|
|
Product Numbers:
WS-X5410 (Base Unit)
GBICs:
1000Base-SX WS-G5484
1000Base-LX/LH WS-G5486
This document describes the nine-port Gigabit EtherChannel Switching Module (GEM) and includes hardware installation and software configuration information.
The GEM (see Figure 1) provides nine Gigabit Ethernet ports that you can configure with any combination of shortwave (SX) and longwave/long-haul (LX/LH) Gigabit Interface Carriers (GBICs). The module has a 3.6-Gbps connection to the Catalyst 5000 series switch backplane. All the front panel ports are connected through a local Gigabit Ethernet switch engine that provides a nonblocking switch fabric between the onboard connections. Traffic between ingress and egress ports on the module is switched entirely on the module without going through the Catalyst backplane. This local switching allows for multiple modules, as well as the backplane, to switch simultaneously. Although traffic can be switched locally, all management and configuration of the module is through the supervisor engine.
This module is used primarily for backbone interconnection of other high-performance
Catalyst 5000 series switches and Cisco routers through Gigabit EtherChannel.
This document contains the following sections:
Cisco documentation and additional literature are available in a CD-ROM package, which ships with your product. The Documentation CD-ROM, a member of the Cisco Connection Family, is updated monthly. Therefore, it might be more current than printed documentation. To order additional copies of the Documentation CD-ROM, contact your local sales representative or call customer service. The CD-ROM package is available as a single package or as an annual subscription. You can also access Cisco documentation on the World Wide Web at http://www.cisco.com, http://www-china.cisco.com, or http://www-europe.cisco.com.
If you are reading Cisco product documentation on the World Wide Web, you can submit comments electronically. Click Feedback in the toolbar, select Documentation. After you complete the form, click Submit to send it to Cisco. We appreciate your comments.
Safety recommendations appear in this note in procedures that, if performed incorrectly, may harm you. A warning symbol precedes each warning statement. Warning symbols indicate the following:
 | Warning This warning symbol means danger. You are in a situation that could cause bodily injury. Before you work on any equipment, be aware of the hazards involved with electrical circuitry and be familiar with standard practices for preventing accidents. To see translations of the warnings that appear in this publication, refer to the appendix "Translated Safety Warnings" in the Catalyst 5000 Series Module Installation Guide. |
This section presents a high-level functional overview of the GEM's interaction with the
Catalyst 5000 series switch.
At the top of Figure 2 are the components that reside on all Catalyst 5000 series switches. These are the Enhanced Address Recognition Logic (EARL), the Network Management Processor (NMP), and the Catalyst backplane buses (the Catalyst 5505, Catalyst 5500, and Catalyst 5509 switches have three buses, the Catalyst 5000 switch has one bus). The middle and lower sections of Figure 2 show the major hardware blocks of the GEM. The switch bus interface connects the Gigabit Ethernet switch to the Catalyst 5000 series backplane buses.
From a hardware perspective, the GEM is a Gigabit Ethernet switch (the switch engine) connected to a three-port Gigabit Ethernet module (the switch bus interface) as shown in Figure 2. When the switch bus interface receives a frame from the Catalyst 5000 series bus, it strips the bus header, adds IEEE 802.1Q virtual LAN (VLAN) encoding (if applicable), and delivers the frame to the switch engine. In the reverse direction, the switch bus interface receives a frame from the switch engine, strips the 802.1Q VLAN tagging (if applicable), prefixes the Catalyst bus header, and delivers the frame to the Catalyst bus.
From a configuration and management perspective, the GEM front panel ports are connected directly to the Catalyst 5000 series bus. The NMP and GEM software provide the communication channels that direct management frames sourced by the NMP to specific front panel ports. In the reverse direction, the GEM software is able to receive management frames from the network and deliver them to the NMP tagged with the proper front panel port numbering.
The following section describes how this module differs from other Ethernet modules in the method it uses to connect to the Catalyst 5000 series buses and in its ability to perform local switching.
From an operational perspective, the GEM is an independent switch connected to the Catalyst 5000 series switch with a single trunk link carrying frames that are either untagged or in compliance with the 802.1Q VLAN tagging format. For all packets arriving on this trunk through the backplane links, as with packets arriving on the front panel ports, this module makes all forwarding decisions locally. The module maintains a host table that contains the locations of hosts adjacent to the front panel ports and to the backplane links. Additionally, all port, spanning tree, VLAN, and multicast group state information is maintained locally.
