Linksys SFE2000P, SFE2000 User Manual
SFE2000/SFE2000P Fast Ethernet
Switch Reference Guide
December 2007
LINKSYS ONE COMMUNICATIONS SOLUTION
© 2006-2007 Copyright 2006-2007 Cisco Systems, Inc.
Specifications are subject to change without notice.
Linksys, the Cisco Systems logo, the Linksys Logo, and the Linksys One logo are registered trademarks of Cisco Systems,
Inc. All other trademarks mentioned in this document are the property of their respective owners.
Document Revision History
Revision Date Description
1.0 December 2007 Added switch stacking information.
Contents
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Contents
Chapter 1: Getting Started . . . . . . . . . . . . . . . . . . . 1
Starting the Application 1
Understanding the Interface 4
Device Representation 5
Using the Linksys Management Buttons 6
Using Screen and Table Options 7
Adding Device Information 7
Modifying Device Information 7
Deleting Device Information 8
Resetting the Device 9
Logging Off The Device 9
Chapter 2: Managing Device Information . . . . . . . . . . . . . 11
Understanding the Device Zoom View 11
Defining General System Information 12
SFE 2000P System Information 12
Managing Stacks 13
Building Automatically-Configured Stacks 14
Building a New Stack 14
Adding Units to a Running Stack 14
Building Manually-Configured Stacks 14
Building a New Stack 14
Adding Units to a Running Stack 14
Understanding Stack Resiliency 15
Understanding Advanced Stacking 15
Unit IDs 15
Stack Master 15
Stack Backup Master 15
Stack Members 16
Master-Enabled Units 16
Unit ID Allocation 16
Stack Unit Startup Process 17
Master Discovery Process 17
Master Election Process 17
Unit ID Allocation and Duplicate Unit ID Conflict Resolution 18
Configuring Units and Ports 19
Setting the Unit’s Operational Mode 19
Configuring the Stack Master and Unit ID 19
Resetting the Unit to Factory Default Mode 21
Understanding LED Indicators 21
Stack Troubleshooting and Maintenance 21
Replacing a Failed Member Stack Unit in an Operational Stack 21
Replacing a Failed Stack Master Unit in an Operational Stack 23
Contents
1
Contents
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Splitting a Stack 24
The Stack Master and Backup Master Units Remain in a Group 24
The Stack Master or the Backup Master Unit Remains in a Group 24
Neither the Stack Master Unit or the Backup Master Unit Remains in the Group 25
Merging Two Stacks 26
Understanding Stacking Cable Failure 27
Inserting Too Many Units into a Stack 27
Inserting a Standalone Unit into a Running Stack 28
Viewing Device Health 28
Chapter 3: Managing Power-over-Ethernet Devices . . . . . . . . . 31
Defining PoE System Information 31
Chapter 4: Configuring Device Security . . . . . . . . . . . . . . 37
Passwords Management 38
Defining User Authentication 38
Defining Authentication 40
Defining Profiles 40
Mapping Profiles 42
Defining TACACS+ 43
Defining RADIUS 47
Defining Access Method 51
Defining Access Profiles 51
Defining Profile Rules 54
Defining Traffic Control 60
Defining Storm Control 60
Defining Port Security 62
Defining 802.1x 66
Defining 802.1X Properties 67
Defining Port Authentication 68
Defining Multiple Hosts 71
Defining Authenticated Host 74
Defining Access Control 75
Defining MAC Based ACL 75
Defining IP Based ACL 77
Defining ACL Binding 84
Defining DOS Prevention 85
Global Settings 85
Defining Martian Addresses 87
Chapter 5: Configuring Device Interfaces . . . . . . . . . . . . . 89
Defining Port Settings 89
Defining LAG Management 94
Defining LAG Settings 96
Configuring LACP 99
2
Contents
Contents
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Chapter 6: Configuring VLANs . . . . . . . . . . . . . . . . 103
Defining VLAN Properties 104
Defining VLAN Membership 106
Defining Interface Settings 108
Configuring GVRP Settings 111
Protocol Group 114
Protocol Port 116
