Microsoft Press windows server 2008 tcp ip protocols and services phần 2 doc

18 Part I: The Network Interface Layer ❑ The Token Ring broadcast address (0xC0-00-FF-FF-FF-FF). A frame using the Token Ring broadcast address is designed to remain on a single ring and is not forwarded by Token Ring source-route bridges. ❑ A multicast address. ❑ A Token Ring functional address. A functional address is a type of multicast address that is specific to Token Ring and is typically used by Token Ring MAC management frames. ■ Source Address The Source Address field is 6 bytes long and indicates the sending node’s unicast address. ■ Payload The Payload field for a Token Ring frame consists of a PDU of an upper layer protocol. Unlike Ethernet, there is no minimum frame size and the maximum transmis- sion unit (MTU) for Token Ring is not a defined number, but dependent on factors such as the bit rate and the token holding time. Token Ring MTUs are further complicated by the presence of Token Ring source-routing bridges. More information on Token Ring MTUs for IP datagrams can be found in the section titled “IEEE 802.5 SNAP,” later in this chapter. ■ Frame Check Sequence The FCS field is a 4-byte CRC that uses the same algorithm as Ethernet to provide a bit-level integrity check of all fields in the Token Ring frame, from the Frame Control field to the Payload field. The FCS does not provide bit-level integrity for the Access Control or Frame Status fields. This allows bits in these fields, such as the Monitor bit, to be set without forcing a recalculation of the FCS. The FCS is checked as it passes each node on the ring. If the FCS fails at any node, the Error Detected indicator in the End Delimiter field is set to 1 and the receiving node does not copy the frame. ■ End Delimiter The End Delimiter is a 1-byte field that identifies the end of the frame. Like the Start Delimiter, the End Delimiter contains J and K nondata symbols to provide an explicit postamble. The End Delimiter field also contains the following: ❑ An Intermediate Frame indicator (1 bit), used to indicate whether this frame is the last frame in the sequence (when set to 0) or more frames are to follow (when set to 1). ❑ An Error Detected indicator (1 bit), used to indicate whether this frame has failed the FCS calculation. ❑ Because there is no Length field in the IEEE 802.5 frame, the End Delimiter is used to locate the end of the payload and the position of the FCS and Frame Status fields. Chapter 1: Local Area Network (LAN) Technologies 19 ■ Frame Status The Frame Status field is a 1-byte field that contains the following: Two copies of the Address Recognized indicator. The destination node sets the Address Recognized indicators to indicate that the address in the Destination Address field was recognized. Two copies of the Frame Copied indicator. The destination node sets the Frame Copied indicators to indicate that the frame was successfully copied into a buffer on the net- work adapter. ❑ Two copies of each indicator are needed because the FCS field does not protect the Frame Status field. ❑ The Address Recognized and Frame Copied indicators are not used as acknowl- edgments for reliable data delivery. The sending Token Ring network adapter uses these indicators to retransmit the frame, if necessary. Note The FCS, End Delimiter, and Frame Status fields are not visible with Network Monitor. IEEE 802.2 LLC Header The fields in the IEEE 802.2 LLC header are defined and used in the same way as the IEEE 802.2 LLC header for the IEEE 802.3 frame format, as discussed in the section titled “IEEE 802.3,” earlier in this chapter. IEEE 802.5 SNAP As described earlier in this chapter, the value of 0x06 is defined as the DSAP and SSAP for IP. However, it is not defined for use in RFC 1042 and not used in the industry. Therefore, similar to the case of IEEE 802.3 frames, to send an IP datagram over an IEEE 802.5 network, the IP datagram must be encapsulated using SNAP, as Figure 1-7 shows. For a 10-millisecond (ms) token-holding time, the maximum sizes for IP datagrams are 4464 bytes for 4-Mbps Token Ring network adapters and 17,914 bytes for 16-Mbps Token Ring net- work adapters. If Token Ring source-routing bridges are present, the maximum size of IP datagrams can be 508, 1020, 2044, 4092, and 8188 bytes. For more information on Token Ring MTUs, see RFC 1042. 