The NMP runs the protocols and configures the module as required by VLAN Trunking Protocol (VTP), Spanning-Tree Protocol, Port Aggregation Protocol (PAgP), and other management protocols.
This section describes the front panel features of the GEM (see Figure 3).
This module has one STATUS LED that provides information about the module and one link LED for each of the nine ports (see Figure 4). The LEDs are described in Table 1.
| LED | Color | Description |
|---|---|---|
STATUS
|
Green Red Orange | The Catalyst 5000 series switch performs a series of self-test diagnostics: All tests pass. A test other than an individual port test fails. System boot or diagnostic tests in progress. Module is disabled. |
Port | Green Orange Flashing orange Off | Link is operational. Link is disabled by user. Power-on self-test indicates faulty port. No signal detected, GBIC not installed, or link configuration failure. |
A GBIC (see Figure 5) is a hot-swappable input/output device that plugs into the module, linking the module with the fiber-optic network. The following GBIC media types are supported:
Other GBIC media types may be supported as additional technology becomes available.
For GBIC cabling information, refer to the "Port Cabling Specifications" section.
For GBIC installation information, refer to the "GBIC Installation, Care, and Handling" section.
Read the installation information in this section prior to installing the GEM.
Table 2 provides cabling specifications for the GBICs that you install in the GEM. Note that all GBIC ports have SC-type connectors, and the minimum cable distance for all GBICs listed (MMF and SMF) is 6.5 feet (2 meters).
| GBIC | Wavelength (nm) | Fiber Type | Core Size (micron) | Modal Bandwidth (MHz.km) | Cable Distance |
|---|---|---|---|---|---|
SX1 | 850 | MMF
| 62.5 62.5 50.0 50.0 | 160 200 400 500 | 722 ft (220 m) 902 ft (275 m) 1640 ft (500 m) 1804 ft (550 m) |
LX/LH | 1300 | MMF2
SMF (LX/LH) | 62.5 50.0 50.0
9/10 | 500 400 500
- | 1804 ft (550 m) 1804 ft (550 m) 1804 ft (550 m)
32,810 feet (10 km) |
| 1MMF only. 2Patch cord required (refer to the "Patch Cord" section for details). |
When using the LX/LH GBIC with 62.5-micron diameter multimode fiber (MMF), you must install a mode-conditioning patch cord (Cisco product no. CAB-GELX-625 or equivalent) between the GBIC and the MMF cable on both the transmit and receive ends of the link. The patch cord is required for link distances greater than 984 feet (300 meters).
Figure 6 shows a typical configuration using the patch cord.
Plug the end of the patch cord labeled "To Equipment" into the GBIC (see Figure 7). Plug the end labeled "To Cable Plant" into the patch panel. The patch cord is 9.84 feet (3 meters) long and has duplex SC male connectors at each end.
When an unconditioned laser source (LX/LH GBIC) designed for operation on SMF cable is directly coupled to an MMF cable, an effect known as Differential Mode Delay (DMD) might occur. DMD can result in a degradation of the modal bandwidth of the fiber-optic cable. This degradation causes a decrease in the link span (the distance between the transmitter and the receiver) that can be reliably supported.
The Gigabit Ethernet specification (IEEE 802.3z) outlines parameters for Ethernet communications at a gigabit-per-second rate. Its initial intent is to offer a higher-speed version of Ethernet for backbone and server connectivity using existing deployed MMF cable. To accomplish this, the specification defines the use of laser-based optical components to propagate data over MMF cable.
Lasers are specified because they function at the baud rates and longer distances required for Gigabit Ethernet. The IEEE 802.3z Gigabit Ethernet Task Force has identified the DMD condition that occurs in certain circumstances with particular combinations of lasers and MMF cable. The resulting characteristics create an additional element of "jitter" which limits the reach of Gigabit Ethernet over MMF cable.
