You can upload/download firmware to or from a TFTP server. By saving runtime code to a file on a TFTP server, that file can later be downloaded to the switch to restore operation. You can also set the switch to use new firmware without overwriting the previous version.
TFTP Server IP Address The IP address of a TFTP server. Destination File Name File names are case-sensitive, should be from 1 to 31 characters, not contain slashes (\ or /), and the leading letter of the file name should not be a period (.). (Valid characters: A-Z, a-z, 0-9,., -, _)
Note: The maximum number of runtime code files is 2.
Downloading System Software from a Server
When downloading runtime code, you can specify the Destination File Name to replace the current image, or first download the file using a different name from the current runtime code file, and then set the new file as the startup file. Web Click System/Firmware. Enter the IP address of the TFTP server, enter the file name of the software to download, select a file on the switch to overwrite or specify a new file name, then click Transfer from Server.

MANAGING FIRMWARE

If you download specifying a new destination file name, select the new file from the drop-down box for the operation code used at startup, and then click Apply Changes. To start the new firmware, reboot the system.
CLI Enter the IP address of the TFTP server, select config or opcode file type, then enter the source and destination file names, set the new file to start up the system, and then restart the switch.
Console#copy tftp file TFTP server ip address: 10.1.0.99 Choose file type: 1. config: 2. opcode: <1-2>: 2 Source file name: v10.bix Destination file name: V10000 / Console#config Console(config)#boot system opcode: V10000 Console(config)#exit Console#reload 4-18

4-24 4-15

CONFIGURING THE SWITCH Saving or Restoring Configuration Settings
You can upload/download configuration settings to/from a TFTP server. The configuration file can be later downloaded to restore the switchs settings.
Note: The maximum number of user-defined configuration files is limited only by available Flash memory space. You can save the configuration file under a new file name and then set it as the startup file, or you can specify the current startup configuration file as the destination file to directly replace it. Note that the file Factory_Default_Config.cfg can be copied to the TFTP server, but cannot be used as a destination file name on the switch. Web Click System/Configuration. Enter the IP address of the TFTP server, enter the name of the file to download, select a file on the switch to overwrite or specify a new file name, and then click Transfer from Server.

2003 Foundry Networks, Inc. 3-57
- Range Ethernet: 200,000-20,000,000 Fast Ethernet: 20,000-2,000,000 Gigabit Ethernet: 2,000-200,000 - Defaults Ethernet half duplex: 2,000,000; full duplex: 1,000,000; trunk: 500,000 Fast Ethernet half duplex: 200,000; full duplex: 100,000; trunk: 50,000 Gigabit Ethernet full duplex: 10,000; trunk: 5,000 Note: When the Path Cost Method is set to short, the maximum path cost is 65,535. Admin Link Type The link type attached to this interface. (Default: Auto) - Point-to-Point A connection to exactly one other bridge. - Shared A connection to two or more bridges. - Auto The switch automatically determines if the interface is attached to a point-to-point link or to shared media. Admin Edge Port You can enable this option if an interface is attached to a LAN segment that is at the end of a bridged LAN or to an end node. Since end nodes cannot cause forwarding loops, they can pass directly through to the spanning tree forwarding state. Specifying Edge Ports provides quicker convergence for devices such as workstations or servers, retains the current forwarding database to reduce the amount of frame flooding required to rebuild address tables during reconfiguration events, does not cause the Spanning Tree to initiate reconfiguration when the interface changes state, and also overcomes other STP-related timeout problems. However, remember that Edge Port should only be enabled for ports connected to an end-node device. Migration Re-checks the appropriate BPDU format to send on the selected interface. If at any time the switch detects STP BPDUs,
including Configuration or Topology Change Notification BPDUs, it will automatically set the selected interface to forced STP-compatible mode. However, you can also check this Migration check box to manually re-check the appropriate BPDU format (RSTP or STP-compatible) to send on the selected interfaces. Trunk Specifies if a port is member of a trunk.
Web Click Spanning Tree/STP Port Configuration or STP Trunk Configuration. Modify the required attributes, then click Apply.

2003 Foundry Networks, Inc. 3-73
rejoining the group. (Range: 500-18000 centiseconds; Default: 1000) * GARP timer settings must follow this rule: 2 x (join timer) < leave timer < leaveAll timer Mode Indicates VLAN membership mode for a port. (Configure via CLI, see page 4-126.) - 1Q Trunk Specifies a port as an end-point for a VLAN trunk. A trunk is a direct link between two switches, so the port transmits tagged frames that identify the source VLAN. However, note that frames belonging to the ports default VLAN (i.e., associated with the PVID) are sent untagged. - Hybrid Specifies a hybrid VLAN interface. The port may receive or transmit tagged or untagged frames. Trunk Member Indicates if a port is a member of a trunk. To add a trunk to the selected VLAN, use the last table on the VLAN Static Table page.
Web Click VLAN/VLAN Port Configuration or VLAN Trunk Configuration. Fill in the required settings for each interface, click Apply.
CLI This example sets port 1 to accept only tagged frames, assigns PVID 3 as the native VLAN ID, enables GVRP, sets the GARP timers, and then sets the switchport mode to hybrid.
Console(config)#interface ethernet 1/1 Console(config-if)#switchport acceptable-frame-types tagged Console(config-if)#switchport ingress-filtering Console(config-if)#switchport native vlan 3 Console(config-if)#switchport gvrp Console(config-if)#garp timer join 10 Console(config-if)#garp timer leave 90 Console(config-if)#garp timer leaveall 2000 Console(config-if)#switchport mode hybrid Console(config-if)# 4-86 4-127 4-128 4-129 4-133 4-135 4-135 4-135 4-126
Class of Service Configuration
Class of Service (CoS) allows you to specify which data packets have greater precedence when traffic is buffered in the switch due to congestion. This switch supports CoS with four priority queues for each port. Data packets in a ports high-priority queue will be transmitted before those in the lower-priority queues. You can set the default priority for each interface, and configure the mapping of frame priority tags to the switchs priority queues.
Setting the Default Priority for Interfaces
You can specify the default port priority for each interface on the switch. All untagged packets entering the switch are tagged with the specified default port priority, and then sorted into the appropriate priority queue at the output port.
This switch provides four priority queues for each port. It uses Weighted Round Robin to prevent head-of-queue blockage. The default priority applies if the incoming frame is an untagged frame received from a VLAN trunk or a static-access port. This priority does not apply to IEEE 802.1Q VLAN tagged frames. If the incoming frame is an IEEE 802.1Q VLAN tagged frame, the IEEE 802.1p User Priority bits will be used. If the output port is an untagged member of the associated VLAN, these frames are stripped of all VLAN tags prior to transmission.

Default Priority The priority that is assigned to untagged frames received on the specified port. (Range: 0 - 7, Default: 0)
CLASS OF SERVICE CONFIGURATION
Number of Egress Traffic Classes The number of queue buffers provided for each port.
Web Click Priority/Default Port Priority or Default Trunk Priority. Modify the default priority for any interface, then click Apply.
CLI This example assigns a default priority or 5 to port 3.
Console(config)#interface ethernet 1/3 Console(config-if)#switchport priority default 5 4-86 4-151
Mapping CoS Values to Egress Queues
This switch processes Class of Service (CoS) priority tagged traffic by using four priority queues for each port, with service schedules based on Weighted Round Robin (WRR). Up to eight separate traffic priorities are defined in IEEE 802.1p. The default priority levels are assigned according to recommendations in the IEEE 802.1p standard as shown in the following table.

