on
02-16-2010
10:43 AM
- edited on
03-25-2019
01:23 PM
by
ciscomoderator
CAE:
UCS Quick Start Guide
|
Author: Kristin Duran, kduran@cisco.com
Table of Contents
Setup Wizard (Management IP), etc. 7
Configure Network Connectivity. 9
Perform Initial System Validation. 17
Management IP Address Pools. 19
Create & Associate Service Profiles. 22
Create a Service Profile using the Expert Wizard. 23
Associate Service Profiles to Server Blade. 29
Additional UCS Management Configuration (Optional) 30
Install an Operating System on a Blade. 38
The purpose of this document is to describe the basic steps required to minimally configure Cisco’s Unified Computing System (UCS). Additional references to detailed Configuration Examples and TechNotes are provided within each section. Comprehensive configuration guides can be found at the following location:
http://www.cisco.com/en/US/products/ps10281/products_installation_and_configuration_guides_list.html
This document assumes basic knowledge of the various UCS components, including:
An elementary understanding of Local Area Network (LAN) and Storage Area Network (SAN) technologies is also assumed.
Cisco UCS is a next-generation data center platform that:
Information about UCS features and functionality can be found at http://www.cisco.com/go/ucs/.
The following tasks should be completed, and all information collected prior to beginning:
Below is an example of a typical UCS topology:
Cisco UCS Manager is the management service for all components in a Cisco UCS instance. Cisco UCS Manager runs within the Fabric Interconnects. You can use any of the interfaces available within this management service to access, configure, administer, and monitor the network and server resources for each chassis connected to the Fabric Interconnects.
Cisco UCS Manager includes the following interfaces you can use to manage a Cisco UCS instance:
Almost all tasks can be performed in any of the interfaces, and the results of tasks performed in one interface are automatically displayed in another.
The console port is an RS-232 port with an RJ-45 interface. The console port is an asynchronous (async) serial port; any device connected to this port must be capable of asynchronous transmission. The following figure shows how to connect to the console port on the Cisco UCS 6100 Fabric Interconnect.
The console port parameters on the computer terminal (or console server) attached to the console port are as follows:
Cisco UCS Manager GUI is the Java application that provides a GUI interface to Cisco UCS Manager. You can launch and access Cisco UCS Manager GUI from any computer that runs a supported operating system and has HTTP or HTTPS access to the UCS 6100 series Fabric Interconnect.
This section provides a brief overview of the Cisco UCS Manager GUI. The most commonly used areas include:
Navigation Pane |
The Navigation Pane has six tabs; Equipment, Servers, LAN, SAN, VM, and Admin. Each tab includes these elements:
Work Pane |
The Work Pane includes these elements:
While performing configuration steps in the UCS Manager GUI, you will find that there are often multiple ways to perform the same task. For example, you can right-click on an object in the Navigation Pane and select an action or you may be able to select that same action from the corresponding view in the Work Pane. In many cases, you may also have the ability to drag and drop objects within the Navigation Pane. As you get comfortable with the UCS Manager GUI, you can choose the method that best suits your work habits.
If you would like additional information on the UCS Manager GUI, see the document entitled “Use UCS Manager GUI to Manage Cisco UCS” on CCO.
The UCS Manager CLI is organized into a hierarchy of command modes, which often correspond to managed objects. Managed objects represent physical and logical components within UCS, such as servers, processors, Service Profiles, and policies.
The create, enter, scope, and exit commands are used to navigate the object hierarchy.
More detailed information on navigating the UCS Manager CLI can be found in the Cisco UCS Manager CLI Configuration Guide, Release 1.x.
The first time that you access a Fabric Interconnect in a Cisco UCS instance, a setup wizard prompts you for the following information required to configure the system:
The following procedure is used for the initial setup of the first Cisco UCS 6100 Fabric Interconnect.
If you choose to go through the setup wizard again, it will automatically provide the values you previously entered, and the values will appear in brackets. To accept previously-entered values, press the Enter key.
The following procedure is used for the initial setup of the second Cisco UCS 6100 Fabric Interconnect.