Chapter 7: Configuring IP Information . . . . . . . . . . . . . 119
Domain Name System 120
Defining DNS Server 120
Host Mapping 122
Configuring Layer 2 124
Configuring IP Addressing: 124
IP Interface 124
ARP 125
Configuring Layer 3 129
Configuring IP Addressing 129
IP Interface 129
ARP Proxy 132
UDP Relay 133
DHCP Relay 136
ARP 137
Defining IP Routing 140
Chapter 8: Defining Address Tables . . . . . . . . . . . . . . 143
Defining Static Addresses 144
Defining Dynamic Addresses 146
Chapter 9: Configuring Multicast Forwarding . . . . . . . . . . 149
IGMP Snooping 149
Defining Multicast Bridging Groups 152
Defining Multicast Forwarding 155
Chapter 10: Configuring Spanning Tree . . . . . . . . . . . . . 157
Defining STP on Interfaces 158
Defining Interface Settings 160
Defining Rapid Spanning Tree 164
Defining Multiple Spanning Tree 168
Defining MSTP Properties 169
Instance to VLAN 170
Instance Settings 170
Interface Settings 171
Chapter 11: Configuring SNMP . . . . . . . . . . . . . . . . 177
SNMP v1 and v2 177
Contents
3
Contents
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
SNMP v3 177
Configuring SNMP Security 178
Defining Engine ID 178
Defining SNMP Views 179
Defining SNMP Users 181
Define SNMP Groups 183
Defining SNMP Communities 185
Defining Trap Management 189
Defining Trap Settings 189
Configuring Station Management 190
Defining SNMP Filter Settings 196
Chapter 12: Configuring Quality of Service . . . . . . . . . . . .199
Defining General Settings 200
Defining CoS 201
Defining Queue 202
Mapping CoS to Queue 204
Mapping DSCP to Queue 205
Configuring Bandwidth 205
Defining Advanced Mode 208
Configuring DSCP Mapping 208
Defining Class Mapping 209
Defining Aggregate Policer 211
Configuring Policy Table 214
Defining Policy Binding 217
Defining QoS Basic Mode 219
Chapter 13: Managing System Files . . . . . . . . . . . . . . .221
File Management Overview 221
Firmware Upgrade 222
Save Configuration 223
Copy Files 224
Active Image 225
Chapter 14: Managing System Logs . . . . . . . . . . . . . . .227
Enabling System Logs 227
Viewing the Device Memory Logs 229
Clearing Message Logs 229
Viewing the Flash Logs 230
Clearing Message Logs 230
Viewing Remote Logs 231
Chapter 15: Configuring System Time . . . . . . . . . . . . . .235
Defining System Time 236
Defining SNTP Settings 238
Defining SNTP Authentication 240
4
Contents
Contents
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Chapter 16: Viewing Statistics . . . . . . . . . . . . . . . . . 243
Viewing Ethernet Statistics 243
Defining Ethernet Interface 243
Resetting Interface Statistics Counters 245
Viewing Etherlike Statistics 245
Resetting Etherlike Statistics Counters 247
Viewing GVRP Statistics 247
Resetting GVRP Statistics Counters 249
Viewing EAP Statistics 249
Managing RMON Statistics 252
Viewing RMON Statistics 253
Resetting RMON Statistics Counters 254
Configuring RMON History 255
Defining RMON History Control 255
Viewing the RMON History Table 258
Configuring RMON Events 259
Defining RMON Events Control 259
Viewing the RMON Events Logs 262
Defining RMON Alarms 263
Chapter 17: Managing Device Diagnostics . . . . . . . . . . . . 269
Viewing Integrated Cable Tests 269
Performing Optical Tests 271
Configuring Port Mirroring 272
Defining CPU Utilization 275
Appendix A: Linksys One Contact Information . . . . . . . . . . 277
Contents
5
Contents
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
6
Contents
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Getting Started
This section provides an introduction to the user interface, and includes the following topics:
• Starting the Application
•
Understanding the Interface
• Using the Linksys Management Buttons
• Using Screen and Table Options
• Resetting the Device
• Logging Off The Device
Starting the Application
This section contains information for starting the Linksys User Interface.
