|
|
< Free Open Study > |
|
Label Distribution Protocol ConfigurationThis section gives you a basic understanding of how to configure LDP. The assumption is made that you already know how to configure the IGP routing protocols. Various aspects of configuring MPLS TE are dealt with in the remaining chapters of this book. Configuration tasks for LDP include the following:
Consider the service provider network shown in Figure 2-31. The goal here is to have an MPLS-enabled core. This section goes over the configuration of a core router 12008a to enable MPLS and thus act as an LSR. In order to enable MPLS in the entire network, you of course must enable CEF, MPLS forwarding globally, and MPLS forwarding per interface on every router you want to participate in MPLS. Figure 2-31. Enabling MPLS Forwarding in a Service Provider Network
Configuring CEFExample 2-17 shows how to enable CEF on a router. 12008a is used as an example. Example 2-17 Enabling CEF12008a#configure terminal Enter configuration commands, one per line. End with CNTL/Z. mpls-12008a(config)#ip cef distributed The keyword distributed applies only to distributed platforms such as the 12000 series router and the 7500 series router—not the 7200 series router. NOTE Example 2-17 shows how CEF is configured on a 12000-series router. In reality, 12000 series routers support CEF only as a forwarding mechanism, so there is no need to explicitly enable it. By default, distributed CEF is enabled on 12000 series routers. CEF is a prerequisite for MPLS forwarding, no matter whether you're using LDP, TDP, RSVP, or BGP for label distribution. Example 2-18 shows output generated from the show ip interface command. Example 2-18 Verifying That CEF Is Enabled on Inbound/Outbound Interfaces with the show ip interface Command12008a#show ip interface pos 2/1 POS2/1 is up, line protocol is up Internet address is 10.0.3.5/24 Broadcast address is 255.255.255.255 Address determined by non-volatile memory MTU is 4470 bytes Helper address is not set Directed broadcast forwarding is disabled Multicast reserved groups joined: 224.0.0.5 224.0.0.6 224.0.0.2 224.0.0.14 Outgoing access list is not set Inbound access list is not set Proxy ARP is enabled Security level is default Split horizon is enabled ICMP redirects are always sent ICMP unreachables are always sent ICMP mask replies are never sent IP fast switching is enabled IP fast switching on the same interface is enabled IP Flow switching is disabled IP CEF switching is enabled IP Null turbo vector IP multicast fast switching is enabled IP multicast distributed fast switching is disabled IP route-cache flags are Fast, Distributed, CEF Router Discovery is disabled IP output packet accounting is disabled IP access violation accounting is disabled TCP/IP header compression is disabled RTP/IP header compression is disabled Probe proxy name replies are disabled Policy routing is disabled Network address translation is disabled WCCP Redirect outbound is disabled WCCP Redirect inbound is disabled WCCP Redirect exclude is disabled BGP Policy Mapping is disabled show ip cef summary is another useful command to ensure that CEF is enabled, as demonstrated in Example 2-19. This is useful for global information about CEF, such as whether it is running in distributed mode, how many routing entries are present, and so on. Example 2-19 Verifying That CEF Is Enabled and Obtaining Statistics with the show ip cef summary Command
mpls-12008a#show ip cef summary
IP Distributed CEF with switching (Table Version 170), flags=0x0, bits=8
37 routes, 0 reresolve, 0 unresolved (0 old, 0 new)
37 leaves, 34 nodes, 40724 bytes, 141 inserts, 104 invalidations
8 load sharing elements, 2688 bytes, 8 references
universal per-destination load sharing algorithm, id 5008782C
2 CEF resets, 33 revisions of existing leaves
34 in-place modifications
refcounts: 8999 leaf, 8960 node
Adjacency Table has 9 adjacencies
