Cisco Catalyst 3850 Series and Cisco
Catalyst 3650 Series Switches Best
Practices Guide
First Published: November 30, 2015
Last Updated: December 14, 2015
Cisco Systems, Inc.
www.cisco.com
Cisco has more than 200 offices worldwide.
Addresses, phone numbers, and fax numbers
are listed on the Cisco website at
www.cisco.com/go/offices.
Page 2
Page 3
THE SPECIFICATIONS AND INFORMATION REGARDING THE PRODUCTS IN THIS MANUAL ARE SUBJECT TO CHANGE WITHOUT NOTICE. ALL
STATEMENTS, INFORMATION, AND RECOMMENDATIONS IN THIS MANUAL ARE BELIEVED TO BE ACCURATE BUT ARE PRESENTED WITHOUT
WARRANTY OF ANY KIND, EXPRESS OR IMPLIED. USERS MUST TAKE FULL RESPONSIBILITY FOR THEIR APPLICATION OF ANY PRODUCTS.
THE SOFTWARE LICENSE AND LIMITED WARRANTY FOR THE ACCOMPANYING PRODUCT ARE SET FORTH IN THE INFORMATION PACKET THAT
SHIPPED WITH THE PRODUCT AND ARE INCORPORATED HEREIN BY THIS REFERENCE. IF YOU ARE UNABLE TO LOCATE THE SOFTWARE LICENSE
OR LIMITED WARRANTY, CONTACT YOUR CISCO REPRESENTATIVE FOR A COPY.
NOTWITHSTANDING ANY OTHER WARRANTY HEREIN, ALL DOCUMENT FILES AND SOFTWARE OF THESE SUPPLIERS ARE PROVIDED “AS IS” WITH
ALL FAULTS. CISCO AND THE ABOVE-NAMED SUPPLIERS DISCLAIM ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING, WITHOUT
LIMITATION, THOSE OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OR ARISING FROM A COURSE OF
DEALING, USAGE, OR TRADE PRACTICE.
IN NO EVENT SHALL CISCO OR ITS SUPPLIERS BE LIABLE FOR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, OR INCIDENTAL DAMAGES, INCLUDING,
WITHOUT LIMITATION, LOST PROFITS OR LOSS OR DAMAGE TO DATA ARISING OUT OF THE USE OR INABILITY TO USE THIS MANUAL, EVEN IF CISCO
OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
Cisco and the Cisco logo are trademarks or registered trademarks of Cisco and/or its affiliates in the U.S. and other countries. To view a list of Cisco trademarks, go to this
URL: www.cisco.com/go/trademarks. Third-party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership
relationship between Cisco and any other company. (1110R)
Any Internet Protocol (IP) addresses and phone numbers used in this document are not intended to be actual addresses and phone numbers. Any examples, command display
output, network topology diagrams, and other figures included in the document are shown for illustrative purposes only. Any use of actual IP addresses or phone numbers in
illustrative content is unintentional and coincidental.
Obtaining Documentation and Submitting a Service Requestviii
Ease of Deployment1-1
Purpose1-1
Configuration Tool1-2
Catalyst Switch Configuration Best Practices1-2
LAN Access Switch Topology1-4
Switch Address Plan1-5
Initial Switch Configuration2-7
Purpose2-7
Prerequisites2-7
Identify Configuration Values2-8
Assign Initial Management Information2-8
Configure the Hostname for Switch Identification2-9
Configure Secure HTTPS and Secure Shell for Secure LAN Management2-9
Configure SNMP for Remote Management2-10
Configure Local Login and Password for Switch Access2-10
Configure Centralized User Authentication Through TACACS+2-10
Assign an IP Address to the Switch2-11
Configure the Management IP Address on an Out-of-Band Interface2-12
Configure the Management IP Address on an In-Band Interface2-14
Create a Management VLAN in Hardware2-15
Verify Basic Switch Configuration2-17
Show Running Configuration for Initial Management Information2-17
Switch Stack Update3-21
Purpose3-21
Prerequisites3-21
Identify Configuration Values3-22
Cisco Catalyst 3850 Series and Cisco Catalyst 3650 Series Switches Best Practices Guide
i
Page 6
Contents
LAN Access Switch Topology with Configured FTP Server3-22
Performing the Stack Update3-23
Obtain the Switch Software Image3-23
Check the Software Version on the Stack Members3-23
Configure the Switch to Run in Install Mode3-24
Download the Switch Image from Cisco.com to a FTP Server3-25
Update the Switch Stack Image3-27
Enable Switch Image Auto-Upgrade3-27
Verify that Stack Members Are Running the Same Software Image3-28
Global System Configuration4-29
Purpose4-29
Prerequisites4-29
Identify Configuration Values4-29
Assign Global Configuration Information4-30
Configure High Availability on the Switch Stack4-31
Configure VTP Transparent Mode4-31
Enable Rapid Per-VLAN Spanning Tree4-32
Configure BPDU Guard for Spanning-Tree PortFast Interfaces4-32
Configure UDLD to Detect Link Failure4-33
Configure an Access List to Limit Switch Access4-33
Configure System Clock and Console Timestamps4-34
Configure DHCP Snooping Security Features4-34
Configure ARP Inspection4-34
Configure EtherChannel Load Balancing4-35
Create Access Layer VLANs4-35
Create IPv6 First Hop Security Policies4-35
Increase the TFTP Block Size4-36
Enable New Members to Automatically Update to the Switch Stack Image4-36
Verify Global Switch Configuration4-37
Show Running Configuration For Global Management Information4-37
Uplink Interface Connectivity5-41
Purpose5-41
Prerequisites5-41
Restrictions5-41
Identify Configuration Values5-42
LAN Access Switch Topology with Uplinks to a Distribution Switch or Distribution Router5-43
Configure Uplink Interface Connectivity5-44
Cisco Catalyst 3850 Series and Cisco Catalyst 3650 Series Switches Best Practices Guide
ii
Page 7
Recommendations for Configuring the Uplink Interface to a Router or Switch5-44
Configure QoS on the Uplink EtherChannel Interfaces5-44
Configure the Uplink Interface as an EtherChannel and as a Trunk5-45
Configure the Uplink Interface to Connect to Distribution VSS or VPC Switches5-45
Configure the Uplink Interface to Connect to Distribution Routers (or Standalone Distribution
Switches)
5-46
Configure Security Features on the Uplink EtherChannel Interfaces5-48
Spanning-Tree Recommendations for Uplink Interfaces Connecting to Distribution Switches5-48
Verify Uplink Interface Configurations5-49
Show Running Configuration for Uplink Interface Connectivity5-49
Access Interface Connectivity6-51
Purpose6-51
Prerequisites6-51
Identify Configuration Values6-51
Contents
LAN Access Switch Topology with Connections to End Devices6-53
Configure Access Interface Connectivity6-53
Recommendations for Configuring Access Interfaces6-53
Configure the Interface for Access Mode6-55
Configure VLAN Membership6-55
Create an Interface Description6-55
Configure Security Features on Access Interfaces6-56
Configure QoS on the Access Interfaces6-57
Verify Access Interface Configurations6-58
Show Running Configuration for Access Interface Connectivity6-61
Access Control on the Wired Network7-65
Purpose7-65
Prerequisites7-65
Restrictions7-65
Identify Configuration Values7-66
LAN Access Switch Topology with IEEE 802.1x Secure Access Control7-67
Provision IEEE 802.1x for Wired LAN7-67
Recommendations for Configuring Security on a Wired LAN7-67
Provision Common Wired Security Access7-68
Provision in Monitor Mode7-71
Provision in Low Impact Mode7-72
Provision in High Impact Mode7-73
Verify Secure Access Control on the Switch7-74
Cisco Catalyst 3850 Series and Cisco Catalyst 3650 Series Switches Best Practices Guide
iii
Page 8
Contents
Show Running Configuration for Provisioning Modes7-74
Monitoring IEEE 802.1x Status and Statistics7-77
Converged Wired and Wireless Access8-81
Purpose8-81
Prerequisites8-81
Restrictions8-82
Identify Configuration Values8-82
LAN Access Switch Topology with Wireless Connectivity8-83
Enable the Switch as a Wireless Controller8-84
Install Access Point Licenses on the Switch8-84
Verify AP-Count License Installation8-85
Configure a Wireless Management VLAN8-86
Configure Service Connectivity8-86
Enable Wireless Controller Functionality8-87
Change a Switch to Run in Mobility Controller Mode8-87
Enable the Access Point Connections8-88
Enable a Client VLAN8-89
Provisioning a Small Branch WLAN8-90
Provision in Easy-RADIUS8-90
Disable Authentication to Enable Easy-RADIUS8-90
Configure QoS to Secure the WLAN8-91
Verify Client Connectivity in RADIUS8-91
Provision in Secure Mode8-93
Enable the AAA RADIUS Server8-93
Configure the WLAN with IEEE 802.1x Authentication8-94
Configure QoS Service Policies for an Open WLAN8-94
Obtain WLAN Client IP Addresses8-95
Manage Radio Frequency and Channel Settings8-95
Disable Low Data Rates8-96
Enable Clean Air8-97
Enable Dynamic Channel Assignment8-97
Associate WLAN Clients8-98
Verify WLAN Client Connectivity8-98
Verify the Converged Access Configuration on the Switch8-99
Show Running Configuration for Wireless LAN Converged Access8-99
iv
System Health Monitoring9-103
Purpose9-103
Cisco Catalyst 3850 Series and Cisco Catalyst 3650 Series Switches Best Practices Guide
Page 9
I
NDEX
Contents
Prerequisites9-103
Show Running Status9-103
Run a System Baseline for Core Resources9-104
Obtain CPU and Core Processor Usage9-104
Obtain Switch Memory Usage9-106
Monitor File Systems Usage9-106
Run a System Baseline for Environmental Resources9-107
Other System Monitoring Considerations9-108
Spanning Tree Monitoring9-108
Cisco Catalyst 3850 Series and Cisco Catalyst 3650 Series Switches Best Practices Guide
v
Page 10
Contents
vi
Cisco Catalyst 3850 Series and Cisco Catalyst 3650 Series Switches Best Practices Guide
Page 11
Preface
Audience
This document is written for managing the Cisco Catalyst 3850 Series Switches and the Cisco 3650
Series switches and switch stacks in their network. A basic understanding of Ethernet networking is
expected. Cisco Certified Network Associate level (CCNA) knowledge is helpful, but not required.
Conventions
This document uses the following conventions:
ConventionIndication
italic blue fontExample configuration values that are replaced with reader values.
bold fontCommands and keywords and user-entered CLI appear in bold font.
italic fontDocument titles, new or emphasized terms, and arguments for which you supply values are in italic
font.
[ ]
• Default responses to system prompts are in square brackets.
• Elements in square brackets are optional.
{x | y | z}Required alternative keywords are grouped in braces and separated by vertical bars.
[x | y | z]Optional alternative keywords are grouped in brackets and separated by vertical bars.
stringA nonquoted set of characters. Do not use quotation marks around the string, or the string will include
the quotation marks.
courier fontTerminal sessions and information the system displays appear in courier font.
< >Nonprinting characters such as passwords are in angle brackets.
!, #An exclamation point (!) or a pound sign (#) at the beginning of a line of code indicates a comment line.
NoteMeans reader take note. Notes contain helpful suggestions or references to material that is not covered
in the manual.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
vii
Page 12
Audience
TipThe tips information might not be troubleshooting or even an action, but could be useful information,
similar to a Timesaver.
TimesaverYou can save time by performing the action described in the paragraph.
Obtaining Documentation and Submitting a Service Request
For information on obtaining documentation, using the Cisco Bug Search Tool (BST), submitting a
service request, and gathering additional information, see What’s New in Cisco Product Documentation
at: http://www.cisco.com/c/en/us/td/docs/general/whatsnew/whatsnew.html.
Subscribe to What’s New in Cisco Product Documentation, which lists all new and revised
Cisco technical documentation as an RSS feed and deliver content directly to your desktop using a reader
application. The RSS feeds are a free service.
viii
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
OL-xxxxx-xx
Page 13
Ease of Deployment
This document describes best practices for deploying your Cisco Catalyst 3850 Series and Cisco
Catalyst 3650 Series switches.
NoteUnless otherwise noted, the term switch refers to a standalone Catalyst 3850 switch, a Catalyst 3650
switch, or a switch stack.
A Cisco switch deployment best practice is a preferred configuration method to employ on your Catalyst
switches. It is a proven and tested way to improve network security, performance, and availability.
A best practice configuration includes an explanation of why you should perform a given task and a
sample snapshot of a full running configuration that you can extrapolate for your specific scenario.
TipUse the configuration recommendations in this document as a template for your switch deployments.
NoteMany Cisco documents are available that define best practices for a variety of features and solutions.
There will be some overlap between the information provided in this guide and other best practices and
deployment guides. When relevant, this document references other existing documents so the reader can
get a deeper understanding of an aspect of the 3850 operation. Otherwise, this document is
self-contained, and provides complete best practice configuration.
Configuration Tool
The configuration examples in this document use the Cisco IOS CLI configuration tool, which is the
most common tool used to configure a switch.
However, you do have the flexibility to use a different tool to perform switch configuration. Other
configuration tools are the Express Setup, Device Manager, and Cisco Prime.
The examples provided in this document show the CLI commands that you should execute on your
switch. You must replace the blue italicized example values with your own values.
Cisco Systems, Inc.
www.cisco.com
Page 14
LAN Access Switch Topology
set system location Building 1, San Jose, CA
Dual redundant switches
in distribution layer running
VSS (Cat6500/6800/4500),
or VPC (Nexus 7000)
Desktop user
direct connect
Desktop user
Printer
Wireless access
Catalyst 3850 stack in access
Voice VLAN 11
Data VLAN 10
Data VLAN 10
Data VLAN 10
Switch management
VLAN 100
Access point VLAN 12
Trunk link
Native VLAN 999
All VLANs included
LAN Access Switch Topology
The workflows described in this document assume that a switch is deployed as a LAN access switch.