Queue 2 Priority 3

The priority levels recommended in the IEEE 802.1p standard for various network applications are shown in the following table. However, you can map the priority levels to the switchs output queues in any way that benefits application traffic for your own network.
Priority Level 0 (default) 7 Traffic Type Background (Spare) Best Effort Excellent Effort Controlled Load Video, less than 100 milliseconds latency and jitter Voice, less than 10 milliseconds latency and jitter Network Control
Priority CoS value. (Range: 0 to 7, where 7 is the highest priority) Traffic Class Output queue buffer. (Range: 0 - 3, where 3 is the highest CoS priority queue)
Web Click Priority/Traffic Classes. Assign priorities to the output queues, then click Apply.
CLI The following example shows how to map CoS values 0, 1 and 2 to CoS priority queue 0, value 3 to CoS priority queue 1, values 4 and 5 to CoS priority queue 2, and values 6 and 7 to CoS priority queue 3.
Console(config)#interface ethernet 1/1 Console(config)#queue cos-map Console(config)#queue cos-map Console(config)#queue cos-map 5 Console(config)#queue cos-map 7 Console(config)#exit Console#show queue cos-map ethernet 1/1 Information of Eth 1/1 Queue ID Traffic class -------- ------------Console# 4-86 4-153
CONFIGURING THE SWITCH Setting the Service Weight for Traffic Classes
This switch uses the Weighted Round Robin (WRR) algorithm to determine the frequency at which it services each priority queue. As described in Mapping CoS Values to Egress Queues on page 3-77, the traffic classes are mapped to one of the four egress queues provided for each port. You can assign a weight to each of these queues (and thereby to the corresponding traffic priorities). This weight sets the frequency at which each queue will be polled for service, and subsequently affects the response time for software applications assigned a specific priority value.

CLI This example adds a trap manager and enables authentication traps.
Console(config)#snmp-server host 10.1.19.23 batman Console(config)#snmp-server enable traps authentication 4-62 4-63

SNMP IP Filtering

The switch allows you to create a list of up to 16 IP addresses or IP address groups that are allowed access to the switch via SNMP management software. IP addresses that are permitted SNMP access are specified by an IP address together with a subnet mask that identifies the range of valid addresses. For example:
IP address 192.168.1.1 and mask 255.255.255.0 Specifies a valid IP address group from 192.168.1.0 to 192.168.1.255. IP address 192.168.1.1 and mask 255.255.255.255 Specifies a valid IP address of 192.168.1.1 only. Note: IP filtering does not affect management access to the switch using the Web interface or Telnet.
IP Filter List Displays a list of the IP address/subnet mask entries currently configured for SNMP access. IP address Specifies a new IP address to add to the IP Filter List. Subnet Mask Specifies a single IP address or group of addresses. If the IP is the address of a single management station, the mask should be set to 255.255.255.255. Otherwise, the IP address group is specified by the mask.
Note: The default setting is null, which allows all IP groups SNMP access to the switch. If one IP address is configured, the IP filtering is enabled and only addresses in the IP group will have SNMP access. Web Click SNMP/SNMP IP Filtering. To add an IP address, type the new IP address in the IP Address box, type the appropriate subnet mask in the Subnet Mask box, and then click Add IP Filtering Entry. To delete an IP address, click the entry in the P Filter List, and then click Remove IP Filtering Entry.
CLI The following is an example of configuring an SNMP IP filter.
Console(config)#snmp ip filter 10.1.2.3 255.255.255.255 Console(config)# 4-65

Multicast Configuration

Multicasting is used to support real-time applications such as video conferencing or streaming audio. A multicast server does not have to establish a separate connection with each client. It merely broadcasts its service to the network, and any hosts that want to receive the multicast register with their local multicast switch/router. Although this approach reduces the network overhead required by a multicast server, the broadcast traffic must be carefully pruned at every multicast switch/router it passes through to ensure that traffic is only passed on the hosts which subscribed to this service.

Console(config)#tacacs-server key green Console(config)#

show tacacs-server

Use this command to display the current settings for the TACACS+ server. Default Setting None Command Mode Global Configuration Example
Console#show tacacs-server Remote TACACS server configuration: Server IP address: 10.11.12.13 Communication key with radius server: Server port number: 49 Console#

SNMP COMMANDS

SNMP Commands
Controls access to this switch from SNMP management stations, as well as the error types sent to trap managers.
Command snmp-server community Function Sets up the community access string to permit access to SNMP commands Mode Page GC GC GC GC GC GC 4-59 4-60 4-61 4-62 4-63 4-65
snmp-server contact Sets the system contact string snmp-server location Sets the system location string snmp-server host snmp-server enable traps snmp ip filter Specifies the recipient of an SNMP notification operation Enables the device to send SNMP notifications (i.e., SNMP traps) Sets IP addresses of clients allowed to management access to the switch via SNMP. Displays the status of SNMP communications

show snmp

NE, PE

snmp-server community

Use this command to define the community access string for the Simple Network Management Protocol. Use the no form to remove the specified community string. Syntax snmp-server community string [ro|rw] no snmp-server community string string - Community string that acts like a password and permits access to the SNMP protocol. (Maximum length: 32 characters, case sensitive; Maximum number of strings: 5) ro - Specifies read-only access. Authorized management stations are only able to retrieve MIB objects.
rw - Specifies read-write access. Authorized management stations are able to both retrieve and modify MIB objects. Default Setting public - Read-only access. Authorized management stations are only able to retrieve MIB objects. private - Read-write access. Authorized management stations are able to both retrieve and modify MIB objects. Command Mode Global Configuration Command Usage The first snmp-server community command you enter enables SNMP (SNMP v1 and v2c). The no snmp-server community command disables all versions of SNMP. Example
Console(config)#snmp-server community alpha rw Console(config)#

snmp-server contact

Use this command to set the system contact string. Use the no form to remove the system contact information. Syntax snmp-server contact string no snmp-server contact string - String that describes the system contact information. (Maximum length: 255 characters) Default Setting None

Console(config)#interface ethernet 1/5 Console(config-if)#switchport broadcast packet-rate 64 Console(config-if)#
COMMAND LINE INTERFACE port security
Use this command to enable and configure port security on a port. Use the no form to disable port security or reset the intrusion action to the default. Syntax port security [action trap-and-shutdown] no port security [action] action - Indicates the security action to be taken when a port security violation is detected (applies globally to all ports). - trap-and-shutdown - Issue an SNMP trap message and disable the port. Default Setting Status: Disabled Action: None Command Mode Interface Configuration (Ethernet) Command Usage If you enable port security, the switch will stop dynamically learning new addresses on the specified port. Only incoming traffic with source addresses already stored in the dynamic or static address table will be accepted. To use port security, first allow the switch to dynamically learn the <source MAC address, VLAN> pair for frames received on a port for an initial training period, and then enable port security to stop address learning. Be sure you enable the learning function long enough to ensure that all valid VLAN members have been registered on the selected port. To add new VLAN members at a later time, you can manually add secure addresses with the mac-address-table static command, or turn off port security to reenable the learning function long enough for new VLAN members to be registered. Learning may then be disabled again, if desired, for security.
A secure port has the following restrictions: - Cannot be connected to a network interconnection device. - Cannot be a member of a static trunk. If a port is disabled due to a security violation, it must be manually re-enabled by using the no shutdown command. Although the port security action command is an Interface Configuration command, it applies globally to all switch ports. Example This example sets the port security action for the switch and enables port security for port 5.

