DSL

DSL takes advantage of unused bandwidth on traditional phone lines, working at a higher frequency than a voice conversation. It allows you to use the high-speed access at the same time a voice conversation is occurring. Typically, a DSL connection is always on; however, some service providers have adopted techniques much like PPP. In fact, they are PPPoE or PPPoA, which we discuss later. Some of the challenges with DSL are distance requirements no greater than 18,000 feet, your CO might not support DSL, and the fact that any fiber lines that are in use will severely limit DSL options.

DSL operates between the customer premises equipment (CPE), such as a Cisco 827 router, and the DSLAM. The DSLAM terminates DSL connections at the CO. DSL has a number of different flavors or variants, each designed for a different reason, such as the level of service, speed, and distance. ADSL, for example, can coexist with the plain old telephone system (POTS), whereas SDLS cannot. There are a variety of DSL technologies, ranging from those designed for the home user to those for large corporations. DSL technologies are anything but static; there is a large amount of research and development for the betterment of DSL. Some of the varying xDSL specifications follow:

  • ADSL Designed for residential use, it is the most popular form of DSL technology. The key to ADSL is that the upstream and downstream bandwidth is asymmetric, or uneven. In practice, the bandwidth from the provider to the user (downstream) will be the higher speed path. This difference is in part due to the limitation of the telephone cabling system, but it also accommodates typical Internet usage, where the majority of data is sent to the user downstream. ADSL is rated for distances up to 18,000 feet and lets you place traditional phone calls while a DSL connection is active.

  • SDSL Designed more for business, the line speed is the same in both directions. SDSL allows for greater upstream speeds than ADSL; because of the greater available bandwidth for upstream communication, services can be hosted at the customer's site. Distances can be up to 12,000 feet.

  • G.SHDSL Also known as G.991.2, G.SHDSL is an international standard for SDSL developed by the ITU. This technology is the first DSL technology to be developed from the ground up as an international standard; it supports longer distances (28,000 feet) and is predicted to be the most adopted standard in the future.

  • IDSL IDSL uses 2B1Q line coding and can be configured to use the full bandwidth of two 64Kbps bearer channels plus one 16Kbps delta channel. ISDN lines and the routers' ISDN U interfaces are used for connectivity. Major benefits of switching to IDSL from ISDN are the always-on connection, no call setup, and flat-rate billing instead of per-minute fees. Distances can be up to 18,000 feet.

  • VDSL VDSL transmits data in the 13Mbps to 55Mbps range over short distances, usually between 1,000 and 4,500 feet. The shorter the distance, the faster the connection rate.

  • HDSL Used as a replacement for T1 or E1 services, this service is limited to 12,000 feet, whereas a traditional T1 requires repeaters every 6,000 feet.

The typical tradeoff for DSL is speed versus reach (distance). The longer the loop, the lower the speed.

graphics/alert_icon.gif

A number of other conditions can also slow down a DSL connection: distance, thickness of the wire (gauge), bridge taps, crosstalk, and AM radio.


ADSL

As mentioned before, ADSL is a great solution for residential markets. It has fast download speeds, up to 8Mbps, and adequate upload speeds, close to 1Mbps. One of the biggest selling points is that it can coexist with POTS. Consumers can use the same line as their regular phone line and, more importantly, use them both at the same time. The three modulation techniques are carrierless amplitude and phase (CAP), discrete multitone (DMT), and consumer/mass-market DMT (G.lite).

graphics/alert_icon.gif

G.lite is the most used standard and supports only 1.5Mbps downstream and 640Kbps upstream.


DSL is a Layer 1 transmission protocol. ATM is the Layer 2 protocol used on top of DSL. The purpose of both is to move IP data packets, and there are three methods for doing that:

  • RFC 1483/2684 Bridged You can use two methods here, bridged or routed. We are only concerned with bridged. Ethernet frames are bridged to a router over ATM. There are a number of advantages to bridging: it is simple to use, easy to install, and involves less sophisticated devices, which are ideal for a single user. Some of the disadvantages are that bridging involves heavy use of broadcasts, it is insecure, and both broadcast attacks and IP address hijacking are possible.

