|
|
Our router/bridge combines the advantages of a spanning-tree bridge and a full multiprotocol router. This combination provides the speed and protocol transparency of an adaptive spanning-tree bridge, along with the functionality, reliability, and security of a router.
This chapter describes how to configure transparent bridging and source-route transparent (SRT) bridging. For a complete description of the commands mentioned in this chapter, refer to the Router Products Command Reference publication.
Cisco supports transparent bridging for Ethernet, Fiber Distributed Data Interface (FDDI), and serial media, and supports SRT bridging for Token Ring media. In addition, Cisco supports all of the mandatory Management Information Base (MIB) variables specified for transparent bridging in RFC 1286.
Cisco's transparent bridging software implementation hasthe following features:
Routers can be configured to serve as both multiprotocol routers and MAC-level bridges, bridging any traffic that cannot otherwise be routed. For example, a router/bridge routing the Internet Protocol (IP) can also bridge Digital's LAT protocol or NetBIOS traffic.
Remote bridging over synchronous serial lines is also supported between our routers. As with frames received on all other media types, the dynamic learning and any filtering is applied to frames received on serial lines.
Transit bridging of Ethernet frames across FDDI media is also supported. The term transit refers to the fact that the source or destination of the frame cannot be on the FDDI media itself. This allows FDDI to act as a highly efficient backbone for the interconnection of many bridged networks. The configuration of FDDI transit bridging is identical to the configuration of transparent bridging on all other media types.
Our router/bridges also support transparent bridging on Token Ring interfaces that are capable of supporting SRT bridging.
As with all other media types, all the features that use bridge-group commands can be used on Token Ring interfaces. As with other interface types, the bridge group can be configured to run either the IEEE or Digital spanning-tree protocols. When using the IEEE spanning-tree protocol, the bridge cooperates with other bridges complying to the draft SRT bridging specification and constructs a loop-free topology across the entire extended LAN.
You can run the Digital spanning-tree protocol over Token Ring as well. Use it when you have other non-IEEE bridges on other media and do not have any SRT bridges on Token Ring. Note that in this configuration, all of your Token Ring transparent bridges must be Cisco routers. This is because the Digital spanning-tree protocol has not been standardized on Token Ring.
As specified by the SRT bridging specification, only packets without a routing information field (RIF) (RII = 0 in the SA field) will be transparently bridged. Packets with a RIF (RII = 1) are passed to the source-route bridging module for handling. Note that an SRT-capable Token Ring interface can have both source-route bridging and transparent bridging enabled at the same time. However, when running SRT bridging, frames that did not have a RIF when they were produced by their generating host will never gain a RIF, and frames that did have a RIF when they were produced will never lose that RIF.
When bridging between Token Ring and other media, certain packet transformations must occur. In all cases, the MAC addresses are bit-swapped, because the bit ordering on Token Ring is different from that on other media. In addition, Token Ring supports one higher-layer packet format, logical link control (LLC), while Ethernet supports two (LLC and Ethernet). The transformation of LLC frames between media is quite simple; a length field is either created (when going to non-Token Ring) or removed (when going to Token Ring). When an Ethernet format frame must go to Token Ring, it is translated into a LLC-1 "SNAP" packet; the destination service access point (DSAP) value is AA, the source service access point (SSAP) value is AA, with the organizational unique identifier (OUI) value 0000F8. Likewise, when such a packet in LLC-1 format is going to be bridged onto Ethernet media, it is translated back into Ethernet format. You can determine the OUI type used when transporting Ethernet Type II frames over other media.
Problems currently occur with the following protocols when bridged between Token Ring and other media: Novell IPX, DECnet Phase IV, AppleTalk, Banyan VINES, XNS, and IP. Further, problems can occur with the Novell IPX and XNS protocols when bridged between FDDI and other media. We recommend that these protocols be routed whenever possible.
Perform one or more of the tasks in the following sections to configure transparent bridging or SRT bridging on your router/bridge:
See the "Transparent and SRT Bridging Configuration Examples" section for configuration examples.
To configure transparent and SRT bridging, you must perform the tasks in the following sections:
The first step in setting up your transparent bridging network is to define a spanning-tree protocol and assign a bridge group number. You can choose either the IEEE 802.1D spanning-tree protocol or the earlier Digital protocol upon which this IEEE standard is based.
To assign a bridge group number and define a spanning-tree protocol, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Assign a bridge group number and define a spanning-tree protocol as either IEEE 802.1D standard or Digital. | bridge bridge-group protocol {ieee | dec} |
The IEEE 802.1D spanning-tree protocol is the preferred way of running the bridge. Use the Digital spanning-tree protocol only for backward compatibility.
You can configure the router to route a given protocol among one group of interfaces and concurrently bridge that protocol among a separate group of interfaces, all within one router. The given protocol is not switched between the two groups. Rather, routed traffic is confined to the routed interfaces and bridged traffic is confined to the bridged interfaces. A protocol may be either routed or bridged on a given interface, but not both.
The concurrent routing and bridging capability is, by default, disabled. While concurrent routing and bridging is disabled, the router absorbs and discards bridgeable packets in protocols that are configured for routing on any interface in the router.
When concurrent routing and bridging is first enabled in the presence of existing bridge groups, it will generate a bridge route configuration command for any protocol for which any interface in the bridge group is configured for routing. This is a precaution that applies only wnen CRB is not already enabled, bridge groups exist, and the bridge crb command is encountered.
To enable concurrent routing and bridging on the router, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Enable concurrent routing and bridging. | bridge crb |
Information about which protocols are routed and which are bridged is stored in a table, which can be displayed with the show interfaces crb privileged EXEC command.
Once concurrent routing and bridging has been enabled, you must configure an explicit bridge route command for any protocol that is to be routed on the interfaces in a bridge group in addition to any required protocol-specific interface configuration.