With DMD, a single laser light pulse excites a few modes equally within an MMF cable. These modes, or light pathways, then follow two or more different paths. These paths may be of different lengths and have different transmission delays as the light travels through the cable. With DMD, a distinct pulse propagating down the cable no longer remains a distinct pulse or, in extreme cases, may become two independent pulses. Strings of pulses tend to interfere with each other making it difficult to recover data in a reliable fashion.
DMD is not experienced in all deployed fibers. It occurs with certain combinations of worst-case fibers and worst-case transceivers. Gigabit Ethernet is the first technology to experience this problem due to its very high baud rate and its long MMF cable lengths. SMF cable and copper cable are not affected by DMD.
Historically, MMF cable has only been tested for use with light-emitting diode (LED) sources. LEDs create a condition within a fiber-optic cable referred to as an overfilled launch condition. The overfilled launch condition describes the way LED transmitters couple light into the fiber-optic cable in a broad spread of modes. Similar to a light bulb radiating light into a dark room, the generated light shines in multiple directions in a way that "overfills" the existing cable space. This generated light "excites" a large number of modes (see Figure 8).
Lasers launch light in a more concentrated fashion. A laser transmitter typically couples light into only a fraction of the existing modes or optical pathways present in the fiber-optic cable (see Figure 8).
The solution to DMD is to condition the laser light launched from the source (transmitter) so it spreads the light evenly across the diameter of the fiber-optic cable making the launch look more like an LED source to the cable. The objective is to scramble the modes of light in a way that distributes the power more equally in all modes. This prevents the light from being concentrated in just a few modes. This is in contrast to an unconditioned launch, which, in the worst case, might concentrate all of its light in the center of the fiber-optic cable, thereby exciting only two or more modes equally.
There is a significant variation in the amount of DMD produced from one MMF cable to the next. Furthermore, there is no reasonable test that can be performed to survey an installed cable plant to assess the effect of DMD. Therefore, you must use the mode-conditioning patch cords for all LX/LH GBICs using MMF when the link span exceeds 984 feet (300 meters). For link spans of less than 300 meters, the patch cord can be omitted (using the LX/LH GBIC with MMF and no patch cord for very short link distances (tens of meters) is not recommended. The result could be an elevated bit error rate [BER]).
This section describes how to install, remove, and maintain GBICs.
 | Caution Unnecessary removal/insertion of a GBIC could lead to premature failure of the GBIC. GBICs have a lifetime of 100 to 500 removals/insertions. |
The GEM can be shipped with or without GBICs installed.
| Caution
When removing or inserting a GBIC, always wear an ESD wrist strap connected to the Catalyst 5000 series switch ESD wrist strap connector. |
To install a GBIC, perform these steps:
Step 1 Remove the GBIC from its protective packaging.
Step 2 Verify that the GBIC is the correct type for your network by checking the part number. The number indicates whether it is 1000Base-SX, 1000Base-LX/LH, or other.
Step 3 Grip the sides of the GBIC with your thumb and forefinger; insert the GBIC into the slot on the front of the module (see Figure 9).
Step 4 Slide the GBIC into the slot until you hear a click. The click indicates the GBIC module is locked into the slot.
| Warning Invisible laser radiation may be emitted from the aperture ports of the single-mode fiber-optic modules when no cable is connected. Avoid exposure and do not stare into open apertures. |
Step 5 When you are ready to attach the fiber-optic cable, remove the plugs from the GBIC and save for future use.
| Warning Class 1 laser product. |
To remove a GBIC, perform these steps.
Step 1 Disconnect the fiber-optic cable from the GBIC SC connector.
Step 2 Release the GBIC from the slot by simultaneously squeezing the plastic tabs (one on each side of the GBIC).
Step 3 Slide the GBIC out of the slot.
Step 4 Install the plugs in the GBIC optical bores, and place the GBIC in protective packaging.
GBIC care and handling guidelines follow:
| Warning Ultimate disposal of this product should be handled according to all national laws and regulations. |
| Warning Before working on equipment that is connected to power lines, remove jewelry (including rings, necklaces, and watches). Metal objects will heat up when connected to power and ground and can cause serious burns or weld the metal object to the terminals. |
| Warning Only trained and qualified personnel should be allowed to install or replace this equipment. |
This section describes how to install the GEM into a Catalyst 5000 series switch. Prior to installing the module, read the "Preinstallation Information" section to determine cabling requirements and, if necessary, install your GBICs.
