CCNP 03 - WANS

WAN Problem Areas

Certification Objectives

• Point-to-Point Protocols
• Switched Protocols
  

    In this chapter, we will discuss troubleshooting point-to-point and switched protocols. It’s very important to understand the concepts and functionality of each protocol. We will discuss the HDLC and SDLC encapsulation methods, as well as both CHAP and PAP authentications via PPP. We will identify ISDN, Frame Relay, X.25, SMDS, and ATM transmissions and common problems associated with each of them.

Point-to-Point Protocol

    In the 1980s, the Internet really started to come to life due to its usage by the U.S. government, universities, and large corporations. These hosts were being supported by the Internet Protocol (IP) and connected via a series of LANs. The most common of these LANs was the Ethernet. Other hosts were connected through WANs, using technologies such as X.25. This eventually evolved into the need to transmit data via the original method used before networking was ever established: the point-to-point or serial connection. To accommodate this need, a standard Internet encapsulation protocol, the Point-to-Point Protocol (PPP), was designed. Not only did PPP standardize Internet encapsulation of IP over point-to-point links, it also addressed other issues such as encapsulation and error detection.

PPP Frame

    The PPP frame consists of three primary parts (see Figure 3-1). It contains a header, which is subdivided into an 8-bit address field, an 8-bit control field and a 16-bit protocol field. The body or payload of the packet contains a variable length (up to 1500 bytes) of reserved bandwidth for user data. Finally, there is a trailer that contains a 16-bit frame check sequence (FCS). Following are the roles of these fields:

• Address The datagram’s destination
• Control The binary sequence that calls for transmission type of user data in an unsequenced frame. 
• Protocol These two bytes identify the protocol encapsulated. 
• Data Contains between zero and 1500 bytes for user information. 
• Frame check sequence Provides error detection.
  

Figure 1: A sample PPP frame

LCP/NCP

    PPP operates primarily on the second (Data Link) and third (Network) layers of the OSI model (see Figure 3-2). It defines how bits transmitted and received by the Physical layer are recognized as bytes and frames. PPP also defines procedures for error detection and correction, sequencing, and flow control. It does this by using two protocols, LCP (Link Control Protocol) and NCPs (Network Control Protocols):

• LCP The Link Control Protocol is used to establish, configure, test, and terminate the data-link connection.
• NCP The family of Network Control Protocols is used for establishing and configuring the various Network layer protocols.

 

Figure 2: The seven OSI layers

    Once these connections have been established, packets from each Network layer protocol can be sent over the link. The logical connection remains up for point-to-point communications until the link is closed, either directly via a closure code presented by either of the connecting protocols, or from an unknown cause (for example, a modem disconnection). Flow control is generally be dictated by the physical medium or actual hardware you are using; this is sometimes actually the root cause of some unknown closures. Obtaining the lists of closure codes and logging them is an excellent step in detecting and analyzing your network’s trend and for troubleshooting day to day problems.

    PPP was also designed to support multiplexing of many higher-level protocols. Although the Internet community is compiled mainly of IP users, it is not be limited to them. On top of that, IP networks must often support other protocols as well. These protocols are not discussed in this chapter, but it is important to note that PPP has identified parameters using a family of NCPs for establishing and configuring Network layer protocols such as IP, IPX, and AppleTalk.

Encapsulation

    PPP encapsulation is used to decipher multiprotocol datagrams. The family of NCPs will negotiate the protocol being used. It utilizes framing to indicate the beginning and end of the encapsulation. To turn encapsulation on, use the ENCAPSULATION PPP command on the interface. By default, this will enable HDLC for IP on synchronous serial interfaces.

Authentication

    To establish a connection using the PPP, there are two standard methods of authentication (if necessary). They are CHAP (Challenged Handshake Authentication Protocol) and PAP (Password Authentication Protocol). These authentication types are utilized in Layer 2 by the LCP.

    PAP is the more common of the two and only works to establish an initial link. It is quite vulnerable to attack (or compromise) because it sends its authentication packets throughout the network. Once the link has been established, PAP’s role has been completed.

    CHAP is used to verify the credibility of a connection using a three-way handshake. After the initial link has been established, it also periodically (and randomly) verifies the validity of the connection by sending a “challenge” message. The connected device will then respond with a calculated hash value that is matched by the challenger. If the values do not match, then the connection is terminated.

    To enable these types of authentications, you must add a PPP AUTHENTICATION statement into the interface section of the configuration. Follow this statement with one of the following:

• chap Enables CHAP on the interface.
• pap Enables PAP on the interface. 
• chap pap Enables both CHAP and PAP, and performs CHAP before PAP. 
• pap chap Enables both CHAP and PAP, and performs PAP before CHAP. 
• You may also need to set your router’s hostname and password for call verification. This configuration command syntax is 
• username hostname password secret 

By default, no PPP authentication type is enabled. Here is a sample configuration:

!
interface serial0
encapsulation ppp
ppp authentication chap
username hostname password secret

Link States

    When a PPP connection is not functioning properly, it must be declared "down" to prevent the routing of packets from higher-level protocols. A link understood as "up" would be capable of successfully exchanging packets, both to transmit and receive. Another indicator that the PPP connection is not functioning properly is “loopback” detection. It is important to note that PPP is capable of automatically detecting a looped-back link without administrator assistance. Communications equipment often places itself into loopback during troubleshooting to help diagnosis problems. It is therefore very important to identify these situations without losing critical information.

    If both the serial interface and protocol are down, there are a number of possible problems, including disconnected or faulty physical cabling, provider issues, or hardware failures. If the interface is up and the protocol is down, it may be that keepalives are not incrementing properly, or there may be a misconfiguration, a carrier service problem, timing issues, or hardware failures. These indicators are found using the SHOW INTERFACE command, discussed in the next section.

