Objective 1 Identify and describe the functions of each of the seven layers of the OSI reference model.
Objective 2 Describe connection-oriented network service and connectionless network service, and identify the key differences between them.
Objective 4 Identify at least 3 reasons why the industry uses a layered model.
Objective 5 Define and explain the 5 converstion steps of data encapsulation.
Objective 7 List the key internetworking functions of the OSI Network layer and how they are performed in a router.
Application, presentation, session, transport, network, data link, and physical.
2. What is the main purpose of Layer 7?
Provides standardized services to applications. The definition for this layer is typically ambiguous because it varies. The key is that it does not define a user interface, but instead a toolbox used by application developers. For example, a web browser is an application that uses HTML format text, as defined by the TCP/IP application layer, to describe the graphics to be displayed on the screen.
3. What is the main purpose of Layer 6?
Defines data formats and possibly encryption.
4. What is the main purpose of Layer 5?
The session layer controls the conversation between two endpoints. Although the term used is "session," the term "conversation" more accurately describes what is accomplished. The session layer ensures that not only communication, but useful sets of communication between endpoints is accomplished.
5. What is the main purpose of Layer 4?
To provide error recovery if requested.
6. What is the main purpose of Layer 3?
The network layer defines logical addressing and routing, as a means to deliver data across an entire network.
7. What is the main purpose of Layer 2?
The data link layer defines addressing specific to a particular medium as part of the means to provide delivery of data across that medium.
8. What is the main purpose of Layer 1?
The physical layer is responsible for encoding of energy onto the medium and interpretation of a received energy signal. It also defines the connector and cabling details.
9. Describe the process of data encapsulation as data is processed from creation until it exits a physical interface to a network. Use the OSI model as an example.
Data encapsulation represents the process of a layer adding a header, and possibly a trailer, to the data as it is processed by progressively lower layers in the protocol specification. In the context of OSI, each layer could add a header so that-other than the true application data-there would be six other headers (Layers 2-7) and a trailer for Layer 2, with this L2-PDU being encoded by the physical layer onto the network interface.
10. Describe the services provided in most connectionless protocol services.
By avoiding features like error correction and windowing, connectionless protocols require fewer bytes in their headers, meaning less overhead. Also, without windowing, there is no artificial slowing of the rate at which data can be sent.
11. Name at least three connectionless protocols.
LLC type 1, UDP, IPX, IP, PPP. Remember, Frame Relay, X.25, and ATM are connection oriented, regardless of whether they define error recovery.
12. Describe the services provided in most connection-oriented protocol services.
Error recovery is provided in most; others simply provide a pre-established path. Error recovery implies that the protocol is connection oriented. Connection oriented, however, does not necessarily imply that the protocol supplies error recovery.
13. In a particular error recovering protocol, the sender sends three frames, labeled 2, 3, and 4. The receiver of these frames, on its next sent frame, sets an acknowledgment field to "4". What does this typically imply?
That frames up through number 3 were received successfully. Most windowing, error recovery protocols use forward acknowledgment.
14. Name three connection-oriented protocols.
TCP, SPX, LLC Type 2, X.25. All the protocols in the answer provide error recovery. ATM and Frame Relay are also connection oriented, but without error recovery.
15. What does MAC stand for?
Media Access Control.
16. Name three terms popularly used as a synonym for MAC Address.
NIC address, card address, LAN address, hardware address, Ethernet address, Token Ring address, FDDI address, burned-in address. All of these names are used casually and in formal documents, and refer to the same 6 byte MAC address concept as defined by IEEE.
17. Are IP addresses defined by a Layer 2 or Layer 3 protocol?
Layer 3. There is a catch. If you examine the TCP/IP protocol stack, technically, it is Layer2. However, compared to OSI, IP most closely matches Layer 3, 50 the popular (and CCNA Exam) answer is Layer 3.
18. Are IPX addresses defined by a Layer 2 or Layer 3 protocols?
Layer 3. See the answer to question 17.
19. Are OSI NSAP addresses defined by a Layer 2 or a Layer 3 protocol?
Layer 3. These are, of course, truly Layer 3 because they are defined by OSI. The number of bits in the address is variable. However, I consider it highly unlikely that questions about NSAPs would be on the exam because they are not mentioned in any objective or covered in any class.