The number of GEMs that can be installed per chassis is as follows:
To maximize GEM operation, connect the module to all three 1.2-Gb buses on the Catalyst 5505, Catalyst 5500, and Catalyst 5509 switch backplanes (the Catalyst 5000 switch has only one 1.2-Gb bus). See Table 3 for slot requirements.
Catalyst 5000 switch | Slots 3 through 5---installing the module in slots 3 through 5 provides a 1.2-Gb connection to the switch backplane. |
Catalyst 5505 switch | Slots 3 through 5---installing the module in slots 3 through 5 provides three 1.2-Gb connections to the switch backplane.1 |
Catalyst 5500 switch | Slots 3 through 5---installing the module in slots 3 through 5 provides three 1.2-Gb connections to the switch backplane.2 Slots 6 through 8---installing the module in slots 6 through 8 provides a 1.2-Gb connection to the switch backplane. Slots 9 through 12 are reserved for LightStream 1010 or Catalyst 8510 modules. |
Catalyst 5509 switch | Slots 3 through 9---installing the module in slots 3 through 9 provides three 1.2-Gb connections to the switch backplane.3 |
| 1In order to utilize all three of the 1.2-Gb buses, you must use a Supervisor Engine III. 2See note 1. 3See note 1. |
To maximize available bandwidth between the GEM and other modules in the chassis, you must use a Supervisor Engine III, which supports all three 1.2-Gb buses on the Catalyst 5505, Catalyst 5500, and Catalyst 5509 switch backplanes.
| Caution
When removing or inserting a module, always wear an ESD wrist strap connected to the Catalyst 5000 series switch ESD wrist strap connector. |
| Caution Always use the ejector levers when installing or removing modules. A module that is partially seated in the backplane causes the system to halt and subsequently crash. |
 | Tips |
Due to the GEM's two-slot design, it is more difficult to insert and remove than single slot modules. Use care when inserting and removing the GEM. Insertion and removal should be done in a single, smooth motion.
To install the GEM, perform these steps:
Step 1 The module is secured with two captive installation screws on each side. Use a 1/4-inch flat-blade screwdriver to loosen the captive installation screws; remove the module filler plates or the existing modules from the two slots you want to use.
Step 2 Guide the module into the slot, aligning the sides of the module with the guides in the lower slot (avoid touching the components on the board).
Step 3 While keeping the module oriented horizontally, carefully slide it into the slot until its front panel contacts the ejector levers (see Figure 10).
Step 4 Using the thumb and forefinger of each hand, simultaneously push the left lever and the right lever in to fully seat the module in the backplane connectors.
Step 5 Use a screwdriver to tighten the captive installation screws on the left and right sides of the module (see Figure 10).
Step 6 Check the status of the module as follows:
Step 7 If the module is not operational, reseat it. If the module is still not operational, refer to the Catalyst 5000 Series Installation Guide.
To connect to the Gigabit EtherChannel ports, perform these steps:
Step 1 Remove the plugs from the GBIC optical bores; store them for future use.
Step 2 Remove the plugs from the SC connector (see Figure 11) on the fiber-optic cable. Insert the connector into the GBIC.
| Caution Do not remove the plugs from the GBIC optical bores or the fiber-optic cable until you are ready to connect the cable. The plugs protect the GBIC optical bores and cable from contamination. |
The following configuration information is available for configuring your GEM:
This section provides configuration information for the GEM and consists of the following sections:
Follow the configuration guidelines in this section when configuring the GEM.
If the GEM is inadvertently reset during a normal software download to the module (using download host file [mod_num]), the supervisor engine will usually report that the Flash image on the module is corrupted, and further attempts to download using the normal download method will fail. The workaround is to run the following two commands on the supervisor engine in privileged mode:
Console> (enable) reload mod_num Console> (enable) download [tftp_server] [file_name] [mod_num] scp
The GEM has the same default settings as the Ethernet and Fast Ethernet modules, except for the following:
The following features are supported on the GEM:
| Caution Autonegotiation with Gigabit Ethernet behaves differently than autonegotiation with Ethernet or Fast Ethernet. To avoid link configuration problems it is important that you read and understand the set port negotiation command description in the "set port negotiation" section and the set port flowcontrol command description in the "set port flowcontrol" section. |
The following features are not supported on the GEM:
The following commands are not supported on the GEM:
This section lists the Catalyst 5000 series switch CLI commands that have been created to support 1000Base-X switching. These commands were introduced in Catalyst 5000 series supervisor engine software release 4.1 to support the three-port Gigabit Ethernet module (WS-X5403).