clear counters

Use this command to clear statistics on an interface. Syntax

clear counters interface

interface ethernet unit/port - unit - This is device 1. - port - Port number. port-channel channel-id (Range: 1-6) Default Setting None Command Mode Privileged Exec
Command Usage Statistics are only initialized for a power reset. This command sets the base value for displayed statistics to zero for the current management session. However, if you log out and back into the management interface, the statistics displayed will show the absolute value accumulated since the last power reset. Example The following example clears statistics on port 5.
Console#clear counters ethernet 1/5 Console#

show interfaces status

INTERFACE COMMANDS show interfaces switchport
Use this command to display advanced interface configuration settings. Syntax show interfaces switchport [interface] interface ethernet unit/port - unit - This is device 1. - port - Port number. port-channel channel-id (Range: 1-6) Default Setting Shows all interfaces. Command Mode Normal Exec, Privileged Exec Command Usage If no interface is specified, information on all interfaces is displayed. The items displayed by this command include: Broadcast threshold Shows if broadcast storm suppression is enabled or disabled; if enabled it also shows the threshold level (page 4-93). Lacp status Shows if Link Aggregation Control Protocol has been enabled or disabled (page 4-168). VLAN membership mode Indicates membership mode as Trunk or Hybrid (page 4-126). Ingress rule Shows if ingress filtering is enabled or disabled (page 4-128). Acceptable frame type Shows if acceptable VLAN frames include all types or tagged frames only (page 4-127). Native VLAN Indicates the default Port VLAN ID (page 4-129).
Priority for untagged traffic Indicates the default priority for untagged frames (page 4-150). Gvrp status Shows if GARP VLAN Registration Protocol is enabled or disabled (page 4-133). Allowed Vlan Shows the VLANs this interface has joined, where (u) indicates untagged and (t) indicates tagged (page 4-130). Forbidden Vlan Shows the VLANs this interface can not dynamically join via GVRP (page 4-131). Example This example shows the configuration setting for Ethernet port 15.
Console#show interfaces switchport ethernet 1/15 Information of Eth 1/15 broadcast threshold: Enabled, 256 packets/second Lacp status: Disabled VLAN membership mode: Hybrid Ingress rule: Disabled Acceptable frame type: All frames Native VLAN: 1 Priority for untagged traffic: 0 Gvrp status: Enabled Allowed Vlan: 1(u), Forbidden Vlan: Console#

Address Table Commands

These commands are used to configure the address table for filtering specified addresses, displaying current entries, clearing the table, or setting the aging time.
Command mac-address-table static show mac-address-table Function Mode Page 4-101 4-102
Maps a static address to a port in a VLAN GC Displays entries in the forwarding database PE

ADDRESS TABLE COMMANDS

Command clear mac-address-table dynamic mac-address-table aging-time show mac-address-table aging-time Function Removes any learned entries from the forwarding database Sets the aging time of the address table Showsthe aging time for the address table Mode Page PE 4-104

COMMAND LINE INTERFACE switchport allowed vlan
Use this command to configure VLAN groups on the selected interface. Use the no form to restore the default. Syntax switchport allowed vlan {add vlan-list [tagged | untagged] | remove vlan-list} no switchport allowed vlan add vlan-list - List of VLAN identifiers to add. remove vlan-list - List of VLAN identifiers to remove. vlan-list - Separate nonconsecutive VLAN identifiers with a comma and no spaces; use a hyphen to designate a range of IDs. Do not enter leading zeros. (Range: 1-4094) Default Setting All ports are assigned to VLAN 1 by default. The default frame type is untagged. Command Mode Interface Configuration (Ethernet, Port Channel) Command Usage If switchport mode is set to trunk, then you can only assign an interface to VLAN groups as a tagged member. Frames are always tagged within the switch. The tagged/untagged parameter used when adding a VLAN to an interface tells the switch whether to keep or remove the tag from a frame on egress. If none of the intermediate network devices nor the host at the other end of the connection supports VLANs, the interface should be added to these VLANs as an untagged member. Otherwise, it is only necessary to add at most one VLAN as untagged, and this should correspond to the native VLAN for the interface. If a VLAN on the forbidden list for an interface is manually added to that interface, the VLAN is automatically removed from the forbidden list for that interface.
Example The following example shows how to add VLANs 2, 5 and 6 to the allowed list as tagged VLANs for port 1:
Console(config)#interface ethernet 1/1 Console(config-if)#switchport allowed vlan add 2,5,6 tagged Console(config-if)#
switchport forbidden vlan
Use this command to configure forbidden VLANs. Use the no form to remove the list of forbidden VLANs. Syntax switchport forbidden vlan {add vlan | remove vlan} no switchport forbidden vlan add vlan - VLAN ID to add. remove vlan - VLAN ID to remove. Do not enter leading zeroes. (Range: 1-4094) Default Setting No VLANs are included in the forbidden list. Command Mode Interface Configuration (Ethernet, Port Channel) Command Usage This command prevents a VLAN from being automatically added to the specified interface via GVRP. If a VLAN has been added to the set of allowed VLANs for an interface, then you cannot add it to the set of forbidden VLANs for that same interface.
Example The following example shows how to prevent port 1 from being added to VLAN 3:
Console(config)#interface ethernet 1/1 Console(config-if)#switchport forbidden vlan add 3 Console(config-if)#

show vlan

Use this command to show VLAN information. Syntax show vlan [id vlan-id | name vlan-name] id - Keyword to be followed by the VLAN ID. - vlan-id - ID of the configured VLAN. (Range: 1-4094, no leading zeroes) name - Keyword to be followed by the VLAN name. - vlan-name - ASCII string from 1 to 32 characters. Default Setting Shows all VLANs. Command Mode Normal Exec, Privileged Exec Example The following example shows how to display information for VLAN 1:

Layer 3 and 4 Priority Commands
Enables IP DSCP class of service mapping GC IC PE PE
switchport priority default
Use this command to set a priority for incoming untagged frames, or the priority of frames received by the device connected to the specified interface. Use the no form to restore the default value. Syntax switchport priority default default-priority-id no switchport priority default default-priority-id - The priority number for untagged ingress traffic. The priority is a number from 0 to 7. Seven is the highest priority. Default Setting The priority is not set, and the default value for untagged frames received on the interface is zero. Command Mode Interface Configuration (Ethernet, Port Channel) Command Usage The precedence for priority mapping is IP Precedence or IP DSCP, and default switchport priority.
The default priority applies for an untagged frame received on a port set to accept all frame types (i.e, receives both untagged and tagged frames). This priority does not apply to IEEE 802.1Q VLAN tagged frames. If the incoming frame is an IEEE 802.1Q VLAN tagged frame, the IEEE 802.1p User Priority bits will be used. This switch provides four priority queues for each port. It is configured to use Weighted Round Robin, which can be viewed with the queue bandwidth command. Inbound frames that do not have VLAN tags are tagged with the input ports default ingress user priority, and then placed in the appropriate priority queue at the output port. The default priority for all ingress ports is zero. Therefore, any inbound frames that do not have priority tags will be placed in queue 0 of the output port. (Note that if the output port is an untagged member of the associated VLAN, these frames are stripped of all VLAN tags prior to transmission.) Example The following example shows how to set a default priority on port 3 to 5:
Console(config)#interface ethernet 1/3 Console(config-if)#switchport priority default 5

queue bandwidth

Use this command to assign Weighted Round-Robin (WRR) weights to the four class of service (CoS) priority queues. Use the no form to restore the default weights. Syntax queue bandwidth weight1.weight4 no queue bandwidth weight1.weight4 - The ratio of weights for queues 0 - 3 determines the weights used by the WRR scheduler. (Range: 1 - 255) Default Setting Weights 16, 64, 128 and 240 are assigned to queue 0, 1, 2 and 3 respectively.
Command Mode Global Configuration Command Usage WRR allows bandwidth sharing at the egress port by defining scheduling weights. Example The following example shows how to assign WRR weights of 1, 3, 5 and 7 to the CoS priority queues 0, 1, 2 and 3:
Console(config)#queue bandwidth Console(config)#
Related Commands show queue bandwith