20 Part I: The Network Interface Layer Figure 1-7 The IEEE 802.5 SNAP frame format showing the SNAP header and an IP datagram Special Bits on Token Ring MAC Addresses Within the Source Address and Destination Address fields of the IEEE 802.5 frame format, special bits are defined, as Figure 1-8 shows. The Individual/Group Bit Identical to Ethernet, the I/G bit for Token Ring addresses is used to indicate whether the address is a unicast (individual) or multicast (group) address. For unicast addresses, the I/G bit is set to 0. For multicast addresses, the I/G bit is set to 1. The Universal/Locally Administered Bit Identical to Ethernet, the U/L Administered bit for Token Ring addresses is used to indicate whether the IEEE has allocated the address. For universal addresses allocated by the IEEE, the U/L bit is set to 0. For locally administered addresses, the U/L bit is set to 1. The U/L bit is relevant for both the Source Address and Destination Address fields. . . . IEEE 802.2 LLC Header IEEE 802.5 Header IEEE 802.5 Trailer = 0xAA = 0xAA = 0x03 Start Delimiter Access Control Frame Control Destination Address Source Address DSAP SSAP Control Organization Code Ether Type IP Datagram Frame Check Sequence End Delimiter Frame Status = 0x00-00-00 SNAP Header = 0x08-00 Chapter 1: Local Area Network (LAN) Technologies 21 Figure 1-8 The special bits defined on Token Ring source and destination MAC addresses Functional Address Bit The Functional Address bit indicates whether the destination address is a functional address (when set to 0) or a nonfunctional address (when set to 1). Token Ring defines the following two types of multicast addresses: ■ Functional addresses Multicast addresses that are specific to Token Ring. There are spe- cific functional addresses for identifying the ring monitor, the ring-parameter server, and a source-routing bridge. ■ Nonfunctional addresses General multicast addresses that are not specific to Token Ring. The Functional Address bit is significant only if the I/G bit is set to 1. Routing Information Indicator Bit The Routing Information Indicator bit indicates whether MAC-level routing information is present. In the case of Token Ring, the Routing Information Indicator bit indicates the pres- ence of a source-routing header between the IEEE 802.5 header and the IEEE 802.2 LLC header. Token Ring source routing is not OSI Network Layer routing, but rather a MAC sub- layer routing scheme that allows a sending node to discover and specify a route through a defined series of rings and bridges within a Token Ring network segment. FDDI FDDI is a network technology developed by the American National Standards Institute (ANSI). FDDI is an optical fiber-based token passing ring with a bit rate of 100 Mbps. It was Destination Address Source Address 0 - Individual 1 - Group 0 - Universal Admin 1 - Local Admin 0 - No Routing 1 - Routing Present 0 - Universal Admin 1 - Local Admin 0 - Functional 1 - Nonfunctional 22 Part I: The Network Interface Layer designed to span long distances and, in most implementations, it acts as a campus-wide high- speed backbone. FDDI offers advanced features beyond Token Ring, such as the ability to self- heal a break in the ring and the use of guaranteed bandwidth. Although not developed by the IEEE as part of the 802 standards, the FDDI specification is quite similar to the IEEE 802.3 and 802.5 specifications; it defines the MAC sublayer of the OSI Data Link Layer and the Physical Layer, and it uses the IEEE 802.2 LLC sublayer. Copper Data Distributed Interface (CDDI) is a version of FDDI that operates over twisted-pair copper wire. RFC 1188 describes IP encapsulation over FDDI networks. FDDI Frame Format The FDDI frame format is the result of the IEEE 802.2 and ANSI FDDI specifications, and con- sists of an FDDI header and trailer and an IEEE 802.2 LLC header. Figure 1-9 shows the FDDI frame format. Figure 1-9 The FDDI frame format showing the FDDI header and trailer and IEEE 802.2 LLC header FDDI Header and Trailer The fields in the FDDI header and trailer are defined as follows: ■ Preamble The Preamble field is 2 bytes long and provides receiver synchronization. ■ Start Delimiter The Start Delimiter field is 1 byte long and identifies the start of the frame. Like Token Ring, the Start Delimiter field contains nondata symbols known as J . . . IEEE 802.2 LLC Header FDDI Header Preamble Start Delimiter Frame Control Destination Address Source Address DSAP SSAP Control Payload Frame Check Sequence End Delimiter Frame Status FDDI Trailer Chapter 1: Local Area Network (LAN) Technologies 23 and K symbols that are deliberate violations of the FDDI signal encoding scheme. The J symbol is an encoding violation of a 1 and the K symbol is an encoding violation of a 0. Note The Preamble and Start Delimiter fields are not visible with Network Monitor. ■ Frame Control The Frame Control field is 1 byte long and contains bits for the following: ❑ Setting the class of the frame (1 bit). FDDI frames can be sent as synchronous or asynchronous frames. Synchronous frames are used for guaranteed bandwidth and response time. Asynchronous frames are used