Ports can be characterized by their ability to generate and respond to flow control frames (PAUSE frames) as well as the pause behavior they require of their link partner.
The module's GBIC ports respond to received pause frames. Upon configuration, these ports advertise a pause capability. The ports at the two ends of the link negotiate a mutually acceptable flow-control configuration.
A pause frame is never generated by the module's GBIC ports.
In all cases, pause frames received on the module are processed internally and are not switched through the system.
Enter the set port flowcontrol command to set the receive flow-control value for a particular port:
set port flowcontrol {receive | send} [mod_num/port_num] {off | on | desired}
The default for receive is off, the default for send is desired, and the user-selected values are off, on, or desired.
Perform one or more of these tasks to administer flow control:
| Task | Command |
|---|---|
Tell a local port to advertise that it will send flow-control frames.1 | set port flowcontrol send mod_num/port_num on |
Tell a local port to advertise that it will send flow-control frames if the attached device elects to receive them.2 | set port flowcontrol send mod_num/port_num desired |
Tell a local port to advertise that it will never send flow-control frames. | set port flowcontrol send mod_num/port_num off |
Allow a local port to operate with an attached device that is required to send flow-control packets, or with an attached device that is not required to but may send flow-control packets. | set port flowcontrol receive mod_num/port_num on |
Same as task description for set port flowcontrol receive mod_num/port_num on. | set port flowcontrol receive mod_num/port_num desired |
Tell a local port to discard received flow-control frames without processing them. If negotiation is enabled, the local port advertises that it refuses to accept flow-control frames. If negotiation is enabled, this can cause the attached device not to send flow-control frames, or if the attached device is configured to send flow-control frames unconditionally, it can prevent the link from coming up. | set port flowcontrol receive mod_num/port_num off |
After entering this command, you see these displays:
Console> (enable) set port flowcontrol send 5/1 on Port 5/1 flow control send administration status set to on (port will send flowcontrol to far end) Console> (enable) Console> (enable) set port flowcontrol send 5/1 desired Port 5/1 flow control send administration status set to desired (port will send flowcontrol to far end if far end supports it) Console> (enable) Console> (enable) set port flowcontrol send 5/1 off Port 5/1 flow control send administration status set to off (port will not send flowcontrol to far end) Console> (enable) Console> (enable) set port flowcontrol receive 5/1 on Port 5/1 flow control receive administration status set to on (port will require far end to send flowcontrol) Console> (enable) Console> (enable) set port flowcontrol receive 5/1 desired Port 5/1 flow control receive administration status set to desired (port will allow far end to send flowcontrol if far end supports it) Console> (enable) Console> (enable) set port flowcontrol receive 5/1 off Port 5/1 flow control receive administration status set to off (port will not allow far end to send flowcontrol) Console> (enable)
To display the current flow-control status and statistics, perform this task:
| Task | Command |
|---|---|
Display the current flow-control status and statistics. | show port flowcontrol |
After entering this command, you see this display:
Console> (enable) show port flowcontrol
Port Send FlowControl Receive FlowControl RxPause TxPause Unsupported
admin oper admin oper opcodes
----- -------- -------- -------- -------- ------- ------- -----------
5/1 desired off off off 0 0 0
5/2 desired off off off 0 0 0
5/3 desired off off off 0 0 0
.
.