To configure specific protocols to be routed in a bridge group, perform the following task in interface configuation mode:
| Task | Command |
|---|---|
| Specify a protocol to be routed on a bridge group. | bridge bridge-group route [apollo | appletalk | clns | decnet | ip | ipx | vines | xns] |
A bridge group is an internal organization of network interfaces on a router. Bridge groups cannot be used outside the router on which it is defined to identify traffic switched within the bridge group. Bridge groups within the same router function as distinct bridges; that is, bridged traffic and BPDUs cannot be exchanged between different bridge groups on a router. Furthermore, bridge groups cannot be used to multiplex or demultiplex different streams of bridged traffic on a LAN. An interface can only be a member of one bridge group. Use a bridge group for each separately bridged (topologically distinct) network connected to the router. Typically, only one such network exists in a configuration.
The purpose of placing network interfaces into a bridge group is twofold:
For SRT bridging, if the Token Ring and serial interfaces are in the same bridge group, changing the serial encapsulation method causes the state of the corresponding Token Ring interface to be reinitialized. Its state will change from "up" to "initializing" to "up" again within a few seconds.
After you assign a bridge group number and define a spanning-tree protocol, assign each network interface to a bridge group by performing the following task in interface configuration mode:
| Task | Command |
|---|---|
| Assign a network interface to a bridge group. | bridge-group bridge-group |
For SRT bridging networks, you must choose the OUI code that will be used in the encapsulation of Ethernet Type II frames across Token Ring backbone networks. To choose the OUI, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Select the Ethernet Type II OUI encapsulation code. | ethernet-transit-oui [90-compatible | standard | cisco] |
Traditionally, a bridge group is an independently bridged subnetwork. In this definition, bridge groups cannot exchange traffic with other bridge groups, nor can they multiplex or demultiplex different streams of bridged traffic. Our transparently bridged virtual LAN feature permits a bridge group to extend outside the router to identify traffic switched within the bridge group.
While bridge groups remain internal organizations of network interfaces functioning as distinct bridges within a router, transparent bridging on subinterfaces permits bridge groups to be used to multiplex different streams of bridged traffic on a LAN or HDLC serial interface. In this way, bridged traffic may be switched out of one bridge group on one router, multiplexed across a subinterface, and demultiplexed into a second bridge group on a second router. Together, the first bridge group and the second bridge group form a transparently bridged virtual LAN. This approach can be extended to impose logical topologies upon transparently bridged networks.
The primary application of transparently bridged virtual LANs constructed in this way is to separate traffic between bridge groups of local network interfaces, to multiplex bridged traffic from several bridge groups on a shared interface (LAN or HDLC serial), and to form virtual LANs composed of collections of bridge groups on several routers. These virtual LANs improve performance because they reduce the propagation of locally bridged traffic, and they improve security benefits because they completely separate traffic.
In Figure 23-1, different bridge groups on different routers are configured into three virtual LANs that span the bridged network. Each bridge group consists of conventionally bridged local interfaces and a subinterface on the backbone FDDI LAN. Bridged traffic on the subinterface is encapsulated and "colored" with a virtual LAN identifier known as a Security Association Identifier common to all bridge groups participating in the virtual LAN. In addition, bridges only accept packets bearing Security Association IDs for which they have a configured subinterface. Thus, a bridge group is configured to participate in a virtual LAN if it contains a subinterface configured with the virtual LAN's characteristic Security Association ID. See the section "Transparently Bridged Virtual LANs Configuration Example" later in this chapter for an example configuration of the topology shown in Figure 23-1.
To configure a virtual LAN on a transparently bridged network, perform the following tasks, beginning in interface configuration mode:
| Task | Command |
|---|---|
| Specify a subinterface. | interface type slot/port.subinterface-number1 |
| Specify the IEEE 802.10 Security Data Exchange (SDE) Security Association ID. (In other words, specify the "color.") | encapsulation sde said |
| Associate the subinterface with an existing bridge group. | bridge-group bridge-group |
We support transparent bridging over the following types of networks:
This section describes how to configure bridging on these networks.
Fast-switched transparent bridging over ATM supports AAL5-SNAP encapsulated packets only. All bridged AAL5-SNAP encapsulated packets are fast switched. Fast-switched transparent bridging supports Ethernet, FDDI, and Token Ring packets sent in AAL5-SNAP encapsulation over ATM. See the section "Fast-Switched Transparent Bridging over ATM Example (Cisco 7000)" for an example configuration of fast-switched transparent bridging over ATM.
For more information on configuring ATM, refer to the "Configuring ATM" chapter of this publication.
Our software implements transparent bridging over multiprotocol LAPB encapsulation on serial interfaces. To configure transparent bridging over multiprotocol LAPB, perform the following tasks, beginning in global configuration mode:
| Task | Command |
|---|---|
| Specify the serial interface. | interface serial number1 |
| Specify no IP address to the interface. | no ip address2 |
| Configure multiprotocol LAPB encapsulation. | encapsulation lapb multi3 |
| Assign the interface to a bridge group. | bridge-group bridge-group |
| Specify the type of spanning-tree protocol. | bridge bridge-group protocol [ieee | dec] |
For an example of configuring transparent bridging over multiprotocol LAPB, see the section "Transparent Bridging over Multiprotocol LAPB Example" later in this chapter.
The transparent bridging software supports bridging of packets in X.25 frames. This ability is useful for such tasks as transmitting packets from proprietary protocols across an X.25 network.
The X.25 bridging software uses the same spanning-tree algorithm as the other bridging functions, but allows packets to be encapsulated in X.25 frames and transmitted across X.25 media. You specify the IP-to-X.121 address mapping, and the system maintains a table of both the Ethernet and X.121 addresses. To configure X.25 transparent bridging, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Specify IP-to-X.121 mapping. | x25 map bridge x.121-address broadcast [options-keywords] |
Configuring X.25 is discussed in more detail in the chapter "Configuring X.25 and LAPB."
The transparent bridging software supports bridging of packets over Frame Relay networks. This ability is useful for such tasks as transmitting packets from proprietary protocols across a Frame Relay network. Bridging over a Frame Relay network is supported both on networks that support a multicast facility and those that do not. Both cases are described in this section.
