Android Network Programming-TCP/IP Protocol

In the article Android Network Programming - Computer Network Basics, we learned that the IP protocol belongs to the network layer, and the TCP and UDP protocols belong to the transport layer.

The IP protocol is the driving force of the TCP/IP protocol suite, providing stateless, connectionless, and unreliable services for upper-layer protocols.

The TCP protocol is a connection-oriented transport layer protocol that provides a connection-oriented, reliable byte stream service.

The UDP protocol is a connectionless transport layer protocol that provides simple, transaction-oriented, unreliable information transmission services.

Datagram

The data units transmitted at different layers have different names. The data units transmitted at the network layer are called datagrams, and the data units transmitted at the transport layer are called segments.

IP Datagram

The IP datagram format is as follows:

Detailed description of each field: Name Length Description Version 4 bits The version of the IP protocol. The current IP protocol version number is 4, and the next generation IP protocol version number is 6. Header length 4 bits The length of the IP header. The maximum length is 60 bytes (15*4).
It is divided into the length of the fixed part (20 bytes) and the length of the variable part. Service Type 8 bits Type Of Service Total length 16 bits The total length of the IP message. The maximum length of a datagram is 65535 bytes Identifier 16 bits It is a counter used to generate the identifier of the datagram.

When the length of an IP packet exceeds the MTU (maximum transmission unit) of the transmission network, it must be fragmented. This flag indicates the fragmentation of the same datagram. The flag has three bits: R, DF, and MF. Currently, only the last two bits are valid.

• DF bit: 1 means no fragmentation, 0 means fragmentation.
• MF: 1 means "more fragments", 0 means this is the last fragment. Fragment offset 13 bits The offset of this fragment relative to the first bit in the original data message.

The fragment offset is in units of 8 bytes. The 8-bit TTL (Time To Live) indicates the lifetime of a datagram in the network, in seconds.

In current practical applications, the unit is often "hop". The protocol 8 bit indicates which protocol the data carried by the IP message uses, so that the IP layer of the destination host can know which process to hand over the datagram to.

The protocol number of TCP is 6 and the protocol number of UDP is 17.

The protocol number of ICMP is 1, and the protocol number of IGMP is 2. The header checksum 16 bits calculates the checksum of the IP header and checks the integrity of the IP header. The source address 32 bits identifies the source device of the IP datagram. The destination address 32 bits identifies the destination address of the IP datagram. The optional field length is variable from 1 to 40 bytes and is used to increase the control function of the IP datagram. Padding ensures that the IP header length is an integer multiple of 4 bytes.

TCP Message

Name Length Description Source port 16 bits Port number of the data sender Destination port 16 bits Port number of the data receiver Sequence number 32 bits Sequence number of the first byte in this data message
(Each byte in the data stream corresponds to a sequence number) Confirmation number 32 bits The sequence number of the first byte in the next data message you want to receive Data offset 4 bits Indicates how far away the data segment of this message is from the message segment Reserved field 6 bits are reserved for future use, but should be set to 0 at present Urgent bit URG When the value is 1, it means that there is something in the next message segment that needs urgent processing Confirmation bit ACK When the value is 1, the confirmation number is valid, and when the value is 0, the confirmation number is invalid Reset bit RST When the value is 1, it means that there is a serious error in the TCP connection and it is necessary to reconnect Synchronization bit SYN When the value is 1, it means that this is a connection request or connection acceptance End of message bit FIN When the value is 1, it means that the datagram to be sent has been sent and needs to be released Transmission connection window 16 bits One end of the TCP connection determines the size of its own receiving window based on the size of the cache space Limit the window upper limit sent and the 16-bit checksum is used to verify the correctness of the header and data parts Urgent pointer field 16 bits The urgent pointer indicates the sequence number of the last byte of urgent data in this message segment The option field is of variable length The TCP header can have up to 40 bytes of optional information.
Used to pass additional information to the endpoint, or to align other options

UDP Message

Compared with TCP messages, UDP messages are much simpler.

