MT: Mikrotik as switch and VLAN configuration

Diagram with GNS3 software

Router R1 configuration

configuration of Router R2

bridge configuration on SW1

port settings on SW1

VLAN settings on SW1

bridge configuration on SW2

port settings on SW2

VLAN settings on SW2

DHCP test on end devices

CONCLUSIONS:

Managed devices (e.g. switches), operating according to the 802.1Q standard, use special frame marking (tagging). This is not additional encapsulation, but merely marking frames that they belong to a VLAN, communicating which network they belong to and that they are operating under the 802.1Q standard. For Ethernet frames using Ethernet II framing, the marking consists of inserting an extra 4 bytes between the source MAC address and the EtherType field. This is referred to as TPID (Tag Protocol Identifier). The first 2 additional bytes are a tag and always have the value 0x8100 to identify the frame as 802.1Q tagged.

The remaining 2 bytes called TCI (English: Tag Control Information) contain the following information:

Ethernet frame
Preamble (8 bytes) Destination MAC (6 bytes) Source MAC (6 bytes) EtherType (2 bytes) Data field (46-1500 bytes) CRC (4 bytes)
Ethernet frame marked 802.1Q
Preamble (8 bytes) Destination MAC (6 bytes) Source MAC (6 bytes) TPID 0x8100 TCI (2 bytes) EtherType (2 bytes) Data field (46-1500 bytes) CRC (4 bytes)

For frames using IEEE 802.2/SNAP encapsulation with the OUI field set to 00-00-00 (i.e. the ID field in the SNAP header set to EtherType), as may happen on LANs using a technique other than Ethernet, you set the EtherType field in the SNAP header to 0x8100 and add the above 4 bytes to the end of the SNAP header.

Changing the header and adding the extra bytes changes the FCS checksum, which must be recalculated.