Unless noted otherwise, a switch that is in the LAN access layer is configured as a Layer 2 switch, with
all Layer 3 services provided by the directly connected distribution switch or router.
This document assumes that the switches are stacked together to form a switch stack (a common
switching unit). We recommend that you use switch stacks because of built-in redundancy. We also
recommend the use of using switch stacks when deploying switches in converged access mode (wireless
mode) and connecting access points to different stack members.
A switch deployed at the LAN access layer provides high-bandwidth connections to devices through
10/100/1000 Ethernet, with both Gigabit and 10-Gigabit uplink connectivity options.
When a switch is deployed in access mode, it enables end devices, such as IP phones, wireless access
points, and desktops to gain access to the network. The Power over Ethernet (PoE) switch models
support PoE+ (30 W) and UPoE (60 W) to power IP phones, wireless access points, and IP cameras. The
field-replaceable uplink module from the switch enables different uplink connectivity types.
Ease of Deployment
Figure 1 shows an enterprise campus deployment, where the switch is connected to a distribution layer
switch (such as a Catalyst 6500,6800,4500 or a Nexus 7000 switch).
Figure 1LAN Access Switch Topology with Distribution Switch
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
2
Page 15
Ease of Deployment
Cisco Catalyst Switch Configuration Workflow
Figure 2 shows a branch deployment, where the switch is connected to a router (ISR). Because the switch
operates as a Layer 2 switch, not many differences occur in the configuration between the campus or
branch deployment cases. Differences in the configuration are noted in the best practice procedures.
Figure 2LAN Access Switch Topology with Distribution Router
Voice VLAN 11
Data VLAN 10
Desktop user
behind IP phone
Data VLAN 10
Desktop user
direct connect
Data VLAN 10
Printer
Wireless access
Catalyst 3850 stack in access
Switch management
VLAN 100
Trunk link
Native VLAN 999
All VLANs included
Access point VLAN 12
Cisco Catalyst Switch Configuration Workflow
This document focuses on configuring a switch network and is organized in a workflow pattern,
beginning with the initial configuration of a switch after it is racked, mounted, connected, and powered
on, and ending with monitoring system health.
Figure 3 shows the best-practice configurations described in this document.
See the Switch Hardware Installation Guide for information on how to install a switch.
Dual redundant
routers running
HSRP
391638
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
4
Page 17
Ease of Deployment
Switch Address Plan
The VLAN IDs and IP addresses designated for a switch and used throughout this document are not a
component of practices; they are only specified for the configuration examples. Your deployment will
have an IP address plan that suits your specific network.
In this document, all IP address ranges are /24 for the sake of simplicity. We recommend that VLAN IDs
be reused across the access switches deployed.
For example, in the access layer, VLAN 10 is always used for data, and VLAN 11 is always used for
voice. The IP subnets for those VLANs are different across the access switches, but the VLAN IDs are
the same. This type of address plan makes it easier to operate the network because the same VLAN IDs
are consistent.
Table 1IP Address Plan
VLAN IDIP Address ServerDescription
100192.168.1.0/24—Switch in-band management VLAN.
10192.168.10.0/24Upstream deviceAccess data VLAN for end devices
11192.168.11.0/24Upstream deviceAccess voice VLAN for IP phones
12192.168.12.0/24Catalyst 3850 switchAccess point VLAN and subnet.
200192.168.13.0/24Upstream deviceWireless client VLAN and subnet.
—192.168.254.0—IP address range for all central
Switch Address Plan
and subnet.
and subnet.
services. The services are not
physically adjacent to the switch.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
5
Page 18
Switch Address Plan
Ease of Deployment
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
6
Page 19
Switch Stack Update
This workflow explains how to update all members of a switch stack with the same software image.
Before proceeding with global and advanced configurations on a switch stack, all stack members must
be running the same Cisco IOS XE release to avoid mismatch issues. In addition, any new switch that
needs to join the switch stack must also be running the same Cisco IOS XE release; otherwise, the switch
stack will not converge and the new switch will remain in a standalone state.
NoteUpdating a Catalyst 3850 or 3650 switch stack is different from updating a Catalyst 3750 switch stack.
Simply changing the boot statement to the desired .bin file is not recommended for Catalyst 3850 and
3650 switch stacks. The update process for Catalyst 3850 and 3650 switch stacks includes a series of
package files, which are extracted from the .bin file and loaded into flash.
Prerequisites
• Obtain a valid Cisco Connection Online (CCO) account with entitled credentials.
• The process to install the new IOS version will use either FTP or TFTP. This requires a FTP or TFTP
• Install and configure the TFTP or FTP before you begin.
• Verify that the TFTP block size is set at the maximum value of 8192, as described in the “Increase
server be available to host the 3850 IOS Software, and the server reachable over an IP network.
the TFTP Block Size” section.
Cisco Systems, Inc.
www.cisco.com
Page 20
Identify Configuration Values
Identify Configuration Values
We recommend that you identify certain switch configuration values in advance so that you are ready to
proceed with this section without interruption. As you follow the configuration sequence, replace the
values in column B with your values in column C.
NoteIn the configuration examples, you must replace the blue italicized example values with your own values.
Table 1 Switch Stack Update Configuration Values
A. Value NameB. Example Value NamesC. Your Value
hostname
TFTP server
Flash file
Switch Stack Update
3850-access-Bld1Flr1
192.168.254.12
cat3k_caa-universalk9.SSA
.16.1.0.EFT3-1.bin
NoteConfiguration examples begin in global configuration mode unless noted otherwise.
LAN Access Switch Topology with Configured TFTP Server
Figure 1LAN Access Switch Topology with Configured TFTP Server
Voice VLAN 11
Data VLAN 10
Desktop user
Data VLAN 10
Desktop user
direct connect
Data VLAN 10
Printer
Catalyst 3850 stack in access
Switch management
VLAN 100
Dual redundant switches
in distribution layer running
VSS (Cat6500/6800/4500),
or VPC (Nexus 7000)
Trunk link
Native VLAN 999
All VLANs included
FTP SERVER
172.18.121.121
Access point VLAN 12
Wireless access
Performing the Stack Update
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
23
391700
Page 21
Switch Stack Update
• Obtain the Switch Software Image
• Check the Software Version on the Stack Members
• Configure the Switch to Run in Install Mode
• Installing IOS image from local TFTP/FTP server
• Update the Switch Stack Image
•
NoteThe following tasks are to be performed in a sequence that is listed here.
Obtain the Switch Software Image
We recommend that you review the appropriate switch release notes before installation to ensure
compatibility with your network topology. Each platform on Cisco.com has a Cisco-suggested release
based on software quality, stability, and longevity, which is designated by the symbol, as displayed
in Appendix 2, “Cisco Catalyst 3850-48P-S Switch”
Performing the Stack Update
Step 1Download the desired .bin file from Cisco.com to the switch flash storage.
NoteThe purpose of this example is only to show you how the Cisco-suggested release symbol is designated,
and not to give you recommended release versions because those change over time.
.
Figure 2Cisco Catalyst 3850-48P-S Switch
Check the Software Version on the Stack Members
Step 2Verify the running software version.
Configure the Switch to Run in Install Mode
Your switches should run in install mode while in production. This mode is not a requirement, but the
update procedure is different if your switches are running in a mode other than install mode.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Step 3If your switch stack is running in bundle mode, use the request platform software package expand
switch file to flash command to convert it to install mode.
Step 4After the .bin file has successfully extracted to flash, change the boot statement and boot to the
packages.conf file.
Switch Stack Update
NoteSince the format of the pacakges.conf file has changed in Cisco IOS XE Release Denali 16.1, overwrite
the old packages.conf with the new packages.conf file. Perform the above step for eachswitch in your
stack. If you have a 3 member stack, it will need to be done on flash:, flash-2:, and flash-3.
NoteMake sure the tftp server is reachable. To improve performance, increase the tftp block size to 8192. Use
the ip tftp blocksize bytes command in global configuration mode.
25
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 23
Switch Stack Update
Switch# show version
Cisco IOS Software, Catalyst L3 Switch Software (CAT3K_CAA-UNIVERSALK9-M), Version
Denali 16.1.1, RELEASE SOFTWARE (fc1)
Technical Support: http://www.cisco.com/techsupport
Copyright (c) 1986-2015 by Cisco Systems, Inc.
Compiled Thu 12-Nov-15 16:23 by mcpre
Step 5Confirm that the switch stack is now running in install mode.
Performing the Stack Update
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
26
Page 24
Performing the Stack Update
# show run | inc block
ip tftp blocksize 8192
ping 192.168.254.12
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.254.12, timeout is 2 seconds: !!!!
Installing IOS image from local TFTP/FTP server
You can use any file transfer method that you are familiar with, but we recommend TFTP or FTP.
Step 6Confirm the block size config using the following command:
We recommend that you use a TFTP block size of 8192 (maximum allowed value) before attempting to
use TFTP or FTP to transfer a file to the switch. Refer to the “Increase the TFTP Block Size” section in
the “Global System Configuration” workflow for details.
Step 7Make sure that there is connectivity to the TFTP server.
In this example, a TFTP server is used that is accessible through the in-band network.
Switch Stack Update
Step 8After verifying connectivity, make sure that there is enough room in flash on all the switch stack
members.
Step 9If you determine that files must be purged from flash, run the request platform clean switch command
to erase unneeded files within flash on all the stack members.
We recommend using the request platform clean switch command instead of individually deleting
files. The command provides a list of the files to purge so that you understand what files are deleted when
you confirm deletion.
NoteUse switch all option to clean up all switches in your stack.
NoteThe request platform clean switch command also deletes the .bin file that is used to install the new
Cisco IOS software. After the .bin is extracted, you no longer need it.
27
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 25
Switch Stack Update
Device# request platform software package clean switch all file flash:
Running command on switch 1
Cleaning up unnecessary package files
Scanning boot directory for packages ... done.
Preparing packages list to delete ...
done.
Running command on switch 2
Cleaning up unnecessary package files
Scanning boot directory for packages ... done.
Preparing packages list to delete ...
done.
The following files will be deleted:
[1]:
/flash/cat3k_caa-rpbase.BLD_V161_0_THROTTLE_LATEST_20151116_230450.SSA.pkg
/flash/cat3k_caa-srdriver.BLD_V161_0_THROTTLE_LATEST_20151116_230450.SSA.pkg
/flash/cat3k_caa-universalk9.BLD_V161_0_THROTTLE_LATEST_20151116_230450.SSA.bin
/flash/cat3k_caa-wcm.BLD_V161_0_THROTTLE_LATEST_20151116_230450.SSA.pkg
/flash/cat3k_caa-webui.BLD_V161_0_THROTTLE_LATEST_20151116_230450.SSA.pkg
/flash/packages.conf
/flash/packages.conf.00/flash/packages.conf.01/flash/packages.conf.02-
[1 2]: Do you want to proceed with reload? [yes/no]
Switch Stack Update
Step 10Copy the switch image to the TFTP server using the copy tftp://flash command.
The following example shows that the TFTP server (192.168.254.12) requires a user name (admin) and
password (cisco), which can easily be integrated into the copy command:
Update the Switch Stack Image
29
Step 11Upload the image to the stack members, and then reload the switch.
The image download and installation can be performed while the stack is in-service, but to complete the
update install, you must perform a switch reload, which causes a service outage.
Step 12After the reload completes, run the request platform software package clean switch all file flash
command.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 27
Switch Stack Update
request platform software package clean switch all file flash
Device# request platform software package clean switch all file flash:
Running command on switch 1
Cleaning up unnecessary package files
Scanning boot directory for packages ... done.
Preparing packages list to delete ...
done.
Running command on switch 2
Cleaning up unnecessary package files
Scanning boot directory for packages ... done.
Preparing packages list to delete ...
done.
software auto-upgrade enable
end
Performing the Stack Update
.
To verify that stack members are using the same software, use the show version command on all
members of the switch stack.
Enable Switch Image Auto-Upgrade
Step 13Enable auto-upgrade so that new or replacement stack members are automatically upgraded with the
software running on the switch stack.
If you are adding a new member, or replacing a stack member, we recommend that you enable the auto
upgrade feature within the stack. This feature helps to avoid stack mismatch issues and ensures that any
new switches are upgraded to the version currently running on the stack and also converts a member in
bundle mode to install mode.
The auto-upgrade feature automatically installs the software packages from an existing stack member to
the stack member that is running incompatible software.
NoteAuto-upgrade is disabled by default.
NoteThe rolling-upgrade feature is not supported.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
30
Page 28
Performing the Stack Update
Switch Stack Update
31
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 29
Initial Switch Configuration
This workflow explains how to configure the basic settings on a switch.
Whether the configuration deployment of a switch is completed all at once or done in phases, the basic
switch settings must first be configured. The initial management configuration includes setting IP
addresses, passwords, and VLANs, which the prerequisites for future feature configuration.
Prerequisites for Initial Switch configuration
Refer to the switch Hardware Installation Guide to complete the following tasks:
1. Rack-mount the switch.
2. Connect the StackWise cables.
3. Connect the switch ports.
4. Perform power on.
5. Provision your upstream switch.
6. Connect at least one Ethernet cable from the uplink interface on the switch to the upstream switch
or router.
Identify Configuration Values
We recommend that you identify certain switch configuration values in advance so that you can proceed
with this section without interruption. We recommend that you take a print out of Table 2, and, as you
follow the configuration sequence, replace the values in column B with your values in column C.
NoteReplace the blue italicized example values with your own values.