Configuring MPLS Forwarding GloballyExample 2-20 shows the router-level configuration to enable MPLS forwarding. Example 2-20 Configuring MPLS Forwarding Globallympls-12008a#configure terminal Enter configuration commands, one per line. End with CNTL/Z. mpls-12008a(config)#mpls ip mpls-12008a(config)#mpls label protocol ldp mpls-12008a(config)#mpls ldp router-id loopback 0 The command mpls label protocol ldp specifies which label protocol to use. The choices are ldp and tdp. The command mpls ldp router-id loopback 0 ties the MPLS LSR ID to the IP address of the virtual interface Loopback0. This is recommended in order to avoid unpredictable results during session establishment. Interface-Level ConfigurationApart from the router-level configuration presented in Example 2-20, you also have to explicitly configure each interface on which you want to run LDP by using the command mpls ip at the interface level. Frame-Mode Interface ConfigurationExample 2-21 shows the MPLS configuration options available at the interface level for a frame-mode interface. Example 2-21 Interface-Level MPLS Configuration12008a#configure terminal Enter configuration commands, one per line. End with CNTL/Z. mpls-12008a(config)#interface ethernet 0 mpls-12008a(config-if)#mpls ip mpls-12008a(config-if)#mpls mtu ? <64-65536> MTU (bytes) mpls-12008a(config-if)#mpls mtu 1504 mpls-12008a(config-if)#end mpls label protocol can also be specified at the interface level. This is particularly useful when you have Cisco LSRs as neighbors on one set of interfaces and non-Cisco neighbors on other interfaces. This way, you can leave the label protocol as tdp (the default) on all interfaces connecting to other Cisco devices while specifying LDP on others. The MPLS MTU size can also be set to values between 64 and 65,536 in order to accommodate situations in which label sizes have to be accounted for during MTU computation. This typically is an issue only on Ethernet interfaces. For example, on an Ethernet interface, you'd normally have a maximum data size of 1500 bytes. After the Ethernet encapsulation has been added, this equals 1518 bytes (the maximum allowable size of an Ethernet frame). However, because of the fact that certain end-user applications don't use Path MTU Discovery to discover the maximum MTU size, you can receive 1500-byte IP packets with the Don't Fragment (DF) bit set. When this arrives at an ingress LER, the LER adds 4 or more bytes of MPLS header(s). The LER needs to fragment the packet, but it cannot do so because the DF bit is set. In such situations, if the LSR and the Layer 2 switches that are downstream from the LSR accommodate jumbo packets (also called baby giants), it is possible to transmit 1504-byte packets (with one MPLS label) or more on the Ethernet by setting the MPLS MTU to 1504. You compute the IP MTU by subtracting the number of labels from the MPLS MTU:
The maximum number of labels in your network depends on several factors. For example, if an LSR is a PE in an MPLS-VPN network, it requires at least two labels. A third label is also imposed if the LSR is the headend of an MPLS TE tunnel. Using Fast ReRoute (see Chapter 7, "Protection and Restoration") might add another label to the mix, making your maximum label size 4. If the IP MTU on an interface is 1500, your best bet is to set the MPLS MTU to 1516 (1500 + (4 * 4)). Setting the MPLS MTU higher than you need doesn't hurt anything, so you might want to set the MPLS MTU to 1516 or 1520 on all Ethernet interfaces as a general practice. The show mpls forwarding a.b.c.d detail command displays the label stack used to forward a data packet to subnet a.b.c.d. Example 2-22 shows the output of show mpls forwarding detail on 7200a when directed LDP is run on top of TE tunnel tunnel1. In this example, you can see the label stack containing two labels, 12335 and 43. 