	0	Designated	Forwarding
SW-CORE -> SW-F2	Direct	0	Designated	Forwarding
SW-CORE -> SW-F3	Direct	0	Designated	Forwarding
SW-F1 -> SW-CORE	Direct	4	Root Port	Forwarding
SW-F2 -> SW-CORE	Direct	4	Root Port	Forwarding
SW-F2 -> SW-F3	Via SW-F3 -> SW-CORE = 4+4 = 8	8	Alternate (Blocked)	Blocking
SW-F3 -> SW-CORE	Direct	4	Root Port	Forwarding
SW-F3 -> SW-F2	Via SW-F2 -> SW-CORE = 4+4 = 8	8	Alternate (Blocked)	Blocking

The link between SW-F2 and SW-F3 is blocked on both sides (or one side is Alternate, the other is a backup port)
If the SW-F2 -> SW-CORE link fails, STP unblocks the SW-F2 -> SW-F3 path (convergence: ~30s for classic STP, <10s for RSTP)

Task E -- Inter-VLAN Routing

Option 1: Router-on-a-Stick (if using a dedicated router)

Router connects to SW-CORE via trunk
Configure sub-interfaces:

interface Gi0/0.10
 encapsulation dot1Q 10
 ip address 10.0.10.1 255.255.255.0

interface Gi0/0.20
 encapsulation dot1Q 20
 ip address 10.0.20.1 255.255.255.0

interface Gi0/0.30
 encapsulation dot1Q 30
 ip address 10.0.30.1 255.255.255.0

interface Gi0/0.99
 encapsulation dot1Q 99
 ip address 10.0.99.1 255.255.255.0

Option 2: Layer 3 Switch (SVI)

Configure SVIs on SW-CORE:

ip routing
interface vlan 10
 ip address 10.0.10.1 255.255.255.0
 no shutdown
interface vlan 20
 ip address 10.0.20.1 255.255.255.0
 no shutdown
interface vlan 30
 ip address 10.0.30.1 255.255.255.0
 no shutdown
interface vlan 99
 ip address 10.0.99.1 255.255.255.0
 no shutdown

PC Configuration:

PC	IP Address	Subnet Mask	Default Gateway
Management PC (VLAN 10)	10.0.10.x	255.255.255.0	10.0.10.1
Engineering PC (VLAN 20)	10.0.20.x	255.255.255.0	10.0.20.1
Sales PC (VLAN 30)	10.0.30.x	255.255.255.0	10.0.30.1
Server (VLAN 99)	10.0.99.x	255.255.255.0	10.0.99.1

The gateway IP for each VLAN is the router/L3 switch interface IP for that VLAN.

Switching in Detail

Switching in Detail: MAC Tables, VLANs, Trunking, and Spanning Tree Protocol Explained

Table of Contents

1. Switch Basics

What is a Switch?

Switch vs. Hub vs. Router

Types of Switches

Switch Port Types

2. Switching Process

The Core Switching Steps

Step-by-Step Example

Switching Methods (Forwarding Modes)

Frame Forwarding Decision Table

3. MAC Table

What is the MAC Table?

Anatomy of a MAC Table Entry

MAC Entry Types

MAC Table Aging

Viewing the MAC Table

Cisco Switch

Linux (as a switch)

Sample Output

Why MAC Table Size Matters

MAC Table vs. Routing Table

4. VLAN (Virtual Local Area Network)

What is a VLAN?

Why Use VLANs?

VLAN Ranges

VLAN Configuration

Creating a VLAN on a Cisco Switch

Verification Commands

Sample Output

VLAN Communication Rules

Inter-VLAN Routing

VLAN Design Best Practices

5. Trunking

What is Trunking?

Trunking vs. Access Ports

How Trunking Works -- 802.1Q Tagging

Native VLAN

Trunking Configuration

Cisco Switch

Allowed VLANs on Trunk

Verification

Sample Output

DTP (Dynamic Trunking Protocol)

6. STP (Spanning Tree Protocol)

What is STP?

The Problem: Bridging Loops

Three Deadly Problems from Loops

How STP Solves This

STP Port Roles

STP Port States

STP Election Process

STP Path Costs (Old vs. New Standard)

BPDU (Bridge Protocol Data Unit)

STP Variants

RSTP (Rapid Spanning Tree Protocol -- 802.1w)

RSTP Port Roles

STP Configuration (Cisco)

Sample show spanning-tree Output

STP Protection Features

STP Troubleshooting

7. Exercises

Exercise 1: MAC Table Population & Frame Forwarding

Scenario

Trace the following sequence of events

Questions

Exercise 2: VLAN & STP Topology Planning

Scenario

Requirements

Tasks

Sign in to track progress

Choose your next step

Sample `show spanning-tree` Output