Chapter 26

26.1	Identify all the delayed ACKs in Figure 26.5.
A:	We expect segments 2, 4, and 9 from the server to be delayed. The time difference between segments 2 and 4 is 190.7 ms and the time difference between segments 2 and 9 is 400.7 ms. All the ACKs from the client to the server appear to be delayed: segments 6, 11, 13, 15, 17, and 19. The time differences of the last five from segment 6 are 400.0, 600.0, 800.0, 1000.0, and 2600 ms.
26.2	Why was segment 12 in Figure 26.7 sent?
A:	If one end of a connection is in TCP's urgent mode, then every time a segment is received, one is sent. This segment does not tell the receiver anything new (it is not acknowledging new data, for example), and it contains no data, it just reiterates that urgent mode has been entered.
26.3	We said that the Rlogin client must use a reserved port (Section 1.9). (Normally the Rlogin client only uses reserved ports in the range 512-1023.) What limitation does this present to a host? Is there a way around this?
A:	There are only 512 of these reserved ports (512-1023), limiting a host to 512 Rlogin clients . The limit is normally less than 512 in real life, since some of the port numbers in this range are used as well-known ports by various servers, such as the Rlogin server. TCP's limitation is that the socket pair defining a connection (the 4-tuple) must be unique. Since the Rlogin server always uses the same well-known port (513) multiple Rlogin clients on a given host can use the same reserved port only if they're connected to different server hosts . Rlogin clients, however, don't use this technique of trying to reuse reserved ports. If this technique were used, the theoretical limit is a maximum of 512 Rlogin clients at any one time that are all connected to the same server host.