SHOW INTERFACE and DEBUG Commands

    Cisco routers maintain data structures that represent the state of all installed interfaces. To view the interface status, use the SHOW INTERFACE (or SH INT) command. In the following example you will find two examples of this information taken from actual Cisco routers (and slightly masked). This will display the LCP and NCP link states and may identify problems such as input/output errors, CRC errors, and interface resets. Here is an example of this output for a BRI: 

 

# show interface BRI0
BRI0: B-Channel is up, line protocol is up
Hardware is BRI
MTU 1500 bytes, BW 64 Kbit, DLY 20000 usec, rely 255/255, load 1/255
Encapsulation PPP, loopback not set, keepalive not set
lcp state = OPEN
ncp state = OPEN ncp ipcp state = OPEN
ncp osicp state = NOT NEGOTIATED ncp ipxcp state = NOT NEGOTIATED
ncp xnscp state = NOT NEGOTIATED ncp vinescp state = NOT NEGOTIATED
ncp deccp state = NOT NEGOTIATED ncp bridgecp state = NOT NEGOTIATED
ncp atalkcp state = NOT NEGOTIATED ncp lex state = NOT NEGOTIATED
Last input 0:00:00, output 0:00:00, output hang never
Last clearing of "show interface" counters never
Output queue 0/40, 0 drops; input queue 0/75, 208 drops
5 minute input rate 0 bits/sec, 1 packets/sec
5 minute output rate 0 bits/sec, 1 packets/sec
111021403 packets input, 4204243744 bytes, 208 no buffer
Received 0 broadcasts, 0 runts, 0 giants
136 input errors, 87 CRC, 14 frame, 17 overrun, 17 ignored, 18 abort
161882984 packets output, 3434960991 bytes, 0 underruns
0 output errors, 0 collisions, 35 interface resets, 0 restarts
0 output buffer failures, 0 output buffers swapped out
34 carrier transitions

    Note that the family of NCPs are showing either “open” or “not negotiated.” In the event that you were utilizing NCPs rather than IP, they would be negotiated accordingly.

    Other important information that can be taken from the show interface command is as follows:

• BW Indicates bandwidth.
• keepalive Shows whether keepalives are sent or not. 
• packets input Total number of error-free packets received by the system.
• no buffer Packets that were received but discarded due to buffer allocation problems.
• input errors Total number of “no buffer,” runts, giants, CRCs, frame, overrun, ignored, and abort counts.
• interface resets The number of times the interface was completely reset.
• alarm indications Number of CSU alarms.
  

    If no problems are evident in the show interface data, it may be necessary to use Cisco’s debug commands.

    Note: Be very careful in using the debug commands because they are taxing to the router. Router problems because of CPU utilization have been accidentally caused by leaving this option enabled in the past. When you are finished, be sure to turn off all debugging by typing the u al (un-debug all) command. With data scrolling across your monitor, this is the quickest way to ensure that all logging is terminated properly.

With this in mind, here are several debug options to choose from: 

• debug serial interface Verifies that HDLC keepalive packets are being passed.
• debug ppp negotiation Shows PPP packets being transferred during startup. 
• debug ppp packet Identifies low-level PPP packet dumps. (See Figure 3-3.) 
• debug ppp errors Identifies LCP errors during connection and link operation. 
• debug ppp chap/pap Shows authentication exchanges.
  

Figure 3: Output from debug ppp packet command

HDLC

    HDLC (High-level Data Link Control) was derived from SDLC (Synchronous Data Link Control). Although commonly referred to as a protocol, HDLC isn’t a protocol at all. It’s actually a Data Link layer bit-stuffing algorithm that specifies a data encapsulation method for synchronous serial links.

    HDLC is the frame-level definition for a number of protocols such as SDLC, BDLC, and X.25, by which frames are identified with special characters called flags. HDLC defines a 0 insertion and deletion process (commonly referred to as bit stuffing) to ensure that the bit pattern of the delimiter flag does not occur in the fields between flags. Bit stuffing is used to introduce extra 0s in any sequence of more than five 1s. This technique ensures that actual data never appears as flag characters. Therefore, the data in HDLC frames appear to be transparent. HDLC operates at the second Data Link) layer of the OSI reference model (see previous illustration). It utilizes the Physical layer to provide clocking and synchronizing of the transmission and reception of frames. 

    
    Data is encapsulated in the (application specific) HDLC frame with embedded information, in an 8- or 16-bit header/trailer. The actual data (Layer 3 frames) is transmitted into the data field, which varies in length depending upon the protocol using the frame. The first two bytes of each frame are defined as the address and control bytes. Each frame also includes two CRC bytes for error checking. The HDLC application decodes the address and control bytes and returns a representation of the contents. The frames are separated by flag sequences that are transmitted between the frames.

    Since HDLC is primarily used in a peer-to-peer environment, one station is designated to be the primary and the others become secondary. A session can use one of the following connection modes, which determine how the primary and secondary stations interact: 

• Normal Unbalanced Occurs when the secondary station responds only to the primary station.
• Asynchronous When the secondary station initiates the message. 
• Asynchronous Balanced Both stations send and/or receive over part of a duplex line. 
• HDLC is used on both point-to-point and multipoint (multidrop) data links. It also supports full-duplex transparent-mode operation.
  

SDLC

    SDLC is an acronym for Synchronous Data Link Control, and just like HDLC, it is an ISO standard. The SDLC standard was originally invented by IBM to replace the Bisynchronous protocol for WANs, which is known as BDLC. It is structured very similarly to a variation of HDLC called HDLC NRM (Normal Response Mode).

    SDLC differs from HDLC in that it is not used for peer-to-peer communications. It is used primarily for multipoint networking. Instead of utilizing the previously mentioned primary and secondary relationships, SDLC is made up of a primary station that controls communications via one or more secondary stations. This primary is usually some type of mainframe or central computer. The secondary nodes act as controllers for all local terminals that transmit the necessary data.

    SDLC uses a framing format similar to that of the HDLC frame. An address control field is added, to distinguish conversations between the primary and each of its secondaries, with each secondary having a different address. To streamline transmissions, each secondary reviews all transmissions from the primary but only responds to frames with the secondary’s own address. Unlike X.25 or Frame Relay, SDLC is based on dedicated lines with permanent physical connections.

    SDLC is capable of full-duplexing but more often than not relies on half-duplexing. This simply means that either the primary or any one of the secondary nodes may be able to transmit simultaneously, but never both. Interference can be created when two or more secondaries transmit at the same time. Often data from both stations is lost as a result. So it becomes important that secondary transmissions are controlled by the primary. In other words, a secondary can only transmit when told to. Further, a limited number of frames may be transmitted before link control is passed back to the primary. SDLC also provides link integrity by using CRC bytes for error detection. Frame acknowledgment is encoded into the control fields. Up to seven frames can be sent from either side before acknowledgment is required.

    SDLC typically supports two types of line encoding, either Non-Return to Zero Inverted (NRZi) or non-NRZi. More commonly used, NRZi encoding ensures that one zero is transmitted out of every 5 bit times at a minimum, which aides synchronization.

    In fact, one of the most typical problems associated with SDLC is its error recovery. In situations where problems exist, for example when noisy lines create disturbances, the error-recovery mechanism causes many frames to be continually retransmitted and causes latency.

PPP

    Several different flavors of PPP are accepted today. The standard, as discussed in the opening section of this chapter, is often referred to as ISPPP (Internet Standard for Point to Point Protocol). Here we will briefly discuss Cisco’s variation of PPP.