Chapter
1
Note By default, the IP address of the device is assigned dynamically. The IP address can be changed.
It is recommended to configure the IP address statically, if the system is in stack mode, in order to
prevent the user from disconnecting from the network in the event of master switchover.
To open the User Interface:
1. Open a web browser.
2. Enter the device’s IP address in the address bar and press Enter. An Appendix 1, "Enter Network
Password Page" opens:
Chapter 1:
1
1
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Enter Network Password Page
3. Enter a user name and password. The default user name is “admin.”
with a default password, and can be configured without entering a password. Passwords are both
case sensitive and alpha-numeric.
4. Click Login The Embedded Web System Home Page opens:
Note If you have logged in automatically via the Service Router user interface, the Tree and Device
views appear and allow you to navigate through the various areas of the web interface. However,
the following page will appear within the frame provided by the Service Router user interface.
The device is not configured
2
Chapter 1:
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Embedded Web System Home Page
Chapter
1
Chapter 1:
3
1
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Understanding the Interface
The following table lists the interface components with their corresponding numbers:
Interface Components
Component Description
1
Tree View The Tree View provides easy navigation through the configurable device
features.The main branches expand to provide the subfeatures.
2 Device View The device view provides information about device ports, current
configuration and status, table information, and feature components.The
device view also displays other device information and dialog boxes for
configuring parameters.
3 Table Area The Table area enables navigating through the different device features.
Click the tabs to view all the components under a specific feature.
4 EWS Information The EWS information tabs provide access to the online help, contains
information about the EWS.
Linksys User Interface Components
4
Chapter 1:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
This section provides the following additional information:
•
Device Representation — Provides an explanation of the Linksys user interface buttons, including both
management buttons and task icons.
•
Using the Linksys Management Buttons — Provides instructions for adding, modifying, and deleting device
parameters.
Device Representation
The Linksys home page displays a graphical representation of the device:
Device Representation
1
The Linksys home page contains a graphical SFE2000 and SFE2000P front panel illustration.
Chapter 1:
5
1
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Using the Linksys Management Buttons
Device Management buttons and icons provide an easy method of configuring device information, and
include the following:
Device Management Buttons
Button Name Button Description
Apply Applies changes to the device.
Clear All Counters Clears statistic counters
Clear Logs Clears log files
Add Opens an Add page
Delete Removes entries from tables
Reset the settings of
Selected Port to
Default
Test Now Performs cable tests.
Resets the settlings of a selected port
to the default settings
6
Chapter 1:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Using Screen and Table Options
Linksys contains screens and tables for configuring devices. This section contains the following topics:
• Appendix 1, "Adding Device Information"
• Appendix 1, "Modifying Device Information"
• Appendix 1, "Deleting Device Information"
Adding Device Information
User defined information can be added to specific EWS pages, by opening a new Add page. To add
information to tables or EWS pages:
1. Open an EWS page.
2. Click the Add button. An add page opens, for example, the Add SNTP Server Page:
Add SNTP Server
1
3. Define the fields.
4. Click Apply. The configuration information is saved, and the device is updated.
Modifying Device Information
1. Open the EWS page.
2. Select a table entry.
3. Click the Edit Button. A Modify page opens, for example, the Interface priority Page opens:
Chapter 1:
7
1
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Edit Interface Priority
4. Define the fields.
5. Click Apply. The fields are modified, and the information is saved to the device.
Deleting Device Information
1. Open the EWS page.
2. Select a table row.
3. Check the Remove checkbox.
4. Click the Delete button. The information is deleted, and the device is updated.
8
Chapter 1:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Resetting the Device
The Reset page enables the device to be reset from a remote location. Save all changes to the Running
Configuration file before resetting the device. This prevents the current device configuration from being
lost. To reset the device:
1. Click System > General > Reset. The Reset page opens.
Reset Page
1
2. Click the Reset button.
resetting the entire stack. If the master unit is reset, the device is reset, and a prompt for a user name
and password is displayed.