The transparent bridging software provides fast-switched transparent bridging for Frame Relay encapsulated serial and HSSI interfaces.
SVCs are not supported for transparent bridging in this release. All the PVCs configured on a subinterfaces must belong to the same bridge group.
The Frame Relay bridging software uses the same spanning-tree algorithm as the other bridging functions, but allows packets to be encapsulated for transmission across a Frame Relay network. You specify IP-to-DLCI (data-link connection identifier) address mapping, and the system maintains a table of both the Ethernet address and the DLCIs.
To configure bridging in a network not supporting a multicast facility, define the mapping between an address and the DLCI used to connect to the address. To bridge with no multicasts, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Define the mapping between an address and the DLCI used to connect to the address. | frame-relay map bridge dlci broadcast |
An example configuration is provided in the section "Frame Relay Transparent Bridging Examples" at the end of this chapter. Frame Relay is discussed in more detail in the chapter "Configuring Frame Relay."
The multicast facility is used to learn about the other bridges on the network, eliminating the need for you to specify any mappings with the frame-relay map bridge broadcast command. An example configuration is provided in the section "Frame Relay Transparent Bridging Examples" at the end of the chapter for use as a configuration guide. Frame Relay is discussed in more detail in the chapter "Configuring Frame Relay."
We support transparent bridging over SMDS networks. Standard bridging commands are used to enable bridging on an SMDS interface. To enable transparent bridging over SMDS, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Enable transparent bridging of packets across an SMDS network. | smds multicast bridge smds-address1 |
When transparently bridging over an SMDS network, the broadcast Address Resolution Protocol (ARP) packets are treated differently from other encapsulation methods. For SMDS, two packets are sent to the multicast address. One is sent using a standard (SMDS) ARP encapsulation; the other is sent with the ARP packet encapsulated in an 802.3 MAC header. The native ARP is sent as a regular ARP broadcast.
In addition, our implementation of 802.6 bridging only supports the transmission and reception of 802.3 encapsulated bridge packets. However, other encapsulations will be supported in a future release. Software Release 9.21 does not support bridging over multiple logical IP subnets (MultiLIS) because bridging of IP packets in a MultiLIS environment is unpredictable. It also does not support bridging from SMDS to SMDS; that is, packets from one SMDS network will not be forwarded to another SMDS network interface through the bridging mechanism. SMDS is discussed in more detail in the chapter "Configuring SMDS."
You can configure one or more transparent bridging options. To configure transparent bridging options, perform one or more of the tasks in the following sections:
If you want to bridge IP, you must disable IP routing because IP routing is enabled by default on all routers/bridges. You can enable IP routing when you decide to route IP packets. To disable or enable IP routing, perform one of the following tasks in global configuration mode:
| Task | Command |
|---|---|
| Disable IP routing. | no ip routing |
| Enable IP routing. | ip routing |
All interfaces in the bridge group that are bridging IP should have the same IP address. However, if you have more than one bridge group, each bridge group should have its own IP address.
Normally, bridging takes place on the processor card at the interrupt level. When autonomous bridging is enabled, bridging takes place entirely on the ciscoBus II controller, significantly improving performance. Autonomous bridging is a high-speed switching feature that allows bridged traffic to be forwarded and flooded on the ciscoBus II controller between resident interfaces. If you are using the ciscoBus II controller, you can maximize performance by enabling autonomous bridging on the following ciscoBus II interfaces:
Although performance improvements will be seen most in the resident interfaces, the autonomous bridging feature can also be used in bridge groups that include interfaces that are not on the ciscoBus II controller. These interfaces include the CTR, FCI with encapsulation bridging, and HSSI with encapsulation other than HDLC, such as X.25, Frame Relay, or SMDS, MCI, STR, or SBE16.
If you enable autonomous bridging for a bridge group that includes a combination of interfaces that are resident on the ciscoBus II controller and some that are not, the ciscoBus II controller forwards only packets between resident interfaces. Forwarding between nonresident and resident interfaces is done in either the fast or process paths. Flooding between resident interfaces is done by the ciscoBus II controller. Flooding between nonresident interfaces is done conventionally. If a packet is forwarded from a nonresident to a resident interface, the packet is conventionally forwarded. If packets are flooded from a nonresident interface to a resident interface, the packet is autonomously flooded.
To enable autonomous bridging on a per-interface basis, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Enable autonomous bridging (if using the ciscoBus II controller). | bridge-group bridge-group cbus-bridging |
The LAT protocol used by Digital and Digital-compatible terminal servers is one of the common protocols that lacks a well-defined network layer (Layer 3) and so always must be bridged.
To reduce the amount of bandwidth that LAT traffic consumes on serial interfaces, you can specify a LAT-specific form of compression. Doing so applies compression to LAT frames being sent out the router/bridge through the interface in question. To configure LAT compression, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Reduce the amount of bandwidth that LAT traffic consumes on a serial interface. | bridge-group bridge-group lat-compression |
LAT compression can be specified only for serial interfaces. For the most common LAT operations (user keystrokes and acknowledgment packets), LAT compression reduces LAT's bandwidth requirements by nearly a factor of two.
The Cisco IEEE 802.1D bridging software supports spanning-tree domains of bridge groups. Domains are a feature specific to Cisco. This feature is only available if you have specified IEEE as the spanning-tree protocol. A domain establishes an external identification of the BPDUs sent from a bridge group. The purpose of this identification is as follows:
You can place any number of router/bridges within the domain. The devices in the domain, and only those devices, will then share spanning-tree information.
When multiple routers share the same cable and you want to use only certain discrete subsets of those routers to share spanning-tree information with each other, establish spanning-tree domains. This function is most useful when running other router applications, such as IP User Datagram Protocol (UDP) flooding, that use the IEEE spanning tree. You also can use this feature to reduce the number of global reconfigurations in large bridged networks.