Cisco Systems, Inc.
www.cisco.com
Page 30
Assign Initial Management Information
Table 3 Initial Configuration Values
A. Value NameB. Example Value NamesC. Your Value
Hostname3850-access-Bld1Flr1
SNMP community strings for
read-only and read-write access
Management VLAN ID100
In-band management IP address
and mask
Default gateway192.168.1.1
Secret passwordmy-secret-password
TACAS server IP address192.168.254.10
TACAS server secret keycisco123
Uplink interface ID
Management VRF IP address for
out-of-band interface
Mgmt-VRF default route next
hop
Native VLAN 999, dummy
Initial Switch Configuration
my-SNMP-RO-name
my-SNMP-RW-name
192.168.1.2 255.255.255.0
GigabitEthernet 1/1/1
Mgmt-vrf 192.168.128.5
255.255.255.0
192.168.128.1
NoteThe configuration examples provided in this document begin in global configuration mode, unless noted
otherwise.
Assign Initial Management Information
• The following configurations should be performed in the same sequence in which they are listed
here.
• Users can now proceed to the Configure Secure HTTPS ans Secure Shell for Secure LAN
Management section.
• Configure SNMP for Remote Management
• Configure Local Login and Password for Switch Access
• Configure Centralized User Authentication Through TACACS+
• Configure a Management IP Address on an Out-of-Band Interface
• Configure a Management IP Address on an In-Band Interface
• Create a Management VLAN in Hardware
• Enter the show running-configuration command to display the initial management information for
the switch.
18
NoteThe following configurations should be performed in the same sequence in which they are listed here.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 31
Initial Switch Configuration
hostname 3850-access-Bld1Flr1
no ip http server
ip http secure-server
ip ssh version 2
!
line vty 0 15
transport input ssh
transport preferred none
Configure the Hostname for Switch Identification
Step 1Configure the hostname on a switch to identify the switch in your network. By default, the system name
and prompt are Switch.
Set the hostname for the switch product family, the role of the switch in your network, and the switch
location.
Note that the system name is also used as the system prompt.
If you have not configured a system prompt, the first 20 characters of the system name are used as the
system prompt. A greater-than symbol [>] is appended. The prompt is updated whenever the system
name changes.
This example is for the switch serving as an access layer switch located on the first floor of Building 1
.
NoteUsers can now proceed to the Configure Secure HTTPS ans Secure Shell for Secure LAN Management
section.
Assign Initial Management Information
Configure Secure HTTPS and Secure Shell for Secure LAN Management
Step 2Disable the HTTP and Telnet unencrypted protocols on the switch.
Step 3Configure Secure HTTP (HTTPS) and Secure Shell (SSH) to enable secure management of the switch.
Enabling HTTPS automatically generates a cryptographic key to use the service. When SSH is
configured after HTTPS, you do not have to explicitly generate the cryptographic key that SSH requires,
unless you want to change the default key size.
We recommend that you use the transport preferred none command on the VTY lines to prevent
connection attempt errors from the CLI prompt. Without this command, your IP name server may
become unreachable, and long timeout delays may occur..
NoteIf the switch acts as a Web authentication server or as an authentication proxy, then do not disable the
HTTP server by executing the no ip http server command.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
19
Page 32
Assign Initial Management Information
snmp-server community my-SNMP-RO-name RO
snmp-server community my-SNMP-RW-name RW
Step 4Enable Simple Network Management Protocol (SNMP) to allow the network infrastructure devices to
be managed by a remote Network Management System (NMS). Configure SNMPv2c read-only and
read-write community strings, as shown in the following example. Once SNMP community strings are
configured, then SNMP tools can be used to monitor the 3850 which includes statistics.
Configure Local Login and Password for Switch Access
Step 5Configure a local user ID and password to secure access to the switch.
We recommend that you encrypt passwords to secure access to the device configuration mode and
prevent the display of plain text passwords in configuration files.
Initial Switch Configuration
Configure Centralized User Authentication Through TACACS+
NoteConfiguring the TACACS+ protocol is optional and recommended only when using TACACS to manage
all of your network devices.
Step 6Configure centralized user authentication through the TACACS+ protocol.
As networks increase the number of devices to maintain, there is an operational burden to maintain local
user accounts on every device. A centralized authentication, authorization, and accounting (AAA)
service reduces operational tasks on each device and provides an audit log of user access for security
compliance and root-cause analysis. When AAA is enabled for access control, all management access to
the network infrastructure devices (SSH and HTTPS) is controlled by the AAA service.
TACACS+ is the primary protocol used to authenticate management infrastructure devices to determine
whether access can be allowed to the AAA server. A local AAA user database defined on each network
infrastructure device to provide a fallback authentication source in case the centralized TACACS+ server
is unavailable.
This example shows how to configure the switch for TACACS administrative access.
20
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 33
Initial Switch Configuration
aaa new-model
tacacs server TACACS-SERVER-1
address ipv4 192.168.254.10
key cisco123
exit
!
aaa group server tacacs+ TACACS-SERVERS
server name TACACS-SERVER-1
exit
!
aaa authentication login default group TACACS-SERVERS local
aaa authorization exec default group TACACS-SERVERS local
ip http authentication local
write memory
Step 7To save your configuration, use the write memory EXEC command in privileged mode.
Assign Initial Management Information
Assign an IP Address to the Switch
Assign an IP to the switch, so that the switch can be managed remotely instead of being restricted to
management through a direct connection to the console port.
Although the switch supports multiple IP addresses for switch management, only primary IP address is
responsible for switch management.
Two types of IP addresses are used for switch management—in-band and out-of-band.
An in-band IP address is an address assigned to an interface that is reached through the production
network. Examples of in-band interfaces that have assigned IP addresses are VLAN, Ethernet, and
loopback interfaces.
An out-of-band IP address is an address assigned to an interface that is unreachable through the
production network. Out-of-band networks are more common in large network deployments. If you do
not have an Out-of-band network, use only an in-band network for management.
On the switch, the out-of-band interface is GigabitEthernet 0/0. The GigabitEthernet 0/0 interface is not
connected to the internal switching hardware, but directly to the CPU. IP traffic on GigabitEthernet 0/0
does not use the operating network. If the physical topology of the switch deployment does not support
out-of-band, then the switch can be managed with an in-band IP address.
We recommend that the switch be assigned multiple IP addresses for high availability; one IP address on
the out-of-band interface, and one on the in-band interface. High availability for switch management
ensures that the most available switch on the switch stack is the active switch and that it has a
management IP address so that all the stack members are accessible for management. You can have both
an in-band and out-of-band IP addresses as long as they are not in the same subnet. The preferred method
for management is out-of-band, because it is highly available and less likely to be impacted by DOS and
broadcast storms. The GigabitEthernet 0/0 interface on the switch is used for out-of-band management.
Configure the management IP addressees, as described in these sections:
• Configure a Management IP Address on an Out-of-Band Interface
• Configure a Management IP Address on an In-Band Interface
• Create a Management VLAN in Hardware
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
21
Page 34
Assign Initial Management Information
interface GigabitEthernet 0/0
ip address 192.168.128.5 255.255.255.0
exit
ip default-gateway 192.168.2.1
ip route vrf Mgmt-vrf 192.168.128.5 255.255.255.0 192.168.128.1
exit
Configure a Management IP Address on an Out-of-Band Interface
Step 8Assign an IP address to an out-of-band interface.
Out-of-band management is managing the switch and all other networking devices through a physical
network, which is separate from the production network that carries end-user traffic. To manage the
switch with an out-of-band network, the switch uses the GigabitEthernet 0/0 interface. The
GigabitEthernet0/0 interface is physically located on the rear of the switch, next to the blue console port.
The following are the advantages of a GigabitEthernet 0/0 interface:
• The interface is not susceptible to network outages, such as broadcast storms or other potential
issues on the production network because it is separated from the data plane.
• The interface is out-of-band and allows the switch and all other networking devices to always be
manageable so that you can quickly respond whenever there is a network issue.
Step 9Configure a Virtual Routing and Forwarding (VRF) instance.
The out-of-band management interface is in its own VRF instance. This means that the routing database
and protocol exchange are also separate for this interface from the other data network interfaces.
Initial Switch Configuration
The following are the limitations of a GigabitEthernet 0/0 interface.
• Management traffic originating from the switch must be associated with the GigabitEthernet 0/0
VRF instance. A Mgmt-vrf is used to segment management traffic from the global routing table of
the switch.
• A default route for the Mgmt-vrf is required.
• This interface cannot be used as the source interface for sending SNMP traps. Sending traps to an
SNMP trap server requires an IP address on a VLAN interface, see the “Configure a Management
IP Address on an In-Band Interface” section.
NoteUse the IP address value that you listed in the print-out (Table 3 ) for the out-of-band management
configuration.
In the following example, the GigabitEthernet 0/0 interface is not on the switch data plane. This interface
(also referred to as the service port) is terminated on the CPU of the switch as opposed to a logical
interface of the forwarding ASIC. The GigabitEthernet 0/0 differs from the Ethernet interfaces on the
front of the switch because it is only a Layer 3 interface (also referred to as a routable interface). The
Ethernet interfaces on the front of the switch default to Layer 2 mode and are used for bridging.
The Ethernet interfaces on the front can be configured to be a routable interface using the no switchport
interface command. The GigabitEthernet 0/0 interface will not function without an IP address assigned
to it.
Mgmt-vrf is built-in; you do not have to create one for out-of-band management.
22
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 35
Initial Switch Configuration
show ip route vrf Mgmt-vrf
Routing Table: Mgmt-vrf
C- IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static
route
o - ODR, P - periodic downloaded static route, H - NHRP, l - LISP
+ - replicated route, % - next hop override
Gateway of last resort is not set
192.168.128.5/16 is variably subnetted, 3 subnets, 2 masks
S 192.168.128.5/24 [1/0] via 192.168.128.1
C 192.168.128.5/24 is directly connected, GigabitEthernet0/0
L 192.168.128.2/32 is directly connected, GigabitEthernet0/0
ping vrf Mgmt-vrf 192.168.128.1
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.128.1, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/10 ms
odes: L - local, C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i
Step 10Following is the example for show ip route vrf command.
Assign Initial Management Information
Configure a Management IP Address on an In-Band Interface
Step 11Assign your management IP address to a VLAN interface that is used only for management, and not used
to carry other network traffic.
A VLAN interface is a Layer 3 endpoint on the subnet assigned to the corresponding VLAN.
NoteDo not use VLAN 1 as the management VLAN for security purposes.
The management VLAN is a separate VLAN for managing the switch and all other network devices in
the same subnet. You should assign an in-band IP address to a VLAN interface regardless of whether an
IP address is assigned to the out-of-band interface.
With in-band management, the IP address can be reached through the production network. For
management purposes, the in-band IP address can be used the same way as the out-of-band IP address.
There is no functional difference. However, the in-band IP address has more capabilities because this is
the source IP address for some of the auto-generated traffic that comes from the switch, for instance,
SNMP traps use the in-band IP address.
You can assign an IP address to your VLAN interface before you configure the VLAN on the switch.
The VLAN interface is not operational until the VLAN is created in hardware, and at least one physical
interface, which is a member of the VLAN, is in a forwarding state.
This example shows a VLAN created for management and indicates that the IP address is reachable.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
23
Page 36
Assign Initial Management Information
interface vlan 100
ip address 192.168.1.2 255.255.255.0
no shutdown
exit
ip default-gateway 192.168.1.1
NoteThe switch supports IP address assignments to physical Ethernet interfaces that have been configured to
operate in Layer 3 mode.
Step 12Configure the default gateway, as shown in the following example. This gateway functions as the default
route.
When using a VLAN interface, a default route is not required.
Create a Management VLAN in Hardware
Initial Switch Configuration
Earlier you assigned an IP address to the interface for VLAN 100. Refer to the "Appendix 3, “Configure
a Management IP Address on an In-Band Interface” section to assign an IP address to the interface.
However, merely assigning the IP address to VLAN 100 does not create the VLAN in hardware. Perform
the below step to make the switch reachable through the assigned IP address.
Step 13Configure a management VLAN in hardware and configure an uplink interface as a member of this
VLAN.
NoteThis is an intermediate step required only to make the switch Layer 3 reachable and manageable from
SSH or HTTPS as well as the console or Express Setup. You can skip this step if you continue to use the
console to complete the configuration, but required if you use another tool to complete the configuration
of the switch. The complete best-practice configuration for uplink connectivity is explained in the
“Uplink Interface Connectivity” workflow.
We recommend that you use a dummy VLAN as the native VLAN on trunk interfaces instead of the
default VLAN 1. Because all interfaces are assigned to VLAN 1 by default on the switch, this step limits
the traffic associated with potential user configuration and possible connection errors propagating across
the trunk.
All other VLANs on the uplink interfaces are tagged with IEEE 802.1q which encapsulates the Layer 2
head of the Frame packet.
The following example shows how to configure VLAN IDs in hardware and assign the names. The
upstream interfaces to the switch or router are modified to make them members of the new VLANs. You
must have the same VLAN ID on both ends of the Ethernet link to properly configure the management
VLAN in hardware. A “dummy” VLAN is used as the native VLAN on trunk interfaces. A dummy
VLAN is not used for data or management traffic.
24
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 37
Initial Switch Configuration
vlan 100
name switch_mgmt
exit
vlan 999
name dummy
exit
!
! The next step assumes the uplink interface is GigabitEthernet 1/1/1, but
! your uplink interface may be different.
!
interface GigabitEthernet 1/1/1
Switchport mode trunk
Switchport trunk native vlan 999
NoteThe Shortest Path Tree (SPT) and ping command used in this example require that the upstream layer
Assign Initial Management Information
device (switch or router) to be configured to operate in a production network, and without any additional
configuration changes being required.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
25
Page 38
Assign Initial Management Information
! Use “show spanning-tree vlan 100” to confirm VLAN 100 FWD on the uplink
! interface.