Note: The Fabric Interconnect should detect the peer Fabric Interconnect in the cluster. If it does not, check the physical connections between the L1 and L2 ports, and verify that the peer Fabric Interconnect has been enabled for a cluster configuration.
If you choose to go through the setup wizard again, it will automatically provide the values you previously entered, and the values will appear in brackets. To accept previously-entered values, press the Enter key.
Complete the following steps to access the UCS Manager GUI through HTTPS:
If a Security dialog box displays, click Yes in order to accept the certificate and continue. If desired, you can check the box in order to accept all content from Cisco.
After launching UCS Manager for the first time, you will see the Fabric Interconnects appear in the Navigation Pane. In order to have a view into the chassis, you will need to properly configure the Server Ports. Once you have properly configured the Server Ports using the information in this document, the next step is configuring your VLANs and VSANs, and finally the Uplink Ports to the rest of your LAN.
The type of network connectivity that is supported is dependent on the adaptor that is installed on your individual blade servers. Table 1 describes the available adaptors and their capabilities.
Table 1: UCS Network Adaptors and Capabilities
Adaptor | Maximum vNICs | Maximum vHBAs | Failover Supported |
82598KR | 2 | - | No |
M71KR-E/M71KR-Q | 2 | 2 | Yes |
M81KR | 15 * (min. acknowledged links between Fabric Extender and Fabric Interconnect) – 2 Mix and match of Ethernet and Fibre Channel | Yes |
With respect to the initial setup, the Chassis Discovery Policy serves one main purpose, which is to tell the system how many active links between the Fabric Extenders and Fabric Interconnects should be connected upon initial discovery of a chassis. For example, if the Chassis Discovery Policy is set to “4-link” and a newly discovered chassis has anything other than four links, the chassis will be visible with the "unsupported-connectivity" configuration, and otherwise unusable until the correct number of links are added, or the policy is changed.
This is a mandatory step in the initial UCS configuration.
To configure the Chassis Discovery Policy through the UCS Manager GUI:
To configure the Chassis Discovery Policy through the UCS Manager CLI:
FabricInterconnect-A# scope org
FabricInterconnect-A /org # scope chassis-disc-policy
FabricInterconnect-A /org/chassis-disc-policy # set action 4-link
FabricInterconnect-A /org/chassis-disc-policy* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /org/chassis-disc-policy # show
Chassis Discovery Policy:
Description Qualifier Action Rebalance
----------- ---------- ----------------- ---------
none 4 Link User Acknowledged
FabricInterconnect-A /org/chassis-disc-policy #
The Fabric Interconnect has two types of ports: Server and Uplink Ports. The Server Ports are what connect to the Fabric Extender on the chassis, and the Uplink ports are what connect to the upstream switch (the LAN). For more detailed information on establishing this connectivity, please see the CCO TechNote entitled “Set up Connectivity between Fabric Extender and Fabric Interconnect in UCS”.
To configure Server Ports through the UCS Manager GUI:
To configure Server Ports through the UCS Manager CLI:
FabricInterconnect-A# scope eth-server
FabricInterconnect-A /eth-server # scope fabric a
FabricInterconnect-A /eth-server/fabric # create interface 1 12
FabricInterconnect-A /eth-server/fabric/interface* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /eth-server/fabric # show interface
Interface:
Slot Id Port Id Admin State Oper State State Reason Chassis
------- ------- ------------ ------------ ----------------------- -------
1 1 Enabled Up 3
1 12 Enabled Up 1
1 2 Enabled Up 3
1 3 Enabled Up 3
1 4 Enabled Up 3
1 5 Enabled Up 1
1 6 Enabled Up 1
1 7 Enabled Up 1
1 8 Enabled Up 1
FabricInterconnect-A /eth-server/fabric #
In order for a blade server to pass Ethernet traffic over a particular VLAN, that VLAN must be created within UCS Manager, and then later associated with a vNIC. It is possible to create VLANs that are unique to either Fabric Interconnect. However, for the purposes of this document, we will create a Global VLAN, which is a VLAN that resides on both Fabric Interconnects.