Console> (enable)
Output field descriptions are as follows:
| Field | Description |
|---|---|
Port | Module and port number. |
Send-Flowcontrol- | Flow-control administrative state. Possible settings: on indicates the local port sends flow control to the far end; off indicates the local port does not send flow control to the far end; desired indicates the local end sends flow control to the far end if the far end supports it. |
Send-Flowcontrol- | Flow-control operation. Possible indications: disagree indicates the two ports could not agree on a link protocol. |
Receive-Flowcntl- | Flow-control administrative state. Possible settings: on indicates the local port requires the far end to send flow control; off indicates the local port does not allow the far end to send flow control; desired indicates the local end allows the far end to send flow control. |
Receive-Flowcntl- | Flow-control operation. Possible indications: disagree indicates the two ports could not agree on a link protocol. |
RxPause | Count of pause frames received. |
TxPause | Count of pause frames transmitted. |
Unsupported opcodes | Count of pause frames with unsupported opcodes. These are frames with a valid destination address (01:80:c2:00:00:01) and a valid Ethernet type (0x8808) but an invalid opcode. Currently we recognize pause frames with an opcode of 1 (Xoff). All others are unsupported. Note that all frames received at this multicast address are discarded by the bridge. |
| Caution Autonegotiation with Gigabit Ethernet behaves differently than autonegotiation with Ethernet or Fast Ethernet. To avoid link configuration problems it is important that you read and understand the information in this section and the information in the "set port flowcontrol" section. |
Unlike 10/100 ports, autonegotiation with Gigabit Ethernet does not involve negotiating port speed. You cannot disable autonegotiation by setting the port speed and duplex state. In Gigabit Ethernet, the link negotiation protocol is used to exchange flow-control behavior, remote fault information, and duplex information (even though the Catalyst 5000 series Gigabit ports only support full-duplex operation). In Gigabit Ethernet, the only way to control whether the link negotiation protocol runs is with the set port negotiation command.
Table 4 shows the four possible autonegotiation configurations for a link and the resulting link status for each configuration.
| Autonegotiation State | Link Status | ||
|---|---|---|---|
| Near End1 | Far End2 | Near End | Far End |
Off | Off | Up | Up |
Off | On | Up | Down |
On | Off | Down | Up |
On | On | Up | Up |
| 1Near End refers to the GEM's front panel port. 2Far End refers to the remote port at the other end of the Gigabit link. |
Both ends of the link must have the same setting. The link will not come up if the two ends of the link are set inconsistently (link negotiation enabled on one end and disabled on the other).
The default is link negotiation protocol enabled. To enable or disable the link negotiation protocol on the specified port, perform this task:
| Task | Command |
|---|---|
Enable or disable link negotiation protocol. | set port negotiation mod_num/port_num {enable | disable} |
After entering this command, you see this display:
Console> (enable) set port negotiation 5/1 disable Port 5/1 negotiation disabled. Console> (enable)
The default is link negotiation protocol enabled. To display the link negotiation protocol setting for the specified port, perform this task:
| Task | Command |
|---|---|
Display the link negotiation protocol setting. | show port negotiation [mod_num/port_num] |
After entering this command, you see this display:
Console> (enable) show port negotiation 5/1 Port Link Negotiation ------- -------------------- 5/1 disabled Console> (enable)
Configuring the GEM is similar to configuring Ethernet or Fast Ethernet modules. This section describes those commands whose implementation has changed to support the GEM.
The following show port commands have been modified to include GEM information:
Refer to the Catalyst 5000 Series Command Reference publication for complete command descriptions.
Enter the show port broadcast command to display statistics related to broadcast suppression on the Catalyst 5000 series backplane.
On the GEM, when you specify a port for broadcast suppression, the traffic is suppressed only in the network-to-Catalyst 5000 series switch bus direction.
After entering this command you see this example:
Console> (enable) show port broadcast Port Broadcast-Limit Broadcast-Drop -------- --------------- -------------- 1/1 - 0 1/2 - 0 5/1 - 0 5/2 - 0 5/3 - 0 . . Console> (enable)
Enter the set port broadcast command to set the broadcast/multicast suppression for one or more ports. This command sets the broadcast threshold that limits the backplane traffic received from the module. The default value for the threshold is 100 percent.
Although bandwidth-based broadcast/multicast suppression applies to all ports on a module, you must still specify a port number according to the syntax rules of the set port broadcast mod_num/port_num threshold [%] command. For example, if you specify port 3 on module 4 (4/3), broadcast/multicast suppression is applied to every port on module 4. You can specify any port number between 1 and 9.
On the GEM, broadcast suppression takes place at the interface to the Catalyst switch backplane. This command only limits the amount of broadcast/multicast traffic that the module can send to the switch backplane for forwarding to other modules. Traffic switched locally on the module (traffic between front panel ports) is never suppressed. Traffic from the backplane to the front panel ports is also never suppressed.