for dynamic bandwidth shar- ing. This Class bit is set to 1 for synchronous frames and 0 for asynchronous frames. ❑ Setting the length of the Destination Address and the Source Address fields (1 bit). Like IEEE 802.3, FDDI supports 2-byte and 6-byte addresses. The Address bit is set to 1 for 6-byte addresses and 0 for 2-byte addresses. ❑ Indicating that what follows is a token (either nonrestricted or restricted), a station management frame, a MAC frame, an LLC frame, or an LLC frame with a specific priority (6 bits). ■ Destination Address The Destination Address field is either 2 bytes or 6 bytes long and indicates the address of the destination (2-byte addresses are seldom used). For 6-byte addresses, FDDI Destination Address fields are defined the same as Ethernet Destina- tion Address fields to provide easy interoperability between bridged or Layer 2 switched Ethernet and FDDI segments. The destination address is a unicast, multicast, or broad- cast address. ■ Source Address The Source Address field is either 2 bytes or 6 bytes long and indicates the unicast address of the sending node (2-byte addresses are seldom used). ■ Frame Check Sequence The FCS field is a 4-byte CRC that uses the same algorithm as Ethernet to provide a bit-level integrity check of all fields in the FDDI frame, from the Frame Control field to the Payload field. The FCS is checked as it passes each node on the ring. If the FCS fails at any node, the Error bit in the Frame Status field is set to 1 and the receiving node does not copy the frame. ■ End Delimiter The End Delimiter field is 1 byte long and identifies the end of the frame. Like the Start Delimiter field, the End Delimiter field contains J and K nondata symbols to provide an explicit postamble. Because there is no Length field in the FDDI frame, the End Delimiter field is also used to locate the end of the payload, and the position of the FCS and Frame Status fields. 24 Part I: The Network Interface Layer ■ Frame Status The Frame Status field is typically 2 bytes long and contains bits for the following: The Address Recognized indicator ❑ The destination node sets the Address Recognized indicator to show that the address in the Destination Address field was recognized. The Frame Copied indicator ❑ The destination node sets the Frame Copied indicator to show that the frame was successfully copied into a buffer on the network adapter. The Error indicator ❑ Any FDDI station sets the Error indicator to 1 when the FCS field is invalid. ❑ Similar to Token Ring, the Address Recognized and Frame Copied indicators are not used as acknowledgments for reliable data delivery. Rather, the sending FDDI network adapter uses these indicators to retransmit the frame if necessary. IEEE 802.2 LLC Header The fields in the IEEE 802.2 LLC header are defined and used in the same way as the IEEE 802.2 LLC header for the IEEE 802.3 and IEEE 802.5 frame format discussed earlier in this chapter. Payload The payload for an FDDI frame consists of a PDU of an upper layer protocol. The entire FDDI frame from the Preamble field to the Frame Status field can be a maximum size of 4500 bytes. Once you subtract the FDDI and IEEE 802.2 LLC headers, the maximum payload size is 4474 bytes with a 3-byte LLC header, and 4473 bytes with a 4-byte LLC header. FDDI SNAP As described earlier in this chapter, the value of 0x06 is defined as the SAP for IP. However, it is not defined for use in RFC 1188 and not used in the industry. Therefore, similar to the case of IEEE 802.3 frames and IEEE 802.5 frames, to send an IP datagram over an FDDI network, the IP datagram must be encapsulated using the SNAP header, as shown in Figure 1-10. The maximum-sized IP datagram that can be sent on an FDDI network is 4352 bytes. This number of bytes is the result of taking the maximum FDDI frame size of 4500 bytes and sub- tracting the FDDI header and trailer (23 bytes), the LLC header (3 bytes), and the SNAP header (5 bytes) and reserving 117 bytes for future purposes. Chapter 1: Local Area Network (LAN) Technologies 25 Figure 1-10 The FDDI SNAP frame format showing the SNAP header and an IP datagram IP datagrams and ARP messages sent over FDDI networks also have the following constraints: ■ Only 6-byte FDDI source and destination addresses can be used. ■ All IP and ARP frames are transmitted as asynchronous class LLC frames using unre- stricted tokens. RFC 1188 does not define how frame priorities are used or how the FDDI node deals with the values of the Address Recognized and Frame Copied indicators. FDDI nodes send ARP Requests using the Ethernet ARP Hardware Type value of 0x00-01, but can receive ARP Requests using the ARP Hardware Types of 0x00-01 and 0x00-06 (IEEE net- works). The use of the Ethernet ARP