  • PPPoE Covered in RFC 2516, PPPoE is a bridged solution. Ethernet frames are bridged over ATM, as with RFC 1483, but this time, the Ethernet packets encapsulate PPP inside. Because PPP is in use, we have all the advantages of PPP, such as IP negotiation, authentication, and AAA server support. Also, because of the encapsulation, the maximum receive unit (MRU) must be negotiated at 1492 bytes or less so that with the additional headers, we do not exceed 1500 bytes. PPPoE goes through four steps in the discovery stage:

    1. The PPPoE client broadcasts a PPPoE active discovery initiation (PADI) packet.

    2. The PPPoE server responds with a PPPoE active discovery offer (PADO) packet.

    3. The client switches to unicast and sends a PPPoE active discovery request (PADR) packet back to the server.

    4. The server responds with a PPPoE active discovery session-confirmation (PADS) packet to finish the process.

    At this point, PPP can then negotiate its parameters to establish connectivity. After a successful conversation, when the devices are ready to terminate the session, either the client or the server sends a PPPoE active discovery terminate (PADT) packet.

  • PPPoA This solution is a routed solution. No host-based software is required as with PPPoE; the CPE routes packets to the aggregation router. The steps are much simpler because it is able to jump right to PPP negotiation. You need hostnames and passwords for Password Authentication Protocol (PAP) or Challenge Handshake Authentication Protocol (CHAP). The aggregation router sends an IP address negotiated through IP Control Protocol (IPCP). The CPE performs NAT to give host workstations connectivity out.

Configuring the PPPoE Client

Cisco IOS 12.1(3)XG introduced a PPPoE client feature for the Cisco 827 router. It lets the router, rather than the workstation, handle the PPPoE functionality. Now, multiple PCs can be behind the router and their traffic is sent via the PPPoE session. It can be encrypted and filtered and run by NAT. There are six steps to configuring the CPE as a PPPoE client. We examine just the basics, enough to get the router up and running:

  1. The first group of commands configures a PPPoE virtual private dialup network (VPDN) group. The first thing to do is enable VPDN with

     Router(config)#vpdn enable 

    Then, you can build the group with the vpdn-group name command.

     Router(config)#vpdn-group dsl-isp Router(config-vpdn)# 

    Notice that the configuration mode has shifted to config-vpdn. Any commands entered at this point are specific to that group. You then specify request-dialin, which creates a subgroup inside your VPDN. In the subgroup, you define which protocols are going to be supported by entering protocol pppoe. The code would look something like Listing 10.1.

    Listing 10.1 Configuring a VPDN group
     Router(config)#vpdn enable Router(config)#vpdn-group dsl-isp Router(config-vpdn)#request-dialin Router(config-vpdn-req-in)#protocol pppoe Router(config-vpdn-req-in)#exit Router(config-vpdn)#exit 
  2. After the VPDN group is built, you need to configure the ATM interface. You specify which ATM interface you are using with the interface atm number command. Next, you identify the permanent virtual circuits (PVCs), virtual path identifiers (VPIs), and virtual circuit identifiers (VCIs) with pvc vpi/vci. That command moves the router into the atm-vc config mode, where you specify the dialer-pool number, which binds the dialer interface and configures PPPoE client encapsulation:

     Router(config)#interface atm 0 Router(config-if)#pvc 9/8 Router(config-if-atm-vc)#pppoe-client dial-pool-number 7 
  3. Now you are ready to build the dialer interface that was referenced in the preceding step. The dialer-interface configuration is the same as any configuration you used earlier; however, you need to check a few items. You need to allow for dynamic IP address assignments through IPCP, which you do with ip address negotiated. Another key parameter is that the packet sizes should not exceed 1492 bytes so that when the PPPoE headers are added, you stay at or below 1500 bytes. To define the packet size, use the ip mtu 1492 command. The rest of the commands are ones that you should know already. One interesting item is that we call a dialer list with the dialer-group command; remember, the purpose of the dialer list is to specify interesting traffic. Because DSL is an always-on technology, you should not need a dialer list, but it does show up in most configurations and examples (better safe than sorry). Listing 10.2 builds a dialer interface and configures it for PPPoE.

    Listing 10.2 Dialer interface configured for PPPoE
     Router(config)#int dialer 0 Router(config-if)#ip address negotiated Router(config-if)#encapsulation ppp Router(config-if)#dialer pool 7 Router(config-if)#dialer-group 1 Router(config-if)#ip mtu 1492 Router(config-if)#ppp chap hostname user@isp.com Router(config-if)#ppp chap password abc123 Router(config-if)#exit Router(config)# 

    graphics/alert_icon.gif

    The maximum packet size must be 1492 bytes or smaller so that with the additional 8 bytes from the PPPoE header, the packet stays at or below the 1500 byte total.