! Use “show interface trunk” to confirm GigabitEthernet 1/1/1 is
! operating in Trunk mode correctly.
show spanning-tree vlan 100
VLAN0100
Spanning tree enabled protocol rstp
Root ID Priority 32868
Address 0022.bdd9.4c00
Cost 4
Port 49 (GigabitEthernet1/1/1)
Hello Time 2 sec Max Age 20 sec Forward Delay 15 sec
Bridge ID Priority 32868 (priority 32768 sys-id-ext 100)
Address 20bb.c05f.b300
Hello Time 2 sec Max Age 20 sec Forward Delay 15 sec
Aging Time 300 sec
Port Mode Encapsulation Status Native vlan
Gi1/1/1 on 802.1q trunking 999
Gi1/1/2 on 802.1q trunking 999
Port Vlans allowed on trunk
Gi1/1/1 1-4094
Gi1/1/2 1-4094
Port Vlans allowed and active in management domain
Gi1/1/1 1,100,999
Gi1/1/2 1,100,999
Port Vlans in spanning tree forwarding state and not pruned
Gi1/1/1 1,100,999
Gi1/1/2 none
!
! Now the default gateway will respond to pings
!
ping 182.168.1.1
Initial Switch Configuration
NoteEnter the show running-configuration command to display the initial management information for the
switch.
26
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 39
Initial Switch Configuration
Assign Initial Management Information
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
27
Page 40
Assign Initial Management Information
Initial Switch Configuration
28
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 41
Global System Configuration
This workflow describes common global configurations for all switch deployments in the access layer.
Prerequisites for Global System Configuration
• Complete the task described in “Initial Switch Configuration” workflow.
• If you have not completed the task described in the “Uplink Interface Connectivity” workflow, the
switch might not be IP reachable. If that is the case, use only the switch console to perform the
Global System Configuration workflow.
If you have completed the “Uplink Interface Connectivity” workflow, you can perform the Global
System Configuration workflow using the switch console, SSH, or any management tool. Using
tools other than the console requires you to log in using user names and passwords configured, as
described in the section the “Initial Switch Configuration” workflow.
Identify Configuration Values
We recommend that you identify certain switch configuration values in advance so that you can proceed
with this workflow without interruption. We recommend that you take a print out of Table 4, and, as you
follow the configuration sequence, you should replace the values in column B with your values in
column C.
NoteReplace the blue italicized example values with your own values.
Table 4 Global System: Setting Values
A. Value NameB. Example ValueC. Your Value
Management subnets allowed
NTP server IP address
Cisco Systems, Inc.
www.cisco.com
192.168.128.5/0.0.0.255
192.168.0.0/0.0.0.255
192.168.254.0/0.0.0.255
192.168.254.11
Page 42
Assign Global Configuration Information
Table 4 Global System: Setting Values
A. Value NameB. Example ValueC. Your Value
Data VLAN
Voice VLAN
Access points VLAN
Management VLAN ID
Wireless clients VLAN
VLAN name for data
VLAN name for voice
VLAN name for access points
VLAN name for wireless clients
SNMP community strings for
read-only and read-write access
IPv6 Router Advertisement
Guard policy for access
interfaces
IPv6 Router Advertisement
Guard policy for upstream router
interfaces
IPv6 Router Advertisement
Guard policy for upstream
switch interfaces
IPv6 DHCP guard policy for
access interfaces
IPv6 DHCP guard policy for
uplink interfaces
Global System Configuration
10
11
12
100
200
Data
Voice
Access_Points
Wireless_Client
my-SNMP-RO-name,
my-SNMP-RW-name
endhost_ipv6_raguard
router_ipv6_raguard
switch_ipv6_raguard
endhost_ipv6_dhcp_guard
uplink_ipv6_dhcp_guard
NoteConfiguration examples begin in global configuration mode, unless noted otherwise.
Assign Global Configuration Information
NoteThe following tasks should be performed in the same sequence in which they are listed here.
• Configure High Availability on the Switch Stack
• Configure the Switch to run in VTP Transparent Mode
• Enable Rapid Per-VLAN Spanning Tree Plus
• Configure BPDU Guard for Spanning-Tree PortFast Interfaces
• Configure UDLD to Detect Link Failure
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
32
Page 43
Global System Configuration
switch 1 priority 15
switch 2 priority 14
• Configure an Access List to Limit Switch Access
• Configure System Clock and Console Timestamps
• Configure DHCP Snooping Security Features
• Configure ARP Inspection
• Configure EtherChannel Load Balancing
• Create Access Layer VLANs
• Create IPv6 First-Hop Security Policies
• Increase the TFTP Block Size
• Enable New Members to Automatically Update to the Switch Stack Image
•
Configure High Availability on the Switch Stack
Step 1Assign the active switch and standby switch with high stack-member priority values, so that network
operations are not affected during a stack-member failure.
Assign Global Configuration Information
Recommendation: For consistency, configure the stack-member priority used to determine the active
stack member. By configuring one member to be the active stack member, you ensure that this member
is always the active member through all stack elections, for the lifetime of the stack. The member with
the highest configured priority becomes the active member.
In a switch stack, the member most likely to fail is the active member. Therefore, in a switch stack with
three or more members, we recommend that you configure uplink connectivity on more than one stack
member and do not configure uplink connectivity on the active member. This way, uplink connectivity
is not affected if the active member fails.
In this document, the stack refers to a two-member stack, and the example here shows how to assign the
highest priority to member 1. Assign a secondary member by giving it a slightly lower priority. The
default priority is 1.
NoteFor additional information about managing switch stacks and configuring high availability features on
the switch, see the Stack Manager and High Availability Configuration Guide, Cisco IOS XE Release.
Configure the Switch to run in VTP Transparent Mode
Step 2Configure your switch to run in VTP transparent mode in order to avoid the VLAN configuration updates
coming from the network, since they have the potential for unexpected behavior due to error operations.
Typically, VLANs are defined once during your initial switch configuration and do not require
continuous VTP updates after the switch is operational.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
33
Page 44
Assign Global Configuration Information
vtp mode transparent
spanning-tree mode rapid-pvst
spanning-tree portfast bpduguard default
A switch in VTP transparent mode can create, modify, and delete VLANs (the same way as VTP
servers), but the switch does not send dynamic propagation of VLAN information across the network
and does not synchronize its VLAN configuration based on advertisements received. Configuration
changes made when the switch is in this mode are saved in the switch’s running configuration, and can
be saved to the switch’s startup configuration file.
NoteThe default VTP mode for the switch is VTP server mode. This mode allows you to create, modify, and
delete VLANs and specify other configuration parameters for the entire VTP domain. VTP servers
advertise their VLAN configuration to other switches in the same VTP domain and synchronize their
VLAN configuration with other switches based on advertisements received over trunk links.
Enable Rapid Per-VLAN Spanning Tree Plus
Step 3Enable Rapid Per-VLAN Spanning Tree Plus (PVST+), to improve the detection of indirect failures or
linkup restoration events over classic spanning tree.
Rapid PVST+ provides an instance of RSTP (IEEE 802.1w) for each VLAN, and PVST+ improves the
detection of indirect failures or linkup restoration events over the classic spanning tree (IEEE 802.1D).
Recommendation: Enable spanning tree even if your deployment is created without any Layer 2 loops.
By enabling spanning tree, you ensure that if physical or logical loops are accidentally configured, no
actual Layer 2 loops occur.
Global System Configuration
Configure BPDU Guard for Spanning-Tree PortFast Interfaces
Step 4Configure the Bridge Protocol Data Unit (BPDU) guard globally to protect all Spanning-Tree
PortFast-enabled interfaces.
The BPDU guard protects against a user plugging a switch into an access port, which many cause a
catastrophic, undetected spanning-tree loop.
If a Spanning-Tree PortFast-configured interface receives a BPDU, an invalid configuration exists, such
as the connection of an unauthorized device. The BPDU Guard feature prevents loops by moving a
nontrunking interface into an errdisable state when a BPDU is received on an interface when STPF is
enabled.
The BPDU configuration protects STPF-enabled interfaces by disabling the port if another switch is
plugged into the port.
This command should configured globally, not at the interface level.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
exit
snmp-server community sample-READONLY RO 55
snmp-server community sampe-READWRITE RW 55
!
Configure UDLD to Detect Link Failure
Step 5Configure Unidirectional Link Detection (UDLD) in aggressive mode, not normal mode.
UDLD detects a unidirectional link, and then disables the affected interface and alerts you.
Unidirectional links can cause a variety of problems, including spanning-tree loops, black holes, and
nondeterministic forwarding. In addition, UDLD enables faster link-failure detection and quick
reconvergence of interface trunks, especially with fiber, which can be susceptible to unidirectional
failures.
In aggressive mode, if the link state of a port is determined to be bidirectional and the UDLD information
times out while the link on the port is still in UP state, UDLD tries to re-establish the state of the port.
If this not successful, the port is put into errdisable state. In normal mode, the port state for UDLD is
marked as undetermined, and operates according to its Spanning Tree Protocol state.
Do not change UDLD aggressive timers.
NoteUDLD in aggressive mode is not needed when the upstream device is a switch operating in VSS mode.
Assign Global Configuration Information
For more information about VSS-enabled campus design, see theCampus 3.0 Virtual Switching System
Design Guide.
Configure an Access List to Limit Switch Access
Step 6If your network operation support is centralized, you can increase network security by using an access
list to limit the networks that can access your switch.
We recommend that you use an access list to permit IP addresses from known source management
locations.
In this example, only the hosts on the 192.168.128.0, 192.168.0.0, and 192.168.254.0 networks can
access your switch using SSH or SNMP. The following example shows an ACL that permits three
subnets. your network may have more subnets or fewer subnets. configure the ACL that best fits your
network. You can continue to add to the list, as required for your network deployment.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
service timestamps debug datetime msec localtime
service timestamps log datetime msec localtime
ip dhcp snooping vlan 10,11,12,100
no ip dhcp snooping information option
ip dhcp snooping
ip dhcp snooping wireless bootp-broadcast enable
ip arp inspection vlan 10,11,100
Configure System Clock and Console Timestamps
Step 7Configure a synchronized clock by programming your network devices to synchronize to a local NTP
server in the network.
The local NTP server typically references a more accurate clock feed from an outside source.
Step 8Configure console messages, logs, and debug output to provide timestamps on output, which allows
cross-referencing of events in a network.
Global System Configuration
Configure DHCP Snooping Security Features
Step 9Enable Dynamic Host Configuration Protocol (DHCP) snooping on the data, voice, and wireless AP
VLANs.
The switch intercepts and safeguards DHCP messages within the VLAN. This configuration ensures that
an unauthorized DHCP server cannot allocate addresses to end-user devices.
Configure ARP Inspection
ARP inspection is a security feature that prevents ARP spoofing.
Step 10Enable Address Resolution Protocol (ARP) inspection on the data, voice, and management VLANs.
Configure EtherChannel Load Balancing
Step 11Set EtherChannels to use the traffic source and destination IP address when calculating which link to
send traffic to.
EtherChannel traffic should be balanced across all physical interfaces. The default load-balancing
scheme for EtherChannels is based on the source MAC address.
36
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 47
Global System Configuration
port-channel load-balance src-dst-ip
vlan 10
name Data
vlan 11
name Voice
vlan 12
name Access_Points
vlan 200
name Wireless_Client
This configuration normalizes the method in which traffic is load-shared across the member links of an
EtherChannel. EtherChannels are used extensively in this design because of their resilience.
Create Access Layer VLANs
Step 12Create VLANs to separate traffic based on end-user devices.
When VLANs are created, they automatically join any interface that is configured for trunk mode.
Earlier, the uplink interface was configured for trunk mode. Therefore, the uplink interface should now
be a member of these VLANs.
Use consistent VLAN IDs and VLAN names in the access layer. Consistent IDs and names help with
consistency, and network operation becomes more efficient.
NoteDo not use VLAN 1.
Assign Global Configuration Information
NoteUse VLAN 200 for wireless clients only if the switch operates as a wireless controller in the converged
access mode.
Create IPv6 First-Hop Security Policies
Step 13Create and apply global IPv6 security policies on the uplink interfaces to define the trust and roles on
the connected distribution switches or routers.
Blocking router advertisements with Router Advertisement Guard and DHCP responses from untrusted
sources are an easy way to secure against the most common IPv6 problems.
NoteAccess interfaces to end devices should not be trusted for router advertisements and IPv6 DHCP
response.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
This example configuration shows how to create global policies that are applied to the interfaces
described in the “Access Control on the Wired Network” workflow.
Global System Configuration
Increase the TFTP Block Size
Step 14Increase the TFTP block size to the maximum allowed value of 8192.
By default, the switch uses a TFTP block size value of 512, which is the lowest possible value. Increasing
this global value significantly improves the TFTP file transfer time.
Enable New Members to Automatically Update to the Switch Stack Image
Step 15Enable the Auto Upgrade feature so that new switch members automatically update to the Cisco IOS
version that is running on the switch stack.
When new members join an existing switch stack, the Cisco IOS version of the new members must match
the Cisco IOS version of the existing members. The Auto Upgrade feature provides the ability to
NoteThe switch stack must be running Cisco IOS XE Release 3.3.1 or higher, or later in install mode.
automatically update new members when they join. However, this feature is not enabled by default.
38
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
For detailed information about the Auto Upgrade feature, see the Using the Auto-Upgrade feature on the
Cisco Catalyst 3850 document.
Page 49
Uplink Interface Connectivity
This workflow describes how to configure the Ethernet interfaces that connect a switch or switch stack
to distribution switches or routers. These interfaces are uplink interfaces. They are different from access
interfaces that connect to non-networking end devices such as IP phones, personal computers, wireless
access points, printers, and IP cameras.
The switch interface configuration recommendations are based on a switch stack deployed in the campus
or branch of the access layer.
When stacking two or more physical switches into one logical switch, we recommend that the uplink
interfaces are configured across the physical members to ensure that an active uplink interface always
available for switch-stack members.