    On the Job: Watch out, don’t make the same mistakes that others have made in the past. For instance, I was configuring a network using PPP across several Wellfleet routers and could not get either end to communicate with the other properly. Examining both configurations eventually uncovered that, although I was running PPP on both ends, one was mis-optioned for the default Wellfleet PPP.

Cisco Systems Point-to-Point Protocols

    Cisco systems will support asynchronous links using SLIP (Serial Link IP), as well as synchronous links using either simple HDLC framing or X.25. It is important to note that SLIP will only support IP traffic. The HDLC framing procedure adds four bytes to the header. The first octet (the address field) is designated as either unicast or multicast. The second octet (the control byte) is left at 0 and simply not checked. Cisco’s HDLC is incompatible with other vendors’ because of the included proprietary type field. You will also encounter problems when directly connecting Cisco equipment over clear-channel or leased connections to non-Cisco boxes.

    Cisco uses a keepalive protocol to ensure connectivity of its serial lines, especially those that do not provide an NBMA (Non-Broadcast MultiAccess) service. NBMA provides access to multiple devices which do not broadcast. Therefore, split-horizon is enabled by default on HDLC interfaces. Each end of the link periodically sends two 32-bit sequence numbers to the other side. These numbers are the identifiers of each side. If the numbers are not correct or if no keepalive is sent, the link is considered to be down.

    Several other diagnostic tools can be used to help in troubleshooting serial line outages.

• SHOW CONTROLLERS interface/all 
• This command determines which cable is connected to the interface. Here is an example:
  

# show controllers serial
HD unit 0, idb = 0x7F9C0, driver structure at 0x81480
buffer size 1524 HD unit 0, RS-530 DTE cable
cpb = 0x42, eda = 0x2064, cda = 0x2078
RX ring with 16 entries at 0x422000
00 bd_ptr=0x2000 pak=0x0C866C ds=0x461444 status=80 pak_size=54
TX ring with 4 entries at 0x422800
00 bd_ptr=0x2800 pak=0x000000 ds=0x40C044 status=80 pak_size=44
85 missed datagrams, 43 overruns
0 bad datagram encapsulations, 0 memory errors
0 transmitter underruns

• SHOW BUFFERS
• This command displays statistics for the router’s buffer pools on the entire router or a specific interface. 
• PING 
• This command sends ICMP requests to the specified IP address. If a response is received, the connection is up. If not, there may be any of several problems occurring, ranging from hardware failures to firewall issues. Here is an example of the output:
  

# ping 204.71.177.35
Sending 5, 100-byte ICMP Echos to 204.71.177.35, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 76/78/80 ms

Switched Protocols

    Switched protocols are the set of rules and conventions that govern how devices on a switching network exchange information. These types of protocols include ATM, SMDS, and Frame Relay, among others. In this chapter, we will discuss utilizing and troubleshooting switched-protocol networks.

Frame Relay

    The Frame Relay protocol (see Figure 3-4), due to its efficiency, has widely replaced X.25. Frame Relay is a Layer 2 (Data Link) encapsulation method that uses packet switching. This means data is taken and encapsulated with a formatted Frame Relay header (most often IETF) and sent to its destination, stripped of the encapsulation, and then routed accordingly. 

Figure 4: Frame Relay

Frame Format

    The Frame Relay header is made up of a 10-bit DLCI (Data Link Connection Identifier), a C/R (Command/Response) bit, FECN and BECN (Forward/Backward Explicit Congestion Notification) bits, and a DE (Discard Eligible) bit. The data section of this frame immediately follows the header. Finally, a trailer concludes the individual frame. It contains an FCS (Frame Check Sequence) for error detection and frame delineation. Frame Relay standards recommend a maximum frame size of 1600 octets.
The Frame Relay Cloud

    The DLCI is actually a locally significant identification to the UNI (User to Network Interface) switch in the frame network. Upon entering the Frame Relay network, this DLCI helps the switch identify the path that the frame circuit, known as a PVC (permanent virtual circuit), is provisioned to take. PVCs are responsible for carrying the signal via the configured switches through the Frame Relay network.

    This type of network is commonly referred to as a Frame Relay Cloud, because the provider only guarantees the entry and exit points of the connection. Therefore, the actual path the PVC takes is transparent to everybody except the actual provider. A Frame Relay cloud network contains switches that form dedicated but shared paths for information to pass through (thus the term permanent virtual circuit). Therefore, if a physical path is not being logically used by a specific connection, another PVC may use the path’s available bandwidth.

Congestion

    Along the PVC, the frame may experience latency due to congestion. This is where the FECNs and BECNs can be useful. These bits explicitly notify a subscriber's device of congestion on the network; they help that device determine whether or not to withhold traffic or reduce its transmission rate until the congestion has cleared. Also, by identifying possible problematic spots in the cloud, some circuits may be reprovisioned to ease connectivity.

    Another field that was mentioned but not discussed is the DE (discard eligible) field. If the DE bit is set, then the frame is marked as “discardable” during high-use periods. Although this is not a popular option, providers can encourage customers to use this by assessing lower tariffs.

Error Correction

    We will discuss the roles played by CRC error detection in the ISDN section of this chapter. Conceptually similar, the FCS is passed in the trailer of the frame. The destination receives the FCS and computes it. The source then sends the “answer” in the next frame. If the FCS does not match the expected “answer,” then the frame is simply discarded and the drop is registered. You may also notice that the interface registers also show runts and giants. These are simply frames that are discarded because they don’t fall within the proper size parameters.

Burst Rate and CIR

    Transmitting information does not mean that you will be sending out a smooth and steady stream of data. However, you should have some idea of the amount of bandwidth you will be using. The CIR (Committed Information Rate) describes the bandwidth supported during normal operations, but it’s possible that you’ll go above this amount of bandwidth at any given time. Frame Relay also allows for a committed burst rate, which describes the maximum amount of data a user can put through the network during a specific time interval. Cisco allows you to provide bandwidth statements that solidify your CIR. This is particularly useful if you have dual paths for redundant purposes. In this scenario, you can “traffic shape” by altering the bandwidth statements to primarily utilize a preferred PVC.

Inverse ARP

InARP (Inverse Address Resolution Protocol) is a basic Frame Relay protocol that allows routers on the Frame network to learn the protocol addresses of other routers. Cisco IOS employs InARP to dynamically determine the network information associated with the system’s assigned DLCI. If your Cisco router is communicating with a non-Cisco router that does not use InARP, this feature must be eliminated.