12335 is the tunnel label, and 43 is the result of LDP over it. Example 2-22 Label Stack Information Obtained Through show mpls forwarding detail7200a#show mpls forwarding 13.13.13.13 detail Local Outgoing Prefix Bytes tag Outgoing Next Hop tag tag or VC or Tunnel Id switched interface 45 12326 13.13.13.13/32 0 PO3/0 point2point MAC/Encaps=4/8, MTU=4470, Tag Stack{12326} FF030281 03026000 No output feature configured Per-destination load-sharing, slots: 0 2 4 6 8 10 12 14 43 13.13.13.13/32 0 Tu1 point2point MAC/Encaps=4/12, MTU=4466, Tag Stack{12335 43}, via PO3/0 FF030281 0302F0000002B000 No output feature configured Per-destination load-sharing, slots: 1 3 5 7 9 11 13 15 Cell-Mode Interface ConfigurationExample 2-23 shows the interface-level MPLS configuration for an ATM subinterface on a router for cell-mode operation. Example 2-23 Cell-Mode ATM Configuration for an ATM Subinterface on a Router12008a#configure terminal Enter configuration commands, one per line. End with CNTL/Z. mpls-12008a(config)#interface atm 3/0 mpls-12008a(config-if)#no ip address mpls-12008a(config-if)#no atm ilmi-keepalive mpls-12008a(config-if)#exit mpls-12008a(config)#interface atm 3/0.1 mpls mpls-12008a(config-subif)#ip address 10.0.4.5 255.255.255.0 mpls-12008a(config-subif)#mpls ip mpls-12008a(config-subif)#mpls ldp address-message mpls-12008a(config)#mpls ldp atm ? control-mode Select LSP setup control mode for MPLS VCs vc-merge Select VC merge capability Configuring MPLS forwarding and LDP on routers for cell-mode ATM is done on ATM subinterfaces. In addition to the commands done at the subinterface level, the following three router-level commands are specific to ATM:
Verifying Your LDP ConfigurationTo verify the configuration, the following commands are useful:
The discovery command option is useful to verify that hellos are being received from neighbors. Example 2-24 shows the output from the show mpls ldp discovery command. Example 2-24 show mpls ldp discovery Command Output Shows a List of Interfaces Over Which the LDP Discovery Process Is Running12008a#show mpls ldp discovery Local LDP Identifier: 5.5.5.5:0 Discovery Sources: Interfaces: POS1/0 (ldp): xmit/recv LDP Id: 11.11.11.11:0 POS2/0 (ldp): xmit/recv LDP Id: 3.3.3.3:0 POS2/1 (ldp): xmit/recv LDP Id: 4.4.4.4:0 ATM3/0.1 (ldp): xmit/recv LDP Id: 11.11.11.11:1; IP addr: 10.0.4.11 mpls-12008a# If you don't see recv for an interface, hello packets were not received. In the context of serial interfaces using PPP encapsulation, this could mean that there is a PPP negotiation issue. The show mpls ldp neighbor command, shown in Example 2-25, is useful to get more details, such as the addresses bound to a neighbor and the neighbor's LSR ID. If you have not explicitly configured the LDP identifier address using the mpls ldp router-id command, LDP might pick a different interface address than you intended. In such cases, the Peer LDP Ident field helps identify the problem. Example 2-25 show mpls ldp neighbor Command Output Shows the Status of LDP Sessions12008a#show mpls ldp neighbor Peer LDP Ident: 4.4.4.4:0; Local LDP Ident 5.5.5.5:0 TCP connection: 4.4.4.4.646 - 5.5.5.5.11000 State: Oper; Msgs sent/rcvd: 67/65; Downstream Up time: 00:39:40 LDP discovery sources: POS2/1 Addresses bound to peer LDP Ident: 7.1.5.110 10.0.1.4 4.4.4.4 10.0.3.4 Peer LDP Ident: 3.3.3.3:0; Local LDP Ident 5.5.5.5:0 TCP connection: 3.3.3.3.646 - 5.5.5.5.11001 State: Oper; Msgs sent/rcvd: 66/65; Downstream Up time: 00:39:40 LDP discovery sources: POS2/0 Addresses bound to peer LDP Ident: 7.1.5.113 10.0.1.3 3.3.3.3 10.0.2.3 Peer LDP Ident: 11.11.11.11:0; Local LDP Ident 5.5.5.5:0 TCP connection: 11.11.11.11.11011 - 5.5.5.5.646 State: Oper; Msgs sent/rcvd: 65/63; Downstream Up time: 00:38:24 LDP discovery sources: POS1/0 Addresses bound to peer LDP Ident: 10.0.19.11 10.0.17.11 10.0.4.11 7.1.5.100 11.11.11.11 10.0.5.11 Peer LDP Ident: 11.11.11.11:1; Local LDP Ident 5.5.5.5:1 TCP connection: 10.0.4.11.11013 - 10.0.4.5.646 State: Oper; Msgs sent/rcvd: 43/43; Downstream on demand Up time: 00:24:20 LDP discovery sources: ATM3/0.1 The TCP connection information comes in handy when you are having problems with your session. The State field is yet another place to check to see if everything is all right with the neighbor in question as far as your LDP session is concerned. Obviously, Oper (Operational) is