  4. NAT overloading or port address translation (PAT) gives your inside hosts a valid address to the outside. With PPPoE, you negotiate or receive just one address, and you most likely want to allow all the inside machines access out. You must configure NAT for overloading and define an access list for the address allowed to be translated. Refer to Chapter 5, "Using Network Address Translation," for detailed explanations of the commands, but it would look something like this:

     Router(config)#ip nat inside source list 100 interface dialer0 overload Router(config)#access-list 100 permit ip any any 
  5. Next, you need to configure a DHCP server; fortunately, the Cisco IOS DHCP server has all the features you need. We could spend an entire chapter on configuring the DHCP server with different options and scenarios, but here we are only interested in a quick working solution. The first command is to build a DHCP pool: enter ip dhcp pool name to build a pool and enter its configuration mode. Next, enter import all, which tells this DHCP server to get additional information from a central DHCP server. Next, you specify the network of the pool with network network-address subnet-mask, and the final step is to specify the default gateway/router with a default-router host-address command. In Listing 10.3, you can see the steps combined to configure the DHCP pool.

    Listing 10.3 DHCP configuration example
     Router(config)#ip dhcp pool home Router(dhcp-config)#import all Router(dhcp-config)#network 192.168.1.0 255.255.255.0 Router(dhcp-config)#default-router 192.168.1.1 Router(dhcp-config)#exit Router(config)# 
  6. The last step in the process is to configure a default route that will direct all traffic out the dialer interface, which is configured for PPPoE:

     Router(config)# ip route 0.0.0.0 0.0.0.0 dialer0 

Configuring DSL for PPPoA

There are five steps to configuring the CPE for PPPoA. The process is almost the same as that for PPPoE, but there are a few differences and they all occur in the first step, the configuration of the ATM interface:

  1. To configure the ATM interface for PPPoA, you need to set the line mode. The Cisco 827 ADSL interface is in auto-detect mode by default and will auto-detect the line encoding. If you need to change it, you use the dsl operating-mode command. Some of the other options are shown as well:.

     Router(config)#interface atm 0 Router(config-if)#dsl operating-mode ? ansi-dmt ANSI full rate mode auto auto detect mode itu-dmt ITU full rate mode splitterless G.lite mode ansi-dmt         ANSI full rate mode auto             auto detect mode itu-dmt          ITU full rate mode splitterless     G.lite mode 

    The other unique part is to set encapsulation for ATM with the encapsulation aal5mux ppp dialer command and to associate it with the appropriate dialer interface with the dialer pool-member number command:

     Router(config)#interface atm 0 Router(config-if)#pvc 9/17 Router(config-if-atm-vc)#encapsulation aal5mux ppp dialer Router(config-if-atm-vc)#dialer pool-member 1 

    The remainder of the steps are identical to those of PPPoE with the exception of the numbering.

  2. Configure a dialer interface.

  3. Configure PAT.

  4. Configure a DHCP server.

  5. Configure a default route.

Troubleshooting DSL

DSL connectivity is a multilayer process; as such, it helps in troubleshooting because we can break it down on a per-layer basis. To troubleshoot at Layer 1, the show dsl interface atm 0 command can check whether the router has trained to the DSLAM. If you receive a screen full of statistics, chances are that Layer 1 is good; if the display indicates that the line is not activated, then you need to troubleshoot at Layer 1. One item to check is the carrier-detect light. If it is on, then Layer 1 is not the issue; if it is off, then the next thing to check is your DSLAM modulation with your provider. Also check, of course, the obvious things such as the plug. The 827 uses a different power supply from other 800 series routers. Check whether the DSL router is cabled to the wall jack. A 6-pin cable uses pins 3(Tip) and 4(Ring); a 4-pin cable uses pins 2(Tip) and 3(Ring).

You usually track down Layer 2 issues with debug commands. debug atm events shows you whether there is any activity. Some of the traffic should indicate the VPIs and VCIs that are in use. There should be some ATM maintenance traffic flowing at regular intervals; if no output is detected, you should contact your provider. As with any debug commands, you should use them sparingly because they can have an adverse effect on performance. Use the undebug all or no debug all command to disable debugging. You can also execute a show int atm0 command to see that input and output packet counters are increasing.

At this point, look at PPP and make sure it is negotiating properly. You can use debug ppp authentication to see just the authentication process or use the debug ppp negotiation to see the entire setup and initialization steps. Refer to the Chapter 4, "PPP Authentication with PAP and CHAP," for additional help.



CCNP BCRAN Remote Access Exam Cram 2 (Exam Cram 640 - XXX)
CCNP BCRAN Remote Access Exam Cram 2 (Exam Cram 640 - XXX)
ISBN: N/A
EAN: N/A
Year: 2003
Pages: 183

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net