802.3 and Ethernet frame structure-routing switching principle 1-"Hcna notes"

1. RAW 802.3 frame Format

leading field	Frame start character		Length	Data	FCS
7 bytes	1 bytes	6 bytes	6 bytes	2 bytes	46-1500 bytes		4 bytes

Leading fields: each frame begins with a 7-byte leading field, consisting of 0-, 1-spaced code that notifies the target to receive preparation (10101010 10101010 10101010 10101010 10101010 10101010 10101010).

frame start: followed by a leading field with a value of 10101011 and the last two bits ' 11 ' representing the actual start of the frame. The leading field and the start character are automatically added by the physical device to synchronize the two communication devices, so the data received by the peer does not contain the two fields, so the frame length calculation and frame Check also do not contain these two fields.

DMAC: destination MAC address.

SMAC: source MAC address.

Length : The lengths of the data fields, recording the actual length of the subsequent field data.

Data : A field that holds frame data. The minimum length of a frame is 64 bytes specified in 802.3, and the frame length is calculated as follows:

Frame length = DMAC + SMAC + length + Data + FCS

Where Dmac, SMAC, length, and FCS are 18 bytes, the minimum length of the data field is 64-18 = 46 bytes.

If the frame length is indeed less than 64 bytes, the sender automatically populates the data field with 0 until the frame length equals 64 bytes.

FCS: frame Check sequence, the receiver is used to determine whether the transmission process is wrong, if the error is found to discard this frame.

2. Ethernetii Frame Format

Leading fields		DMAC	SMAC	Type	Data	Fcs
8 bytes		6 bytes	6 bytes	2 bytes	46-1500 bytes	4 bytes

Ethernetii is similar to the 802.3-frame format, except that the ethernetii does not have a frame start character, and the leading field occupies 8 bytes, and the function is consistent with the 802.3 leading field + frame start character, which realizes the communication synchronization between the two devices.

Another difference is that the length field of 802.3 is replaced by the Type field to indicate the protocol type of the frame. This is because ETHERNETII frames support multiple protocol types, while raw 802.3 frames support only one.

similarities and differences of 3, 802.3 and Ethernetii

For the convenience of 802.3 and ethernetii two frame formats can be used in the network at the same time, it is stipulated that the maximum length of the data field in 802.3 and ethernetii must be less than 1500 bytes, that is, the value of the length field in 802.3 is necessarily less than 1500. At the same time, the protocol type in Ethernetii is assigned from the digital 1536 (0x0600), so that if the value of the type or length field is less than 1536 then it is a 802.3 frame, and a value greater than 1536 proves that it is a ethernetii frame.

In use, ethernetii is more suitable for transmitting large amounts of data than 802.3, while the 802.3 advantage is its strict transmission control. So most application Ethernet packets use Ethernetii frame format (such as HTTP, FTP, SMTP, POP3, etc.), while bpdus packets between switches, Cisco's discovery protocol uses 802.3 frame format

4, the development of Ethernet frame format

Ethernet V1 (1980): This is the most primitive format, the 3Mbps Ethernet standard packaging format presented by Xerox, which was later standardized in 1980 by Dec, Intel and Xerox to form the Ethernet V1 standard.

Ethernet V2 (1982): That's what we're talking about ethernetii, the standards announced by DEC, Intel, and Xerox in 1982 largely changed the electrical and physical interfaces of V1, with no change in frame format. After the advent of V2, the replacement of V1 became the standard format for Ethernet frames.

ethernetii frame format

preamble field /td>		DMAC			Data	FCS
8 bytes		6 bytes	6 bytes	2 bytes	46-1500 bytes	4 bytes

RAW 802.3 (Novell, 1983):1983 Novell uses this Ethernet frame format when it publishes its NETWARE/86 network kit, which draws on the IEEE802.3 standards that have not yet been published Ethernet The Type field in V2 is changed to the length field, because the raw 802.3 frame used at that time only needs to support the Ipx/spx type of protocol. Two years later, the officially released IEEE802.3 joined the LLC header, leading to a non-compliant standard with the IEEE official standards. In order to be compatible with the new IEEE Standard, the Raw 802.3 data field is specified as the first two bytes of the raw field and the value is 0xFFFF to differentiate between Raw and later published LLC standards.

802.3 Raw Frame format

Leading fields	Frame Start character	DMAC	SMAC	Length	Data (46-1500)			Fcs
7 bytes	1 bytes	6 bytes	6 bytes	2 bytes	RAW (0xFFFF)		Data	4 bytes

IEEE802.3 LLC (1985): This is the official IEEE 802.3 standard, developed by Ethernet V2, which replaces the Type field of Ethernet V2 with the length field, The first 3 bytes in the data field are also named the LLC header to identify the upper-level protocol type. The LLC contains DSAP, SSAP, and control three fields, each of which has a length of 1 bytes. DSAP is the protocol type on the target side, and SSAP is the source-side protocol type.

802.3 LLC Frame format

leading field	frame start character	DMAC	SMAC /td>	Length	Data (46-1500)			FCS
7 bytes	1 bytes	6 bytes	6 bytes	2 bytes	LLC (Dsap/ssap/control)		Data	4 bytes

IEEE802.3 SNAP (1985): In order to enable 802.3 to support more upper-level protocol types other than the LLC and to better support the IP protocol, the published standard, based on the inclusion of the LLC header, divides the 5 bytes after the LLC header in the data field into SNAP headers, which contains a 3-byte oui field used to represent different organizations (usually equal to the MAC address of the network Adapter manufacturer's code) and a 2-byte type field to support more upper-layer protocol types. In order to differentiate from LLC frames, SNAP will cure the value of the LLC header, the value of DSAP, SSAP fixed to 0xaa,control is fixed to 0x03 (that is, the first 3 bytes value of the data field is 0xaaaa03 to indicate that it is a snap frame).