queue cos-map

Example The following example shows how to enable IP precedence mapping globally:
Console(config)#map ip precedence Console(config)#
map ip precedence (Interface Configuration)
Use this command to set IP precedence priority (i.e., IP Type of Service priority). Use the no form to restore the default table. Syntax map ip precedence ip-precedence-value cos cos-value no map ip precedence precedence-value - 3-bit precedence value. (Range: 0-7) cos-value - Class-of-Service value (Range: 0-7) Default Setting The list below shows the default priority mapping.
IP Precedence Value CoS Value 6 7
Command Mode Interface Configuration (Ethernet, Port Channel)
Command Usage The precedence for priority mapping is IP Precedence or IP DSCP, and default switchport priority. IP Precedence values are mapped to default Class of Service values on a one-to-one basis according to recommendations in the IEEE 802.1p standard, and then mapped to the queue defaults. This command sets the IP Precedence for all interfaces. Example The following example shows how to map IP precedence value 1 to CoS value 0:
Console(config)#interface ethernet 1/5 Console(config-if)#map ip precedence 1 cos 0 Console(config-if)#
map ip dscp (Global Configuration)
Use this command to enable IP DSCP mapping (i.e., Differentiated Services Code Point mapping). Use the no form to disable IP DSCP mapping. Syntax map ip dscp no map ip dscp Default Setting Disabled Command Mode Global Configuration Command Usage The precedence for priority mapping is IP Precedence or IP DSCP, and default switchport priority.
IP Precedence and IP DSCP cannot both be enabled. Enabling one of these priority types will automatically disable the other type. Example The following example shows how to enable IP DSCP mapping globally:
Console(config)#map ip dscp Console(config)#
map ip dscp (Interface Configuration)
Use this command to set IP DSCP priority (i.e., Differentiated Services Code Point priority). Use the no form to restore the default table. Syntax map ip dscp dscp-value cos cos-value no map ip dscp dscp-value - 8-bit DSCP value. (Range: 0-255) cos-value - Class-of-Service value (Range: 0-7) Default Setting The list below shows the default priority mapping. Note that all the DSCP values that are not specified are mapped to CoS value 0.
IP DSCP Value 10, 12, 14, 16 18, 20, 22, 24 26, 28, 30, 32, 34, 36 38, 40, 46, 56 CoS Value 6 7

Bandwidth

The difference between the highest and lowest frequencies available for network signals. Also synonymous with wire speed, the actual speed of the data transmission along the cable.
Boot protocol used to load the operating system for devices connected to the network.

Collision

A condition in which packets transmitted over the cable interfere with each other. Their interference makes both signals unintelligible. February Foundry Networks, Inc. Glossary-1

Collision Domain

Single CSMA/CD LAN segment.

CSMA/CD

Carrier Sense Multiple Access/Collision Detect is the communication method employed by Ethernet and Fast Ethernet.
Dynamic Host Control Protocol (DHCP)
Provides a framework for passing configuration information to hosts on a TCP/IP network. DHCP is based on the Bootstrap Protocol (BOOTP), adding the capability of automatic allocation of reusable network addresses and additional configuration options.

End Station

A workstation, server, or other device that does not act as a network interconnection.

Ethernet

A network communication system developed and standardized by DEC, Intel, and Xerox, using baseband transmission, CSMA/CD access, logical bus topology, and coaxial cable. The successor IEEE 802.3 standard provides for integration into the OSI model and extends the physical layer and media with repeaters and implementations that operate on fiber, thin coax and twisted-pair cable.

Fast Ethernet

A 100 Mbps network communication system based on Ethernet and the CSMA/ CD access method.

Full Duplex

Transmission method that allows switch and network card to transmit and receive concurrently, effectively doubling the bandwidth of that link.
GARP VLAN Registration Protocol (GVRP)
Defines a way for switches to exchange VLAN information in order to register necessary VLAN members on ports along the Spanning Tree so that VLANs defined in each switch can work automatically over a Spanning Tree network. Glossary-Foundry Networks, Inc. February 2003
Generic Attribute Registration Protocol (GARP)
GARP is a protocol that can be used by endstations and switches to register and propagate multicast group membership information in a switched environment so that multicast data frames are propagated only to those parts of a switched LAN containing registered endstations. Formerly called Group Address Registration Protocol.
Generic Multicast Registration Protocol (GMRP)
GMRP allows network devices to register endstations with multicast groups. GMRP requires that any participating network devices or endstations comply with the IEEE 802.1p standard.

Gigabit Ethernet

A 1000 Mbps network communication system based on Ethernet and the CSMA/ CD access method.

TACACS 3-15, 4-50 TACACS, logon authentication 3-15 Technical Support 1-1 e-mail 1-2 telephone 1-2 Web 1-2 trap manager 3-93 troubleshooting A-1 trunk configuration 3-87 LACP 3-88 static 3-90
RADIUS 3-15, 4-50 RADIUS, logon authentication 3-15 RSTP 4-108 global configuration 4-108
Secure Socket Layer <Emphasis> See SSL serial port configuring 4-74 XModem downloads A-2 SNMP community string 3-92 enabling traps 3-93 trap manager 3-93 software downloads 3-24, A-2 software version, displaying 3-31 Spanning Tree Protocol 3-46 SSL 4-32 Index-2
upgrading software 3-24, A-2 user password 3-13
VLANs configuring 3-60 February 2003
Web interface access requirements 3-1 configuration buttons 3-3 home page 3-2 menu list 3-5 panel display 3-4

XModem downloads A-2

Index-3

Index-4

Safety Compliance

Warning: Fiber Optic Port Safety
When using a fiber optic port, never look at the transmit laser while it is powered on. Also, never look directly at the fiber TX port and fiber cable ends when they are powered on.

CLASS I LASER DEVICE

Avertissment: Ports pour fibres optiques - scurit sur le plan optique
Ne regardez jamais le laser tant quil est sous tension. Ne regardez jamais directement le port TX (Transmission) fibres optiques et les embouts de cbles fibres optiques tant quils sont sous tension.
DISPOSITIF LASER DE CLASSE I
Warnhinweis: Faseroptikanschlsse - Optische Sicherheit
Niemals ein bertragungslaser betrachten, whrend dieses eingeschaltet ist. Niemals direkt auf den Faser-TX-Anschlu und auf die Faserkabelenden schauen, whrend diese eingeschaltet sind.

LASERGERT DER KLASSE I

Please read the following safety information carefully before installing the switch:
Warning: Installation and removal of the unit must be carried out by qualified personnel only. This guide is for system administrators with a working knowledge of network management.You should be familiar with switching and networking concepts. The unit must be connected to an earthed (grounded) outlet to comply with international safety standards. Do not connect the unit to an A.C. outlet (power supply) without an earth (ground) connection. The appliance coupler (the connector to the unit and not the wall plug) must have a configuration for mating with an EN 60320/IEC 320 appliance inlet. The socket outlet must be near to the unit and easily accessible. You can only remove power from the unit by disconnecting the power cord from the outlet. This unit operates under SELV (Safety Extra Low Voltage) conditions according to IEC 60950. The conditions are only maintained if the equipment to which it is connected also operates under SELV conditions.
France and Peru only This unit cannot be powered from IT supplies. If your supplies are of IT type, this unit must be powered by 230 V (2P+T) via an isolation transformer ratio 1:1, with the secondary connection point labelled Neutral, connected directly to earth (ground).