Prerequisites for Uplink Interface Connectivity
Ensure that the best-practice configurations are set, as described in the Global System Configuration
workflow.
Restrictions for Uplink Interface Connectivity
• A maximum of only eight physical links can be active in a single EtherChannel group.
• All the ports in an EtherChannel must be assigned to the same VLAN, or must be configured as trunk
ports.
• All the interfaces in an EtherChannel must be of the same type, for example, Gigabit Ethernet
interfaces cannot be mixed with 10-Gbps interfaces.
Identify Configuration Values
We recommend that you identify certain switch configuration values in advance so that you can proceed
with this workflow without interruption. We recommend that you take a print out of Table 5, and, as you
follow the configuration sequence, replace the values in column B with your values in column C.
Cisco Systems, Inc.
www.cisco.com
Page 50
Identify Configuration Values
NoteReplace the blue italicized example values with your own values.
NoteConfiguration examples begin in global configuration mode, unless noted otherwise.
42
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 51
Uplink Interface Connectivity
Dual redundant switches
in distribution layer running
VSS (Cat6500/6800/4500),
or VPC (Nexus 7000)
Desktop user
direct connect
Desktop user
Printer
Wireless access
Catalyst 3850 stack in access
Voice VLAN 11
Data VLAN 10
Data VLAN 10
Data VLAN 10
Switch management
VLAN 100
Access point VLAN 12
Trunk link
Native VLAN 999
All VLANs included
LAN Access Switch Topology with Uplinks to a Distribution Switch or Distribution Router
LAN Access Switch Topology with Uplinks to a Distribution
Switch or Distribution Router
The following illustration displays the LAN Access Switch Topology with Uplinks to a distribution
switch or distribution router:
Figure 6LAN Access Switch Topology with Uplinks to a Distribution Switch
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
43
Page 52
LAN Access Switch Topology with Uplinks to a Distribution Switch or Distribution Router
Figure 7Uplinks for a Distribution Router
Voice VLAN 11
Data VLAN 10
Uplink Interface Connectivity
Desktop user
behind IP phone
Data VLAN 10
Desktop user
direct connect
Data VLAN 10
Printer
Wireless access
Catalyst 3850 stack in access
Switch management
VLAN 100
Dual redundant
routers running
HSRP
Trunk link
Native VLAN 999
All VLANs included
Access point VLAN 12
391936
44
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 53
Uplink Interface Connectivity
Configure Uplink Interface Connectivity
Configure Uplink Interface Connectivity
• Recommendations for Configuring an Uplink Interface to a Router or Switch
• Configure QoS on an Uplink EtherChannel Interfaces
• Configure an Uplink Interface as an EtherChannel and as a Trunk
• Configure Security Features on an Uplink EtherChannel Interface
• Spanning-Tree Recommendations for an Uplink Interface Connecting to a Distribution Switch
• Verify Uplink Interface Configurations
Recommendations for Configuring an Uplink Interface to a Router or Switch
When configuring your uplink interface, follow the below recommendations to guide you through the
configuration from interface to upstream router or switch:
• Make sure that the uplink connections from the switch stack to the distribution switches have enough
bandwidth to carry the traffic associated with all of the access interfaces on the switch stack.
• Use EtherChannels to increase resilience of in case an uplink interface fails.
• For EtherChannels, use Link Aggregation Control Protocol (LACP) active-active mode, which
adheres to the IEEE 802.3ad standard. The active-active mode implies that both the switch stack as
well as the distribution switch side of the EtherChannel must be configured in LACP active mode.
• Use uplink ports on the different switches in the switch stack to connect back to the distribution
switches. This configuration ensures that there is no single source of failure for the switch stack. If
a switch in the stack owning one of the uplink connections fails, there will still be an uplink port
connection from a remaining member of the switch stack connecting back to the distribution
switches.
• All the interfaces are assigned to VLAN 1 by default. Do not configure VLAN 1 on the trunk; this
is to prevent traffic associated with potential user connection errors from propagating across the
trunk.
Configure QoS on an Uplink EtherChannel Interfaces
NoteThis configuration should be applied to the physical uplink interfaces before adding them to an
EtherChannel.
Step 1Apply the Trust Differentiated Services Code Point (DSCP) service policy on an interface in the ingress
direction, and then apply the 2P6Q3T policy in order to ensure proper congestion management and
egress bandwidth distribution on the interface in the egress direction.
Ethernet traffic that is received from the upstream switch or router contains trusted QoS markings and
is classified to guarantee a type of service.
Additional service policies should be applied after traffic is transmitted in order to ease congestion. For
more information see, “Configure QoS on an Access Interface” on page 56
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Configure an Uplink Interface as an EtherChannel and as a Trunk
Step 1Choose one of the following configurations based on your network topology:
• “Configure an Uplink Interface to Connect to a Distribution VSS or VPC Switch”
• “Configure an Uplink Interface to Connect to a Distribution Router (or Standalone Distribution
Switch)”
Configure an Uplink Interface to Connect to a Distribution VSS or VPC Switch
1. Ensure that the distribution Virtual Switch System (VSS) or Virtual Port Channel (VPC) switch
connections are configured the same way and that the EtherChannel is configured in LACP active
mode.
2. For additional resilience, ensure that the uplink interfaces are located on different switches in the
switch stack.
Figure 6, shows the switch stack that has a single EtherChannel connection to a distribution VSS or VPC
switch pair.
The VSS and VPC systems have an explicit configuration between the Cisco distribution switch pair.
That allows them to act as a single logical switch when connected to the EtherChannel. The
EtherChannel is configured as a trunk with VLANs 10, 11, 12, and 100, with the native VLAN set to 999.
46
NoteUse this switch-stack uplink interface configuration only when connecting the switch stack to a VSS or
VPC distribution switch pair, and not when the distribution switch pair is configured as two standalone
switches.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 55
Uplink Interface Connectivity
interfaceGigabitEthernet 1/1/1descriptionconnection to Distribution VSS or VPC switch 1switchport mode trunkswitchport trunk native vlan999switchport trunk allowed vlan10,11,12,100,200channel-protocol lacpchannel-group1mode active
!
interfaceGigabitEthernet 2/1/1descriptionconnection to Distribution VSS or VPC switch 1switchport mode trunkswitchport trunk native vlan999switchport trunk allowed vlan10,11,12,100,200channel-protocol lacpchannel-group 1 mode active
!
interfaceGigabitEthernet 1/1/2descriptionconnection to Distribution VSS or VPC switch 2switchport mode trunkswitchport trunk native vlan999switchport trunk allowed vlan10,11,12,100,200channel-protocol lacpchannel-group1mode active
!
interfaceGigabitEthernet 2/1/2descriptionconnection to Distribution VSS or VPC switch 2switchport mode trunkswitchport trunk native vlan999switchport trunk allowed vlan10,11,12,100,200channel-protocol lacpchannel-group 1 mode active
Configure Uplink Interface Connectivity
Configure an Uplink Interface to Connect to a Distribution Router (or Standalone Distribution
Switch)
NoteUse this configuration when connecting the switch stack to two standalone distribution switches (not
configured as a VSS or VPC pair). However, do not use the spanning-tree portfast trunk command for
switch configuration.
• Ensure that the distribution VSS or VPC router side of the connections are configured the same and
that the EtherChannel is configured with the LACP active mode.
• For additional resilience, the configured uplink interfaces should be located on different switches in
the switch stack.
• Use the spanning-tree portfast trunk command to allow the switch side of the uplink to
immediately transition to a spanning-tree forwarding state when the link becomes available, because
routers do not participate in a spanning tree.
Figure 7 shows a switch stack having a separate EtherChannel to each distribution router. Each
EtherChannel is configured as a trunk with VLANs 10, 11, 12, 100, 200, and 999, with the native VLAN
set to 999.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
47
Page 56
Configure Uplink Interface Connectivity
interface GigabitEthernet 1/1/1
descriptionconnection to Distribution router 1switchport mode trunkswitchport trunk native vlan 999switchport trunk allowed vlan10,11,12,100,200spanning-tree portfast trunkchannel-protocol lacpchannel-group1mode activeinterfaceGigabitEthernet 2/1/1descriptionconnection to Distribution router 1switchport mode trunkswitchport trunk native vlan999switchport trunk allowed vlan10,11,12,100,200spanning-tree portfast trunkchannel-protocol lacpchannel-group1mode active
interface GigabitEthernet 1/1/2
description connection to Distribution router 2switchport mode trunkswitchport trunk native vlan 999switchport trunk allowed vlan 10,11,12,100,200spanning-tree portfast trunkchannel-protocol lacpchannel-group 2 mode activeinterface GigabitEthernet 2/1/2description connection to Distribution router 2switchport mode trunkswitchport trunk native vlan 999switchport trunk allowed vlan 10,11,12,100,200spanning-tree portfast trunkchannel-protocol lacpchannel-group 2 mode active
EtherChannel Connection to Router 1
Uplink Interface Connectivity
EtherChannel Connection to Router 2
Configure Security Features on an Uplink EtherChannel Interface
48
Step 2Configure IPv4 and IPv6 security features on uplink EtherChannel interfaces.
The uplink EtherChannel interfaces to distribution routers and switches should be configured to trust
router advertisements and IP response, because Layer 3 routing and server functionality resides on the
distribution switches and routers. This step is different from the access interface-to-end device
configuration, which should not be trusted, as specified in the “Access Interface Connectivity” workflow.
The policies that should be applied are defined in the “Global System Configuration” workflow.
In the following example, security is applied to the uplink interfaces connecting to VPC, VSS, or
standalone switch.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 57
Uplink Interface Connectivity
interface Port-channel 1
ip arp inspection trust
ip snooping trustipv6 nd raguard attach-policyswitch_ipv6_raguardipv6 guard attach-policyuplink_ipv6__guard
interface Port-channel 1
ip arp inspection trust
ip snooping trustipv6 nd raguard attach-policyrouter_ipv6_raguardipv6 guard attach-policyuplink_ipv6__guard
exit
!
interface Port-channel 2
ip arp inspection trust
ip snooping trust
ipv6 nd raguard attach-policy router_ipv6_raguard
ipv6 guard attach-policy uplink_ipv6__guard
In the following example, security is applied to the uplink interfaces connecting to routers:
Configure Uplink Interface Connectivity
Spanning-Tree Recommendations for an Uplink Interface Connecting to a
Distribution Switch
NoteComplete this configuration on the distribution switches and not on the switch. The recommendations
listed below are not applicable when routers are used at the distribution layer.
Step 3On uplink interfaces to distribution switches (Figure 6), ensure that the spanning-tree root for the
switch-stack VLANs is configured on the distribution switch pair.
Follow the below recommendations when the standalone distribution switches are used instead of a VSS
or VPC system:
• Make sure that the spanning-tree roots for the VLANs are distributed evenly between two standalone
distribution switches. For example, configure one switch as the spanning-tree root for all the even
VLANs, and configure the other switch as the spanning-tree root for all the odd VLANs. This
distribution configuration ensures that the spanning tree does not block all the VLANs on a single
uplink interface, and results in an even traffic flow on the uplink interfaces.
• If Hot Standby Router Protocol (HSRP) or Virtual Router Redundancy Protocol (VRRP) is
configured for the VLANs located on the standalone distribution switches, make sure that the VLAN
configuration on the active switch is the same on the switch that is the spanning-tree root for that
VLAN.
• Avoid flooding of traffic caused by asymmetric routing of traffic flows, by configuring the arp
timeout interface configuration command. This command adjusts the ARP aging timer to less than
the MAC address table aging timer on the Layer 3 VLAN interfaces of the distribution switches. By
default, the MAC address table aging timer is set to 5 minutes (300 seconds) on the switch.
For more information about spanning tree root configuration on the VSS, see the “Spanning Tree
Configuration Best Practice with VSS” section of the VSS Enabled Campus Design Guide.
For more information about spanning-tree root on distribution switches, see the “Spanning VLANs
across Access Layer Switches” section of the Campus Network for High Availability Design Guide.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
49
Page 58
Display Uplink Interface Connectivity for the Switch
Switch#sh int te2/1/3
TenGigabitEthernet2/1/3 is up, line protocol is up (connected)
Hardware is Ten Gigabit Ethernet, address is 381c.1a24.d537 (bia
381c.1a24.d537)
MTU 1500 bytes, BW 10000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive not set
Full-duplex, 10Gb/s, link type is auto, media type is SFP-10GBase-SR
input flow-control is off, output flow-control is unsupported
ARP type: ARPA, ARP Timeout 04:00:00
Last input never, output 00:00:19, output hang never
Last clearing of "show interface" counters never
Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 1000 bits/sec, 1 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec
2596461 packets input, 426179392 bytes, 0 no buffer
Received 2596461 broadcasts (2596461 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 2596461 multicast, 0 pause input
0 input packets with dribble condition detected
303459 packets output, 45794121 bytes, 0 underruns
0 output errors, 0 collisions, 2 interface resets
0 unknown protocol drops
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out
For more information about spanning-tree root configuration and asymmetric routing, see the “Spanning
VLANs Across Access Layer Switches” and “Asymmetric Routing and Unicast Flooding” sections of
the Campus Network for High Availability Design Guide.
Verify Uplink Interface Configurations
Use the following commands to verify if configurations in this workflow are correctly applied to your
uplink interfaces:
• show etherchannel summary
• show interface
• show interface trunk
• show cdp neighbors
• show auto qos interface
• show policy-map interface
Uplink Interface Connectivity
Display Uplink Interface Connectivity for the Switch
Step 1Enter the show running-configuration command to display uplink interface connectivity for the switch.