Name Length Description Source port 16bit Data sender's port number Destination port 16bit Data receiver's port number Packet length 16bit The sum of the length of the UDP header and the length of the data. The unit is bytes Checksum 16bit Used to check the correctness of the header and data.

TCP three-way handshake and four-way wave

TCP uses three handshakes to create a connection and four handshakes to release the connection.

Three-way handshake

The purpose of the three-way handshake is to synchronize the sequence numbers and acknowledgment numbers of both parties and exchange information about the TCP window size.

Handshake process:

• First handshake: To establish a connection, the client first sends a connection request message, with SYN set to 1 and Sequence Number set to x. The client enters the SYN+SEND state and waits for confirmation from the server.
• Second handshake: The server receives the SYN message. After receiving the SYN message from the client, the server needs to confirm the SYN message and set the Acknowledgment Number to x+1 (Sequence+1). At the same time, it also sends a SYN message with the SYN bit set to 1 and the Sequence Number set to y. The server puts all the above information into a message segment (i.e., the SYN+ACK message segment) and sends it to the client. At this time, the server enters the SYN+RECV state.
• The third handshake: The client receives the SYN+ACK segment from the server. Then it sets the Acknowlegment Number to y+1 and sends an ACK segment to the server. After this segment is sent, both the client and the server enter the ESTABLISHED state, completing the TCP three-way handshake.

After completing the three-way handshake, the client and server can start transmitting data.

Four waves

When the client and server have finished transmitting data, they need to disconnect the TCP connection. The TCP disconnection process is four waves.

• First wave: The client (or server) sets the Sequence Number and Acknowledgment Number and sends a FIN segment to the server. At this time, the client enters the FIN_WAIT_1 state, which means that the client has no data to send to the host.
• Second wave: The server receives the FIN segment sent by the client and returns an ACK segment to the client. The Acknowledgement Number is the Sequence Number plus 1. The client enters the FIN_WAIT_2 state and the server enters the CLOSE_WAIT state. The server tells the client that I agree to your "close" request.
• The third wave: The server sends a FIN segment to the client, requesting to close the connection, and the server enters the LAST_ACK state.
• The fourth wave: The client receives the FIN segment sent by the server, sends an ACK segment to the host, and then the client enters the TIME_WAIT state. After the server receives the ACK segment from the client, it closes the connection. At this time, if the client does not receive a reply after waiting for 2MSL, it proves that the server has been closed normally. Well, the client can also close the connection.

The necessity of TCP three-way handshake

Prevent the server from waiting for client requests due to receiving connection request messages that have long expired, which will eventually lead to deadlock and waste of resources.

The necessity of TCP four-wave

In order to ensure that both communicating parties can notify each other, the connection needs to be released and disconnected.

Why does the client wait 2MSL before closing the connection?

• MSL: Maximum Segment Lifetime

After sending the four messages, you can directly enter the CLOSE state, but the network may be unreliable and anything may happen, such as the last ACK may be lost. Therefore, the TIME_WAIT state is used to resend the ACK message that may be lost. To be more specific:

• In order to ensure that the last connection release confirmation message sent by the client can reach the server, so that the server can release the connection normally.
• Prevent long-expired connection request messages from appearing in this connection. After the client sends the last connection release request confirmation message, after 2MSL time, all message segments generated during the duration of this connection will disappear from the network.

Comparison between TCP and UDP

• TCPUDP Reliability Reliable Unreliable Connectivity Connection-oriented Connectionless Message-oriented Byte stream-oriented Message-oriented Efficiency Low efficiency High efficiency Duplexity Full duplex One-to-one, one-to-many, many-to-one, many-to-many
• Supports multicast and broadcast flow control sliding window mechanism congestion control slow start/congestion avoidance
• Fast retransmission/fast recovery The transmission speed is slow and fast. The efficiency requirement is relatively low, and the accuracy requirement is relatively high.
• Scenarios requiring connection High efficiency requirements, low accuracy requirements Applications SMTP, TELNET, HTTP, FTPDNS, RIP, NFS, SNMP,
• IP Telephony, Streaming Media