x 2007 Foundry Networks, Inc. December 2007
SFP Slots.4-6 Xenpak Module Slots.4-6 Stacking Ports.4-6 Status LEDs.4-7 Optional Redundant Power Supply.4-9 Power Supply Receptacles.4-9 Optional Media Extender Module.4-10 Xenpak Transceiver Support.4-10 Extender Module LEDs.4-11 Features and Benefits.4-12 Connectivity.4-12 Expandability.4-12 Performance.4-12 Management.4-13 Chapter 5. About the EdgeIron 8X10G.5-1 Overview.5-1 Package Contents.5-1 Supported Features.5-2 Features not Supported.5-3 Switch Architecture.5-4 Switching Method.5-4 Management Options.5-4 Description of Hardware.5-4 10GBASE Slots.5-4 XFP Transceiver Support.5-5 100BASE-T Port.5-5 Status LEDs.5-5 Optional Redundant Power Supply.5-7 Power Supply Receptacles.5-7 Features and Benefits.5-8 Connectivity.5-8 Performance.5-8 Management.5-8 Chapter 6. About the EdgeIron Redundant Power Supplies.6-1 Overview.6-1 Features and Benefits.6-1 Front and Rear Panels.6-2 Package Contents.6-2
LEDs.6-3 Chapter 7. Network Planning.7-1 Introduction to Switching.7-1 Application Examples.7-2 Collapsed Backbone.7-2 Central Wiring Closet.7-3 Remote Connections with Fiber Cable.7-4 Making VLAN Connections.7-6 Connectivity Rules.7-Gbps Ethernet Collision Domain.7-Mbps Gigabit Ethernet Collision Domain.7-Mbps Fast Ethernet Collision Domain.7-Mbps Ethernet Collision Domain.7-8 Application Notes.7-8 Chapter 8. Installing an EdgeIron Redundant Power Supply.8-1 Installation.8-1 Selecting a Site.8-1 Equipment Checklist.8-1 Optional Rack-Mounting Equipment.8-2 Mounting.8-2 Rack Mounting.8-2 Desktop or Shelf Mounting.8-3 Connecting Switches to the RPS.8-4 Ports Pin-Out (RPS-EIF and RPS2-EIF).8-6 Chapter 9. Installing an EdgeIron Switch.9-1 Selecting a Site.9-1 Equipment Checklist.9-1 Optional Rack-Mounting Equipment.9-1 Mounting.9-2 Rack Mounting.9-2 Desktop or Shelf Mounting.9-4 Installing an SFP or XFP Transceiver.9-4 Installing an Optional Module into the EdgeIron 24GS or EdgeIron 48GS.9-5 Installing an Optional Xenpak Transceiver into the EdgeIron 24GS or EdgeIron 48GS.9-6 Stacking Switches.9-6 Operational Concepts.9-7
Functionalities.9-8 Connecting to a Power Source.9-10 Chapter 10. Making Network Connections.10-1 Connecting Network Devices.10-1 Twisted-Pair Devices.10-1 Cabling Guidelines.10-1 Connecting to PCs, Servers, Hubs and Switches.10-2 Wiring Closet Connections.10-2 Fiber Optic Devices.10-Gbps Fiber Optic Connections.10-4 Chapter 11. Managing the Device.11-1 Accessing the CLI.11-1 Connecting to the Console Port.11-1 Wiring Map for Serial Cable.11-2 Logging on through the CLI.11-2 Console Connection.11-2 Telnet Connection.11-3 Logging On Through the Web Management Interface.11-4 Navigating the Web Management Interface.11-4 Using the Web Management Interface.11-5 Setting Passwords.11-6 Setting an IP Address.11-6 Enabling SNMP Management Access.11-8 Saving Configuration Settings.11-10 Entering CLI Commands.11-11 Keywords and Arguments.11-11 Minimum Abbreviation.11-11 Command Completion.11-11 Getting Help on Commands.11-11 Partial Keyword Lookup.11-12 Negating the Effect of Commands.11-12 Using Command History.11-13 Understanding Command Modes.11-13 Exec Commands.11-14 Configuration Commands.11-14 Command Line Processing.11-15 Chapter 12. Troubleshooting.12-1

Each unit includes two stacking ports that provide a 40 Gbps high-speed serial stack backplane connection. Up to eight 24-port or eight 48-port switches can be connected together using the stacking cables. Or, 24-port and 48-port switches can be mixed together in a stack of eight.
The LEDs, which are located on the front panel for easy viewing, are shown below and described in the following table. Figure 4-3. EdgeIron 24GS Port and System LEDs

Edgelron 24GS

Combo Port Status Indicators (21~24)
Figure 4-4. EdgeIron 48GS Port and System LEDs
Port Status Indicators System Indicators
Combo Port Status Indicators (45~48)
Port Status LEDs - EdgeIron 24GS and EdgeIron 48GS LED Condition Status
Gigabit Ethernet Ports (Ports 1~24, or 1~48) Link/Act On Yellow On Green Flashing Yellow Flashing Green Off Port has established a valid 10/100 Mbps network connection. Port has established a valid 1000 Mbps network connection. Indicates network activity at 10/100 Mbps. Indicates network activity at 1000 Mbps. There is no valid link on the port.
System Status LEDs - EdgeIron 24GS and EdgeIron 48GS LED Power Condition On Green On Yellow Off RPU On Green On Yellow Off Diag On Green Flashing Green On Amber Stack Master On Green Flashing Green or Flashing Amber On Amber Off Stack Link On Green On Amber Flashing Green Flashing Amber Off Module On Green On Amber Flashing Green Off Status The units internal power supply is operating normally. The units internal power supply has failed. The unit has no power connected. The redundant power supply (RPS2-EIF) is operating normally. The redundant power supply (RPS2-EIF) has failed. No redundant power supply is connected. The system diagnostic test has completed successfully. The system diagnostic test is in progress. The system diagnostic test has detected a fault. Switch is operating as the Master unit in the stack. Switch in arbitration/election state. Switch is operating as a Slave unit in the stack. System in standalone mode. Uplink and downlink operating normally on the master unit. Uplink and downlink operating normally on the slave unit. Uplink has failed. Downlink has failed. No stacking link present. An expansion module is installed in the slot. An expansion module is installed, but cannot communicate with the switch. An installed expansion module has been disabled. There is no module installed.

10 Gbps XFP Ports Management Port

Edgelron 8X10G

PWR 1 Link/Act XFP Module Link/Act XFP Module 2 Link/Act XFP Module 3 Link/Act XFP Module 4 Link/Act XFP Module 5 Link/Act XFP Module 6 Link/Act XFP Module 7 Link/Act XFP Module 8 RPS Diag Link/Act Console Management
System Indicators Console Port
EdgeIron 8X10G switch EdgeIron Release Notes Four adhesive foot pads Bracket Mounting Kit containing two brackets and eight screws for attaching the brackets to the switch Power Cordeither US, Continental Europe or UK
RS-232 console cable User Agreement Envelope Registration Card
Features System Size Uplink Management Message of the Day Banner Port-channel load balance Redundant Power Jumbo Frames Flash Memory DRAM Buffer Architecture RADIUS and TACACS+ authentication TACACS+ AAA Accounting Remote Logging HTTPS SSH (version 2.0) SNMP (v1 and v2c)
EdgeIron 8X10G 1U XFP Fiber Module 10/100BASE-TX Yes No External Yes (up to 9216 bytes) 32 Mbytes 128 Mbytes 4 Mbytes per system Yes Yes Yes Yes Yes Yes

About the EdgeIron 8X10G

Features Layer 2 IEEE 802.1q Port-based VLANs Private VLANs IEEE 802.1D (STP) IEEE 802.1w (RSTP) Foundry PVST* GVRP 10 Gigabit Trunks IEEE 802.3ad IEEE 802.1X Rate Limiting - ingress and egress Layer 3 Multicast ToS/DSCP Honoring IGMP Snooping
EdgeIron 8X10G Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
EdgeIron does not support Layer 3 features, such as: Border Gateway Protocol (BGP) Foundry Standby Router Protocol IP Load Sharing IP v6 IS-IS Multiprotocol Border Gateway Protocol (MBGP) Multiprotocol Source Discovery Protocol (MSDP) Network Address Translation (NAT) Netflow Outbound ACLs ACL-based Rate Limiting Policy Based Routing Super Aggregated VLANs SuperSpan Port Security
VLAN-based priority Policy-Based Routing
The EdgeIron 8X10G employs a wire-speed, non-blocking switching fabric. This permits simultaneous wire-speed transport of multiple packets at low latency on all ports. This switch also features full-duplex capability on all ports, which effectively doubles the bandwidth of each connection.