To establish multiple spanning-tree domains, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Establish a multiple spanning-tree domain. | bridge bridge-group domain domain-number |
For an example of how to configure domains, see the "Complex Transparent Bridging Network Topology Example" section later in this chapter.
Normally, the system forwards any frames for stations that it has learned about dynamically. By disabling this activity, the bridge will only forward frames whose address have been statically configured into the forwarding cache. To prevent or allow forwarding of dynamically determined stations, perform one of the following task in global configuration mode:
A packet with a RIF, indicated by a source address with the multicast bit turned on, is not usually forwarded. However, you can configure bridging support to allow the forwarding of frames that would otherwise be discarded because they have a RIF. Although you can forward these frames, the bridge table will not be updated to include the source addresses of these frames.
To forward frames with multicast addresses, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Allow the forwarding of frames with multicast source addresses. | bridge bridge-group multicast-source |
A bridge forwards, floods, or drops packets based on the bridge table. The bridge table maintains both static entries and dynamic entries. Static entries are entered by the network manager or by the bridge itself. Dynamic entries are entered by the bridge learning process. A dynamic entry is automatically removed after a specified length of time, known as aging time, from the time the entry was created or last updated.
If hosts on a bridged network are likely to move, decrease the aging-time to enable the bridge to adapt to the change quickly. If hosts do not transmit continuously, increase the aging time to record the dynamic entries for a longer time and thus reduce the possibility of flooding when the hosts transmit again.
To set the aging time, perform the following task in global configuration mode:
| Task | Command |
| Set the bridge table aging time. | bridge-group bridge-group aging-time seconds |
A bridge examines frames and transmits them through the internetwork according to the destination address; a bridge will not forward a frame back to its originating network segment. The bridge software allows you to configure specific administrative filters that filter frames based upon information other than paths to their destinations. You can perform administrative filtering by performing one of the tasks in the following sections:
When configuring transparent bridging access control, keep the following points in mind:
You can filter transmission of frames at the MAC layter by performing tasks in one of the following sections:
When filtering by a MAC-level address, you can use two kinds of access lists: standard access lists that specify a simple address, and extended access lists that specify two addresses. You can also further restrict access by creating filters for these lists. After you have completed one of the preceding tasks, perform the task in the following section:
You can filter frames with a particular MAC-level station source or destination address. Any number of addresses can be configured into the system without a performance penalty. To filter by the MAC-level address, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Filter particular MAC-level station addresses. | bridge bridge-group address mac-address {forward | discard} [interface] |
When filtering specific MAC destination addresses, allow for multicast or broadcast packets that are required by the bridged network protocols. Refer to the example in the section "Multicast or Broadcast Packets Bridging Example" later in this chapter to guide you in building your configuration to allow for multicast or broadcast packets.
The bridging software allows you to create access lists to administratively filter MAC addresses. These access lists can filter groups of MAC addresses, including those with particular vendor codes. There is no noticeable performance loss in using these access lists, and the lists can be of indefinite length. You can filter groups of MAC addresses with particular vendor codes by performing the first task and one or both of the other tasks that follow:
To establish a vendor code access list, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Prepare access control information for filtering of frames by canonical (Ethernet-ordered) MAC address. | access-list access-list-number {permit | deny} address mask |
The vendor code is the first three bytes of the MAC address (left to right).
For an example of how to filter by vendor code, see "Multicast or Broadcast Packets Bridging Example" later in this chapter.
Once you have defined an access list to filter by a particular vendor code, you can assign an access list to a particular interface for filtering on the MAC source addresses of packets received on that interface or the MAC destination addresses of packets that would ordinarily be forwarded out that interface. To filter by source or destination addresses, perform one of the following tasks in interface configuration mode:
You can filter by protocol type by using the access-list mechanism and specifying a protocol type code. To filter by protocol type, perform the first task and one or more of the other tasks that follow:
The order in which you enter access-list commands affects the order in which the access conditions are checked. Each condition is tested in succession. A matching condition is then used to execute a permit or deny decision. If no conditions match, a "deny" decision is reached.
Access lists for Ethernet- and IEEE 802.2-encapsulated packets affect only bridging functions. It is not possible to use such access lists to block frames with protocols that are being routed.
You can establish protocol type access lists. Specify either an Ethernet type code for Ethernet-encapsulated packets or a DSAP/SSAP pair for 802.3 or 802.5-encapsulated packets. Ethernet type codes are listed in the "Ethernet Type Codes" appendix of the Router Products Command Reference publication.
To establish protocol type access lists, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Prepare access control information for filtering frames by protocol type. | access-list access-list-number {permit | deny} type-code wild-mask |
You can filter Ethernet- and SNAP-encapsulated packets on input. For SNAP-encapsulated frames, the access list you create is applied against the two-byte TYPE field given after the DSAP/SSAP/OUI fields in the frame. The access list is applied to all Ethernet and SNAP frames received on that interface prior to the bridge learning process. SNAP frames also must pass any applicable IEEE 802.2 DSAP/SSAP access lists.
You can also filter Ethernet- and SNAP-encapsulated packets on output. The access list you create is applied just before sending out a frame to an interface.
To filter these packets on input or output, perform either or both of the following tasks in interface configuration mode:
You can filter IEEE 802-encapsulated packets on input. The access list you create is applied to all IEEE 802 frames received on that interface prior to the bridge-learning process. SNAP frames also must pass any applicable Ethernet type-code access list.
You can also filter IEEE 802-encapsulated packets on output. SNAP frames also must pass any applicable Ethernet type-code access list. The access list you create is applied just before sending out a frame to an interface.
To filter these packets on input or output, perform one or both of the following tasks in interface configuration mode:
Access lists for Ethernet- and IEEE 802-encapsulated packets affect only bridging functions. You cannot use such access lists to block frames with protocols that are being routed.
If you are filtering by the MAC-level address, whether it is by a specific MAC address, vendor code, or protocol type, you can define and apply extended access lists. Extended access lists allow finer granularity of control. They allow you to specify both source and destination addresses and arbitrary bytes in the packet.