LMI

    LMI (Local Management Interface) is a basic protocol that is passed through the PVC to verify the validity and state of the link. LMI does not congest traffic on the PVC and is optional. LMI messages perform two functions:

• Verifying a "heartbeat" exchange (the link is running normally)
• Showing PVC status
  

    Several different flavors of LMI are available on Cisco router interfaces. The Cisco default is “gang of four”; others include ANSI Annex A and Annex D. The LMI information must be configured directly into the interface. IOS versions 11.2 and later offer an auto-configuration of LMI. You will see examples of this type of configuration later in this chapter.

    The LMI protocol is designed so that the subscriber must initiate information exchange, in order not to tax the network unnecessarily. The exchange consists of a requesting frame comprising a six-octet header and a list of IEs (Information Elements) that carry the status information:

• The first two octets contain the necessary identifiers (similar to the DLCIs).
• The third octet identifies all (LMI) frames as being informational. 
• The fourth octet contains a protocol discriminator. 
• The fifth octet contains a dummy field that's always set to 0. 
• Finally, the sixth octet identifies the message type (either from the subscriber or provider).
  

    And behind the header, LMI recognizes three types of IEs:

• Report type
• Keepalive 
• PVC status
  

Multicasting

    Multicasting is an optional feature of Frame Relay that allows users to either send traffic to or receive traffic from multiple users. With this arrangement, the frame network actually duplicates the incoming signal and broadcasts it to the associated MDLCIs (Multicast DLCIs). This is a duplexed operation because traffic flows in both directions.

PVC States

    If you experience problems with the link state of your PVC, there are two options you should check: the serial status and the line protocol status. You can do this test using one of three commands:

· SHOW INTERFACE

    This command shows the status of both the physical and line-protocol interface. It also identifies LMI connectivity. An example of this type of output is as follows:

Serial0 is up, line protocol is up
Hardware is XXXXXX
MTU 1500 bytes, BW 1544 Kbit, DLY 20000 usec, rely 255/255, load 6/255
Encapsulation FRAME-RELAY IETF, loopback not set, keepalive set (10 sec)
LMI enq sent 1134915, LMI stat recvd 1133891, LMI upd recvd 0, DTE LMI up
LMI enq recvd 0, LMI stat sent 0, LMI upd sent 0
LMI DLCI 0 LMI type is ANSI Annex D Frame Relay DTE
Broadcast queue 0/64, broadcasts sent/dropped 0/0, interface broadcasts 5277765
Last input 0:00:00, output 0:00:00, output hang never
Last clearing of "show interface" counters never
Output queue 0/40, 0 drops; input queue 0/75, 0 drops
5 minute input rate 107000 bits/sec, 32 packets/sec
5 minute output rate 41000 bits/sec, 47 packets/sec
46887557 packets input, 1612456289 bytes, 0 no buffer
Received 0 broadcasts, 0 runts, 0 giants
2563 input errors, 1690 CRC, 2 frame, 43 overrun, 85 ignored, 61 abort
57785411 packets output, 2143340051 bytes, 0 underruns
0 output errors, 0 collisions, 392 interface resets, 0 restarts
0 output buffer failures, 0 output buffers swapped out
40 carrier transitions
DCD=up DSR=up DTR=up RTS=up CTS=up

Serial0.1 is up, line protocol is up
Hardware is XXXXXX
Internet address is 1.1.1.1 255.255.255.255
MTU 1500 bytes, BW 1000 Kbit, DLY 20000 usec, rely 255/255, load 6/255
Encapsulation FRAME-RELAY
Serial0.2 is up, line protocol is up
Hardware is XXXXXX
Internet address is 1.1.1.1 255.255.255.255
MTU 1500 bytes, BW 1544 Kbit, DLY 20000 usec, rely 255/255, load 4/255
Encapsulation FRAME-RELAY

SHOW FRAME LMI

    This commands expands upon the LMI information. Here’s an example of this command:

LMI Statistics for interface Serial0 (Frame Relay DTE) LMI TYPE = ANSI
Invalid Unnumbered info 0 Invalid Prot Disc 0
Invalid dummy Call Ref 0 Invalid Msg Type 0
Invalid Status Message 0 Invalid Lock Shift 0
Invalid Information ID 0 Invalid Report IE Len 0
Invalid Report Request 0 Invalid Keep IE Len 0
Num Status Enq. Sent 1877971 Num Status msgs Rcvd 1877917
Num Update Status Rcvd 0 Num Status Timeouts 54

SHOW FRAME PVC

    This command identifies the DLCI and PVC status. It also shows the create and status-change times on the PVC. Here is an example of this command:

PVC Statistics for interface Serial0 (Frame Relay DTE)
DLCI = 101, DLCI USAGE = LOCAL, PVC STATUS = ACTIVE, INTERFACE = Serial0.1
input pkts 60745337 output pkts 83035363 in bytes 2827350014
out bytes 4097860340 dropped pkts 50 in FECN pkts 0
in BECN pkts 0 out FECN pkts 0 out BECN pkts 0
in DE pkts 2266 out DE pkts 0
pvc create time 31w0d last time pvc status changed 4w1d

    If the serial port is physically down, you will want to check the actual physical connections and verify that your provider is not experiencing a problem. However, if the line protocol is showing “down,” you need to verify LMI and troubleshoot your configurations. Never be afraid to contact your provider for assistance in the event that you are experiencing a problem.

Exam Watch: Ensure that you are troubleshooting the proper problem. The serial port may still be up while the protocol is not functioning properly.

There are a couple of debug commands that you will find particularly useful when troubleshooting Frame Relay problems:

DEBUG FRAME-RELAY LMI

This displays LMI information.

DEBUG FRAME-RELAY EVENTS

This shows the exchanges between a router and UNI device in the cloud as shown in Figure 3-5.

DEBUG FRAME-RELAY PACKET

This command shows the frame packets.

Figure 5: Output from debug frame-relay events command

Configuration Example

    Following is a sample Frame Relay configuration. This specific router configuration is subinterfaced to use two DLCIs in order to pass traffic through the cloud to redundant points. Note that, as discussed above, the bandwidth statements are weighted to pass traffic through a desired path.

interface Serial0
no ip address
encapsulation frame-relay
no ip route-cache
frame-relay lmi-type ansi
!
interface Serial0.1 point-to-point
ip address 1.1.1.1 255.255.255.255
bandwidth 1544
frame-relay interface-dlci 10
!
interface Serial0.2 point-to-point
ip address 1.1.1.1 255.255.255.255
bandwidth 1000
frame-relay interface-dlci 11

X.25

    X.25 is very similar to Frame Relay because it utilizes packet switching and PVCs. The X.25 standard was approved by the CCITT back in 1976, when other WAN standards were poor at best. X.25 answered several transmission problems by introducing error correction and flow control. Although newer, faster, and more efficient protocols have emerged to replace X.25, it is still being used. So we will discuss it briefly.

