Home Page

 *What are the different physical Ethernet network types?

       Some of the physical Ethernet types as defined are:

       10BASE5 - 10BASE5 is the original design of the traditional

                 Ethernet backbone, designed to be left in place

                 permanently or for extended periods.

       10BASE2 - 10BASE2 is the original design for a departmental or

                 workgroup sized Ethernet environment. It is designed to be

                 simple, inexpensive, and flexible as people and stations

                 move.

       10BROAD36 - 10BROAD36 is a seldom used Ethernet specification which

                   uses a physical medium similar to cable television, with

                   CATV-type cables, taps, connectors, and amplifiers.

       1BASE5 - 1BASE5 is a specification of Ethernet that runs at 1 Mb/s

                over twisted pair wiring. This physical topology uses

                centralized hubs to connect the network devices.

       10BASE-T - 10BASET provides Ethernet services over twisted pair

                  copper wire.

       FOIRL - Fiber Optic Inter-Repeater Link - This specification of the

               802.3 standard defines a standard means of connecting

               Ethernet repeaters via optical fiber.

       10BASE-F - 10BASE-F is a set of optical fiber medium specifications

                  which define connectivity between devices.

       100BASE-T - 100BASE-T is a series of specifications that provides

                   100 megabit speeds over copper or fiber. These

                   topologies are often referred to as Fast Ethernet.

       Gigabit Ethernet - Gigabit Ethernet provides speeds of 1000 Mb/s

                          over copper and fiber.

 * What does baseband and broadband mean?

       A baseband network has a single channel that is used for

       communication between stations. Ethernet specifications which use

       BASE in the name refer to baseband networks.

       A broadband network is much like cable television, where different

       services communicate across different frequencies on the same cable.

       Broadband communications would allow a Ethernet network to share the

       same physical cable as voice or video services. 10BROAD36 is an

       example of broadband networking.

* What is the difference between a bus topology and a star topology?

       A bus topology is a networking architecture that is linear, usually

       by using one or more pieces of cable to form a single line, or bus.

       The signals sent by one station extend the length of this cable to

       be heard by other stations.

       A star topology is an architecture that includes a central device or

       hub to connect all stations together. Signals sent by a station must

       pass through (and are usually regenerated) by these central hubs.

       Since the hub sits in the center and all other stations are linked

       through the hub, the architecture resembles a star.

*What physical Ethernet topologies are no longer popular?

       There are a number of physical networking components specified in

       the IEEE 802.3 specification, but many of those early physical

       networking components are not used in most modern Ethernet networks.

       However, there may be instances where an existing legacy network

       still exists which uses these older components. Since these older

       pieces of equipment are still part of the 802.3 specification, there

       are no technical reasons why an Ethernet network would not operate

       properly with these components. The two most popular older Ethernet

       technologies are 10BASE5 and 10BASE2.

       10BASE5

       -------

       10BASE5 is the original Ethernet backbone, and is occasionally

       referred to as thicknet or thick Ethernet because of the thick 50

       ohm coax that was used as the physical medium. 10BASE5 is a bus

       topology that uses transceiver cables to attach stations to the

       central 10BASE5 cable.

       Maximum segment length: 500 meters

       Maximum number of segments connected with repeaters: 5 (2500 meters)

       Maximum attachments per segment: 100

       Minimum separation between attachments: 2.5 meters

       10BASE2

       -------

       10BASE2 is designed as a smaller and less expensive alternative to

       10BASE5, and is sometimes referred to as Thinnet or Thin Ethernet

       because of the much smaller cables. 10BASE2 is also a bus topology,

       but each of the workstations use a 'T' BNC connector to connect

       workstations to the central bus.

       Maximum segment length: 200 meters

       Maximum number of segments connected with repeaters: 5 (1000 meters)

       Maximum attachments per segment: 30

       Minimum separation between attachments: .5 meters

       * What are the most common physical Ethernet networks used today?

       Most modern Ethernet networks use twisted pair copper cabling or

       fiber to attach devices to the network. The 10BASE-T, 100BASE-T, and

       Gigabit Ethernet topologies are well suited for the modern cabling

       and fiber infrastructures.

 * What pin assignments are used in twisted-pair Ethernet cabling?

       Twisted-pair Ethernet (10BASE-T, 100BASE-T, or 1000BASE-T) uses an

       RJ-45 connector, which is an eight-pin modular connector.