To define an extended access list, perform the following task in global configuration mode:
To apply an extended access list to an interface, perform one or both of the following tasks in interface configuration mode:
After an access list is created initially, any subsequent additions (possibly entered from the terminal) are placed at the end of the list. In other words, you cannot selectively add or remove access list command lines from a specific access list.
 | Caution Because of their complexity, only use extended access lists if you are very familiar with the router. Further, do not specify an offset value that is greater than the size of the packet. |
The bridging software allows you to filter LAT frames. LAT bridge filtering allows the selective inclusion or exclusion of LAT multicast service announcements on a per-interface basis.
In the LAT protocol, a group code is defined as a decimal number in the range 0 to 255. Some of the LAT configuration commands take a list of group codes; this is referred to as a group code list. The rules for entering numbers in a group code list follow:
In LAT, each node transmits a periodic service advertisement message that announces its existence and availability for connections. Within the message is a group code list; this is a mask of up to 256 bits. Each bit represents a group number. In the traditional use of LAT group codes, a terminal server only will connect to a host system when there is an overlap between the group code list of the user on the terminal server and the group code list in the service advertisement message. In an environment with many bridges and many LAT hosts, the number of multicast messages that each system has to deal with becomes unreasonable. The 256 group codes might not be enough to allocate local assignment policies, such as giving each DECserver 200 device its own group code in large bridged networks. LAT group code filtering allows you to have very fine control over which multicast messages actually get bridged. Through a combination of input and output permit and deny lists, you can implement many different LAT control policies.
You can filter LAT service advertisements by performing any of the tasks in the following sections:
You can specify LAT group-code filtering to inform the system that LAT service advertisements require special processing. To enable LAT group-code filtering, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Enable LAT service filtering. | bridge bridge-group lat-service-filtering |
You can specify the group codes by which to deny or permit access upon input. Specifying deny conditions causes the system to not bridge any LAT service advertisement that contain any of the specified groups. Specifying permit conditions causes the system to bridge only those service advertisements that match at least one group in the specified group list.
To specify deny or permit conditions for LAT groups on input, perform one of the following tasks in interface configuration mode:
| Task | Command |
|---|---|
| Specify the group codes by which to deny access upon input. | bridge-group bridge-group input-lat-service-deny group-list |
| Specify the group codes with which to permit access upon input. | bridge-group bridge-group input-lat-service-permit group-list |
If a message specifies group codes in both the deny and permit list, the message is not bridged.
You can specify the group codes by which to deny or permit access upon output. Specifying deny conditions causes the system to not bridge onto the output interface any LAT service advertisements that contain any of the specified groups. Specifying permit conditions causes the system to bridge onto the output interface only those service advertisements that match at least one group in the specified group list.
To specify deny or permit conditions for LAT groups on output, perform one of the following tasks in interface configuration mode:
If a message matches both a deny and a permit condition, it will not be bridged.
You might need to adjust certain spanning-tree parameters if the default values are not suitable for your bridge configuration. Parameters affecting the entire spanning tree are configured with variations of the bridge global configuration command. Interface-specific parameters are configured with variations of the bridge-group interface configuration command.
You can adjust spanning-tree parameters by performing any of the tasks in the following sections:
You can globally configure the priority of an individual bridge when two bridges tie for position as the root bridge, or you can configure the likelihood that a bridge will be selected as the root bridge. This priority is determined by default; however, you can change it. To set the bridge priority, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Set the bridge priority. | bridge bridge-group priority number |
You can set a priority for an interface. When two bridges tie for position as the root bridge, you configure an interface priority to break the tie. The bridge with the lowest interface value is elected. To set an interface priority, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Establish a priority for a specified interface. | bridge-group bridge-group priority number |
Each interface has a path cost associated with it. By convention, the path cost is 1000/data rate of the attached LAN, in Mbps. You can set different path costs. Refer to the entry for this command in the Router Products Command Reference publication for the various media defaults. To assign path costs, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Set a different path cost other than the defaults. | bridge-group bridge-group path-cost cost |
You can adjust hello (BPDU intervals as described in the following sections:
You can specify the interval between hello BPDUs. To adjust this interval, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Specify the interval between hello BPDUs. | bridge bridge-group hello-time seconds |
The forward delay interval is the amount of time spent listening for topology change information after an interface has been activated for bridging and before forwarding actually begins. To change the default interval setting, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Set the default of the forward delay interval. | bridge bridge-group forward-time seconds |
If a bridge does not hear BPDUs from the root bridge within a specified interval, it assumes that the network has changed and recomputes the spanning-tree topology. To change the default interval setting, performing the following task in global configuration mode:
| Task | Command |
|---|---|
| Change the amount of time a bridge will wait to hear BPDUs from the root bridge. | bridge bridge-group max-age seconds |
When a loop-free path exists between any two bridged subnetworks, you can prevent BPDUs generated in one transparent bridging subnetwork from impacting nodes in the other transparent bridging subnetwork, yet still permit bridging throughout the bridged network as a whole. For example, when transparently bridged LAN subnetworks are separated by a WAN, BPDUs can be prevented from traveling across the WAN link.
To disable the spanning tree on an interface, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Disable the spanning tree on an interface. | bridge-group bridge-group spanning-disabled |
In the process of loop elimination, the spanning-tree algorithm always blocks all but one of a group of parallel network segments between two bridges. When those segments are of limited bandwidth, it might be preferable to augment the aggregate bandwidth between two bridges by forwarding across multiple parallel network segments. Circuit groups can be used to group multiple parallel network segments between two bridges to distribute the load while still maintaining a loop-free spanning tree.
Deterministic load distribution distributes traffic between two bridges across multiple parallel network segments grouped together into a single circuit group. As long as one port of the circuit group is in the forwarding state, all ports in that circuit group will participate in load distribution regardless of their spanning-tree port states. This process guarantees that the computed spanning tree is still adaptive to any topology change and the load is distributed among the multiple segments. Deterministic load distribution guarantees packet ordering between source-destination pairs, and always forwards traffic for a source-destination pair on the same segment in a circuit group for a given circuit-group configuration.