3. Enter a user name and password to reconnect to the Web Interface, if the stack is not part of a full
Linksys One system. If the stack is part of a Linksys One system, login is automatically done from
the Service Router.
Each unit can be reset individually. Resetting the stack master results in
Logging Off The Device
1. Click . The system logs off. The Embedded Web System Home Page closes.
Chapter 1:
9
1
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
10
Chapter 1:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Managing Device Information
This section provides information for defining both basic and advanced system information. This section
contains the following topics:
• Understanding the Device Zoom View
• Defining General System Information
• Managing Stacks
Understanding the Device Zoom View
The Zoom Page is the main window used for viewing the devices either in stand-alone mode or operating
in a stack. To open the Zoom Page:
1. Click the System > System Management > Zoom. The Zoom Page opens:
Zoom Page
2
The Zoom Page contains the following port indicators:
•
Green — Indicates the port is currently operating.
• Red — indicates the port is not currently operating.
Chapter 2:
11
2
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Defining General System Information
SFE 2000P System Information
The System Information page contains parameters for configuring general device information.
SFE 2000P System Information Page
The System Information page contains the following fields:
• Model Name — Displays the user defined user name.
• System Name — Displays the user configured name of the system. Configured in the Network page.
• System Location — Defines the location where the system is currently running. The field range is
up-to 0-160 Characters.
• System Contact — Defines the name of the contact person.The field range is up to 0-160
Characters.
• System Object ID— Displays the vendor’s authoritative identification of the network management
subsystem contained in the entity.
• System Up Time — Displays the amount of time that has elapsed since the last device reset. The
system time is displayed in the following format: Days, Hours, Minutes and Seconds. For example:
41 days, 2 hours, 22 minutes and 15 seconds.
12
Chapter 2:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
• Base MAC Address — Displays the device MAC address. If the system is in stack mode, the Base
MAC Address of the master unit is displayed.
• Hardware Version — Displays the hardware version number. If the system is in stack mode, the
hardware version of the master unit is displayed.
• Software Version — Displays the software version number. If the system is in stack mode, the
version of the master unit is displayed.
• Boot Version — Indicates the system boot version currently running on the device. If the system is
in stack mode, the version of the master unit is displayed.
• Switch Operation Mode After Reset — Indicates the mode the device operates in after the system
is reset. A switch may operate in one of two modes: stack or standalone. Either mode is selected by
the user during software boot or in the web interface’s System Information page. The new mode
takes effect after the unit is reset. The factory default is stack mode. The possible field values are:
– Standalone — Indicates the device operates as a Standalone device after the system is reset.
A switch operating in standalone mode runs as an independent, single unit. All ports of a
standalone switch operate as normal Ethernet links. A standalone switch does not participate
in a stack even if physically connected to a stack.
2
– Stack — Indicates the device operates as a Stacked unit after the system is reset. A switch
operating in stack mode is not an independent unit, but a member of an organized group of
switches known as a stack. A stack consists of one Stack Master control switch, one Master
Backup switch and up to six Stack Member switches. In some cases, a unit in stack mode
that is not connected to any other units may operate as a “stack–of-one.”
2. Define the relevant fields.
3. Click Apply. The system information is defined, and the device is updated.
Managing Stacks
Stacking allows you to build a switch with many more ports than would be available in a single unit. The
stack is managed by one of the units (called the Stack Master) and all of the other units serve as ports
only.
You can build stacks by building a new stack from a group of switches, or adding new units to an
existing stack. Stacks can be automatically or manually configured.
NOTE: Two ports of each unit in a stack mode (ports 12
and 24 on GE units, and ports G1 and G2 on FE units) are
reserved for stacking links, and cannot be used for regular
network connections.
Chapter 2:
13
2
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Building Automatically-Configured Stacks
Building a New Stack
The easiest way to build a stack is to use a group of switches, each of which is in factory default mode:
NOTE: If the units to be used in building the new stack
have been used previously, we recommend that you reset
them to the factory default by holding the reset button for at
least 10 seconds before using them.