50
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 59
Uplink Interface Connectivity
Display Uplink Interface Connectivity for the Switch
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
51
Page 60
Display Uplink Interface Connectivity for the Switch
Uplink Interface Connectivity
52
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 61
Access Interface Connectivity
This workflow describes how to configure the Ethernet interfaces that connect to the end devices of a
switch. End devices are the non-networking devices that connect to the network, such as IP phones,
personal computers, wireless access points, printers, and IP cameras. The Ethernet interfaces that
connect to end devices are referred to as access interfaces. They differ from uplink interfaces that link
to other networking devices.
The workflow for configuring access interfaces is based on a switch deployed at the access layer in a
campus or branch network (Figure 8). The switch interfaces connected to end devices are the edge of the
network, which network security and QoS begins.
Prerequisites for Access Interface Connectivity
• Complete the procedure described in the Global System Configuration workflow, which includes the
necessary configurations for the access interface configuration.
• Complete the procedure described in the “Configure QoS on an Uplink EtherChannel Interfaces”
workflow, which includes the creation of input services policies for end devices.
Identify Configuration Values
We recommend that you identify certain switch configuration values in advance so that can proceed with
this section without interruption. We also recommend that you take a print out of Table 6, and, as you
follow the configuration sequence, replace the values in column B with your values in column C.
NoteReplace the blue italicized example values with your own values.
Cisco Systems, Inc.
www.cisco.com
Page 62
Identify Configuration Values
Table 6 Access Interface Connectivity Values
A. Value NameB. Example Value NameC. Your Value
Access interface ranges
Data VLAN
Voice VLAN
Access Points VLAN
Management VLAN ID
Wireless Clients VLAN
IPv6 Router Advertisement
Guard policy name
IPv6 Router Advertisement
Guard policy name
QoS service policy input names
(See the “Configure QoS on an
Uplink EtherChannel Interfaces”
section.)
QoS service policy output name
Access Interface Connectivity
interface range
GigabitEthernet1/0/1-48
interface range
GigabitEthernet2/0/1-48
10
11
12
100
200
endhost_ipv6_raguard
endhost_ipv6__guard
IPPhone-Input-Policy
Classify-Police-Input-Policy
Classify-Police-Input-Policy
Trust-Dscp-Input-Policy
SoftPhone-Input-Policy
Trust-Dscp-Input-Policy
Trust-Dscp-Input-Policy
Trust-COS-Input-Policy
No-Trust-Input-Policy
2P6Q3T
52
NoteConfiguration examples begin in global configuration mode, unless noted otherwise.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 63
Access Interface Connectivity
Desktop user
direct connect
Desktop user
Printer
Wireless access point
Catalyst 3850 stack in access
Voice VLAN 11
Data VLAN 10
Data VLAN 10
Switch management
VLAN 100
391689
Access point VLAN 12
Data VLAN 10
LAN Access Switch Topology with Connections to End Devices
LAN Access Switch Topology with Connections to End Devices
The following illustration shows the topology of LAN Access Switch to end devices:
Figure 8LAN Access Switch Topology with Connections to End Devices
Configure Access Interface Connectivity
• Recommendations for Configuring an Access Interface
• Configure an Interface for Access Mode
• Configure VLAN Membership
• Create an Interface Description
• Configure Security Features on an Access Interfaces
• Configure QoS on an Access Interface
• Verify Access Interface Configurations
Recommendations for Configuring an Access Interface
Although some end devices do not require the following access interface configurations, we recommend
that you perform them to ensure consistency. The configurations do not interfere with the operation of
the network or the attached end device, and is considered safe to use.
When configuring your access interface, you should complete the following tasks:
• Configure an Interface for Access Mode
• Configure VLAN Membership
• Create an Interface Description
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
53
Page 64
Configure Access Interface Connectivity
interface GigabitEthernet1/0/1|
nmsp attach suppress
interface GigabitEthernet1/0/1|
ip device tracking maximum 0
• Configure Security Features on an Access Interfaces
• Configure QoS on an Access Interface
• Verify Access Interface Configurations
IP Device Tracking
CautionThe IP Device Tracking (IPDT) feature could have some negative side effects that may impact the normal
day-to-day operation of your switch.
NoteSymptoms as a result of IPDT issues are seen on the end device. For instance on Windows PC, an error
message report for a duplicate IP Address 0.0.0.0 appears.
IPDT is enabled globally, but it cannot be globally disabled. To disable IPDT, you must disable it at the
interface level.
Access Interface Connectivity
NoteTo disable IPDT on a port channel, you must first unbundle the physical Ethernet interfaces from the port
channel.
We recommend that you disable IPDT on all access interfaces except under these situations where a
feature explicitly has IPDT enabled:
• IPDT is required for Centralized Web Authentication with Identity Services Engine (ISE).
• Network Mobility Services communicates with the Mobility Services Engine to track location.
• Device Sensor watches the control packets that ingress from the attached end device and determine
what type of device is attached. Device Sensor uses multiple sources (such as IPDT) to determine
the device type. Device Sensor is critical to other features, such as Auto Smart Ports, and AutoConf.
• Auto Smart Ports and AutoConf are indirectly affected, because they are clients of Device Sensor.
The Device Sensor feature uses IPDT to aid in detection of attached device types.
• Address Resolution Protocol (ARP) snooping will be impacted if IPDT is disabled.
Recommended ways to disable IPDT at the interface levels:
Alternately, you can use the following method:
Configure an Interface for Access Mode
Step 1Use the switchport host command to perform the following configurations for the end devices on your
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
54
switch:
Page 65
Access Interface Connectivity
switchport access vlan 10switchport voice vlan 11
description IP Phone
• Configure the access interface for static access mode, which is single VLAN mode with no
negotiation.
• Configure the interface for Spanning Tree PortFast (STPF), which shortens the time it takes for the
interface to go into forwarding mode. We recommend STPF on interfaces that do not connect to
other bridging devices (Ethernet switches).
The default Administrative mode for Ethernet interfaces on a switch is dynamic auto. Dynamic mode
means the interface will negotiate to trunk mode if the networking device on the side of the link initiates
the negotiation to trunk (administrative mode “dynamic desirable”).
Configure VLAN Membership
Step 2Configure the VLANs for voice and data traffic.
VLAN configuration on an interface is dependent on the end device being used:
• IP phones, IP cameras, and access points are typically configured on separate VLANs.
• VLANs 10 and 11 are defined as the data and voice VLANs, respectively.
Recommendation: Do not use VLAN 1 for data or voice. VLAN 1 is the default VLAN on the 3850.
This is well documented and understood by experienced networking personnel. Thus VLAN 1 will be
more susceptible to attacks. Changing the VLAN IDs to something other than VLAN1 has been a long
standing Cisco recommendation for Ethernet switching
Configure Access Interface Connectivity
Create an Interface Description
Step 3Create a description for the interface to identify the end-device type.
TipWhen you create an interface description, you can quickly scan a long list of interfaces to learn how they
are used in your network.
Configure Security Features on an Access Interfaces
Step 4Enable port security features to protect the network from malicious or troublesome end devices.
The primary purpose of port security is to prevent an end device from overloading the switch with too
many source MAC addresses. Port security controls the MAC addresses remembered from the attached
network device. Port security controls how many MAC addresses are remembered, how long they are
remembered, and what happens when too many are remembered.
The MAC address limit is 11. When the end device exceeds 11 source MAC addresses, the ingress traffic
to the switch on those source MAC addresses is dropped.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
55
Page 66
Configure Access Interface Connectivity
switchport port-security maximum 11switchport port-securityswitchport port-security aging time 2switchport port-security aging type inactivityswitchport port-security violation restrict
ip arp inspection limit rate 100ip snoopping limit rate 100
NoteMAC addresses that are remembered on interfaces with port security do not appear in the dynamic MAC
address table; they appear in the static MAC address table.
Step 5Configure IP ARP inspection and (DHCP, IGMP, and so on) snooping to 100 p/s on the interface.
(Incoming ARP packets exceeding 100 p/s is not typical and is considered malicious. Those packets are
dropped and a syslog message is raised).
Access Interface Connectivity
Step 6Configure IP source guard to prevent IP address spoofing on the interface.
Step 7Enable storm control on broadcast and multicast packets on the interface to protect the network from a
flood of broadcast or multicast packets.
When the configured levels are exceeded, the switch sends an SNMP trap. The interfaces are not put into
a disabled state.
Unicast packets are blocked on egress and not ingress traffic. The switch drops unknown unicast packets
from being egressed to the end device, ensuring that only the packets intended for the end device are
forwarded.
Step 8Configure IPv6 security on the interface to secure the end devices from malicious or unexpected
operation by preventing them from transmitting IPv6 router advertisements, and IPv6 responses.
The applied policies are defined in the “Global System Configuration” workflow.
Configure QoS on an Access Interface
56
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 67
Access Interface Connectivity
auto qos voip cisco-phoneservice-policy input AutoQos-4.0-CiscoPhone-Input-Policyservice-policy output 2P6Q3T
Quality of Service (QoS) provides preferential treatment to certain types of traffic at the expense of
others. Without QoS, the switch offers best-effort service to each packet, regardless of the packet
contents or size. It sends the packets without any assurance of reliability, delay bounds, or throughput.
Aut0 QoS on the switch generates multiple service policies for various end devices. The service policy
that is generated depends with the end device type.
Step 9Apply service policies to a single access interface.
The switch then automatically generates the modular QoS command-line interface (MQC) service
policies needed for access.
This example identifies some of the service policy configurations.
Step 10Apply ingress and egress service policies.
Check the end device-specific configuration to see which service policy is recommended for an end
device.
Configure Access Interface Connectivity
Verify Access Interface Configurations
This following section describes the commands that you should use to use to confirm that your
configurations in this workflow are correctly applied to your switch:
Step 11Use the show running-configurationcommand to verify the operational configuration of the access
interfaces.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
57
Page 68
Configure Access Interface Connectivity
show storm-control
Key: U - Unicast, B - Broadcast, M - Multicast
Interface Filter State Upper Lower Current Action
Type
show ip snooping
Switch snooping is enabled
Switch gleaning is disabled
snooping is configured on following VLANs:
10-13,100
snooping is operational on following VLANs:
10-13,100
snooping is configured on the following L3 Interfaces:
Insertion of option 82 is disabled
circuit-id default format: vlan-mod-port
remote-id: 2037.0653.c800 (MAC)
Option 82 on untrusted port is not allowed
Verification of hwaddr field is enabled
Verification of giaddr field is enabled
snooping trust/rate is configured on the following Interfaces:
Interface Trusted Allow option Rate limit (pps)
----------------------- ------- ------------ ----------------
GigabitEthernet1/0/1 no no 100
Custom circuit-ids:
GigabitEthernet1/0/2 no no 100
Custom circuit-ids:
GigabitEthernet1/0/3 no no 100
Custom circuit-ids:
GigabitEthernet1/0/4 no no 100
Custom circuit-ids:
Use the show storm-control command to confirm that the interfaces are configured for storm control.
Access Interface Connectivity
Use the show ip snooping command to confirm that the interfaces are configured for snooping.
58
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 69
Access Interface Connectivity
show ip verify source
Interface Filter-type Filter-mode IP-address Mac-address Vlan
--------- ----------- ----------- --------------- ----------------- ---Gi1/0/1 ip active deny-all 10-11
Gi1/0/2 ip active deny-all 10
Gi1/0/3 ip active deny-all 12
Gi1/0/4 ip active deny-all 10
show port-security
Secure Port MaxSecureAddr CurrentAddr SecurityViolation Security Action
--------------------------------------------------------------------------Total Addresses in System (excluding one mac per port) : 0
Max Addresses limit in System (excluding one mac per port) : 4096
Policy endhost_ipv6_raguard configuration:
device-role host
Policy endhost_ipv6_raguard is applied on the following targets:
Target Type Policy Feature Target range
Gi1/0/1 PORT endhost_ipv6_raguard RA guard vlan all
Gi1/0/2 PORT endhost_ipv6_raguard RA guard vlan all
Gi1/0/3 PORT endhost_ipv6_raguard RA guard vlan all
Gi1/0/4 PORT endhost_ipv6_raguard RA guard vlan all
Use the show ip verify source command to confirm that the IP source guard is configured and working.
Use the show port-security command to confirm that access interfaces are configured for port security.
Configure Access Interface Connectivity
Use the show ip arp inspection interfaces command to confirm the rate and untrusted state of access
interfaces.
Use the show ipv6 nd raguard policy command to confirm that access interfaces are configured for
Router Advertisement Guard with specific policies.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Display Running Configuration for Access Interface Connectivity
Use the show ipv6 guard policy command to confirm the guard on access interfaces.
Use the show policy-map interface command to confirm the input and output service policies applied
to access interfaces.
Display Running Configuration for Access Interface
Connectivity
Step 1Show the recommended configuration for each end device type described in the beginning of this
workflow.
TipTo use the same interface configuration for multiple interfaces on the switch, use the interface range
command. This command allows you to issue a command once and have it apply to many interfaces.
Because most of the interfaces in the access layer are configured identically, using this command can
save a lot of time. For example, the following command allows you to enter commands simultaneously
on all 48 interfaces (GigabitEthernet 1/0/1 to GigabitEthernet 1/0/48).
.
NoteApply the interface range command to every switch stack member. This range command will work for
all interfaces on a single switch member. Enter the range command for each member.
IP Phone Access Interface
The following example displays the IP phone Access Interface information:
60
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 71
Access Interface Connectivity
show running-configuration
.
.
.