If a circuit-group port goes down and up as a result of configuration or a line protocol change, the spanning-tree algorithm will bypass port transition and will time out necessary timers to force the eligible circuit-group ports to enter the forwarding state. This avoids the long disruption time caused by spanning-tree topology recomputation and therefore resumes the load distribution as quickly as possible.
To tune the transparently bridged network, perform the following tasks:
Step 1 Define a circuit group.
Step 2 Optionally, configure a transmission pause interval.
Step 3 Modify the load distribution strategy.
To define a circuit group, perform the following task in interface configuration mode:
| Task | Command |
|---|---|
| Add a serial interface to a circuit group. | bridge-group bridge-group circuit-group circuit-group |
For circuit groups of mixed-bandwidth serial interfaces, it might be necessary to configure a pause interval during which transmission is suspended to avoid misordering packets following changes in the composition of a circuit group. Changes in the composition of a circuit group include the addition or deletion of an interface and interface state changes. To configure a transmission pause interval, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Configure a transmission pause interval. | bridge bridge-group circuit-group circuit-group pause milliseconds |
For applications that depend on the ordering of mixed unicast and multicast traffic from a given source, load distribution must be based upon the source MAC address only. To modify the load distribution strategy to accommodate such applications, perform the following task in global configuration mode:
| Task | Command |
|---|---|
| Base load distribution on the source MAC address only. | bridge bridge-group circuit-group circuit-group source-based |
For an example of how to configure a circuit group, see the "Complex Transparent Bridging Network Topology Example" section later in this chapter.
This section describes how to monitor and maintain activity on the bridged network. You can perform one or more of the following tasks in privileged EXEC mode:
The following sections provide example configurations that you can use as a guide to configuring your bridging environment:
Figure 23-2 is an example of a basic bridging configuration. The system has two Ethernets, one Token Ring, one FDDI port, and one serial line. The IP is being routed, and everything else is being bridged. The Digital-compatible bridging algorithm with default parameters is being used.
The configuration file for the router/bridge depicted in Figure 23-2 would be as follows:
interface tokenring 0 ip address 131.108.1.1 255.255.255.0 bridge-group 1 ! interface fddi 0 ip address 131.108.2.1 255.255.255.0 bridge-group 1 ! interface ethernet 0 ip address 192.31.7.26 255.255.255.240 bridge-group 1 ! interface serial 0 ip address 192.31.7.34 255.255.255.240 bridge-group 1 ! interface ethernet 1 ip address 192.31.7.65 255.255.255.240 bridge-group 1 ! bridge 1 protocol dec
In the following example DecNet and IPX are concurrently routed and bridged. IP and AppleTalk are routed on all interfaces, DecNet and IP are routed on all interfaces not in the bridge group, and all protocols other than IP and AppleTalk are bridged on all interfaces in the bridge group:
! ipx routing 0000.0c36.7a43 appletalk routing ! decnet routing 9.65 decnet node-type routing-iv ! ! interface Ethernet0/0 ip address 172.19.160.65 255.255.255.0 ipx network 160 appletalk address 160.65 decnet cost 7 ! interface Ethernet0/1 ip address 172.19.161.65 255.255.255.0 ipx network 161 appletalk address 161.65 decnet cost 7 ! interface Ethernet0/2 ip address 172.19.162.65 255.255.255.0 appletalk address 162.65 bridge-group 1 ! interface Ethernet0/3 ip address 172.19.14.65 255.255.255.0 appletalk address 14.65 appletalk zone california bridge-group 1 ! router igrp 666 network 172.19.0.0 ! bridge crb bridge 1 protocol ieee bridge 1 route appletalk bridge 1 route ip !
The following examples show the configuration for the topology in Figure 23-1. The "striped" virtual LAN is identified as Security Association ID 45; the "dot" virtual LAN is identified as Security Association ID 1008; the "sliced" virtual LAN is identified as Security Association ID 4321. Note that the assignment of bridge group, interface, and subinterface numbers is of local significance only. You must coordinate only the configuration of a common Security Association ID across bridges.
bridge 18 protocol ieee interface ethernet 0/1 bridge-group 18 interface ethernet 0/2 bridge group 18 interface ethernet 0/3 bridge-group 18 interface fddi 4/0.8 encapsulation sde 45 bridge-group 18 ! bridge 54 protocol ieee interface ethernet 1/1 bridge group 54 interface ethernet 1/2 bridge group 54 interface ethernet 1/3 bridge group 54 interface fddi 4/0.13 encapsulation sde 1008 bridge group 54 ! bridge 3 protocol ieee interface ethernet 2/1 bridge-group 3 interface ethernet 2/2 bridge-group 3 interface ethernet 2/3 bridge-group 3 interface fddi 4/0.30 encapsulation sde 4321 bridge-group 3
bridge 7 protocol ieee interface ethernet 0/1 bridge-group 7 interface ethernet 0/2 bridge-group 7 interface ethernet 0/3 bridge-group 7 interface ethernet 0/4 bridge-group 7 interface fddi 2/0.11 encapsulation sde 4321 bridge-group 7 ! bridge 8 protocol ieee interface ethernet 1/1 bridge-group 8 interface ethernet 1/2 bridge-group 8 interface ethernet 1/3 bridge-group 8 interface ethernet 1/4 bridge-group 8 interface fddi 2/0.14 encapsulation sde 1008 bridge-group 8
bridge 1 protocol ieee interface ethernet 0/1 bridge-group 1 interface ethernet 0/2 bridge-group 1 interface ethernet 0/3 bridge-group 1 interface fddi 2/0.5 encapsulation sde 4321 bridge-group 1 ! bridge 6 protocol ieee interface ethernet 1/1 bridge-group 6 interface ethernet 1/2 bridge-group 6 interface ethernet 1/3 bridge-group 6 interface fddi 2/0.3 encapsulation sde 45 bridge-group 6
The following configuration example shows the configuration commands that enable transparent bridging between Ethernet and FDDI interfaces. Transparent bridging on an FDDI interface is allowed only on the CSC-C2FCIT interface card.