Hardware Type value is designed to allow FDDI hosts and Ethernet hosts in a bridged or Layer 2 switched environment to send and receive ARP messages. Special Bits on FDDI MAC Addresses Because FDDI MAC addresses are defined in the same way as Ethernet MAC addresses, the special bits on FDDI MAC addresses are the same as those defined for Ethernet MAC addresses. IEEE 802.2 LLC Header FDDI Header = 0xAA = 0xAA = 0x03 Preamble Start Delimiter Frame Control Destination Address Source Address DSAP SSAP Control Organization Code Ether Type IP Datagram Frame Check Sequence End Delimiter Frame Status = 0x00-00-00 . . . Up to 4352 bytes SNAP Header FDDI Trailer = 0x08-00 26 Part I: The Network Interface Layer IEEE 802.11 IEEE 802.11 is a set of standards for wireless LAN technologies. The original 802.11 standard defines wireless networking using either 1-Mbps or 2-Mbps bit rates in the Industrial, Scien- tific, and Medical (ISM) 2.54-gigahertz (GHz) frequency band. IEEE 802.11b defines a maxi- mum bit rate of 11 Mbps in the 2.54-GHz ISM band. IEEE 802.11a defines a maximum bit rate of 54 Mbps in the 5.8-GHz band. 802.11g defines a maximum bit rate of 54 Mbps in the 2.54- GHz band. IEEE 802.11b is the most widely deployed of the IEEE 802.11 standards. At the MAC sublayer, IEEE 802.11 (all versions) uses a combination of congestion avoidance and Request to Send (RTS), Clear to Send (CTS), and Acknowledgment (ACK) frames to ensure that only one wireless node is transmitting at a time and that the sent frame is success- fully received. IEEE 802.11 wireless nodes can communicate in the following ways: ■ Directly with each other using an operating mode known as ad hoc mode. ■ With a wireless access point (AP) using an operating mode known as infrastructure mode. In infrastructure mode, the wireless AP acts as a transparent bridge connecting wireless nodes to a wired network. To identify a wireless network in either operating mode, IEEE 802.11 uses a Service Set Iden- tifier (SSID), also known as a wireless network name. Because wireless networking uses broadcast radio waves, a wireless node within range of a transmitting wireless node can capture IEEE 802.11 frames and interpret the data. To provide data confidentiality (encryption) for IEEE 802.11 payloads, IEEE 802.11 networks can use Wi-Fi Protected Access 2 (WPA2), Wi-Fi Protected Access (WPA), or Wired Equivalent Privacy (WEP). IEEE 802.11 Frame Format The IEEE 802.11 frame format consists of an IEEE 802.11 header and trailer and an IEEE 802.2 LLC header. Figure 1-11 shows the IEEE 802.11 frame format. IEEE 802.11 Header and Trailer The fields in the IEEE 802.11 header and trailer for a data frame sent by wireless nodes or by a wireless AP to a wireless node are defined as follows: ■ Frame Control A 2-byte field that contains control information that defines the type of frame and how to process the frame. For more information, see the section titled “Frame Control Field,” later in this chapter. ■ Duration/ID Field A 2-byte field that is used to indicate the duration of time in micro- seconds needed to transmit the frame and the acknowledgment. Chapter 1: Local Area Network (LAN) Technologies 27 Figure 1-11 The IEEE 802.11 frame format showing the IEEE 802.11 header and trailer and the IEEE 802.2 LLC header ■ Address 1 A 6-byte field that contains either the destination MAC address of a wireless node (when sent by a wireless node to another wireless node in ad hoc mode or sent by the wireless AP to the wireless node) or the SSID (when sent by a wireless node to a wireless AP). ■ Address 2 A 6-byte field that contains either the MAC address of the sending node (when sent to another wireless node in ad hoc mode or sent to the wireless AP) or the SSID (when sent by the wireless AP to a wireless node). ■ Address 3 A 6-byte field that contains the SSID for frames sent to another wireless node in ad hoc mode, the source address for frames sent from the wireless AP to a wireless node, or the destination address for frames sent from a wireless node to a wireless AP. ■ Sequence Control A 2-byte field that contains a 4-bit Fragment Number field and a 12-bit Sequence Number field that, when used together, allow the receiver to discard duplicate frames. When a frame is fragmented, the Fragment Number field is used to indicate the number of the fragment. Otherwise, the Fragment Number field is set to 0. The Sequence Number field indicates the number of the frame starting at 0, incrementing to 4095, and then starting again at 0. All fragments of a frame have the same sequence number. . . . IEEE 802.2 LLC Header IEEE 802.11 Header Frame Control Duration/ID Address 1 Address 2 Address 3 Sequence Control Address 4 DSAP SSAP Control Organization Code Frame Check Sequence IEEE 802.11 Trailer [...]