Description IP Phone
switchport host
switchport access vlan 10
switchport voice vlan 11
switchport port-security maximum 11
switchport port-security
switchport port-security aging time 2
switchport port-security aging type inactivity
switchport port-security violation restrict
ip arp inspection limit rate 100
ip snooping limit rate 100
ip verify source
switchport block unicast
storm-control broadcast level pps 1k
storm-control multicast level pps 2k
storm-control action trap
ipv6 nd raguard attach-policy endhost_ipv6_raguard
ipv6 guard attach-policy endhost_ipv6__guard
auto qos voip cisco-phone
service-policy input AutoQos-4.0-CiscoPhone-Input-Policy
service-policy output 2P6Q3T
Display Running Configuration for Access Interface Connectivity
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
61
Page 72
Display Running Configuration for Access Interface Connectivity
Display Running Configuration for Access Interface Connectivity
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
63
Page 74
Display Running Configuration for Access Interface Connectivity
Access Interface Connectivity
64
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 75
Access Control on the Wired Network
This workflow describes a phased approach to deploy IEEE 802.1x port-based authentication to provide
secure and identity-based access control at the edge of the switch stack network.
Prerequisites for Access Control on the Wired Network
• Before globally enabling IEEE 802.1x authentication, remove the EtherChannel configuration from
all of the interfaces.
• Define the authenticator (switch) to RADIUS server communication.
• Initiate Extensible Authentication Protocol (EAP) over LAN (EAPoL) messaging to successfully
authenticate the end device (or supplicant).
• Based on your requirements, choose an appropriate EAP method. For information, see the Wired
802.1x Deployment Guide.
• Automate the certificate enrollment process for supplicants, as described in the Certificate
Autoenrollment in Windows Server 2003.
• Enable machine authentication for end points, such as printers, to ensure that user login is supported.
Restrictions for Access Control on the Wired Network
• You cannot configure an IEEE 802.1x port that is a member of an EtherChannel.
• Destination ports configured with Switched Port Analyzer (SPAN) and remote SPAN (RSPAN)
cannot be enabled with IEEE 802.1x authentication.
• You cannot enable an IEEE 802.1x port on trunk or dynamic ports. Dynamic ports can negotiate with
its neighbors to become a trunk.
• Do not use port security with IEEE 802.1x. When IEEE 802.1x is enabled, port security then
becomes redundant and might interfere with the IEEE 802.1x functionality.
Identify Configuration Values
Cisco Systems, Inc.
www.cisco.com
Page 76
Identify Configuration Values
We recommend that you identify certain switch configuration values in advance so that you can proceed
without interruption. We recommend that you take a print out of Table 7, and, as you follow the
configuration sequence, replace the values in column B with your values in column C.
NoteDepending on your authentication server settings, the authentication and accounting ports could be
assigned the values 1812 and 1813 respectively.
NoteReplace the blue italicized example values with your own values.
Table 7 Secure Access Control for Wired Network Values
A. Value NameB. Example Value NamesC. Your Value
Interface range
RADIUS server
RADIUS server IPv4 address
Auth-port
Acct-port
RADIUS server encryption key
Data VLAN
Voice VLAN
Auth-server dead vlan
Extended IP ACL
Access Control on the Wired Network
GigabitEthernet
1/0/1-1/0/24
AuthServer
192.168.254.14
1656
1646
cisco123
10
11
20
LowImpactSecurity-acl
66
NoteConfiguration examples begin in global configuration mode, unless noted otherwise.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 77
Access Control on the Wired Network
LAN Access Switch Topology with IEEE 802.1x Secure Access Control
LAN Access Switch Topology with IEEE 802.1x Secure Access
Control
Figure 9LAN Access Switch Topology with IEEE 802.1x Secure Access Control
Catalyst 3850 stack in access
Switch management
VLAN 100
Authentication
Data VLAN 10
Printer
Data VLAN 10
Voice VLAN 11
Data VLAN 10
Server
Desktop user
direct connect
Desktop user
391703
Securing Access Using 802.1x on a wired LAN
The following tasks are to be performed in the same order that is listed here.
• Recommendations for Configuring Security on a Wired LAN
• Provision Common Wired Security Access
• Provision in Monitor Mode
• Provision in Low-Impact Mode
• Provision in High-Impact Mode
•
Recommendations for Configuring Security on a Wired LAN
IEEE 802.1x permits or denies network connectivity based on the identity of users and devices. It
provides a link between the user name and IP address, MAC address, and a port on a switch. It also
provides customized network access based on the identity of the end device or user.
The main components of IEEE 802.1x are:
• Supplicant (end device)
• Authenticator (switch)
• Authentication server (RADIUS or ISE)
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
67
Page 78
Securing Access Using 802.1x on a wired LAN
To provide secure access to your wired switch network, we recommend that you first provision your
common wired security features. Provision security modes in phased deployments (monitor mode to
high-security mode) of IEEE 802.1x authentication along with MAC Authentication Bypass (MAB),
which uses the MAC address of the end device (or supplicant) to make decisions about access.
NoteEach phased deployment should occur over time after ensuring that your network is ready to transition
to the next security mode.
Table 8 describes the recommended IEEE 802.1x deployment scenarios that will have limited impact on
network access. Test your network infrastructure while in monitor mode. If you are satisfied, then
transition to low-impact mode and allow a subset of network traffic to pass through. Finally, transition
to high-security mode, requiring authorization from all end devices.
• Multi-authentication mode authenticates all the devices that gain access to the network through a
single switch port, such as devices connected through IP phones.
• Multi-authentication mode is more secure than multi-host mode (which also allows multiple data
devices) because it authenticates all the devices that try to gain access to the network.
Step 1Run the show run command on your switch to ensure that your access interface connections are set up.
This output is what you inherit after performing the “Access Interface Connectivity” workflow
configuration for an interface connected to an IP phone.
Securing Access Using 802.1x on a wired LAN
Step 2(Optional) If you observe excessive timeouts, fine-tune the IEEE 802.1x timers and variables. Timers
and variables are important for controlling the IEEE 802.1x authenticator process on the switch.
We recommend that you do not change the IEEE 802.1x timer and variable default settings, unless
necessary.
Begin in interface configuration mode:
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
69
Page 80
Securing Access Using 802.1x on a wired LAN
dot1x timeout supp-timeout 30dot1x max-req 2
mab
authentication port-control auto
dot1x pae authenticator
!Enable new access control
!
aaa new-model
!
!Set authentication list for 802.1x
!
aaa authentication dot1x default group radius
!
!Enable 802.1x authentication
!
dot1x system-auth-control
Step 3Set the timers on the appropriate interfaces.
These timers and variables control IEEE 802.1x authenticator operations when end devices stop
functioning during authentication.
Begin in interface configuration mode.
Reference
For detailed information about the IEEE 802.1x timers and variables, see the Wired 802.1x Deployment
Guide.
Step 4Enable MAC authentication bypass (MAB) from interface configuration mode to authenticate
supplicants that do not support IEEE 802.1x authentication.
When MAB is enabled, the switch uses the MAC address of the device as its identity. The authentication
has a database of MAC addresses that are allowed network access.
We recommend that you enable MAB to support non-802.1x-compliant devices. MAB also is an
alternate authentication method when end devices fail IEEE 802.1x authentication due to restricted ACL
access.
Begin in interface configuration mode.
Access Control on the Wired Network
Step 5Configure IEEE 802.1x on the appropriate interfaces.
When you configure an IEEE 802.1x parameter on a port, a dot1x authenticator is automatically created
on the port. When that occurs, the dot1x pae authenticator command must also be configured to ensure
that the dot1x authentication will work on legacy configurations.
Begin in interface configuration mode:
Step 6Enable access control and IEEE 802.1x authentications.
Begin in global configuration mode.
70
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
authentication host-mode multi-auth
authentication open
no switchport port-security
no switchport port-security violation
no switchport port-security aging type
no switchport port-security aging time
no switchport port-security maximum
Step 7To establish the radius server, configure the RADIUS server with IP address, UDP port for authentication
and accounting server, and server encryption key.
Provision in Monitor Mode
Monitor mode enables IEEE 802.1x authentication without impacting the access of the end devices
(supplicants) to a switch (authenticator). This mode allows you to continuously gather the following
types of data for all the devices connected to your network:
• List of IEEE 802.1x-capable devices
• List of devices that are not capable of IEEE 802.1x
• Devices with good credentials
• Devices with bad credentials.
• List of valid MAC addresses (for MAB)
Securing Access Using 802.1x on a wired LAN
• List of unknown or invalid MAC addresses (for MAB)
We recommend monitor mode as a first-phase approach to provide secure access with IEEE 802.1x.
Although this mode authenticates the end devices and users (supplicants), traffic is not impacted if
authentication fails.
In monitor mode, IEEE 802.1x and MAB are enabled, but access is open to all users.
Step 8To allow hosts to gain access to a controlled port, configure multi-authentication host mode and open
authentication.
Step 9Disable the Port Security feature, because when IEEE 802.1x is enabled, the Port Security feature
becomes redundant and might interfere with the IEEE 802.1x functionality.
Begin in interface configuration mode.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
71
Page 82
Securing Access Using 802.1x on a wired LAN
authentication host-mode multi-domain
ip access-list extended LowImpactSecurity-acl
permit tcp any any established
permit udp any any eq bootps
permit udp any any eq tftp
permit udp any any eq domain
exit
interface GigabitEthernet1/0/1
ip access-group LowImpactSecurity-acl in
Provision in Low-Impact Mode
The next deployment phase in securing your network is to provision in low impact mode, which allows
differentiated network access to authenticated users while permitting basic network services for all
users.
NoteFor information about configuration of multiple-authentication mode on IEEE 802.1x ports, see
“Provision Common Wired Security Access”.
Minimize the impact to your initial network access settings and add differentiated network access to
authenticated users with low-impact mode provisioning. In low-impact mode, authentication is open and
network access is contained using less restrictive port ACLs. After authentication, dACLs are used to
allow full network access to end devices.
Step 10configure multi-domain mode to prevent unauthorized users from accessing an interface after an
authorized user has been authenticated.
Access Control on the Wired Network
Step 11Add a static ACL to allow basic network access.
Configure a restrictive port ACL that allows access for configuration and a Configured Trust List (CTL).
Begin in global configuration mode.
72
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 83
Access Control on the Wired Network
authentication host-mode multi-auth
authentication open
interface GigabitEthernet 1/0/1-1/0/24
no authentication open
authentication event server dead action authorize vlan 20
authentication dead action authorize voice
Provision in High-Impact Mode
The final deployment phase of securing your wired network is high-impact mode.
This phase goes beyond low-impact mode and provisions tight access control on the network port by
configuring the default IEEE 802.1x authentication mode with dynamic VLAN for differentiated access.
Step 12Configure multi-authentication host mode, and open authentication.
Step 13Disable RADIUS for this deployment phase.
High-impact mode provides no network access to devices and users that fail authentication. In monitor
mode and low-impact mode, we recommend that you identify and resolve the devices and user accounts
that have failed authentication. Transition to high-impact mode when you are confident that end devices
(that need network access) authenticate successfully, and authentication fails for devices and users that
do not need access.
Begin in global configuration mode.
Securing Access Using 802.1x on a wired LAN
Step 14Assign critical VLAN assignments for situations where the authentication server is unavailable.
The following command is used to configure a port to send both new and existing hosts to the critical
VLAN when the RADIUS server is unavailable. Use this command for ports in multiple authentication
(multiauth) mode or if the voice domain of the port is in MDA mode.
Step 15If the authentication server does not respond, authorize voice.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
73
Page 84
Show Running Configuration for Provisioning Modes
show running-configuration
hostname 3850-access-Bld1Flr1
!
!
aaa new-model
!
aaa authentication dot1x default group radius
!
ip device tracking
!
!
dot1x system-auth-control
!
!
interface GigabitEthernet1/0/1
switchport access vlan 10
switchport mode access
switchport block unicast
switchport voice vlan 11
ip arp inspection limit rate 100
trust device cisco-phone
authentication host-mode multi-auth
authentication open
authentication port-control auto
mab
dot1x pae authenticator
ip access-group LowImpactSecurity-acl in
authentication event fail action next-method
authentication host-mode multi-domain
authentication open
authentication port-control auto
mab
dot1x pae authenticator
storm-control broadcast level pps 1k
storm-control multicast level pps 2k
storm-control action trap
Ipv6 nd raguard attach-policy endhost_ipv6_raguard
Ipv6 guard attach-policy endhost_ipv6__guard
auto qos voip cisco-phone
service-policy input AutoQos-4.0-CiscoPhone-Input-Policy
service-policy output AutoQos-4.0-Output-Policy
ip verify source
ip snooping limit rate 100
!
!
ip access-group LowImpactSecurity-acl in
permit tcp any any established
permit udp any any eq bootps
permit udp any any eq tftp
permit udp any any eq domain
!
radius server AuthServer
address ipv4 192.168.254.14 auth-port 1645 acct-port 1646
key cisco123
Figure 11how running-configuration command for Provision in Low-Impact Mode
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
To detect errors, filter the dot1x verbose messages that are enabled by default.
Step 2Use the show dot1x interface statistics command to display IEEE 802.1x statistics for a specific port.
Monitoring IEEE 802.1x Status and Statistics
Step 3Use the show dot1x all command to display the IEEE 802.1x administrative and operational status for
a switch.
.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
77
Page 88
Monitoring IEEE 802.1x Status and Statistics
show dot1x interface g1/0/1
Dot1x Info for GigabitEthernet1/0/1
-----------------------------------
PAE = AUTHENTICATOR
QuietPeriod = 60
ServerTimeout = 0
SuppTimeout = 30
ReAuthMax = 2
MaxReq = 2
TxPeriod = 30
Step 4Use the show dot1x interface command to display the IEEE 802.1x administrative and operational
status for a specific port.
.
Access Control on the Wired Network
78
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 89
Converged Wired and Wireless Access
This workflow explains how to enable the converged access functionality of the switch, and explains how
the switch can operate as the wireless mobility controller (MC) as well as the wireless mobility anchor
(MA) in a small branch deployment.
Wired and wireless features that are enabled in the same platform is referred to as converged access. The
wired plus wireless features are bundled into a single Cisco IOS Software image, which reduces the
number of software images that users have to qualify and certify before enabling them in their network.
Converged access improves wireless bandwidth across the network and the scale of wireless deployment.