good. If for any reason the session is not operational, you do not see an entry for this neighbor. The discovery sources field is useful when you have multiple links between two LSRs. In such cases, losing one link between the LSRs does not result in the session's going down. You can observe this in the output of this show command. Finally, the show mpls interfaces detail command is useful for checking things such as an interface's MPLS MTU size, as demonstrated in Example 2-26. The interesting information in this output is highlighted. Example 2-26 show mpls interfaces detail Command Output Displays Detailed Label Switching Information12008a#show mpls interfaces detail Interface Ethernet0: IP labeling enabled (ldp) LSP Tunnel labeling not enabled MPLS Frame Relay Transport labeling not enabled BGP labeling not enabled MPLS operational MTU = 1504 Interface POS1/0: IP labeling enabled (ldp) LSP Tunnel labeling enabled MPLS Frame Relay Transport labeling not enabled MPLS operational MTU = 4470 Interface POS2/0: IP labeling enabled (ldp) LSP Tunnel labeling not enabled MPLS Frame Relay Transport labeling not enabled MPLS operational MTU = 4470 Interface POS2/1: IP labeling enabled (ldp) LSP Tunnel labeling enabled MPLS Frame Relay Transport labeling not enabled MPLS operational MTU = 4470 Interface ATM3/0.1: IP labeling enabled (ldp) LSP Tunnel labeling not enabled MPLS Frame Relay Transport labeling not enabled MPLS operational MTU = 4470 ATM labels: Label VPI = 1, Control VC = 0/32 IP labeling enabled shows (ldp), meaning that the label imposition and switching are based on label bindings derived from LDP. This could be (tdp) if the TDP protocol were used on interface POS1/0. LSP Tunnel labeling enabled means that MPLS TE has been enabled on that interface. In Example 2-26, you can see that it is enabled for POS1/0 but not for POS2/0. This is a good place to see if MPLS MTU has been bumped up, as in the case of Ethernet/0, which has been increased from the default 1500 to 1504. Finally, for the LC-ATM interface ATM3/0.1, you can see the VPI being used and the control VC. In a multivendor environment, these default values might not be the same between two vendors and might have to be modified. Configuring Session AttributesIn some cases, you might want to change the default LDP Hello and Holdtime parameters. Example 2-27 shows how you can find out what the current LDP parameter values are. Example 2-27 Verifying Hello and Holdtime Intervals12008a#show mpls ldp parameters Protocol version: 1 Downstream label generic region: min label: 12304; max label: 100000 Session hold time: 180 sec; keep alive interval: 60 sec Discovery hello: holdtime: 15 sec; interval: 5 sec Discovery targeted hello: holdtime: 180 sec; interval: 5 sec Downstream on Demand max hop count: 255 TDP for targeted sessions LDP initial/maximum backoff: 15/120 sec LDP loop detection: off Example 2-28 shows how the default discovery hello holdtime can be changed. Example 2-28 Configuring Holdtime and Hello Intervals12008a#configure terminal Enter configuration commands, one per line. End with CNTL/Z. 12008a(config)#mpls ldp dis 12008a(config)#mpls ldp discovery hello ? holdtime LDP discovery Hello holdtime interval LDP discovery Hello interval 12008a(config)#mpls ldp discovery hello holdtime ? <1-2147483> Holdtime in seconds 12008a(config)#mpls ldp discovery hello holdtime 240 Example 2-29 shows how you can verify the configuration changes you made. Example 2-29 Verifying LDP Parameters12008a#show mpls ldp parameters Protocol version: 1 Downstream label generic region: min label: 12304; max label: 100000 Session hold time: 180 sec; keep alive interval: 60 sec Discovery hello: holdtime: 240 sec; interval: 5 sec Discovery targeted hello: holdtime: 180 sec; interval: 5 sec Downstream on Demand max hop count: 255 TDP for targeted sessions LDP initial/maximum backoff: 15/120 sec LDP loop detection: off As you can see from the highlighted text in Example 2-29, the discovery holdtime has been changed to 240. |
|
|
< Free Open Study > |
|
