cisco certification 300-115

cisco certification 300-115.Indicates that "This machine has received all segments up to sequence number 399 and expects to receive segments with sequence number 400".Since each message segment has a sequence number attached to it, the TCP receiver can correctly sort the received data. For example, even if an out-of-sequence or duplicate message segment is received, the TCP receiver can identify it and put it in order correctly.Figure 1-85 shows how TCP detects and recovers from errors.Suppose that a TCP client sends three message segments with sequence numbers 100, 200, and 300. The message segment with sequence number 200 is lost or corrupted because of an error during transmission. If the message segment is corrupted during transmission, the TCP receiver can detect the message segment corruption based on the checksum field value in the TCP header. Because the server fails to correctly receive a message segment with a sequence number of 200, it will only reply with an acknowledgement segment with a maximum acknowledgement number of 200. As a result, the TCP client needs to resend the segment with a sequence number of 200. When it receives a dropped sequence number 200 segment, the server performs a "regular acknowledgement" and sends an acknowledgement number 400 to the TCP client, since the server has already received a sequence number 300 segment. The server thus indicates to the TCP client that it is ready to receive a segment with sequence number 400, and so on. It is worth noting that as long as the server does not receive a segment with a sequence number of 200, it will continue to send an acknowledgement of 200 to the TCP client, asking it to resend the lost segment.1.8 Transport Layer Technologies and Protocols 103TCP Client ServerOops! Something's wrong. The server did not receive the message segment with the sequence number 200, so send this message segment againFigure 1-85 TCP's method for detecting and recovering from errors5. TCP Flow ControlThe TCP protocol performs flow control (flow control) through a "sliding window" mechanism. At the beginning of a TCP session, TCP endpoints use the window size field value contained in the TCP header to inform themselves of the capacity of the receive cache when exchanging TCP message segments, and to control the amount of data they can receive and process accordingly. Each TCP endpoint maintains a local receive window (RWND), and the capacity of that window is the maximum amount of data that the TCP recipient can receive and cache for further processing. The TCP endpoint fills in this RWND value in the window size field in the TCP header when it sends out a TCP message segment. The TCP sender determines the capacity of the sliding window based on the RWND value advertised by the TCP receiver. In the absence of an acknowledgement, the number of TCP message segments that the TCP sender can send is limited by the value of the window size field advertised by the TCP receiver. This mechanism is particularly useful for the TCP client-server communication model. For example, with the "sliding window" mechanism, a TCP client can ask the server to slow down the data, and the server can ask the TCP client to slow down or speed up the data. TCP endpoints can increase or decrease the data transfer rate depending on the network conditions and local processing power to avoid "data flooding" of the TCP receiver. The size of the receive window is communicated through the "window size" field in the TCP header. Figure 1-86 shows how to adjust the TCP receive window according to the processing power of the TCP receiver.