For example, a 48-port Catalyst 3850 switch provides 40 Gbps of wireless throughput. This wireless
capacity increases with the number of members in the stack. This ensures that the network will scale
with current wireless bandwidth requirements, as dictated by IEEE 802.11n-based access points and
with future wireless standards such as IEEE 802.11ac.
Prerequisites
Complete the following tasks before proceeding with wireless configuration:
• Switch stack must function in Stateful Switchover (SSO) mode.
• Interface configuration is completed, as explained in the “Access Interface Connectivity” workflow.
Restrictions
• Lightweight access points are used.
• NTP configuration should be present and operational, as explained in the “Global System
Configuration” workflow.
• A wireless site survey should be completed. The site survey identifies the proper placement of
wireless access points for the best coverage. For detailed information about the site survey process
and the tool to use, see the Wireless Site Survey FAQ.
• Complete the QoS workflow.
• AP-count licenses are supported only on IP Base and IP Services licenses. See the Cisco Catalyst
3850 Switch Right-to-Use Licensing Model.
Cisco Systems, Inc.
www.cisco.com
Page 90
Identify Configuration Values
• A Catalyst 3850 switch stack can support a maximum of 50 access points.
• A Cisco Catalyst 3650 stack can support a maximum of 25 access points.
• WLAN cannot use client VLAN 0.
Identify Configuration Values
We recommend that you identify certain switch configuration values in advance so that you are ready to
proceed with this section without interruption. As you follow the configuration sequence, replace the
values in column B with your values in column C.
NoteThis workflow contains two separate IP subnets that contain VLANs used for access points and wireless
clients. The access points are on VLAN 12, and use IP subnet 192.168.12.x. The wireless clients are on
VLAN 200, and use IP subnet 192.168.13.x.
NoteIn the configuration examples, you must replace the blue italicized example values with your own values.
Converged Wired and Wireless Access
Table 10 Wireless LAN Controller Values
A. Value NameB. Example Value NamesC. Your Value
Number of access point count
10/1, 15/2
licenses and slots
Management VLAN
Management VLAN access point
and description
IP address for VLAN interface
wireless-management-vlan
Wireless VLAN
Wireless Management VLAN
Interface
192.168.12.2 255.255.255.0
managing access points
Access point pool
Access point client pool
Default router for client
excluded address
Wireless management interface
Access interface
Description
WLAN interface for client
APVlan10-Pool
192.168.12.0 255.255.255.0
10.1.1.1
192.168.12.1
vlan12
GigabitEthernet1/0/3
Lightweight Access Point
200
VLAN
WLAN profile and ID
Wireless client VLAN IP address
WLAN for easy-RADIUS and
Wireless_Client
192.168.13.2 255.255.254.0
OPEN_WLAN 1 open_wlan
ID
RADIUS server
AuthServer
82
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
Page 91
Converged Wired and Wireless Access
Table 10 Wireless LAN Controller Values
A. Value NameB. Example Value NamesC. Your Value
IPv4 address for RADIUS
Auth-port
Acct-port
AAA group
RADIUS server dead-criteria
time/tries
RADIUS server deadtime
WLAN with WPA2 and IEEE
802.1x enabled
Input service policy
Output service policy
LAN Access Switch Topology with Wireless Connectivity
192.168.254.14
1645
1646
RADIUS-GROUP
10/3
1
Secure_WLAN1 CISCO_WLAN
wlan-Guest-Client-Input-Policy
wlan-Guest-SSID-Output-Policy
NoteConfiguration examples begin in global configuration mode, unless noted otherwise.
LAN Access Switch Topology with Wireless Connectivity
This topology shows the switch stack connected to multiple routers. The most common deployment of
converged access is in a branch scenario, but this workflow also applies to a campus deployment.
The switch is stacked and acts as both the MC and MA. In a single stack converged access deployment,
the switch can support up to 50 directly connected access points. For converged access, at least one
lightweight access point is required. A maximum of 50 access points can be supported by a switch stack.
We recommend that you distribute the access points equally across the stack to achieve reliability during
switchover scenarios preventing connectivity loss to access points connected to a member or standby
switch.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
83
Page 92
Enable the Switch as a Wireless Controller
Dual redundant
routers running
HSRP
Desktop user
direct connect
Desktop user
behind IP phone
Printer
Wireless access
Catalyst 3850 stack in access
DHCP Server
ISE
Voice VLAN 11
Data VLAN 10
Data VLAN 10
Data VLAN 10
Switch management
VLAN 100
Access point VLAN 12
391701
Trunk link
Native VLAN 999
All VLANs included
Figure 13LAN Access Switch Topology with Wireless Connectivity
Converged Wired and Wireless Access
Enable the Switch as a Wireless Controller
Install Access Point Licenses on the Switch
84
• Install Access Point Licenses on the Switch
• Configure a Wireless Management VLAN
• Configure Service Connectivity
• Enable Wireless Controller Functionality
• Change a Switch to Run in Mobility Controller Mode
• Enable the Access Point Connections
For ease of use, an evaluation license is preinstalled on your switch, but you are required to accept the
End-User-License Agreement (EULA) before the 90-day period expires.
The IP Base and IP Services image-based licenses support wireless functionality. The minimum license
level for wireless functionality is IP Base.
The total AP-count license of a switch stack is equal to the sum of all the individual member AP-count
licenses, up to a maximum of 50 AP-count licenses.
The total AP-count license of the stack is affected when stack members are added or removed:
• When a new member is added to the stack that has an existing AP-count license, then the total
available AP-count license for the switch stack is automatically recalculated.
• When members are removed from the stack, the total AP-count license is decremented from the total
available AP-count license in the stack.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
• If more access points are connected that exceed the total number of accepted AP-count licenses, a
syslog warning message is sent without disconnecting the newly connected access points until a
stack reload.
• After a stack reload, the newly connected access points are removed from the total access point
count.
You can activate permanent RTU licenses after you accept the EULA. The EULA assumes you have
purchased the permanent license. Use AP-count adder type licenses to activate access point licenses. The
adder AP-Count license is an “add as you grow” license. You can add access point licenses as your
network grows. You activate an adder AP-count license by using EXEC commands, and it is activated
without a switch reload.
Step 1Activate a permanent access point license and accept the EULA.
Access point licenses are configured for permanent or for evaluation purposes. To prevent disruptions in
operation, the switch does not change licenses when an evaluation license expires. You get a warning
that your evaluation license will expire and you must disable the evaluation license and purchase a
permanent one.
We recommend that you purchase and activate a permanent license and accept the EULA to avoid an
untimely expiration.
Enable the Switch as a Wireless Controller
The following examples activate 10 access point licenses on member 1 and 15 on member 2.
For more information about RTU licenses, see the “Configuring Right-To-Use Licenses” chapter in the System Management Configuration Guide, Cisco IOS SE Release 3E.
Verify AP-Count License Installation
Step 2Verify the allocation of the access point licenses on the switch.
The following example shows two members in the stack:
Step 3Verify the RTU license summary details.
The example shows that a permanent IP Services license is installed and is available upon switch reboot:
Five AP-count licenses are in use.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
License Level In Use: ipservices
License Level on Reboot: ipservices
Evaluation AP-Count: Disabled
Total AP Count Licenses: 25
AP Count Licenses In-use: 5
AP Count Licenses Remaining: 20
! To activate the VLAN in the database if it does not exist.
Step 4Configure the VLAN and SVI and assign it an IP address.
A wireless management VLAN is used for access point CAPWAP and other CAWAP mobility tunnels.
The creation of a wireless management VLAN is mandatory. First, configure the VLAN in hardware and
then create the SVI and assign it to an IP address. (See the “Create a Management VLAN in Hardware”
section in the Initial Switch Configuration workflow.)
Configure Service Connectivity
Step 5Create a name for the server address pool and specify the subnet network number and mask of the
address pool client, and the default router for the client.
If you want the switch to receive IP address information you must configure the server with the IP
address and subnet mask of the client and a router IP address to provide a default gateway for the switch.
86
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
The server uses the DNS server to resolve the TFTP server name to an IP address, but configuration of
the DNS server IP address is optional.
In small branch deployments in which the MC and MA are combined, we recommend using the switch
as the server for the lightweight access points. In this deployment, the switch operates in Layer 2 mode,
and the upstream router provides all routing functions.
We recommend that you exclude the IP address already used for the default router and the in-use wireless
management SVI address to prevent an upstream router from allocating this IP address to an access
point.
Page 95
Converged Wired and Wireless Access
ip pool APVlan10-Pool
network 192.168.12.0 255.255.255.0
default-router 192.168.12.1
ip excluded-address 192.168.12.1 192.168.12.2
wireless management interface vlan12
wireless mobility controller
%
Mobility role changed to Mobility Controller. Please save config and
reboot the whole stack.
end
write memory
reload
proceed with reload? [confirm] y
Enable Wireless Controller Functionality
Step 6Configure an SVI (rather than a physical interface) as the management VLAN.
The wireless management interface command is used to source the access point CAPWAP and other
CAPWAP mobility tunnels.
An SVI must be configured with an IP address before enabling the wireless controller.
Enable the Switch as a Wireless Controller
Change a Switch to Run in Mobility Controller Mode
Step 7Enable the switch as an MC before the AP-count license installation.
In the wireless licensing model, the MA is the access point enforcer and the MC is the gatekeeper of the
access points. The MC allows an access point to join the switch or not. The default role of the switch
after boot up is an MA.
It is mandatory to save the configuration and reload the switch for the MC role to take effect.
Step 8After the switch reboots, verify that the role of the switch has changed to Mobility Controller.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
87
Page 96
Enable the Switch as a Wireless Controller
show wireless mobility summary
Mobility Controller Summary:
Mobility Role : Mobility Controller
Mobility Protocol Port : 16666
Mobility Group Name : default
Mobility Oracle IP Address : 0.0.0.0
DTLS Mode : Enabled
Mobility Domain ID for 802.11r : 0xac34
Mobility Keepalive Interval : 10
Mobility Keepalive Count : 3
Mobility Control Message DSCP Value : 48
Mobility Domain Member Count : 3
Link Status is Control Link Status : Data Link Status
Step 9Connect the access points directly to the switch ports to complete installation.
It is mandatory that the access point connection port be configured as an access port. The access point
does not register if the port is configured as a trunk.
NoteThe access VLAN on the switch port should be the same as the wireless management VLAN configured
in Step 4 in this workflow.
88
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
!Activate the client VLAN in the VLAN database.
!Configure VLAN 200 if not already configured.
!
vlan 200
name Wireless_Client
end
!
interface vlan 200
description Client VLAN
ip address 192.168.13.2 255.255.254.0
no shutdown
end
Enable the Switch as a Wireless Controller
Enable a Client VLAN
Step 10Configure an external server to allocate IP addresses for clients. Define a client VLAN and activate the
VLAN in the database.
Every WLAN profile must be associated with a client VLAN.
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
89
Page 98
Provisioning a Small Branch WLAN
wlan OPEN_WLAN 1 open_wlan
client vlan 200
no security wpa
no security wpa akm dot1x
no security wpa wpa2
no security wpa wpa2 ciphers aes
no shutdown
Provisioning a Small Branch WLAN
• Provision in Easy-RADIUS—Easiest to configure and does not rely on outside services.
• Provision in Secure Mode—End-users are authenticated by the external RADIUS server or ISE.
• Manage Radio Frequency and Channel Settings
We highly recommend that secure mode be provisioned for security concerns. However, both WLAN
modes can co-exist if the network design requires it. For example, you can provision both WLANs on a
single switch with each WLAN having its own purpose in the network.
NoteIf your network does not permit open access for any wireless device, proceed to the “Provision in Secure
Mode” section and provision your wireless network in secure mode.
NoteGuest Access network deployment is beyond the scope of this document. For detailed information, see
the “Configuring Wireless Guest Access” chapter in the Security Configuration Guide, Cisco IOS XE
Release 3E, (Catalyst 3850 Switches).
Converged Wired and Wireless Access
Provision in Easy-RADIUS
Easy-RADIUS allows access to the network without authentication and is not secure.
• Disable Authentication to Enable Easy-RADIUS
• Configure QoS to Secure the WLAN
• Verify Client Connectivity in RADIUS
NoteIf your network does not permit open access for any wireless device, proceed to the “Provision in Secure
Mode” section and provision your wireless network in secure mode.
Disable Authentication to Enable Easy-RADIUS
Step 1To provision in easy-RADIUS, use the no security EXEC commands to disable authentication for a
WLAN.
By default, the WLAN is enabled for security with Wi-Fi Protected Access (WPA) and Wi-Fi Protected
Access II (WPA2). To make the WLAN open, use the no security wpa wpa2 command.
90
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
NoteBy default, the broadcast SSID is enabled, and the WLAN/SSID information is sent in the beacons. The
no broadcast-ssid command can be used to hide the SSID from being broadcast or made visible to end
clients. When the SSID broadcast is disabled, the end-users will still be able to connect to the SSID by
explicitly entering the SSID information manually in the wireless client network properties.
Configure QoS to Secure the WLAN
Step 2Configure a service policy on the ingress direction to properly classify traffic.
All ingress traffic is classified the same as wired traffic. On egress, the secure WLAN is given the
majority of the available bandwidth.
QoS configuration for a secure WLAN assumes that there is another WLAN with lower priority, such as
a guest or open WLAN. The end users on a secure WLAN should not be impacted by non-critical traffic
on other WLANs.
All WLANs share the default port_child_policy egress service policy. This policy is configured by
default and does not need to be explicitly configured on a WLAN.
Provisioning a Small Branch WLAN
Best Practice User Guide for the Catalyst 3850 and Catalyst 3650 Switch Series
91
Page 100
Provisioning a Small Branch WLAN
pol-edu-3850-mc-12#show wireless client summary
Number of Local Clients : 2