1. Connect the units physically through the stacking ports, using standard Ethernet cables.
2. Power the units on. After a short interval the stack becomes operational, with one of the units
selected as the Stack Master. The unit selected as Stack Master is indicated by a lit green “MST”
LED on its front panel. If a serial console connection is desired, the serial cable should be connected
to the console port of the unit serving as the Stack Master.
Adding Units to a Running Stack
1. Reset the units that will be added by restoring them to the factory default mode. Connect the units
physically to the stack.
2. Power the units on. After a short interval, they will become members of the stack.
Building Manually-Configured Stacks
You can manually configure stacks, including choosing a specific unit as the Stack Master. You must
assign a unique Unit ID (from 1 to 8) to each stack member.
Building a New Stack
1. Reset all relevant units to by restoring them to the factory default mode.
2. Connect the units physically through the stacking ports, using standard Ethernet cables.
3. Assign each unit its desired number, making sure no duplicates exist, and reset the stack.
Adding Units to a Running Stack
1. Reset the units to be added by restoring them to the factory default mode.
2. Connect the units physically to the stack.
14
Chapter 2:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
3. Power the units on. After a short interval, they will become stack members, but will have
automatically-assigned Unit IDs. Assign each such unit its desired Unit ID (using the Stack
Management Interface through the console port, by Telnet, or by using the graphical user interface
(GUI).
4. Reset the units to make this assignment permanent.
NOTE: We recommended that if you manually assign a
Unit ID to one unit, you manually assign Unit IDs to all
units. Using a mix of both system-assigned and manuallyassigned IDs in your network can impact system
performance.
The unit that is assigned the Unit ID 1 is the Stack Master, and its front panel “MST” LED lights green.
The unit assigned the Unit ID 2 is the Backup Master.
Understanding Stack Resiliency
2
Stacks can be configured in ring or chain topologies. We recommend configuring the stack in ring
topology, due to the high resiliency in case of unit failure or stacking links failure.
Additionally, if a redundant power supply is present, we recommend connecting the Stack Master and
Backup Master units to the redundant power supply.
Understanding Advanced Stacking
To understand advanced stacking, you must understand Unit IDs and how they are allocated, and the
stack unit startup process.
Unit IDs
Each unit in a stack has an assigned unique Unit ID number. The following sections describe the Unit
IDs and their characteristics.
Stack Master
The unit assigned the Unit ID number 1 serves as the Stack Master. All other units are stack members.
The Stack Master provides a single point of control, configuration and management for the entire stack,
and stores the configuration for all stack members. (Members do not store any configuration
information.)
Stack Backup Master
The unit assigned the Unit ID number 2 is a special stack member that serves as the stack Backup
Master. A stack Backup Master assumes the role of Stack Master for the remaining stack members if the
stack Master fails or is disconnected.
Chapter 2:
15
2
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
The Stack Master stores a copy of the active configuration on the Backup Master. This copy is used only
if the Backup Master assumes the role of Stack Master.
NOTE: Only the configuration file is copied. Any
dynamically-filled tables (for example, learned addresses)
are not copied from the Stack Master to the Backup Master.
If the Backup Master assumes the role of Stack Master, it
builds its own dynamic tables.
Stack Members
The units assigned the Unit IDs 3 through 8 are called stack members. A stack member operates only as
a member of the stack under the direction of an operational Stack Master (or a Backup Master that has
assumed the Stack Master role). Stack members are not directly manageable and configurable, and must
be managed through the Stack Master. They do not contain any meaningful configuration information,
including their own configuration. If an operational Stack Master is not present and reachable, these
units are not functional.
Master-Enabled Units
Units that are assigned a Unit ID number of 1 or 2 are called master-enabled units. Only master-enabled
units participate in the Master Election process (see below) when they are initialized, are inserted into a
new stack, or lose connectivity with the existing Stack Master. Only master-enabled units participate in
the Master Election process and can become the Stack Master or Backup Master. (Units that are assigned
a Unit ID of 3 through 8 can only become a Stack Master or a Backup Master if they are manually
configured by the system administrator or if they are reset to the factory default mode.)