hostname tester ! buffers small min-free 20 buffers middle min-free 10 buffers big min-free 5 ! no ip routing ! interface ethernet 0 ip address 131.108.7.207 255.255.255.0 no ip route-cache bridge-group 1 ! interface ethernet 2 ip address 131.108.7.208 255.255.255.0 no ip route-cache bridge-group 1 ! interface Fddi 0 ip address 131.108.7.209 255.255.255.0 no ip route-cache no keepalive bridge-group 1 ! bridge 1 protocol ieee
If the other side of the FDDI ring were an FDDI interface running in encapsulation mode rather than in transparent mode, the following additional configuration commands would be needed:
interface fddi 0 fddi encapsulate
In the following example, two buildings have networks that must be connected via a T1 link. For the most part, the systems in each building use either IP or DECnet, and therefore, should be routed. There are some systems in each building that must communicate, but they can use only a proprietary protocol.
The example places two Ethernets in each building. One of the Ethernets is attached to the hosts that use a proprietary protocol, and the other is used to attach to the rest of the building network running IP and DECnet. The Ethernet attached to the hosts using a proprietary protocol is enabled for bridging to the serial line and to the other building.
Figure 23-3 shows an example configuration. The interfaces marked with an asterisk (*) are configured as part of spanning tree 1. The routers are configured to route IP and DECnet. This configuration permits hosts on any Ethernet to communicate with hosts on any other Ethernet using IP or DECnet. In addition, hosts on Ethernet 1 in either building can communicate using protocols not supported for routing.
The configuration file for the router/bridge in Building 1 would be as follows. Note that no bridging takes place over Ethernet 0. Both IP and DECnet routing are enabled on all interfaces.
decnet address 3.34 interface ethernet 0 ip address 128.88.1.6 255.255.255.0 decnet cost 10 ! interface serial 0 ip address 128.88.2.1 255.255.255.0 bridge-group 1 decnet cost 10 ! interface ethernet 1 ip address 128.88.3.1 255.255.255.0 bridge-group 1 decnet cost 10 ! bridge 1 protocol dec
The configuration file for the router/bridge in Building 2 is similar:
decnet address 3.56 ! interface ethernet 0 ip address 128.88.11.9 255.255.255.0 decnet cost 10 ! interface serial 0 ip address 128.88.2.2 255.255.255.0 bridge-group 1 decnet cost 10 ! interface ethernet 1 ip address 128.88.16.8 255.255.255.0 bridge-group 1 decnet cost 10 ! bridge 1 protocol dec
In Figure 23-4, a Token Ring and an Ethernet at a remote sales site in New York City must be configured to pass unroutable bridged traffic across a satellite link to the backbone Token Ring at the corporate headquarters in Thule, Greenland. IP is the only routed protocol. They are running the IEEE spanning-tree protocol to comply with the SRT bridging standard.
If there were source-routed traffic to bridge, the source-bridge command would also be used to configure source routing.
interface tokenring 0 ip address 150.136.1.1 255.255.255.128 bridge-group 1 ! interface ethernet 0 ip address 150.136.2.1 255.255.255.128 bridge-group 1 ! interface serial 0 ip address 150.136.3.1 255.255.255.128 bridge-group 1 ! bridge 1 protocol ieee
interface tokenring 0 ip address 150.136.10.1 255.255.255.128 bridge-group 1 ! interface serial 0 ip address 150.136.11.1 255.255.255.128 bridge-group 1 ! bridge 1 protocol ieee
When filtering specific MAC destination addresses, allow for multicast or broadcast packets that are required by the bridged network protocols.
Assume you are bridging IP in your network as illustrated in Figure 23-5.
The MAC address of Host A is 0800.0907.0207, and the MAC address of Host B is 0260.8c34.0864. The following configuration would work as expected, because input addresses work on the source address on the incoming interface:
access-list 700 permit 0260.8c34.0864 0000.0000.0000 access-list 700 deny 0000.0000.0000 FFFF.FFFF.FFFF interface ethernet 0 bridge-group 1 input-address-list 700
However, the following configuration might work initially but will eventually fail. The failure occurs because the configuration does not allow for an ARP broadcast with a destination address of FFFF.FFFF.FFFF, even though the destination address on the output interface is correct:
access-list 700 permit 0260.8c34.0864 0000.0000.0000 access-list 700 deny 0000.0000.0000 FFFF.FFFF.FFFF interface ethernet 0 bridge-group 1 output-address-list 700
The correct access list would be as follows:
access-list 700 permit 0260.8c34.0864 0000.0000.0000 access-list 700 permit FFFF.FFFF.FFFF 0000.0000.0000 access-list 700 deny 0000.0000.0000 FFFF.FFFF.FFFF interface ethernet 0 bridge-group 1 output-address-list 700
Figure 23-6 is an example configuration illustrating three bridges connected to each other through an X.25 network.
Following are the configuration commands for each of the bridges depicted in Figure 23-6.
interface ethernet 2 bridge-group 5 ip address 128.88.11.9 255.255.255.0 ! interface serial 0 encapsulation x25 x25 address 31370019027 bridge-group 5 x25 map bridge 31370019134 broadcast x25 map bridge 31370019565 broadcast ! bridge 5 protocol ieee
interface serial 1 encapsulation x25 x25 address 31370019134 bridge-group 5 x25 map bridge 31370019027 broadcast x25 map bridge 31370019565 broadcast ! bridge 5 protocol ieee
interface serial 0 encapsulation x25 x25 address 31370019565 bridge-group 5 x25 map bridge 31370019027 broadcast x25 map bridge 31370019134 broadcast ! bridge 5 protocol ieee
Figure 23-7 illustrates three bridges connected to each other through a Frame Relay network.