When the module is plugged into a chassis slot with three backplane links, the command limits the broadcast/multicast traffic the module can send to each of the three backplane links to the percentage specified in the command. The limit is enforced independently on each backplane link, and is not an aggregate limit for all three links.
Therefore if a limit of 5 percent has been set, depending on whether the multicast traffic is entirely from one host to one multicast destination address on a single VLAN (in which case all of the traffic will be on one backplane), or whether the traffic is from a variety of hosts to a variety of destination addresses (in which case the traffic will be spread across all three backplane links), the aggregate traffic limit will range from 5 Mbps (one backplane link active) to 15 Mbps (three backplane links active, with each limited to 5 Mbps).
In the following example, GEMs occupy slots 4 and 6:
Console> (enable) set port broadcast 4/1,6/1 70% Port(s) 4/1-9,6/1-9 broadcast traffic limited to 70%. Suppression of broadcast traffic on Module(s) 4,6 does not apply to locally switched traffic on the module(s). Console> (enable)
Enter the set span command to enable or disable SPAN and to set up the port and VLAN analyzer. The default configuration has port monitoring disabled, port 1/1 as the monitoring port (destination), VLAN 1 as the monitored VLAN (source), and both transmit and receive packets monitored. If the parameter rx, tx, or both is not specified, the default is both. The command syntax is as follows:
set span enable
set span disable
set span {src_mod/src_ports} {dest_mod/dest_port} [rx | tx | both]
set span {src_vlan} {dest_mod/dest_port} [rx | tx | both]
On the GEM, the specified source and destination ports must be on the same module or you get an error indication as shown in the following display:
Console> (enable) set span enable Console> (enable) set span 3/1 4/1 both Source port 3/1 is a gigabit module port. Source port and destination port must be on the same module. Console> (enable)
On the GEM, the specified VLAN must be active on the module or you get an error indication as shown in the following display:
Console> (enable) set span 999 3/2 both Destination port 3/2 is a gigabit module port. Vlan 999 is not active on module 3. Gigabit module ports can only monitor vlans that are active on the module. Console> (enable)
This example shows a successful enabling of SPAN for the monitoring of transmit traffic on port 3/3 through port 3/4:
Console> (enable) set span enable span enabled. Console> (enable) Console> (enable) set span 3/3 3/4 tx Enabled monitoring of ports 3/3 transmit traffic by ports 3/4. Console> (enable)
Enter the set trunk command to configure trunk ports and to add VLANs to the allowed VLAN list for existing trunks. The dot1q syntax specifies an IEEE 802.1Q trunk. The command syntax is as follows:
set trunk mod_num/port_num [on | off | desirable | auto | nonegotiate] [vlan_range] [isl | dot1q | dot10 | lane]
On the GEM, you can configure a trunk only as an 802.1Q trunk.
In the following example, port 5 on module 5 is set as an 802.1Q trunk port:
Console> (enable) set trunk 5/5 nonegotiate dot1q Port(s) 5/5 trunk mode set to nonegotiate. Port(s) 5/5 trunk type set to dot1q. Console> (enable)
Enter the show counters command to show all counters for a port. For the GEM, the command has been modified to include counters for the following:
Configuring Gigabit EtherChannel is very similar to configuring Fast EtherChannel on 100-Mbps modules.
Gigabit EtherChannel provides parallel bandwidth of up to 8 Gbps (4 Gbps full duplex) between a Catalyst 5000 series switch and another switch or host by grouping multiple Gigabit Ethernet interfaces into a single logical transmission path.
You can configure Gigabit Ethernet ports into Gigabit EtherChannel groups containing two or four segments, yielding 4- or 8-Gbps bandwidth (2- or 4-Gbps in each direction).
Inbound broadcast and multicast packets on one segment in a channel are blocked from returning on any other segment of the channel. Outbound broadcast and multicast packets are sent through only one channel segment.
If a segment within a channel fails, traffic previously carried over the failed link switches to the remaining segments within the channel. A trap is sent upon a failure identifying the switch, the channel, and the failed link.
Channels are configured using the standard CLI or SNMP.