Unit ID Allocation
Units are shipped from the factory without an assigned Unit ID, and must be assigned a unique Unit ID
before they can operate as part of a stack. Unit ID numbers are assigned to units in one of two ways:
• Unit ID numbers are assigned by the system administrator, and can be changed only manually by
the system administrator.
• Unit ID numbers are allocated to a stack member unit by the Stack Master during system
initialization.
A unit that was assigned a Unit ID will usually keep this number even after it is rebooted. The Stack
Master may reallocate Unit IDs during system initialization to resolve duplicate Unit ID conflicts (see
below). Manually assigned Unit IDs cannot be changed by the Stack Master, even if there is a conflict.
Unit ID assignment or change takes effect only during system initialization and does not take place
during system runtime. Units of a stack do not have to be numbered in sequence, and can be
interconnected as long as each unit has a unique ID and at least one unit of the stack serves as Stack
Master.
16
Chapter 2:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Stack Unit Startup Process
When a unit in stack mode is initialized (powered up or rebooted), it goes through the following steps:
1. The Master Discovery and Master Election processes.
2. Unit ID allocation by the Stack Master (including duplicate Unit ID conflict resolution).
3. Unit and port configuration by the Stack Master.
Master Discovery Process
When a unit in stack mode initializes, its behavior depends on its Unit ID (if one is configured):
• If the unit does not have a current Unit ID (that is, the unit is in factory default mode) and if there
is a Stack Master, the unit is allocated a Unit ID number from the Stack Master. If there is no
Stack Master, then the unit participates in the Master Election process, and may be chosen as the
new Stack Master or Backup Master.
• If the unit’s current Unit ID is 1 or 2 (that was previously allocated, even if used in a different
stack), then the unit participates in the Master Election process.
2
• If the unit has a current Unit ID (that was previously allocated, even if used in a different stack),
the unit tries to act according to its Unit ID number in the new stack. For example, if the unit’s
current Unit ID is 3 through 8, it will try to connect to the running Stack Master, and will not
proceed to the next stage until contact with the Stack Master is made. These units will not
participate in the Master Election process, and if no Stack Master is present, the units are
effectively shut down.
The Stack Master and all other stack units carry out a continuous process of Master Discovery by
frequently exchanging stack control messages. This allows units to know if another unit fails or becomes
unreachable.
Master Election Process
When units in stacking mode initialize, one of the units is elected as the Stack Master. If a unit in the
stack was set to “Force Master” by the system administrator, that unit is elected as the Stack Master.
Only master-enabled stack units (for example, those with the Unit ID of 1 or 2) can be configured as
“Force Master.”
If the stack contains units whose unique Unit ID is 1 or 2, then one of these two units will be the Stack
Master. It does not matter if the Unit ID was originally assigned automatically or manually. These units
are called master-enabled units. If there is only one master-enabled unit, it will be elected as the Stack
Master (even if its Unit ID is 2).
If there are two master-enabled units, the two units decide which of them is the Master by checking
which one has been running for a longer time (in intervals of 10 minutes). The unit that has been running
for the longer time will be the Stack Master. If they have been running for the same amount of time, the
unit with the Unit ID of 1 will be the Stack Master. If both units have been running for the same amount
of time and both units have the same Unit ID, the unit with a lower MAC (hardware) address will be
selected as the Stack Master.
Chapter 2:
17
2
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
If the stack contains one or more units without a current Unit ID (the units are in factory default mode),
then one of these units will be the Stack Master. The unit selected to be the Stack Master is the one
running for the longest time (in intervals of 10 minutes), or, if all units are running for the same amount
of time, the one with the lowest MAC (hardware) address.
The Master Election process ensures that the stack has a Stack Master. The Stack Master has the Unit ID
of 1 and the Backup Master, if it exists, has the Unit ID of 2. Alternatively, the Stack Master has the Unit
ID of 2 and the Backup Master, if it exists, has the unit ID of 1.