The Frame Relay bridging software uses the same spanning-tree algorithm as the other bridging functions, but allows packets to be encapsulated for transmission across a Frame Relay network. The command specifies IP-to-DLCI address mapping and maintains a table of both the Ethernet and DLCIs.
Following are the configuration commands for each of the bridges in a network that does not support a multicast facility.
interface ethernet 2 bridge-group 5 ip address 128.88.11.9 255.255.255.0 ! interface serial 0 encapsulation frame-relay bridge-group 5 frame-relay map bridge 134 broadcast frame-relay map bridge 565 broadcast ! bridge 5 protocol ieee
interface serial 1 encapsulation frame-relay bridge-group 5 frame-relay map bridge 27 broadcast frame-relay map bridge 565 broadcast ! bridge 5 protocol ieee
interface serial 0 encapsulation frame-relay bridge-group 5 frame-relay map bridge 27 broadcast frame-relay map bridge 134 broadcast ! bridge 5 protocol ieee
The multicast facility is used to learn about the other bridges on the network, eliminating the need for the frame-relay map commands.
Following are the configuration commands for each of the bridges in a network that supports a multicast facility.
interface ethernet 2 bridge-group 5 ip address 128.88.11.9 255.255.255.0 ! interface serial 0 encapsulation frame-relay bridge-group 5 ! bridge 5 protocol ieee
interface serial 1 encapsulation frame-relay bridge-group 5 ! bridge 5 protocol ieee
interface serial 0 encapsulation frame-relay bridge-group 5 ! bridge 5 protocol ieee
The following example illustrates a router configured for transparent bridging over multiprotocol LAPB encapsulation:
! no ip routing ! interface ethernet 1 no ip address no mop enabled bridge-group 1 ! interface serial 0 no ip address encapsulation lapb multi bridge-group 1 ! bridge 1 protocol ieee
The following configuration example enables fast-switched transparent bridging over ATM:
interface atm 4/0 ip address 1.1.1.1 255.0.0.0 atm pvc 1 1 1 aal5snap atm pvc 2 2 2 aal5snap atm pvc 3 3 3 aal5snap bridge-group 1 ! bridge 1 protocol dec
Figure 23-8 shows a network topology made up of four bridged subnetworks. Each bridged subnetwork is defined by the scope of a spanning tree. However, the scope of each spanning tree is not shown in detail because it is unnecessary for purposes of this discussion. Instead, it is shown by a half cloud labeled "To other parts of BSN."
For proper bridging operation, the bridged subnetworks cannot have connections between them, but they can be connected to the same backbone. In this example, three of the four bridged subnetworks are connected to the FDDI backbone and each belongs to a separate domain. Domains used in this topology allow the bridged subnetworks to be independent of one another while still bridging traffic onto the backbone destined for other connected bridged subnetworks. Domains can only be used in this manner if the bridged subnetworks have a single point of attachment to one another. In this case, the connection to the FDDI backbone is that single point of attachment. Each router on which a domain is configured and that has a single point of attachment to the other bridged subnetworks, checks whether a BPDU on the backbone is its own. If the BPDU does not belong to the bridged subnetwork, the router ignores the BPDU.
Separate bridged subnetworks, as in this example, allow spanning-tree reconfiguration of individual bridged subnetworks without disrupting bridging among the other bridged subnetworks.
The routers in this network are configured for bridging and demonstrate some of the bridging features available.
Router A demonstrates multiple bridge groups in one router for bridged traffic separation.
In Router A, the Token Ring interfaces are bridged together entirely independently of the other bridged interfaces in the router and belong to bridge group 1. Bridge group 1 does not use a bridge domain because the interfaces are bridged independently of other bridged subnetworks in the network topology and it has no connection to the FDDI backbone.
Also in Router A, the Ethernet interfaces belong to bridge group 3. Bridge group 3 has a connection to the FDDI backbone and has a domain defined for it so that it can ignore BPDUs for other bridged subnetworks.
interface ethernet 0 bridge-group 3 ! interface ethernet 1 bridge-group 3 ! interface ethernet 2 bridge-group 3 ! interface ethernet 3 bridge-group 3 ! interface fddi 0 bridge-group 3 ! interface tokenring 1 bridge-group 1 ! interface tokenring 2 bridge-group 1 ! bridge 1 protocol ieee bridge 3 domain 1 bridge 3 protocol ieee
Router B demonstrates a simple bridge configuration. It is connected to the FDDI backbone and has domain 2 defined. As such it can bridge traffic with the other FDDI-connected BSNs. Note that bridge group 1 has no relationship to bridge group 1 in Router A; bridge groups are an organization internal to each router.
interface ethernet 1 bridge-group 1 ! interface ethernet 2 bridge-group 1 ! interface fddi 0 bridge-group 1 ! bridge 1 domain 2 bridge 1 protocol ieee
Router C and Router D combine to demonstrate load balancing by means of circuit groups. Circuit groups are used to load balance across multiple parallel serial lines between a pair of routers. The router on each end of the serial lines must have a circuit group defined. The circuit group number can be the same or can be different. In this example, they are different.
Router C and Router D are configured with the same domain, because they must understand one another's BPDUs. If they were configured with separate domains, Router D would ignore Router C's BPDUs and vice versa.
interface fddi 0 bridge-group 2 ! interface serial 0 bridge-group 2 bridge-group 2 circuit-group 7 ! interface serial 1 bridge-group 2 bridge-group 2 circuit-group 7 ! interface serial 2 bridge-group 2 bridge-group 2 circuit-group 7 ! bridge 2 domain 3 bridge 2 protocol ieee
interface ethernet 4 bridge-group 5 ! interface ethernet 5 bridge-group 5 ! interface serial 0 bridge-group 5 bridge-group 5 circuit-group 4 ! interface serial 1 bridge-group 5 bridge-group 5 circuit-group 4 ! interface serial 2 bridge-group 5 bridge-group 5 circuit-group 4 ! bridge 5 domain 3 bridge 5 protocol ieee
|
|