System Status LEDs - RPS2-EIF LED Fan Condition On Green Flashing Yellow Off Thermal On Green Flashing Yellow Off Power On Green Off Status The fans are functioning normally. One or more fans are not operating properly. The unit is powered off. The temperature of the unit is within an acceptable range. Indicates a high temperature that is approaching an overtemperature condition. The thermal detection function has failed, or the unit is powered off. AC power is being supplied to the RPS. No AC power is being supplied to the RPS.
Chapter 7 Network Planning
Introduction to Switching
A network switch allows simultaneous transmission of multiple packets via non-crossbar switching. This means that it can partition a network more efficiently than bridges or routers. The switch has, therefore, been recognized as one of the most important building blocks for todays networking technology. When performance bottlenecks are caused by congestion at the network access point (such as the network card for a high-volume file server), the device experiencing congestion (server, power user or hub) can be attached directly to a switched port. And, by using full-duplex mode, the bandwidth of the dedicated segment can be doubled to maximize throughput. When networks are based on repeater (hub) technology, the distance between end stations is limited by a maximum hop count. However, a switch turns the hop count back to zero, so subdividing the network into smaller and more manageable segments, and linking them to the larger network by means of a switch, removes this limitation. A switch can be easily configured in any Ethernet, Fast Ethernet, Gigabit Ethernet, or 10 Gigabit Ethernet network to significantly boost bandwidth while using conventional cabling and network cards.

Application Examples

The EdgeIron family is not only designed to segment your network, but also to provide a wide range of options in setting up network connections. Some typical applications are described below.

Collapsed Backbone

The EdgeIron family is an excellent choice for mixed Ethernet, Fast Ethernet, Gigabit Ethernet, or 10 Gigabit Ethernet installations where significant growth is expected in the near future. When the time comes for further network expansion, you can easily build on this basic configuration by adding Gigabit Ethernet links directly to one or more workgroup switches. In the figure below, the EdgeIron 48G is operating as a collapsed backbone for a small LAN. It is providing dedicated 10 Mbps full-duplex connections to workstations, 100 Mbps full-duplex connections to power users, and 1 Gbps full-duplex connections to servers. Figure 7-1. Collapsed Backbone
Servers 1 Gbps Full Duplex

Patch Panel

Fiber Optic Devices

An optional 1000BASE-SX, 1000BASE-LX, or 1000BASE-LHA SFP transceiver may be used for backbone or remote connections, or for connecting to a high-speed server. Each single-mode fiber optic port requires 9/125 micron single-mode fiber optic cabling with an LC or SC connector at both ends. Each multimode fiber optic port requires 50/125 or 62.5/125 micron multimode fiber optic cabling with an LC or SC connector at both ends. Warning: This switch uses lasers to transmit signals over fiber optic cable. The lasers are compliant with the requirements of a Class 1 Laser Product and are inherently eye safe in normal operation. However, you should never look directly at a transmit port when it is powered on. 1. Remove and keep any protective port covers. When not connected to a fiber cable, the cover should be replaced to protect the optics.
2. Check that the fiber terminators are clean. You can clean the cable plugs by wiping them gently with a clean tissue or cotton ball moistened with a little ethanol. Dirty fiber terminators on fiber optic cables will impair the quality of the light transmitted through the cable and lead to degraded performance on the port. 3. Connect one end of the cable to the LC or SC port on the switch and the other end to the port on the other device. Since LC and SC connectors are keyed, the cable can be attached in only one orientation. Figure 10-3. Making Fiber Port Connections
As a connection is made, check the Link LED on the switch corresponding to the port to be sure that the connection is valid.
10 Gbps Fiber Optic Connections
An optional 10 Gigabit transceiver (Xenpak or XFP) can be used for a backbone connection between switches. Warning: This switch uses lasers to transmit signals over fiber optic cable. The lasers are compliant with the requirements of a Class 1 Laser Product and are inherently eye safe in normal operation. However, you should never look directly at a transmit port when it is powered on. 1. Remove and keep the ports protective cover. When not connected to a fiber cable, the cover should be replaced to protect the optics. 2. Check that the fiber terminators are clean. You can clean the cable plugs by wiping them gently with a clean tissue or cotton ball moistened with a little ethanol. Dirty fiber terminators on fiber cables will impair the quality of the light transmitted through the cable and lead to degraded performance on the port. 3. Connect one end of the cable to the LC port on the switch and the other end to the LC port on the other device. Since LC connectors are keyed, the cable can be attached in only one orientation.

User Access Verification Username: admin Password: CLI session with the EdgeIron 4802CF is opened. To end the CLI session, enter [Exit]. Console#

Managing the Device

Telnet Connection
Telnet operates over the IP transport protocol. In this environment, your management station and any network device you want to manage over the network must have a valid IP address. Valid IP addresses consist of four numbers, 0 to 255, separated by periods. Each address consists of a network portion and host portion. For example, the IP address assigned to this switch, 10.1.0.1, consists of a network portion (10.1.0) and a host portion (1). To access the switch through a Telnet session, you must first set the IP address for the switch, and set the default gateway if you are managing the switch from a different IP subnet. By default, the switch uses DHCP to assign IP settings to VLAN 1 on the switch. If you wish to manually configure IP settings, you need to change the switchs user-specified defaults to values that are compatible with your network. You may also need to a establish a default gateway between the switch and management stations that exist on another network segment. You must configure at least one IP address using the serial connection to the CLI before you can manage the system using the other management interfaces. Foundry devices support both classful IP network masks (i.e. Class A, B, and C subnet masks, and so on), and Classless Interdomain Routing (CIDR). To enter a classical network mask, enter the mask in IP address format. For example, enter 209.157.22.99 255.255.255.0 for an IP address with a Class-C subnet mask.
Note: If your network uses a BootStrap Protocol (BootP) server or a Dynamic Host Configuration Protocol (DHCP) server, you can allow the Foundry device to obtain an IP address on the network.
Console(config)#interface vlan 1 Console(config-if)#ip address 10.1.0.1 255.255.255.0 Console(config-if)#exit Console(config)#ip default-gateway 10.1.0.254
If your corporate network is connected to another network outside your office or to the Internet, you need to apply for a registered IP address. However, if you are attached to an isolated network, then you can use any IP address that matches the network segment to which you are attached. After you configure the switch with an IP address, you can open a Telnet session by performing these steps. 1. From the remote host, enter the Telnet command and the IP address of the device you want to access. 2. At the prompt, enter the user name and system password. The CLI will display the Vty-n# prompt for the administrator to show that you are using privileged access mode (i.e., Privileged Exec), or Vty-n> for the guest to show that you are using normal access mode (i.e., Normal Exec), where n indicates the number of the current Telnet session. 3. Enter the necessary commands to complete your desired tasks. 4. When finished, exit the session with the quit or exit command.

Note: The only changes that become permanent are when the running config is saved to the Flash memory. Any other elements you enable or disable will go back to their default settings the next time you re-start the switch.

Setting Passwords

Note: If this is your first time to log into the CLI program, you should define new passwords for both default user names using the username command, record them and put them in a safe place. Passwords can consist of up to eight alphanumeric characters and are case sensitive. To prevent unauthorized access to the switch, set the passwords as follows: 1. Open the console interface with the default user name and password admin to access the Privileged Exec level. 2. Type configure and press <Enter>. 3. Type username guest password 0 password, for the Normal Exec level, where password is your new password. Press <Enter>. 4. Type username admin password 0 password, for the Privileged Exec level, where password is your new password. Press <Enter>.
Username: admin Password: CLI session with the EdgeIron 4802CF is opened. To end the CLI session, enter [Exit]. Console#configure Console(config)#username guest password 0 [password] Console(config)#username admin password 0 [password] Console(config)#

Setting an IP Address

You must establish IP address information for the switch to obtain management access through the network. This can be done in either of the following ways: Manual You have to input the information, including IP address and subnet mask. If your management station is not in the same IP subnet as the switch, you will also need to specify the default gateway router. Dynamic The switch sends IP configuration requests to BOOTP or DHCP address allocation servers on the network. Note: Only one VLAN interface can be assigned an IP address (the default is VLAN 1). This defines the management VLAN, the only VLAN through which you can gain management access to the switch. If you assign an IP address to any other VLAN, the new IP address overrides the original IP address and this becomes the new management VLAN.
Manual Configuration You can manually assign an IP address to the switch. You may also need to specify a default gateway that resides between this device and management stations that exist on another network segment. Valid IP addresses consist of four decimal numbers, 0 to 255, separated by periods. Anything outside this format will not be accepted by the CLI program. Note: The IP address for this switch is assigned via DHCP by default. Before you can assign an IP address to the switch, you must obtain the following information from your network administrator: IP address for the switch Default gateway for the network Network mask for this network