       Contact 1   Transmit +

       Contact 2   Transmit -

       Contact 3   Receive +

       Contact 4   Not Used

       Contact 5   Not Used

       Contact 6   Receive -

       Contact 7   Not Used

       Contact 8   Not Used

       When looking at an RJ-45 wall jack (female), contact 1 is on the

       left and contact 8 is to the right. When looking at the RJ-45

       connector on the end of a cable (male) with the tab on the bottom

       and the contacts on the top, contact 8 is on the left and contact 1

       is to the right.

 * Can two Ethernet stations be directly attached with 10BASE-T?

       Two Ethernet stations can be directly attached to each other, but

       the cabling will be wired differently than a normal 10BASE-T

       Ethernet network connection. The 802.3 specification refers to this

       direct connection between two stations as a crossover function.

       The crossover function is accomplished by simply wiring the receive

       pins to the transmit pins:

       Contact 1 - Contact 3

       Contact 2 - Contact 6

       Contact 3 - Contact 1

       Contact 6 - Contact 2

*What is propagation delay?

        The propagation speed of a medium refers to the speed that the data

        travels through that medium. Propagation delays differ between

        mediums, which affect the maximum possible length of the Ethernet

        topology running on that medium.

        In the following table, c refers to the speed of light in a vacuum,

        or 300,000 kilometers per second.

        Medium        Propagation Speed

        ------        -----------------

        Thick Coax    .77c (231,000 km/sec)

        Thin Coax     .65c (195,000 km/sec)

        Twisted Pair  .59c (177,000 km/sec)

        Fiber         .66c (198,000 km/sec)

        AUI Cable     .65c (195,000 km/sec)

        From these values, the size of a bit on 10BaseT can be calculated.

        10BaseT is twisted pair, which has a propagation delay of 177,000

        km/sec. 177,000 km/sec divided by 10 million bits per second is

        17.7 meters, or the size of a single bit on a 10BaseT network.

        The maximum propagation delay through the network can be calculated

        by dividing the maximum length by the speed. For 10Base2 thin coax

        network, this is 185 meters divided by 195,000 km/sec, or 950

        nanoseconds. If the actual propagation delay from one end of the

        network to the other is greater than 950 nanoseconds, late

        collisions may occur. See section [5.4] for more information on

        late collisions.

 * What is an interframe gap?

        The inteframe gap is the amount of time that is specified between

        frames transmitted from a workstation. The designers of the

        Ethernet specification arbitrarily chose 96 bit times to occur

        between frames from a transmitting station.

        This delay is designed to provide the workstations on the Ethernet

        network with some 'breathing time' between frames to perform normal

        Ethernet housekeeping functions on the network interface card.

 Ethernet Data Link Layer

* What are the different Ethernet frame formats?

       Ethernet Version 2 and IEEE 802.3 Frame Formats

       -----------------------------------------------

       The Ethernet Version 2 frame format was designed before the IEEE

       specifications, but is almost identical to the 802.3 frame type.

       With the Ethernet Version 2 frame type, a two-byte Type field

       follows the source station's six-byte MAC address. In the 802.3

       frame type, this two-byte field after the source address is a length

       field specifying the number of bytes in the LLC and data fields. If

       these two bytes are greater than 05DC hex (1500 decimal), the frame

       is a Version 2 frame. Since all type fields are greater than 1500

       decimal (the maximum Ethernet frame size), both frame types can

       easily coexist on the same network. Some network protocol analyzers

       call a Version 2 frame an Ethertype frame because of this two-byte

       Type field.

       This is an Ethernet Version 2 frame:

       +--------------+

       |                      | The preamble consists of 62 bits of alternating

       |   Preamble   | ones and zeros that allows the Ethernet card to

       |   7 bytes      | synchronize with the beginning of a frame.

       |                      |

       +--------------+ The Start Frame Delimiter is the sequence

       | SFD - 1 byte | 10101011, and indicates the start of a frame.

       +--------------+

       |                       | The destination address is a six byte Media Access

       | Destination | Control (MAC) address, usually burned into the

       |   6 bytes       | ROM of the Ethernet card.

       +--------------+

       |                    | The source address is a six byte MAC address, and

       |   Source     | can signify a physical station or a broadcast.

       |   6 bytes    |

       +--------------+

       |     Type     | The Type field (see explanation above).

       |    2 bytes   |

       +--------------+

       |                   |  Any higher layer information is placed in the

       |    Data      |  data field, which could contain protocol

       |                  |  information or user data.

       ~              ~

       ~              ~

       |  46 to 1500  |

       |    bytes       |

       |                     |

       +--------------+

       |     FCS        |  The Frame Check Sequence is a cyclic redundancy

       |   4 bytes    |  check used by the sending and receiving stations

       +--------------+  to verify a successful transmission. The FCS is

                         based on the contents of the destination address,

                         source address, type, and data.