    Data is transferred via X.25 in the first several layers (Physical, Data Link, and Network) of the OSI model. X.25 relies on the HDLC bit-stuffing algorithm, described in the PPP discussion, to format the information into frames. DTEs (Data Terminal Equipment) and DCEs (Data Communications Equipment) establish connections via PADs (Packet Assemblers/Disassemblers) on logical virtual circuits through either SVCs (Switched Virtual Circuits) or PVCs (Permanent Virtual Circuits), over which the frames are passed.

SVCs and PVCs

    Switched Virtual Circuits are only utilized when the connection is established and are lost when the data is finished being transferred. To establish this connection, the DTE sends a Call Request with the necessary information (source/destination addresses) to establish a connection. If the receiving end sends an acknowledgment back, then transmissions begin and remain up until the necessary information has been passed. This is signaled with a Clear Request packet.

    Permanent Virtual Circuits in X.25 work very similarly to those of Frame Relay. The circuit is a permanent physical connection that is only utilized when data is being transferred.

Addressing

    X.25 networks have unique addresses that use the X.121 standard, commonly referred to as NUAs (Network User Addresses). This addressing scheme dictates that a 14-digit address is assigned; however, only two of these digits are optional (for subaddressing). Using the SHOW INTERFACE command (again), we can see an example of this type of addressing:

# show interface serial 1
X25 address 000000010100, state R1, modulo 8, idle 0, timer 0, nvc 1
Window size: input 2, output 2, Packet size: input 128, output 128
Timers: T20 180, T21 200, T22 180, T23 180, TH 0
(configuration on RESTART: modulo 8,
Window size: input 2 output 2, Packet size: input 128, output 128
Channels: Incoming-only none, Two-way 5-1024, Outgoing-only none)
RESTARTs 3/2 CALLs 1000+2/1294+190/0+0/ DIAGs 0/0

Error Detection

    By using the HDLC error detection, X.25 does have primitive troubleshooting capabilities. However, there are some control messages used to set up, maintain, and tear down communications circuits. Two of these types of messages can indicate a problem: the Disconnect and Reset codes.

• Disconnect is used to either reject a connect request or kill an established call.
• Reset is used to clear a hung connection or identify synchronization problems.
  

    Finally, whenever the transport service receives a TPDU (Transport Protocol Data Unit) timeout, the message will automatically be resent. However, if the header contains a protocol error (generally identified in the Transport layer), TPDU will return an ERR cause code and end the connection.

Cisco provides several debugging commands for troubleshooting on X.25:

DEBUG X25 EVENTS

This will detect X.25 events and provide cause and diagnostic information as shown in Figure 3-6.

DEBUG LAPB

This command displays any Level 2 X.25 information.

Figure 6: Output from debug x25 events command

SMDS

    SMDS (Switched Multi-megabit Data Service) was originally released at the beginning of 1992. Throughout this section, you will notice that SMDS is closely related to ATM and Frame Relay technologies. It was engineered to connect LANs to WANS via the local telephone company (telco).

SMDS has a three-tiered architecture:

• A switching infrastructure
• SNIs (Subscriber Network Interfaces) composed of T1 and T3 circuits 
• An infrastructure access-control system
  

Switching Infrastructure

    SMDS is connectionless because it does not require that local telcos utilize their switches to establish call paths. Instead, they pass ATM-like 53-byte cells (IEEE 802.6) to a carrier switch (sometimes even through an ATM switch). This switch identifies the destination address and forwards the uniform cells to that destination via the SMDS cloud. The cloud contains a network of switches that permit data to travel over the least-congested paths. The actual SMDS address ensures that the cells arrive in the proper order. This, in turn, allows SMDS to offer a service that provides a faster end-to-end solution, regardless of the bandwidth requirements.

SNI

    Another advantage of this type of engineering is that SMDS is scalable. Being connectionless, it’s easier for users to expand networks where each site is connected to all other sites. This provides the look and speed of a LAN over a WAN.

    Networks can connect to an SMDS cloud via SNIs (Subscriber Network Interfaces), which are generally composed of T1 and T3 circuits. The latest technology, however, allows for bandwidth including 56K to 64K. This gives SMDS connectivity to networks in a flexible connectivity range from 56Kbps to 45Mbps.

Access Control System

    SMDS uses E.164 addressing, which is very similar to standard telephone numbers. This gives SMDS access-control features such as call screening, verification, and blocking.

Benefits of SMDS

    SMDS includes reasonable availability and increased LAN performance. It also provides some data management features, a degree of flexibility, bandwidth on demand, privacy, multiprotocol support, and technology compatibility. SMDS offers a range of accommodating features, including

• End-to-end connectivity
• High-speed, low-delay connectionless data 
• Multicasting (group addressing) 
• Support of all WAN protocols (for example TCP/IP, Novell, AppleTalk, SNA)
• Scalability 
• Security
  

Compatibility

    ATM (discussed later in this chapter) is compatible with SMDS. They both have similar 53-byte cell-relay packets with 48 bytes for data and a 5-byte address header. In addition, they share compatible headers that simplify transitioning from SMDS to higher ATM speeds.

    Exam Watch: In order to provide Frame Relay users with access to SMDS, a Bellcore standard called the SMDS Interface Protocol Relay Service (a.k.a. SIP Relay) was released. This standard enables SMDS to be encapsulated and transported within Frame Relay, as well.

    The Synchronous Optical Network (SONET) is an international standard, fiber-optic transmission that is used for broadband transport. SONET offers a variety of optical line rates (in multiples of 51.84 Mbps) and gives users the ability to send signals at multigigabit rates. SMDS is also compatible to operate at speeds up to 155 Mbps on the SONET's STS-3c interface (OC-3).

    Note: The number after the OC is actually the number of STS-1 (Synchronous Transport Signal Level 1) frames, which equals 51.84 Mbps of bandwidth. Therefore, an OC-3 is roughly 155 Mbps of bandwidth.

ATM

    ATM stands for Asynchronous Transfer Mode and is often compared to Frame Relay because it is an ISDN (Integrated Services Digital Network) protocol standard. Also, it utilizes both switching and multiplexing technologies. ATM was designed to make Broadband-ISDN (B-ISDN) a reality. B-ISDN was created conceptually as just an extension of ISDN, so it functions as a communications network, providing integrated services. ATM, however, is not guaranteed and utilizes cell switching for much higher speeds and higher volumes over both WANs and LANs. In fact, ATM’s primary advantage is that it can use the higher speeds ranging from DS-3 (45 Mbps) through the OC (Optical Carrier) family.

