Switch

Ethernet Switch that uses VLAN segmentation for network efficiency, Spanning Tree Protocol (STP) for loop prevention, and MAC address learning for fast frame forwarding. By managing VLAN membership, preventing loops, and learning MAC addresses, the implementation ensures reliable communication and facilitates network segmentation for enhanced performance.

Frame Forwarding Process

Frame Arrival: An Ethernet frame arrives at a switch through one of its ports.
Destination MAC Address: The switch examines the destination MAC address in the header frame.
MAC Address Table Lookup: The switch checks its MAC address table to find the port associated with the destination MAC address. If found, the frame is forwarded out of the corresponding port.
Unknown Destination MAC Address: If the MAC address is unknown, the switch floods the frame to all ports except the incoming one to ensure connectivity.
Broadcast and Multicast:
- Broadcast frames are forwarded to all ports except the receiving one, all devices receive the broadcast.
- Multicast frames are forwarded only to ports interested in the multicast group.

VLAN

VLANs (Virtual Local Area Networks) segment a single physical LAN into multiple broadcast domains. The VLANs introduce the concept of VLAN tagging and forwards frames based on VLAN membership, facilitating efficient network segmentation and traffic management.

VLAN Forwarding Process

Frame Reception: When a switch receives a frame with an unknown destination or broadcast, it forwards it to all ports within the same VLAN, including trunks.
IEEE 802.1Q VLAN Tagging: VLAN tagging is introduced using IEEE 802.1Q, adding 4 bytes to the frame. The VID field (12 bits) represents the VLAN identifier.

Switch Behavior:

If on an access port:
- Forwards with 802.1Q header on trunk interfaces.
- Forwards without header if VLAN ID matches that of the incoming interface.
If on a trunk port:
- Removes VLAN tag and forwards:
  - With 802.1Q header (including tag) on trunk interfaces.
  - Without header if VLAN ID matches that of the received frame on access interfaces.

Linux VLAN Filtering: TPID value of 0x8200 is used instead of 0x8100. PCP and DEI are set to 0.
Trunk Links: Links between switches operate in trunk mode, allowing passage of all VLANs.
Configuration: VLANs and trunk configurations are set via a configuration file specified in the API section.

STP (Spanning Tree Protocol)

STP (Spanning Tree Protocol) is a protocol used to prevent loops in network topologies by creating a loop-free logical topology. This implementation provides a simplified version of STP to avoid LAN loops.

Each switch initially considers itself as the root bridge and starts with all ports in the Listening state.
BPDUs are exchanged between switches to elect the root bridge and determine the designated ports.
Trunk ports are crucial for loop prevention, and STP operates only on these ports to avoid potential loops.

Structure of BPDU Frames

BPDU frames utilize encapsulation with the 802.2 Logical Link Control (LLC) header. The Bridge Protocol Data Units (BPDUs) utilize the encapsulation of the 802.2 Logical Link Control (LLC) header. The structure of a BPDU is as follows:

Destination MAC (DST_MAC): 6 bytes
Source MAC (SRC_MAC): 6 bytes
LLC Length (LLC_LENGTH): 2 bytes, indicating the total size of the frame including the BPDU.
LLC Header (LLC_HEADER): 3 bytes, comprising the DSAP (Destination Service Access Point), SSAP (Source Service Access Point), and Control fields.
- DSAP: 1 byte, typically set to 0x42 to identify the STP protocol.
- SSAP: 1 byte, also set to 0x42 to identify the STP protocol.
- Control: 1 byte, often set to 0x03 for control purposes.
BPDU Header (BPDU_HEADER): 4 bytes, containing control information specific to STP.
BPDU Configuration (BPDU_CONFIG): 31 bytes, encompassing various parameters such as flags, root bridge ID, root path cost, sender bridge ID, port ID, message age, max age, hello time, and forward delay.

STP Forwarding Process

Initialization: Trunk ports start in the Blocking state to prevent loops. Switches consider themselves as root bridges, with all ports in the Listening state. If a switch it's the root bridge, all ports are Designated.
BPDU Exchange: Switches exchange BPDUs to elect the root bridge and determine designated ports. BPDUs contain root bridge ID, sender bridge ID, and root path cost, sent regularly on trunk ports.
Root Bridge Election: Upon receiving a BPDU, switches compare root bridge IDs. If received ID is lower, the switch updates its information and forwards the BPDU.
- Switches continuously update root bridge information.
Port States: Ports can be Blocking, Listening, Learning, or Forwarding.
- Blocking prevents loops,
- Listening prepares for Learning state,
- Learning populates MAC address tables, and
- Forwarding fully operates.
Loop Prevention: STP operates on trunk ports to prevent loops. BPDUs determine best paths to:
- root bridge
- block redundant links
Frame Forwarding: Forward frames based on established spanning tree topology, ensuring a loop-free network.

Usage

To simulate a virtual network use Mininet. Mininet is a network simulator that utilizes real kernel, switch, and application code implementations. Mininet can be used on both Linux and WSL2.

sudo apt update
sudo apt install mininet openvswitch-testcontroller tshark python3-click python3-scapy xterm python3-pip
sudo pip3 install mininet

After installing Mininet use the following command to increase the font size in the terminals we open.

echo "xterm*font: *-fixed-*-*-*-18-*" >> ~/.Xresources
xrdb -merge ~/.Xresources

When running the simulation, you may encounter the following error: Exception: Please shut down the controller which is running on port 6653:. To resolve the issue, you will need to run pkill ovs-test.

justin-marian/switch