       The 802.2 Logical Link Control (LLC) Information

       ------------------------------------------------

       The IEEE 802.3 Ethernet specification was intended to be used with

       the 802.2 Logical Link Control (LLC) specification. The LLC

       information envelops the data of the frame, and the 802.3 headers

       envelop this 802.2 LLC protocol data unit (PDU).

       This is the frame structure for an 802.3 Ethernet frame that

       contains the 802.2 LLC information:

       +----------------+

       |                        |

       |    Preamble    |

       |    7 bytes       |

       |                        |

       +----------------+

       |  SFD - 1 byte  |

       +----------------+

       |                        |

       |  Destination  |

       |    6 bytes       |

       +----------------+

       |             

       |     Source     |

       |    6 bytes     |

       +----------------+

       |  Frame Length  |

       |    2 bytes       |

       +----------------+

       |  DSAP - 1 byte |  The Destination and Source Service Access Point

       +----------------+  fields determine the protocol used for the upper

       |  SSAP - 1 byte |  protocol type of the frame.

       +----------------+

       |Control - 1 byte|  The Control field is used for administration by

       +----------------+  certain protocols.

       |      Data      |

       |                    |

       ~                ~

       ~                ~

       |   46 to 1500   |

       |     bytes         |

       |                        |

       +----------------+

       |      FCS           |

       |    4 bytes       |

       +----------------+

       The 802.2 Sub-Network Access Protocol (SNAP)

       --------------------------------------------

       After the 802.2 frame type was defined, many people felt that a

       single byte for DSAP and SSAP would not be sufficient to handle the

       growth of protocols into the future. A single byte DSAP or SSAP can

       only specify 256 separate protocols, and many of those were

       predefined from the beginning of the 802.2 specification.

       To provide future growth, the Sub-Network Access Protocol (SNAP) was

       created as an extension to the 802.2 specification. To differentiate

       this protocol from the original 802.2 specification, 802.2 SNAP uses

       the DSAP and SSAP of 0xAA.

       This is an 802.2 SNAP frame encapsulated in an 802.3 frame:

       +----------------+

       |                        |

       |    Preamble    |

       |    7 bytes       |

       |                        |

       +----------------+

       |  SFD - 1 byte  |

       +----------------+

       |                          |

       |  Destination   |

       |    6 bytes       |

       +----------------+

       |                      |

       |     Source     |

       |    6 bytes     |

       +----------------+

       |  Frame Length  |

       |    2 bytes       |

       +----------------+

       |  DSAP - 1 byte |

       +----------------+

       |  SSAP - 1 byte |

       +----------------+

       |Control - 1 byte|

       +----------------+ The Organizationally Unique ID (OUI) is assigned

       | OUI - 3 bytes  | to unique vendors to help differentiate protocols

       |                          | from different manufacturers.

       +----------------+

       | Type - 2 bytes | The two-byte protocol type defines a specific

       +----------------+ protocol in the SNAP. This also maintains a

       |                        | compatibility with Ethernet v2.

       |      Data         |

       |                       |

       ~                ~

       ~                ~

       |   46 to 1500   |

       |     bytes         |

       |                        |

       +----------------+

       |      FCS          |

       |    4 bytes       |

       +----------------+

       Novell 802.3 'Raw' Frame Format

       -------------------------------

       Before the final 802.2 LLC specifications were finalized, Novell

       implemented IPX/SPX over Ethernet. For this reason, Novell

       originally utilized 802.3 Ethernet without using 802.2 LLC. Because

       of this lack of LLC header, this frame type was nicknamed 802.3

       'raw'. Since Novell created this proprietary frame type for their

       own use, no other manufacturer uses this frame type.

       To implement their 'raw' frame type, Novell used the first two bytes

       of the 802.3 data field as 0xFFFF. Since the DSAP and SSAP values of

       0xFF do not exist, it becomes easy to differentiate between the

       802.3 and 802.3 'raw' frame types.

* What is transparent bridging?

       Transparent bridging is a method to connect two similar network

       segments to each other at the datalink layer. It is done in a way

       that is transparent to end stations, hence end-stations do not

       participate in the bridging algorithm.

       Transparent bridges are sometimes called learning bridges.

       When they are turned on and receive data packets from a network

       segment they:

       1) learn the relation between MAC address and segment/port, and

       2) forward the packet to all (!) other segments/ports.

       The first step in this process is essential to the "learning" aspect

       of the bridge. After some time the bridge has learned that a

       particular MAC address, say MACa, is on a particular segment/port,

       say PORT1. When it receives a packet destined for the MAC address

       MACa (from any port not being PORT1) it will no longer forward the

       packet to all ports (step 2). It knows that MACa is associated with

       PORT1 and will only forward the packet to PORT1.