20. What portion of a MAC address encodes an identifier representing the manufacturer of the card?
The first three bytes.
21. Are MAC addresses defined by a Layer 2 or a Layer 3 protocol?
Layer 2. Ethernet and Token Ring MAC addresses are defined in the 802.3 and 802.5 specifications.
22. Are DLCI addresses defined by a Layer 2 or a Layer 3 protocol?
Layer 2. While not specifically covered in this chapter, Frame Relay protocols do not define a logical addressing structure that can usefully exist outside of a Frame Relay network; by definition, the addresses would be OSI Layer 2 equivalent.
23. Name two differences between Layer 3 addresses and Layer 2 addresses.
Layer 3 addresses can be used regardless of media type, whereas Layer 2 addresses are only useful on a particular media type. Layer 3 addresses are designed to imply a minimum of two parts, the first of which creates a grouping concept. Layer 2 MAC addresses have no grouping concept that implies that the addresses in the same group must reside on the same data link, like network addressing implies in many cases. However, all burned-in addresses from a single vendor can be considered to be "grouped," in that they all use the same value for the first three bytes of their MAC addresses.
24. How many bits in an IP address?
32 bits. A variable number in the network portion, and the rest of the 32 in the host portion. IP Version 6 uses a much larger address. Stay tuned!
25. How many bits in an IPX address?
80 bits. 32 bits in the network portion, and 48 bits in the node portion.
26. How many bits in a MAC address?
48 bits. The first 24 bits for burned-in addresses are a code that identifies the manufacturer.
27. How many bits in a DLCI address?
Typically 10, or up to 12 with extended addressing bits. Most typically 10 bits in length, for decimal 0-1023. Some are reserved for LMI and other purposes.
28. Name the two main parts of an IPX address. Which part identifies which "group" this address is a member of?
Network number and node number. Addresses with the same network number are in the same group. On LAN interfaces, the node number is made to have the same value as the LAN MAC address.
29. Name the two main parts of an IP address. Which part identifies which "group" this address is a member of?
Subnet and host. Addresses with the same subnet number are in the same group. Technically, as described in Chapter 5, there are three portions of the IP address: network, subnet, and host. However, most people think of the network and subnet portions as one portion.
30. Name the two main parts of a MAC address. Which part identifies which "group" this address is a member of?
There are no parts, and nothing defines a grouping concept. This is a trick question. While you might have guessed that the MAC address has two parts, the first part being dictated to the manufacturer, the second part being made up by the manufacturer, there is no grouping concept.
31. Name three benefits to layering networking protocol specifications.
Reduces complexity, standardizes interfaces, facilitates modular engineering, ensures interoperable technology, accelerates evolution, and simplifies teaching and learning. The answers shown here are directly from the ICRC course book.
32. What header and/or trailer does a router discard as a side effect of routing?
The data-link header and trailer. This is because the network layer, where routing is defined, is interested in delivering the network layer (Layer 3) PDU from end-to-end. Routing uses intermediate data links (Layer 2) to transport the data to the next routers and, eventually, the true destination. The data-link header and trailer is only useful to deliver the data to the next router or host; so the header and trailer are discarded by each router.
33. Describe the differences between a routed protocol and a routing protocol.
The routed protocol defines the addressing and Layer 3 header in the packet that is actually forwarded by a router. The routing protocol defines the process of routers exchanging topology data such that the routers know how to forward the data.
34. Name at least three routed protocols.
IP IPX, OSI, DECNET, AppleTalk, VINES
35. Name at least three routing protocols.
RIP, IGRP, EIGRP OSPF, NLSP, RTMP VTP, IS-IS
36. How does an IP host know what router to send a packet to? In which cases does an IP host choose to send a packet to this router instead of directly to the destination host?
By configuring a default route. If the destination of the packet is in another subnet, the host sends the packet to the default router. Otherwise, the host sends the packet directly to the destination host because it is in the same subnet and by definition must be on the same data link.
37. How does an IPX host know which router to send a packet to? In which case does an IPX host choose to send a packet to this router instead of directly to the destination host?
The node will use a RIP request to locate any servers on the attached IPX network that have a route to the destination network. If the destination is an IPX address on the attached network, a router is not needed, and the node will forward the packet directly.
38. Name three items in an entry in any routing table.
The group identifier, the interface to forward the packet out, and the Layer 3 address of the next router to send this packet to.