Cell Structure

    ATM uses alike 53-btye cells in order to divide data into smaller and more manageable packets; this allows quicker cell relay across a better range of high-performance communications networks. The ATM cells are composed of 5 bytes for header information, and 48 bytes for actual data (these 48 bits are typically referred to as the “payload”). The header itself is made up in one of two formats, either UNI (User to Network Interface) or NNI (Network Node Interface).

    Both of these contain an 8- to 12-bit VPI (Virtual Path Identifier), a 16-bit VCI (Virtual Circuit Identifier), a 3-bit PT (Payload Type), a 1-bit CLP (Cell Loss Priority), and an 8-bit HEC (Header Error Check). However, the UNI cell contains a GFC (Generic Flow Control) field, which steals 4 bits from the VPI field. See Figure 3-7 for an example. The use of these fixed fields allows for greater speeds because active devices can be configured for set lengths, permitting hardware switching versus the accessing of software routing tables.

Figure 7: ATM cell examples

Here are descriptions of the fields:

• Virtual Path Identifier/Virtual Circuit Identifier VPI/VCI fields indicate locally significant Virtual Path or Virtual Channel identification numbers, so that the cells belonging to the same connection can be distinguished.
• Payload Type This field provides user information (if present) and shows if the cell has suffered traffic congestion. 
• Cell Loss Priority This field determines whether or not the cell should be dropped by network equipment during periods of congestion.
• Header Error Check This field serves a dual purpose. First, it provides CRC error detection. It also provides some cell delineation. 
• Generic Flow Control This cell header field appears only in UNI systems and bears only local significance. It is intended to provide standardized local functions such as flow control.
  

    Another ATM advantage is that it is the first cell-switched protocol designed specifically to support integrated applications, including voice, video, and data communications. It accommodates transmission speeds from 64 kbps to gigabit speeds.

    The information exchange works via switching across dedicated lines on the network. Circuit paths are allocated to the nodes for data transfer across the network. Channel capacity becomes available and is reserved between the source and destination. Depending on the bandwidth requirements of the application, utilization of the line can vary enormously. At each node, the packet is received, queued in a buffer, and passed on.

    There are two basic approaches to the actual transport:

1. Datagram
   An example of datagram transport is when each packet can take a separate path through the network and get sorted at the destination. This allows for dynamic congestion handling.
2. Virtual Circuit
   This type of transport occurs when all the packets are sent through the same path without having the path actually dedicated. The routes are determined between paths through the ATM cloud by the VPI and VCI sections of the cell header before the data is transferred across VCCs (Virtual Channel Connections). An advantage of this method is that the virtual channels allow for sequencing, error detection, and flow control.
   

Signal Path

    A virtual channel is a segment of the virtual circuit. A virtual path consists of a set of these channels. Each channel and path have locally significant specific identifiers associated with them. The virtual channel and path numbers of a connection may differ from source to destination if the connection is switched at some point within the network. This is because all channels within a single path must have distinct channel identifiers, but may have the same channel identifier as channels in different virtual paths. This makes a VCI uniquely identifiable via both its virtual channel and virtual path number.

    Switched virtual circuits establish paths dynamically for the data transfer and then release them once, upon completion. This is especially convenient at the LAN level. It is important to understand that ATM technologies are not only limited to SVCs. ATM switching structures also support PVCs identical to those used in Frame Relay.

QSAAL & ILMI

    QSAAL (Q.2100-Signalling ATM Adaptation Layer) is used for UNI signaling across virtual circuits from the router to the ATM switch. It is an ATM Forum-recommended standard that the QSAAL VCI value is 5. ILMI (Integrated Local Management Interface) facilitates administration and configuration of ATM address and is necessary for the auto-configuration. This is accomplished by informing the switch of its unique ESI (MAC address), which returns the address prefix that is used to construct the entire ATM address. The ATM forum recommended standard for the ILMI VCI value is 16. These values are necessary in the basic configuration of the router.

Addressing and Encapsulation

    ATM uses an NSAP (Network Service Access Point) format address. These addresses are compilations of a unique end-identifier, prefixes that identify individual switches and collections of switches within specific peer groups. The prefix for the ESI (End Station Identifier) will come from the switch via ILMI. The ATM connection is used to transfer information to these addresses. To do this more efficiently, AALs (ATM Adaptation Layers) are used to encapsulate the traffic. AAL is meant to support many classes of services with different transfer requirements. That is why AALs are service dependent.

There are four types of AAL:

1. AAL1 Used for connection-oriented, delay-sensitive services such as DS-1 or DS-3.
2. AAL2 Supports connection-oriented services that do not require a constant bit rate. This is currently an incomplete standard. 
3. AAL3/4 Supports both connectionless and connection-oriented protocols that utilize a variable bit-rate such as SMDS. The actual encapsulation type is aal34smds. 
4. AAL5 Used for connection-orientated services such as LANE and classic IP over ATM. The actual encapsulation types are aal5snap (Sub-Network Access Protocol), aal5mux (MUX type) and aal5nlpid (Network Layer Protocol Identifier).
   

Sample Configurations

    Although we will not discuss sample configurations in more detail, it’s worth noting that there are typically two types of implementation for ATM: IP over ATM, and LANE (LAN Emulation). So as a result, there are various ways to set up an ATM network via your Cisco router. These variables will depend on your networking needs. Following are several router configurations for ATM WAN solutions.

Basic PVC Sample Configuration

!
interface atm2/0\
ip address 1.1.1.1 255.255.255.255
atm pvc 1 0 100 aal5snap
map-group PVC
!
map-list PVC
ip 1.1.1.2 atm-vc 1 broadcast

Basic SVC Sample Configuration

!
interface atm2/0\
Ip address 1.1.1.1 255.255.255.255
atm pvc 1 0 5 qsaal
atm pvc 2 0 16 ilmi
atm esi-address 2222222222.00
map-group SVC
!
map-list SVC
ip 1.1.1.1 atm-nsap AA.0000000000.111111111.00 broadcast

SHOW Commands

The commands that you will utilize to troubleshoot your ATM network include the following:

• SHOW ATM INTERFACE ATM
• This displays ATM-specific information directly from the ATM interface. 
• SHOW ATM MAP 
• This displays all of the configured ATM static maps to their remote hosts. 
• SHOW ATM TRAFFIC  
• This displays the current, global ATM traffic information connected to the router. 
• SHOW ATM VC
• This command displays all active permanent and switched virtual circuits, as well as traffic information.
  

DEBUG Commands

Following are the debug commands you will utilize to troubleshoot your ATM network:

• DEBUG ATM SIGNAL
• Displays all the ATM signaling, including QSAAL and ILMI. 
• DEBUG ATM ILMI 
• Displays specifically the ILMI signaling. 
• DEBUG SIGNALING 
• Shows all signaling to the router.
  