A unique VLAN ID is required for each named VLAN. Note that VLANs with IDs from 3968 to 4048 are reserved.
To configure VLANs through the UCS Manager GUI:
To configure VLANs through the UCS Manager CLI:
FabricInterconnect-A# scope eth-uplink
FabricInterconnect-A /eth-uplink # create vlan PerfEngineering 199
FabricInterconnect-A /eth-uplink/vlan* # commit-buffer
FabricInterconnect-A /eth-uplink/vlan #
To verify after the change has been made:
FabricInterconnect-A /eth-uplink/vlan # show
VLAN:
Name VLAN ID Fabric ID Native VLAN
---------- ---------- --------- -----------
PerfEngineering
199 Dual No
FabricInterconnect-A /eth-uplink/vlan #
For additional information on VLAN configuration with UCS Manager, see the document entitled “VLAN Configuration for Cisco UCS” on CCO.
If you are utilizing VSANs within your Storage Area Network, you must create a VSAN in UCS Manager, and later associate it to a vHBA. Similar to a VLAN, a VSAN can either be Global or individual to a Fabric Interconnect. VLANs and VSANs differ in that each fiber uplink from UCS to its upstream SAN switch supports only one VSAN. This is specified as a property of the Uplink FC Port. While vNICs can support multiple VLANs, each vHBA can support only one VSAN.
To configure a global VSAN through the UCS Manager GUI:
To configure a global VSAN through the UCS Manager CLI:
FabricInterconnect-A# scope fc-uplink
FabricInterconnect-A /fc-uplink # create vsan PerfEngineering 199 299
FabricInterconnect-A /fc-uplink/vsan* # commit-buffer
FabricInterconnect-A /fc-uplink/vsan #
To verify after the change has been made:
FabricInterconnect-A /fc-uplink/vsan # show
VSAN:
Name Id FCoE VLAN Fabric ID
---------- ---------- ---------- ---------
PerfEngineering
199 299 Dual
FabricInterconnect-A /fc-uplink/vsan #
Please see the TechNote entitled “LAN and SAN Connectivity for a Cisco UCS Blade” on CCO for additional information on configuring SAN connectivity in UCS.
Uplink Ethernet ports connect your Fabric Interconnects to the upstream LAN switches. Uplink Fibre Channel ports connect your Fabric Interconnects to the upstream SAN switches. The procedure to configure Uplink Ports is similar to configuring Server Ports.
To configure Uplink Ethernet Ports through the UCS Manager GUI:
To configure Uplink Ethernet Ports through the UCS Manager CLI:
FabricInterconnect-A# scope eth-uplink
FabricInterconnect-A /eth-uplink # scope fabric a
FabricInterconnect-A /eth-uplink/fabric # create interface 1 20
FabricInterconnect-A /eth-uplink/fabric/interface* # commit-buffer
FabricInterconnect-A /eth-uplink/fabric/interface #
FabricInterconnect-A /eth-uplink/fabric/interface # exit
To verify after the change has been made:
FabricInterconnect-A /eth-uplink/fabric # show interface
Interface:
Slot Id Port Id Admin State Oper State State Reason
---------- ---------- ----------- ---------------- ------------
1 19 Enabled Up
1 20 Enabled Up
FabricInterconnect-A /eth-uplink/fabric #
For additional information on configuring Uplink Ports in a UCS environment, see the document entitled “UCS Uplink Ethernet Connection Configuration Example” on CCO.
To configure Uplink FC Ports through the UCS Manager GUI:
To configure Uplink Ethernet Ports through the UCS Manager CLI:
cae-ucs-rtp-B# scope fc-uplink
cae-ucs-rtp-B /fc-uplink # enter vsan VSAN100 100 100
cae-ucs-rtp-B /fc-uplink/vsan # create member-port a 2 8
cae-ucs-rtp-B /fc-uplink/vsan/member-port* #
cae-ucs-rtp-B /fc-uplink/vsan/member-port* # commit-buffer
cae-ucs-rtp-B /fc-uplink/vsan/member-port #
To verify after the change has been made:
cae-ucs-rtp-B /fc-uplink/vsan/member-port # show
Member Port:
Fabric ID Slot ID Port ID Oper State State Reason Oper Speed
--------- ---------- ---------- ---------------- ------------ ----------
A 2 8 Sfp Not Present SFP not present
Indeterminate
For additional information on configuring Uplink Ports in a UCS environment, see the document entitled “UCS Uplink Ethernet Connection Configuration Example” on CCO.