802.3 snap Frame format

leading field	frame start character	DMAC	SMAC /td>	Length	Data (46-1500)			FCS
7 bytes	1 bytes	6 bytes	6 bytes	2 bytes	LLC (0XAAAA03)	SNAP (Oui/type)	Data	4 bytes

5, different frame format of the distinction

ethernetii and 802.3: If the value of the length or type field in the frame is less than 1536, this is a ethernetii frame, and the reverse is a 802.3 frame.

Raw, LLC, SNAP: This is a 802.3 (raw) frame if the first two bytes of the data field are 0xFFFF, and if the first three bytes in the data field are 0XAAAA03, this is a 802.3 (SNAP) frame The remainder is the 802.3 (LLC) frame.

"Identification of upper-level protocol types"802.3 more than 3 standards because they use different upper-layer protocol type identification methods, then what is the identity of the upper-layer protocol type?

IEEE divides the network into the OSI 7 layer model, where the first layer is the physical layer, the second layer is the data link layer, the third layer is the network layer, and the data link layer is subdivided into the LLC layer and the MAC layer. We know that the MAC layer is used for physical addressing, the data frame is reflected in the Dmac and Smac two fields, respectively, the target MAC address and the source MAC address, the network card with these two fields to know that the packet is from that device destination is that device. Who should handle the frame when it arrives? The different protocols in the device are handled by different processes, and the role of the LLC layer is to indicate the protocol type of the process, that is, the Type field of the Ethernetii (the specific protocol will be further specified in the IP message, and the receiving process for locating the data). In raw 802.3 only need to support ipx/spx this kind of protocol so there is no Type field, 802.3 LLC uses the DSAP and SSAP in the LLC header to identify the target end of the protocol type and the source-side protocol type, 802.3 Snap uses the Type field in the snap header to identify the protocol type similar to ethernetii.

"Frame Check" data frame transmission in the network will inevitably occur data loss or bit error, etc., the network card after receiving the frame is how to determine that the data frame in the process of transmission did not happen unexpectedly.

There is no real school recipe (too complicated), just a simple example. For example, a set of binary numbers 0111 1101 1001 0011 from one device to another, the receiving device does not prevent errors in the transmission process, only the set of data can be verified, if it does not conform to a certain rule is judged as an error will be discarded this frame. The validation rules are defined by the protocol that different frame protocols may use different validation rules, and the FCS field holds the results of this rule execution. For example, the sender can add the binary number to 22 per 8 bits: 0111 1101 + 1001 0011 = 0001 0001 0000, take the latter eight bits 0001 0000, and store the latter eight bits in the FCS field. The receiver receives the frame and repeats the verification rule of the sending end, and the result is compared with the result saved in the FCS field, and if it is consistent, the transmission process is correct and the frame is not used.

"802.3 names" we can understand that Ethernetii is the V2 version of the Ethernet protocol, but why 802.3 is called 802.3.

After the advent of Ethernet, the network was first standardized by DEC, Intel and Xerox to establish the Ethernet V1 protocol, which specifies the physical layer of Ethernet connectivity, electrical signals and media, and so on, the definition of frame format is only part of it. The Ethernet V2, now the most commonly used ethernetii protocol, was soon introduced. Two years later, due to the development of Ethernet, IEEE Association of Ethernetii on the basis of a more detailed regulation of Ethernet, that is, the IEEE802.3 protocol. The IEEE Association consists of a number of different professional associations and committees, such as the IEEE Aeronautical and Electronic Systems Association, IEEE Communications Association, IEEE Biometrics Commission, etc., in which the Commission for the development of the 802.3 agreement is called the IEEE LAN/Metropolitan Area Network Standards Committee, It was founded in February 1980. So also called the 802 Committee, the association issued a series of standard names are the beginning of the IEEE 802, 802.3 Indicates that this is a series of standard number three. Let's take a look at some of the standard names in this series:

IEEE 802.1: LAN architecture, addressing, network interconnection, and networking

IEEE 802.1A: Overview and System architecture

IEEE 802.1B: Network Management and network interconnection

IEEE 802.2: Definition of a logical link Control sublayer (LLC).

IEEE 802.3: Ethernet Media Access Control Protocol (CSMA/CD) and physical layer technical specifications.

IEEE 802.4: The Media access control Protocol and physical layer technical specifications for the token bus network (Token-bus).

IEEE 802.5: Token Ring Network (token-ring) Media access Control Protocol and physical layer technical specifications.

"Ethernet name origin" in the early years, scientists believe that the world is a kind of invisible touch of the media "ether" flooding, and then the existence of the ether is falsified, then Ethernet and the ether is what is the connection?

There are online jokes that the father of the Ethernet Metcalfe in his doctoral dissertation is the study of APRA Net, but his thesis defense has not been, is considered to be lack of theoretical support, but his technology has been a lot of manufacturers, and then Metcalfe think of their papers did not have a very uncomfortable, So I designed the local area network technology called Ethernet, ridicule this is a kind of invisible, not touch the technology. The above satin shows that Ethernet is a local area network, the inventor himself blind, no special meaning.

802.3 and Ethernet frame structure-routing switching principle 1-"Hcna notes"