(RPS2-EIF) Link LED is Flashing Yellow (RPS2-EIF) Activity LED is Flashing Green
Troubleshooting Chart Symptom (RPS2-EIF) Fan LED is Flashing Yellow (RPS2-EIF) Thermal LED is Flashing Yellow Action One or more fans are not operating properly. Contact Technical Support. The unit is approaching an over-temperature condition. Power cycle the switch to try and clear the condition. If the condition persists, contact Technical Support.
Power and Cooling Problems
If the power indicator does not turn on when the power cord is plugged in, you may have a problem with the power outlet, power cord, or internal power supply. However, if the unit powers off after running for a while, check for loose power connections, power losses or surges at the power outlet, and verify that the fans on the unit are unobstructed and running prior to shutdown. If you still cannot isolate the problem, then the internal power supply may be defective. In this case, contact Technical Support for assistance.
Verify that all system components have been properly installed. If one or more components appear to be malfunctioning (such as the power cord or network cabling), test them in an alternate environment where you are sure that all the other components are functioning properly.

In-Band Access

You can access the management agent in the switch from anywhere within the attached network using Telnet, a Web browser, or other network management software. However, you must first configure the switch with a valid IP address, subnet mask, and default gateway. If you have trouble establishing a link to the management agent, check to see if you have a valid network connection. Then verify that you entered the correct IP address. Also, be sure the port through which you are connecting to the switch has not been disabled. If it has not been disabled, then check the network cabling that runs between your remote location and the switch. Note: The management agent can accept up to four simultaneous Telnet sessions. If the maximum number of sessions already exists, an additional Telnet connection will not be able to log into the system.

Troubleshooting

Stack Troubleshooting (EdgeIron 24GS and 48GS)
If a stack fails to initialize or function, first check the following items: Check that all stacking cables are properly connected. Check that all stacking modules are inserted in their slots correctly. Check if any stacking cables appear damaged. Check that all switches in the stack are powered on.
After checking all items, reboot all the switches in the stack. If the problem is still not resolved, contact Technical Support.

Chapter 13 Cables

Operating: 0 to 50 C (32 to 122 F) Storage: -40 to 70 C (-40 to 158 F) Humidity Operating: 10% to 90% Power Supply Internal, auto-ranging transformer: 100 to 240 VAC, 47 to 63 Hz External, supports connection for 14-pin redundant power unit Power Consumption EdgeIron 2402CF: 45 Watts maximum EdgeIron 4802CF: 48 Watts maximum Heat Dissipation EdgeIron 2402CF: 238.7 BTU/hr maximum EdgeIron 4802CF: 163.68 BTU/hr maximum Maximum Current EdgeIron 2402CF: 0.8 A @ 110 VAC, 0.5 A @ 240 VAC EdgeIron 4802CF: 1.2 A @ 110 VAC, 0.6 A @ 240 VAC Noise 48 dB CPU Type and Speed EdgeIron 2402CF: MPC8241/166MHz (8MB Flash, 32MB SDRAM) EdgeIron 4802CF: MPC8241/266MHz (8MB Flash, 32MB SDRAM) MTBF EdgeIron 2402CF: 122,109 hours EdgeIron 4802CF: 46,799 hours
EdgeIron 24G-A and EdgeIron 48G
Ports 20 or 44 10/100/1000BASE-T, with auto-negotiation 4 10/100/1000BASE-T shared with four SFP transceiver slots Network Interface 10BASE-T: RJ-45 (100-ohm, UTP cable; Categories 3, 4, 5) 100BASE-TX: RJ-45 (100-ohm, UTP cable; Category 5) 1000BASE-T: RJ-45 (100-ohm Category 5, 5e, or 6 UTP or STP cable) RJ-45 connector, auto MDI/MDI-X System Memory 8 Mbytes Flash memory 32 Mbytes DRAM Buffer Architecture EdgeIron 24GA: 2 Mbyte per system EdgeIron 48G: 4 Mbytes per system Switching Database 16 K MAC address entries LEDs
-Foundry Networks, Inc. December 2007
EdgeIron 24GA: System: Power, RPS, Diag., FDX (Full Duplex/Half Duplex) Port: Link, Act EdgeIron 48G: System: Power, RPS, Diag., Port: Link, Act. Weight EdgeIron 24G-A: 3.24 kg (7.17 lbs) EdgeIron 48G: 5.88 kg (12.97 lbs) Size EdgeIron 24G-A: 44.0 x 31 x 4.4 cm (17.4 x 12.2 x 1.8 in.) EdgeIron 48G: 44.0 x 41.5 x 4.4 cm (17.4 x 16.4 x 1.8 in.) Temperature Operating: 0 to 50 C (32 to 122 F) Storage: -40 to 70 C (-40 to 158 F) Humidity Operating: 5% to 95% AC Input 100 to 240 V, 50 to 60 Hz Power Supply Internal, auto-ranging transformer: 100 to 240 VAC, 50 to 60 Hz External, supports connection for 14-pin redundant power unit Power Consumption EdgeIron 24GA: 70 Watts maximum EdgeIron 48G: 105 Watts maximum Heat Dissipation EdgeIron 24GA: 238.7 BTU/hr maximum EdgeIron 48G: 358.05 BTU/hr maximum Maximum Current EdgeIron 24GA: 0.7 A @ 110 VAC, 0.3 A @ 240 VAC EdgeIron 48G: 1.2 A @ 110 VAC, 0.6 A @ 240 VAC Noise 56 dB MTBF EdgeIron 24G-A: 148,333 hours EdgeIron 48G: 159,149 hours CPU Type and Speed: EdgeIron 24G-A: MPC8245/266MHz (8MB Flash, 32MB SDRAM) EdgeIron 48G: MPC8245/266MHz (8MB Flash, 32MB SDRAM)

December Foundry Networks, Inc. -5
Temperature Operating: 0 C to 50 C (32 F to 122 F) Storage: -40 C to 70 C (-40 F to 158 F) Power Supply Internal, auto-ranging transformer: 100 to 240 VAC, 50 to 60 Hz Power Consumption 150 Watts maximum Heat Dissipation 511.5 BTU/hr maximum Maximum Current 2 A @ 110 VAC 1 A @ 240 VAC Noise 45 dB MTBF 113,990 hours CPU Type and Speed: Edgeiron 8X10G: MPC8245 466MHz (32MB Flash, 128MB SDRAM)
EdgeIron External Redundant Power Supplies (RPS-EIF and RPS2-EIF)
Ports 4 RPS DC power outlets (IEC socket 216C743-07) LEDs EdgeIron RPS-EIF: Ports 1 to 4, and fan status LEDs. EdgeIron RPS2-EIF: Ports 1 to 4, Fan, Thermal and Power LEDs. Weight EdgeIron RPS-EIF: 5 kg (11.02 lbs) EdgeIron RPS2-EIF: 5.5 kg (12.1 lbs) Size 44.0 x 28.4 x 4.3 cm (17.34 x 11.19 x 1.69 in.) Temperature EdgeIron RPS-EIF: 0 to 40 C (32 to 104F); Storage: -40 to 70 C (-40 to 158 F) Operating: 0 to 50 C (32 to 122F) Storage: -40 to 70 C (-40 to 158 F) Humidity EdgeIron RPS-EIF: Operating: 5% to 95% EdgeIron RPS2-EIF: Operating: 20% to 90% AC Input 100 to 240 V, 50 to 60 Hz EIA Ratings EdgeIron RPS-EIF: 115V/10A, 230V/10A EdgeIron RPS2-EIF: 125V/10A, 250V/10A
DC Power Cable Connector 14-pin P-Type plastic type Power Supply Internal, auto-ranging transformer: 100 to 240 VAC, 50 to 60 Hz Output 12V DC Maximum Power Consumption EdgeIron EIF-RPS: 253 Watts EdgeIron EIF-RPS2: 292 Watts Heat Dissipation EdgeIron EIF-RPS: 865 BTU/hr maximum EdgeIron EIF-RPS2: 998 BTU/hr maximum Input Power Consumption 70 Watts maximum per port Current: 0 A (minimum), 14 A (maximum) Line regulation: +/- 1% Load regulation: +/- 10% Overshoot and undershoot: 5% Maximum Current EdgeIron RPS-EIF: 10 A @ 110 VAC; 5 A @ 230 VAC EdgeIron RPS2-EIF: 8 A @ 110 VAC; 3.5 A @ 240 VAC Overload Protection Reduces output to safe levels when output exceeds 110% Fan Voltage +12V DC Operating voltage range: 7V to 13.8V Speed: 8200 RPM Air delivery: 9.43 CFM Noise: 30.5 dB Noise 51.4 dB MTBF 50,050 hours