ISDN

    Back in 1984, the POTS (Plain Old Telephone System) could not handle the overwhelming amount of traffic that was necessary to transfer information. So ISDN was brought to life. It allows digital data transfer at a rate of 64 kilobits per second over the analog network infrastructure of today’s telephone companies. Today, this need to pass data has only increased, and there are several different ways for it to be accommodated. The ISDN technologies include BRI, MBRI, and PRI.

    Before getting started, it is important that you can identify some of the quirks you may encounter with ISDN. First, the ISDN provider will need to give you some necessary information such as the switch type, framing/encoding types, and SPID numbers. The switch information relates to the equipment with which the provider will be connecting you. If you set the wrong option for this parameter, then nothing will come up.

ISDN BRI

    The most commonly utilized form of ISDN connectivity is called ISDN BRI (Basic Rate Interface). The reason for its popularity is because it allows quicker Internet connectivity at greater data transfer rates for individual users without tying up a telephone line. It is also reasonably priced for dedicated digital access.

    ISDN BRI is often called “2B+D” or “bonded ISDN” because it utilizes two bearer channels with 64 Kbps for data, and one signaling channel with a 16-Kbps bandwidth. See Figure 3-8. The data channels do exactly what they insinuate: pass traffic such as data, voice, video, and so on. The signaling channel communicates with the switch by passing any connection-related traffic. 

Figure 8: Basic Rate Interface

MBRI

    Multiple channels or BRI lines can be multiplexed together (usually in increments of 128 Kbps), depending on your bandwidth needs and the terminal adapter that you possess. This type of technology is known as MBRI (Multiple Basic Rate Interface).

DDR

    BRI can be configured to be one type of DDR (Dial on Demand Routing) because data is only being transmitted when the connection is established. In other words, the call is placed when data transfer is necessary, and hangs up when the channels are idle. Setting this configuration can be tricky because important roles are placed on idle timeouts and on identifying interesting traffic.

ISDN PRI

    ISDN PRI (Primary Rate Interface) is a service provided for companies and others who have a greater need to transfer information. In North America and Japan, this particular type of connection multiplexes 23 bearer channels for actual data, often called the B channels, and one signaling channel called a D (or data) channel. In Europe, PRIs contain 31 bearer channels multiplexed with one data channel. The bandwidth of each of these channels, as previously mentioned, is 64 Kbps (kilobits per second).

ISDN PRI differs from BRI as follows:

• No provisions for multipoint connections. This means the connection will be point-to point, directly from the network onto the CPE (Customer Premise Equipment).
• No available DDR. 
• Obviously, the available bandwidth will be different.
  

Layer 2 of ISDN

    LAP D (Link Access Procedure, D Channel) is the data-link control protocol used by ISDN. This protocol allows the end devices to communicate via the D channel, which is used primarily for signaling. Signaling is a technique where the provider’s equipment (for example, switches such as DMS-100s, Seimans, and 5ESS) pass their standard set of control signals or messages to the receiving communications equipment. This signaling is used to report incoming connections, call terminations, and any other link-related information. The specific signaling messages are dependent upon the type of ISDN service being utilized.

Layer 3 of ISDN

    PRI is capable of supporting the following protocols:

• ATM (See Chapter 12) This acronym stands for Asynchronous Transfer Mode and represents the international standard for cell relay.
• Frame Relay (See Chapter 14) This is the protocol at the Data Link layer that encapsulates data and transmits it over PVCs (permanent virtual circuits) between logically connected devices. 
• HDLC (See Chapter 12) This acronym stands for High-Level Data Link Control and specifies the Data Link layer data-encapsulation method on synchronous serial links. 
• PPP (See Chapter 12) This acronym stands for Point-to-Point Protocol, discussed in this chapter. 
• SMDS (See Chapter 12) This acronym stands for Switched Multimegabit Data Service. It is a high-speed, packet-switched WAN technology and is commonly used in DSL (Digital Subscriber Link). This is the telco’s answer to the cable modem. 
• X.25 (See Chapter 12) This is the standard that defines connections for remote terminal access.
  

    PRIs are particularly useful to ISPs (Internet Service Providers) because they can multiplex incoming individual analog connections via the local telco, digitalize the signal, and back-haul the information to their NASs (Network Access Servers). In this scenario, each connection would be assigned to one channel (or DS-0) on the PRI and authenticated as a PPP connection.

T1 WAN Alarms

    This brings us to the topic of intrusive and nonintrusive circuit evaluations. There are many ways of testing for possible problems. What can be confusing is identifying exactly what these problems can mean.

    There are two types of problems a circuit can experience. Actual WAN alarms are specific signals sent out by the device that sees the problem. They are meant to help troubleshooters identify from which direction the problem is coming. These alarms are:

• Red Alarm identifies that the signal is out of frame or out of synchronization. It generally indicates a break in the line prior to the device that detects or reports the alarm. In other words, the device either does not receive any signal or it is unrecognizable.
• Yellow Alarm is a signal sent back in the direction from which a red alarm comes. Because this alarm is sent intentionally, it surfaces in one of two signal patterns: Either the second bit in every byte is a 0 (in D4), or it’s an alternating pattern of 1s and 0s (in ESF). 
• Blue Alarm is sometimes referred to as “All Ones.” This is also a signal that is sent from a device that detects a red alarm. Because it has not received any passable traffic, the device sends a steady pattern of all 1s to the next device in the circuit path. The receiving device will in turn continue relaying this signal until the red alarm is cleared and traffic is normal. 
• D Channel Failure is another alarm that falls into the “WAN alarm” category. This alarm indicates that the channel reserved for signaling is failing or unrecognizable. Usually this is a problem is associated with the source or destination of the signal.
  

    Figure 3-9 illustrates the process by which a signal is generated but does not reach its destination because of a physical media problem. The receiving device is therefore unable to interpret the incoming signal (resulting in a RA, red alarm) and in turn transmits the appropriate signals to the closest active devices in the span of the signal.

Figure 9: Alarms

    Identifying problems across T1 spans can be done by examining actual errors within the data or problems with synchronization. Here is a list of these types of circuit errors.

• Bit Errors This is the basic performance evaluator and can only be tested intrusively by taking the circuit down. By sending predetermined and stressing test patterns through the circuit, these errors are basically logic errors—in other words, 1s that should have been 0s and vice versa.
• BPVs (Bi-Polar Violations) This type of transmission error occurs every time consecutive pulses of the same polarity are transmitted in violation of the bipolar signal format. This type of signal violation can be monitored without taking the circuit down or during pattern testing. 
• Frame Errors These errors are measured by the number of times an incorrect value appears in the framing bit position. The circuit can be monitored (not effecting live traffic) to identify these errors, but they are usually difficult to uncover. This is generally because the framing bit is only the 193rd bit in each frame and therefore appears less than one percent of the time. 
• CRC Errors (Cyclic Redundancy Check) This method of error detection is both very interesting and very accurate (estimated at over 98 percent). It also is incorporated in any block of live data. The only drawback is that CRC requires the ESF framing format to be utilized. CRC works by building a mathematical equation and transmitting it across the control bits in an ESF frame. The result of the calculation is then transmitted in the next frame. The distant end performs the calculation and compares answers. If the resulting answers do not match, a CRC is incremented.
  