Catalyst 5000 series software includes an enhancement to the Gigabit EtherChannel feature called the Port Aggregation Protocol (PAgP). PAgP facilitates the automatic creation of Gigabit EtherChannel links by sending packets between Gigabit EtherChannel-capable ports. The protocol learns and informs the neighbors and their group capabilities dynamically. Once PAgP identifies correctly paired Gigabit EtherChannel links, it groups the ports into a channel. The channel is then added to the spanning tree as a single bridge port.
PAgP includes four user-configurable channel modes: on, off, auto, and desirable. Each mode affects the way a port handles PAgP packets. By default, ports are in auto mode. Table 5 describes each mode.
| Mode | Description |
|---|---|
On | Forces the port to channel without negotiation. |
Off | Prevents the port from channeling without negotiation. |
Auto | Places a port into a passive negotiating state; the port responds to received PAgP packets but does not initiate PAgP packet negotiation. (Default) |
Desirable | Places a port into an active negotiating state, in which the port initiates negotiations with other ports by sending PAgP packets. |
Both the auto and desirable modes allow ports to negotiate with connected ports to determine if they can form a channel, based on criteria such as port speed, trunking state, VLAN numbers, and so on.
Channel ports can be in different channel modes as long as the modes are compatible. For instance, a port in desirable mode can form a channel successfully with another port that is in desirable or auto mode. Similarly, a port in auto mode can form a channel with another port in desirable mode. However, a port in auto mode cannot form a channel with another port that is also in auto mode, since neither port will initiate the needed negotiations.
If improperly configured, some Gigabit EtherChannel ports are disabled automatically to avoid network loops and other problems. Use the following guidelines to avoid configuration problems:
| Acceptable Channel Segments | ||||||
|---|---|---|---|---|---|---|
1 + 2 | 3 + 4 | 1 + 2 + 3 + 4 | 5 + 6 | 7 + 8 | 5 + 6 + 7 + 8 |
|
Use the set port channel port_list {on | off | auto | desirable} command to configure Gigabit EtherChannel.
To configure a channel, complete these steps:
| Task | Command |
|---|---|
Step 1 Make sure that the ports you want to channel are configured correctly. (Refer to the "Gigabit EtherChannel Configuration Guidelines" section.) |
|
Step 2 Ensure a loop-free topology for all channeled VLANs. |
|
Step 3 Create a channel. | set port channel port_list on |
To set a channel port to auto mode, enter this command in privileged mode:
| Task | Command |
|---|---|
Set a channel port to auto mode. | set port channel port_list auto |
To set a channel port to desirable mode, enter this command in privileged mode:
| Task | Command |
|---|---|
Set a channel port to desirable mode. | set port channel port_list desirable |
To remove a channel, enter this command in privileged mode:
| Task | Command |
|---|---|
Remove a channel. | set port channel port_list off |
This example shows how to enable Gigabit EtherChannel on ports 5 through 8 of the GEM in slot 3:
Console> (enable) set port channel 3/5-8 on Port(s) 3/5-8 channel mode set to on. Console> (enable)
To verify that the Gigabit EtherChannel is configured correctly, enter the show port channel [mod_num[port_num]] command.
When using the channel as a trunk, follow these guidelines:
Table 7 lists standards compliance specifications.
| ||
Cisco Connection Online (CCO) is Cisco Systems' primary, real-time support channel. Maintenance customers and partners can self-register on CCO to obtain additional information and services.
Available 24 hours a day, 7 days a week, CCO provides a wealth of standard and value-added services to Cisco's customers and business partners. CCO services include product information, product documentation, software updates, release notes, technical tips, the Bug Navigator, configuration notes, brochures, descriptions of service offerings, and download access to public and authorized files.
CCO serves a wide variety of users through two interfaces that are updated and enhanced simultaneously: a character-based version and a multimedia version that resides on the World Wide Web (WWW). The character-based CCO supports Zmodem, Kermit, Xmodem, FTP, and Internet e-mail, and it is excellent for quick access to information over lower bandwidths. The WWW version of CCO provides richly formatted documents with photographs, figures, graphics, and video, as well as hyperlinks to related information.
You can access CCO in the following ways:
For a copy of CCO's Frequently Asked Questions (FAQ), contact cco-help@cisco.com. For additional information, contact cco-team@cisco.com.
|
|