Switch Features

Spanning Tree Protocol Forwarding Mode Store-and-forward Flow Control Full Duplex: IEEE 802.3x Half Duplex: Back pressure Broadcast Storm Suppression Traffic throttled above a critical threshold
VLAN Support Up to 255 groups; port-based or with 802.1Q VLAN tagging GVRP for automatic VLAN learning Multicast Switching IGMP Snooping Class of Service EdgeIron 2402CF and 4802CF Supports four levels of priority and Weighted Round Robin queueing EdgeIron 24GS, 48GS, 24G-A, 48G and 8X10G Supports eight levels of priority and Weighted Round Robin queueing

Management Features

In-Band Management Web, Telnet, SSH or SNMP manager Out-of-Band Management RS-232 DB-9 console port Software Loading TFTP in-band or XModem out-of-band MIB Support MIB II (RFC 1213), Bridge MIB (RFC 1493), Interfaces Evolution MIB (RFC 2863), Ethernet MIB (RFC 2665), Extended Bridge MIB (RFC 2674), RMON MIB (RFC 2819), Entity MIB (RFC 2737), RADIUS authentication client MIB (RFC 2621), Foundrys private MIB RMON Support Groups 1, 2, 3, 9 (Statistics, History, Alarm, Event) Additional Features Port Trunks (static - Cisco EtherChannel compliant, dynamic - LACP) Port Mirroring BOOTP/DHCP Client

Standards

IEEE Std 802.3-2002 (Ethernet, Fast Ethernet and Gigabit Ethernet) IEEE 802.3ae 10 Gigabit Ethernet, IEEE 802.3af Power-over-Ethernet, IEEE 802.1D Spanning Tree Protocol and traffic priorities, IEEE 802.1p priority tags, IEEE 802.1Q VLAN, IEEE 802.3ac VLAN tagging, IEEE 802.ad Link Aggregation Control Protocol IEEE 802.3x full-duplex flow control (ISO/IEC 8802-3) IEEE 802.1q Ethernet Prioritization IEEE 802.1p Class of Service (CoS) IEEE 802.1d Spanning Tree Protocol (STP) (Not supported in GS or 8X10G) IEEE 802.1w Rapid Spanning Tree Protocol (RSTP) (Not supported in GS or 8X10G) SNMPv2 (RFC 2571) SNMPv3 (RFC Draft 3414, 2570, 2273, 3411, 3415)
CE Mark Emissions FCC Class A Industry Canada Class A EN 50081-1 EN 50082-1 EN 55022 (CISPR 22) Class A EN 60555-3 EN 61000-3-2/3 IEC 1000-4-2/3/4/5/6 VCCI Class A C-Tick - AS/NZS 3548 (1995) Class A Immunity EN 55024 Safety CSA/ C US (CSA22.2 NO 60950-00 & UL60950) CB: IE 60950 EN 60950 (TV/GS)
Appendix B Warnings and Cautionary Messages

Warnings

A warning calls your attention to a possible hazard that can cause injury or death. The following are the warnings used in this manual. "Achtung" weist auf eine mgliche Gefhrdung hin, die zu Verletzungen oder Tod fhren knnen. Sie finden die folgenden Warnhinweise in diesem Handbuch: Un avertissement attire votre attention sur un risque possible de blessure ou de dcs. Ci-dessous, vous trouverez les avertissements utiliss dans ce manuel. Una advertencia le llama la atencin sobre cualquier posible peligro que pueda ocasionar daos personales o la muerte. A continuacin se dan las advertencias utilizadas en este manual.

Glossary-1

1000BASE-T IEEE 802.3ab specification for Gigabit Ethernet over 100-ohm Category 5 or 5e twisted-pair cable (using all four wire pairs). Auto-Negotiation Signalling method allowing each node to select its optimum operational mode (e.g., 10, 100, or 1000 Mbps and half or full duplex) based on the capabilities of the node to which it is connected. Bandwidth The difference between the highest and lowest frequencies available for network signals. Also synonymous with wire speed, the actual speed of the data transmission along the cable. Collision A condition in which packets transmitted over the cable interfere with each other. Their interference makes both signals unintelligible. Collision Domain Single CSMA/CD LAN segment. CSMA/CD CSMA/CD (Carrier Sense Multiple Access/Collision Detect) is the communication method employed by Ethernet, Fast Ethernet, or Gigabit Ethernet. End Station A workstation, server, or other device that does not forward traffic. Ethernet A network communication system developed and standardized by DEC, Intel, and Xerox, using baseband transmission, CSMA/CD access, logical bus topology, and coaxial cable. The successor IEEE 802.3 standard provides for integration into the OSI model and extends the physical layer and media with repeaters and implementations that operate on fiber, thin coax and twisted-pair cable. Fast Ethernet A 100 Mbps network communication system based on Ethernet and the CSMA/CD access method. Full Duplex Transmission method that allows two network devices to transmit and receive concurrently, effectively doubling the bandwidth of that link. Gigabit Ethernet A 1000 Mbps network communication system based on Ethernet and the CSMA/CD access method. IEEE Institute of Electrical and Electronic Engineers. IEEE 802.3 Defines carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications.

Glossary-2

IEEE 802.3ab Defines CSMA/CD access method and physical layer specifications for 1000BASE-T Gigabit Ethernet. (Now incorporated in IEEE 802.3-2002.) IEEE 802.3ae Defines access method and physical layer specifications for 10 Gigabit Ethernet. IEEE 802.3u Defines CSMA/CD access method and physical layer specifications for 100BASE-TX and 100BASE-FX Fast Ethernet. (Now incorporated in IEEE 802.3-2002.) IEEE 802.3x Defines Ethernet frame start/stop requests and timers used for flow control on full-duplex links. (Now incorporated in IEEE 802.3-2002.) IEEE 802.3z Defines CSMA/CD access method and physical layer specifications for 1000BASE Gigabit Ethernet. (Now incorporated in IEEE 802.3-2002.) LAN Segment Separate LAN or collision domain. LED Light emitting diode used for monitoring a device or network condition. Local Area Network A group of interconnected computers and support devices. Media Access Control (MAC) A portion of the networking protocol that governs access to the transmission medium, facilitating the exchange of data between network nodes. MIB An acronym for Management Information Base. It is a set of database objects that contains information about the device. Modal Bandwidth Bandwidth for multimode fiber is referred to as modal bandwidth because it varies with the modal field (or core diameter) of the fiber. Modal bandwidth is specified in units of MHz per km, which indicates the amount of bandwidth supported by the fiber for a one km distance. Network Diameter Wire distance between two end stations in the same collision domain. Redundant Power Supply (RPS) A backup power supply unit that automatically takes over in case the primary power supply should fail. RJ-45 Connector A connector for twisted-pair wiring.