    To identify the errors listed just above, you should be able to request that the provider run various tests across your spans. Most providers will require that you have their circuit identification ready. (For example, a World Comm circuit looks something like this: WZ3-123456; and some LECs utilize CLLI codes.) This does not ensure that the technician is actually testing the correct span. So to ensure it’s being done properly, monitor the circuit and look for signs that the testers have actually accessed it. If you suspect they might not have, don’t be afraid to throw the circuit into alarm by changing either the framing, the encoding, or the switch type, and see if they notice.

The testing patterns most often utilized to test T1 spans are as follows:

• ALL ONES This test sends out an unframed pattern of continuous 1s in an attempt to see if any network devices are dropping bits.
• QUASI This test is often referred to as a STRESS TEST. It’s a type of testing that randomly changes the bit pattern (using the pulses – 1s) around in an attempt to create a signal that physically stresses the circuit past the level that normal data transmissions would cause. Often this is an unframed signal. This pattern is meant to test the actual physical media, but it can also show equipment failures. 
• THREE & TWENTY-FOUR This is a test pattern that injects three pulses (1s) in every 24 bits (note that this is the minimum 1s density). It is generally used to test active devices in the span. When other testing patterns pass and this one fails, there’s often a problem with an active device (a repeater or DAC, for example) along the span. 
• ONE in EIGHT This test is a less stressful version of the THREE & TWENTY-FOUR pattern. It basically sends one pulse in every eight, and identifies the same problems. 
• ALL ZEROS This test sends out a pattern of all 0s. It is meant to test active network devices in an ESF environment by forcing B8ZS. Any device that does not substitute the zeros pattern with the intentional BPVs would be identified as mis-optioned. 
• FRAMED ONES This is a low stress test that inserts 1s into all bits except those reserved for framing. It has a low functionality in the grand scheme and is rarely used to identify less than obvious problems.
  

Certification Summary

    Your success at troubleshooting WANs will depend on your understanding of how they work conceptually. In this chapter we discussed both switched and point-to-point protocols and their many attributes.

    The PPP standardized Internet encapsulation of IP over point-to-point links, and addressed other previously neglected issues such as encapsulation and error detection. PPP operates on dual layers because it basically comprises two protocols, LCP (Link Control Protocol) and a family of NCPs (Network Control Protocols). PPP supports multiplexing and many higher-level protocols. By utilizing SDLC and HDLC, PPP is able to encapsulate datagrams over point-to-point links. Finally, there are two standard methods of authentication: CHAP (Challenged Handshake Authentication Protocol) and PAP (Password Authentication Protocol).

    Switched protocols describe the set of rules and conventions that govern exchange of information among devices on a switching network. In this chapter we discussed, in detail, LAN-to-WAN protocols such as SMDS; higher-speed protocols such as ATM; the older, reliable, pioneering X.25 and its predecessors, including ISDN and Frame Relay.

Two-Minute Drill

• PPP standardized Internet encapsulation of IP over point-to-point links, it also addressed other issues such as encapsulation and error detection.
• The PPP frame consists of three primary parts: 
• A header 
• A body or payload of the packet 
• A trailer 
• PPP operates primarily on the second (Data Link) and third (Network) layers of the OSI model. 
• The Link Control Protocol (LCP) is used to establish, configure, test, and terminate the data-link connection. 
• The family of Network Control Protocols (NCP) is used for establishing and configuring the various Network layer protocols. 
• PPP was also designed to support multiplexing of many higher-level protocols. 
• PPP encapsulation is used to decipher multiprotocol datagrams. 
• To establish a connection using the PPP, there are two standard methods of authentication; CHAP (Challenged Handshake Authentication Protocol) and PAP (Password Authentication Protocol). 
• PAP is the more common of the two and only works to establish an initial link.  
• CHAP is used to verify the credibility of a connection using a three-way handshake. 
• It is important to note that PPP is capable of automatically detecting a looped-back link without administrator assistance. 
• Cisco routers maintain data structures that represent the state of all installed interfaces. To view the interface status, use the SHOW INTERFACE (or SH INT) command. 
• Be very careful in using the debug commands because they are taxing to the router. 
• HDLC (High-level Data Link Control) is a Data Link layer bit-stuffing algorithm that specifies a data encapsulation method for synchronous serial links. 
• SDLC (Synchronous Data Link Control) is used primarily for multipoint networking. 
• Several different flavors of PPP are accepted today. 
• Switched protocols are the set of rules and conventions that govern how devices on a switching network exchange information.  
• Frame Relay is a Layer 2 (Data Link) encapsulation method that uses packet switching. 
• LMI (Local Management Interface) is a basic protocol that is passed through the PVC to verify the validity and state of the link.
• Multicasting is an optional feature of Frame Relay that allows users to either send traffic to or receive traffic from multiple users. 
• Ensure that you are troubleshooting the proper problem. The serial port may still be up while the protocol is not functioning properly. 
• X.25 is very similar to Frame Relay because it utilizes packet switching and PVCs. X.25 answered several transmission problems by introducing error correction and flow control. 
• SMDS (Switched Multi-megabit Data Service) has a three-tiered architecture: 
• A switching infrastructure 
• SNIs (Subscriber Network Interfaces) composed of T1 and T3 circuits 
• An infrastructure access-control system 
• In order to provide Frame Relay users with access to SMDS, a Bellcore standard called the SMDS Interface Protocol Relay Service (a.k.a. SIP Relay) was released. This standard enables SMDS to be encapsulated and transported within Frame Relay, as well. 
• ATM (Asynchronous Transfer Mode) is often compared to Frame Relay because it is an ISDN (Integrated Services Digital Network) protocol standard. It utilizes both switching and multiplexing technologies. 
• ISDN allows digital data transfer at a rate of 64 kilobits per second over the analog network infrastructure of today’s telephone companies. 
• LAP D (Link Access Procedure, D Channel) is the data-link control protocol used by ISDN. 
• PRIs are particularly useful to ISPs (Internet Service Providers) because they can multiplex incoming individual analog connections via the local telco, digitalize the signal, and back-haul the information to their NASs (Network Access Servers).