* What is the spanning tree protocol?

       Spanning tree is a protocol defined in IEEE 802.1q to prevent

       bridges from creating network loops. Using the spanning tree

       protocol, bridges communicate to each other and disable certain

       ports/segments to prevent looping of packets.

       Many implementations of the spanning tree protocol are configured so

       an alternate path is available to network traffic, should the

       original path become disabled.

* What is Ethernet switching?

       From a functional point of view, switching is exactly the same as

       bridging. However switches use specially designed hardware called

       Application Specific Integrated Circuits (ASICs) to perform the

       bridging and packet-forwarding functionality (as supposed to

       implementations using a central CPU and special software).

       Consequently, switches are much faster than bridges.

       Ethernet switches also offer additional capabilities such as virtual

       LANs (VLANs) and full duplex connectivity.

* Ethernet Errors and Troubleshooting

 *What is a collision, and how many collisions are bad?

       Ethernet networking uses collisions as one of the contention access

       methods. When the network carrier is not active, any station can

       send information. If two stations attempt to send information at the

       same time, the signals overlap with each other, creating a

       collision.

       Collisions are not errors! Many people misinterpret a flashing

       collision light or a collision counter as a network problem!

       Although the term 'collision' may bring to mind a terrible crash, be

       assured that a collision is a normal part of Ethernet networking.

       The total number of collisions that occur on a network may be

       related to traffic patterns or utilization. Because of this

       variability of collisions, it is not applicable to define a 'good'

       or 'bad' level of collisions. In most cases, detailed analysis of

       collisions alone yields very little qualitative network health

       information.

 *What is the Signal Quality Error (SQE) Test?

       The SQE Test is used to test for the collision present circuit

       between a transceiver and a network interface card (NIC). After data

       is successfully transmitted, the Ethernet transceiver asserts the

       SQE signal on the collision presence circuit of the NIC. The NIC

       sees this test signal as a verification that the transceiver will

       inform the NIC when a collision occurs.

       In most modern Ethernet networks, the SQE test is not used or

       applicable. Most NICs now have an integrated transceiver and

       therefore have a hard-wired AUI, so a test for the collision

       presence circuit is unnecessary.

 *What is jam?

       When a collision is recognized by a transmitting station, a bit

       sequence called jam is transmitted. This jam is 32 bits long, which

       is long enough to traverse the entire collision domain so that all

       transmitting stations can detect the collision.

       Interestingly enough, the actual format of jam is unspecified in

       the 802.3 specifications. Most manufacturers have used alternating

       1s and 0s as jam, which is displayed as 0x5 (0101) or 0xA (1010)

       depending on when the jam is captured in the data stream.

       In many Fast Ethernet implementations, the jam has been seen as

       other arbitrary values, such as 1101000 (0xD0) or 10000110 (0x43).

       The reasoning for this particular jam pattern isn't very obvious. If

       anyone has more information on this jam sequence, please email

 *What is a late collision, and why is it bad?

       A collision is considered late if the jam occurs after 512

       bit-times, or 64 bytes. Collisions that occur after the first 64

       bytes of a frame may be indicative of a network design problem (the

       network is so large the jam cannot traverse the entire length in 32

       bit-times), or a hardware or Ethernet firmware issue.

       When collisions do not propagate the network quickly enough, a

       collision could occur between two stations without the stations

       aware that the packets collided. In this situation, the frames are

       simply lost, and the upper-layer protocols must begin a

       retransmission process to retransmit the information. These

       retransmissions can cause large delays, especially at the

       application layer.

*What is a runt?

       In Ethernet networks, any frame shorter than the minimum 64 bytes

       but with a valid CRC is considered a runt. Other frame-length errors

       in Ethernet are long frames, which are longer than 1518 bytes yet

       have a valid CRC.

* What is jabber?

       Jabber is described most often as a frame greater than the maximum

       of 1518 bytes with a bad CRC. A jabbering NIC is often indicative of

       a hardware problem with a NIC or transceiver.

* What is a CRC/Alignment error?

       When a station sends a frame, it appends a Cyclical Redundancy

       Check to the end of the frame. This CRC has been generated from an

       algorithm and is based on the data in the frame. If the frame is

       altered between the source and destination, the receiving station

       will recognize that the CRC does not match the actual contents of

       the packet.

       All frames should end on an 8-bit boundary, but problems on the

       network could cause the number of bits to deviate from the multiple

       of 8.

       Both CRC errors and alignment errors are grouped together as the

       single CRC/Alignment error counter.