If a master-enabled unit is added to a stack and powered on, when it comes up it invokes the Master
Election process, even though the rest of the stack already has an elected master. Because the unit is new,
it loses the election and joins as a stack member or Backup Master.
Unit ID Allocation and Duplicate Unit ID Conflict Resolution
After a Stack Master is elected, it allocates Unit IDs to units that do not have a currently assigned Unit ID
(units that are in factory default mode). The Stack Master also attempts to resolve all cases of units with
duplicate Unit IDs. The Stack Master changes the Unit IDs of units that have a duplicate current Unit ID,
provided that there are available, unused Unit IDs. In a merged stack, if the Stack Master unit remains as
the Stack Master, units that were in its group will keep their unit IDs. Members of other groups are
renumbered.
If the conflict occurs after the units reboot, the conflict is resolved as follows:
• If both duplicate units are in auto (self ordering) mode, then the unit ID with the lower MAC
(hardware) address will keep its unit ID. The other unit is assigned a new unit ID.
• If one of the duplicates is in auto (self ordering) mode, and the other unit is in manual mode,
then the manual mode unit will keep its ID and the other is assigned a new unit ID.
• If both duplicate units are in manual mode, then both of them are shut down.
If the Stack Master is able to allocate a unique Unit ID to each unit, then all units can operate as a stack.
If the Stack Master is unable to allocate a Unit ID to a unit, that unit is effectively shut down and will not
participate in the stack. For example, units with a conflicting manually-set Unit ID number are shut
down because the Stack Master cannot override the system administrator’s assignment and resolve the
conflict.
If there are more units than the maximum number allowed in a stack, and the incoming units are already
in factory default mode (they do not have unit ID assigned), then a Stack Master is elected following the
Master Discovery and Master Election processes. All other units remain shut down.
18
Chapter 2:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Occasionally, due to a race condition during the boot process, some of the units might be connected and
join the stack. If the incoming units already have a unit ID, then none of them will join the stack and all
are left in shutdown mode because there is no way for the Stack Master to determine their Unit ID
preference.
NOTE: If a unit is shut down, its stacking links are inactive.
If the stacking units are connected in a chain topology, the
shutdown of one unit breaks the chain and can cause other
units to be shut down if they have no active link to the Stack
Master unit.
Configuring Units and Ports
After the Master Discovery and Master Election processes, each unit in the stack has a unique Unit ID,
one of the units is the Stack Master, and one of the units may serve as the Backup Master. The Stack
Master then configures each of the member units and its ports according to the configuration file present
on the Stack Master. If the stack has a Backup Master, the configuration file is copied to the Backup
Master.
2
After all the units and ports are configured, the stack enters normal operational mode. If a change is
made to the system configuration, the change is stored by the Stack Master and is copied to the Backup
Master if one exists.
You can use the command-line interface (CLI) or GUI to configure the stack units.
Setting the Unit’s Operational Mode
Use the GUI to set the unit’s operational mode to standalone or stack. This configuration takes effect
after the next reboot.
Configuring the Stack Master and Unit ID
The Stack Management Page allows network managers to either reset the entire stack or a specific
device. Device configuration changes that are not saved before the device is reset are not saved. If the
Stack Master is reset, the entire stack is reset.
To open the Stack Management Page:
1.
Click System > System Management > Stack Management. The Stack Management Page opens:
Chapter 2:
19
2
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Stack Management Page
The Stack Management Page contains the following fields:
• Master Election — Indicates the method of electing the Stack Master device. The possible values
are:
– Automatically — The master is selected automatically by software.
– Force Master — The unit is forced to be master of the stack. Note that only units with the
Unit ID of 1 or 2 can be the stack master.
• Stacking Ports After Reset — Allows the user to decide what cable type is in use. The possible
values are:
– Combo Ports — Indicates that the combo port is used as the stacking port.
– Copper Ports — Indicates that the copper port is used as the stacking port.
• Unit No.— Displays the current Stacking Master.