... common in many TCP/ IP implementations, TCP/ IP for Windows Server 20 08 and Windows Vista maintains a RAM-based table of IP and MAC address mappings Historically known as the ARP cache, in Windows Server 20 08 and Windows Vista, it is also known as the neighbor cache When an ARP exchange for address resolution is complete, both the ARP requester and the ARP responder have each other’s IP address-to-MAC... Interface Layer ARP in Windows Server 20 08 and Windows Vista Unlike ARP in previous versions of Windows, ARP in Windows Server 20 08 and Windows Vista is designed to work in the same way as Neighbor Discovery in IP version 6 (IPv6), as described in RFC 4861 Neighbor Discovery in IPv6 is the replacement for ARP, router discovery, and the redirect function in IP version 4 (IPv4) IPv6 nodes use a neighbor... Server 20 08 and Windows Vista and the Routing and Remote Access service for Windows Server 20 08 uses a variation of ARP called proxy ARP to forward IP datagrams between nodes on separate segments of a subdivided subnet Note This chapter assumes prior knowledge of the route determination process for IP hosts and routers in Microsoft Windows For more information, see Chapter 5, IP Routing,” of the TCP/ IP. .. allows you to bundle or aggregate the bandwidth of multiple physical connections It is supported by Windows Server 20 08 and Windows Vista Network Connections and the Windows Server 20 08 Routing and Remote Access service MP takes multiple physical connections and makes them appear as a single logical link For example, with MP, two analog phone lines operating at 28 .8 Kbps appear as a single connection... (ARP) 45 ARP for Windows Server 20 08 and Windows Vista supports the broadcast ARP Request and unicast ARP Reply exchange described in RFC 826 to perform address resolution As described in the “Duplicate Address Detection” and “Neighbor Unreachability Detection” sections of this chapter, Windows Server 20 08 and Windows Vista also support a unicast ARP Request and unicast ARP Reply exchange and a broadcast... Cache Similar to IPv6 nodes, ARP in Windows Server 20 08 and Windows Vista use a neighbor cache to store recently resolved IP address-to-MAC address mappings This was known as an ARP cache in previous versions of Windows You can view the neighbor cache in Windows Server 20 08 and Windows Vista with the following commands: Chapter 3: Address Resolution Protocol (ARP) 51 ■ netsh interface ipv4 show neighbors... forwarding IP datagrams on a local area network (LAN) link, it is important to understand how TCP/ IP uses Address Resolution Protocol (ARP) to resolve a next-hop IP address to its corresponding Network Interface Layer address TCP/ IP for Windows Server 20 08 and Windows Vista uses ARP for address resolution, duplicate address detection, and neighbor unreachability detection The Network Bridge for Windows Server. .. IEEE 8 02. 11 frame, from the Frame Control field to the Payload field Frame Check Sequence IEEE 8 02. 2 LLC Header The fields in the IEEE 8 02. 2 LLC header are defined and used in the same way as the IEEE 8 02. 2 LLC header for the IEEE 8 02. 3, IEEE 8 02. 5, and FDDI frame formats discussed earlier in this chapter Payload The payload for an IEEE 8 02. 11 frame can be a maximum size of 23 12 bytes IEEE 8 02. 11 payloads... maximum frame size of 1600 bytes and a 2- byte address field, the IP MTU for Frame Relay is 15 92 Summary Typical WAN technology encapsulations used by Windows Server 20 08 and Windows Vista provide delimitation, addressing, protocol identification, and bit-level integrity services IP datagrams sent over point-to-point WAN links can be encapsulated using PPP or MP IP datagrams and ARP messages sent over Frame... the SPA and TPA set to an existing entry in the neighbor cache (as sent by previous versions of Windows) , ARP does not update the entry with the offending node’s MAC address With Windows Server 20 08 and Windows Vista, there are two different exchanges when there is an IP address conflict, depending on the version of Windows running on the offending node Offending Node Runs Windows Server 20 08 or Windows . or aggregate the bandwidth of multiple physical connections. It is supported by Windows Server 20 08 and Windows Vista Network Connections and the Windows Server 20 08 Routing and Remote Access. broadband Internet connections. Figure 2- 2 shows a typical PPP encapsulation for an IP datagram when using Address and Control field suppression and Protocol field compression. Figure 2- 2 Typical. if necessary. IEEE 8 02. 2 LLC Header The fields in the IEEE 8 02. 2 LLC header are defined and used in the same way as the IEEE 8 02. 2 LLC header for the IEEE 8 02. 3 and IEEE 8 02. 5 frame format discussed