• Unit No. After Reset — Indicates the stacking member elected Stacking Master after the device is
reset.
2. Define the relevant fields.
3. Click Apply. Stack management is defined, and the device is updated.
20
Chapter 2:
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
Resetting the Unit to Factory Default Mode
To reset the unit to the factory default settings, press the front panel RESET button (see Figure 3.) The
unit is set to Stack mode with a Unit ID of 0.
Understanding LED Indicators
Each unit contains a Master LED indicator and eight unit LEDs. The LED status definitions are shown in
the table below.
2
LED Mode Color Description
Master Solid Green The switch is the Stack Master.
Off N/A The switch is not the Stack Master or the switch is
not stacked.
ID n Solid Green The switch is Unit ID n.
Off N/A The switch is not Unit ID n or the switch is not
stacked.
All ports Solid Red The switch is powered on, but not operational.
Stack Troubleshooting and Maintenance
Replacing a Failed Member Stack Unit in an Operational Stack
If a unit that is not the Stack Master fails in an operational stack, the Stack Master discovers that the unit
is no longer responding during the Master Discovery process. The Stack Master directs all other stack
members to route unit-to-unit traffic around the failed unit using the ring topology of the stacking
connections. Concurrently, the Stack Master notifies the system administrator of the failure by sending
SYSLOG messages and SNMP traps.
Because all traffic has been routed around the failed unit, when it is disconnected from the stack, the
stack continues to run as long as all other stacking connections are left intact.
Chapter 2:
21
2
Chapter
SFE2000/SFE2000P Gigabit Ethernet Switch Reference Guide
When a new unit is inserted in the stack and powered on, the following occurs:
1. The incoming unit, which is in stack mode, performs the Master Discovery process, and may
participate in the Master Election process.
• If the incoming unit has a Unit ID of 1 or 2 (it is a master-enabled unit) it initiates the Master
Election process. However, because the running Stack Master has a longer runtime, the current
Stack Master retains its position and the incoming unit does not become the new Stack Master.
• If the incoming unit has a Unit ID of 3 through 8, it attempts to become a member unit of the
stack, subject to control by the already running Stack Master, and the Master Election process
does not occur.
2. The Stack Master performs Unit ID allocation and the conflict resolution process.
• If the incoming unit did not have an assigned Unit ID (that is, it is in factory default mode), it is
assigned the lowest available Unit ID by the Stack Master. We recommend that you use the
automatically-assigned unit ID mode because it provides better resiliency to the stack.
• If the incoming unit already has an assigned Unit ID, and that Unit ID is unused in the current
stack, the incoming unit keeps its assigned Unit ID and the Stack Master applies any
configuration relevant to that Unit ID to the incoming unit.
• If the incoming unit already has an assigned Unit ID, and that Unit ID conflicts with a unit ID in
the current stack, the Stack Master allocates a new Unit ID to the incoming unit, giving it the
lowest available Unit ID. However, if the incoming unit has a manually assigned Unit ID, the
Stack Master cannot change it. If the incoming unit cannot be assigned an available Unit ID,
then it is shut down and is not joined to the stack.
3. The Stack Master performs unit and port configuration for the incoming unit.
• Any configuration information that is relevant to the number assigned to the incoming unit is
applied by the Stack Master. For example, if the incoming unit is assigned the same Unit ID of
the unit it replaced, then when possible, it receives the same configuration as the failed unit.
• If the incoming unit is identical to the replaced unit, the entire configuration of the replaced unit
is applied to the incoming unit and the stack returns to the state it was in before unit failure.
If the incoming unit is not identical to the unit that failed, the Stack Master applies the configuration in
the following manner:
• If a 24-port unit replaces a failed 48-port unit, the ports of the incoming unit are configured
according to the configuration of the first 24 ports of the failed unit.
NOTE: The configuration of all 48 ports of the failed unit is
kept in memory, even though the first 24 are currently
applied. If, in the future, a 48 port unit is inserted and
assigned the same Unit ID, it is configured the same as the
original failed 48-port unit.
22
Chapter 2:
Loading...