Once you have configured Server and Uplink ports, it is a good idea to validate the hardware has been properly discovered and that you are running the desired version of firmware.
The Hybrid Display provides an excellent visual depiction of the components within your UCS deployment, including the connected ports.
To validate the system hardware through the UCS Manager GUI:
Chassis |
Fabric Interconnect |
Fabric Extender |
Blade Server |
To view the hardware details of a blade server through the UCS Manager GUI:
Look at the Inventory tab to see what type of hardware is installed in your system. Make note of the Adaptor.
While not mandatory, it may be desirable to note the firmware versions that are running on the various components within UCS.
To validate system software through the UCS Manager GUI:
To validate the system software through the UCS Manager CLI:
FabricInterconnect-A # scope firmware
FabricInterconnect-A /firmware # show image
Name Type Version State
----------------------------------- -------------------- --------------- -----
diag-nuova-or-4.0.0.062.bin Unspecified 4.0(1a) Active
ucs-2100.1.0.1.6.gbin Iom 1.0(1.6) Active
ucs-2100.1.1.0.158i.gbin Iom 1.1(0.158i) Active
ucs-2100.1.1.0.215.bin Iom 1.1(0.215) Active
ucs-2100.1.1.0.217a.bin Iom 1.1(0.217a) Active
ucs-2100.1.1.0.282.bin Iom 1.1(0.282) Active
ucs-2100.1.1.0.89c.gbin Iom 1.1(0.89c) Active
...
FabricInterconnect-A /firmware #
For additional information on software management and validation in UCS, see the following documents on Cisco.com:
One of the key benefits of UCS is stateless computing. Stateless computing refers to the ability to move the “identity” of one server to another using the concept of Service Profiles. In order to take advantage of stateless computing, traditionally derived (burned-in) addresses would need to be individually assigned to each Service Profile, which functions as a type of abstraction layer between the hardware and the Operating System. This assignment can either be done on a per-Service Profile basis or by creating a Pool of addresses and allowing UCS to decide which address to assign.
Examples of resources that can be pooled include Management IP Addresses, MAC Addresses, WWNN, WWPN, and UUID values. Additional information about pools can be found in the CCO TechNote entitled “Create Pools to Simplify Blade Management in Cisco UCS”.
The Management IP Address pool facilities the assignment of a management IP address to an individual blade server. The Management IP Address is used for Serial Over LAN (SOL) or IPMI access. Note that the Management IP Address pool currently has to be on the same subnet as the Management interface of the UCS Manager.
To configure Management IP Address Pools through the UCS Manager GUI:
To configure Management IP Address Pools through the UCS Manager CLI:
FabricInterconnect-A# scope org
FabricInterconnect-A /org # scope ip-pool ext-mgmt
FabricInterconnect-A /org/ip-pool # create block 172.18.100.10 172.18.100.60 172.18.100.1 255.255.255.0
FabricInterconnect-A /org/ip-pool # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /org/ip-pool # show
IP Pool:
Name Size Assigned
-------------------- ---------- --------
ext-mgmt 50 10
FabricInterconnect-A /org/ip-pool #
A MAC Address Pool eases the administrative burden of assigning individual MAC addresses to each Service Profile.
To configure MAC Address Pools through the UCS Manager GUI:
To configure MAC Address Pools through the UCS Manager CLI:
FabricInterconnect-A# scope org
FabricInterconnect-A /org # create mac-pool MyMACPool
FabricInterconnect-A /org/mac-pool* # create block 00:25:B5:40:00:00 00:25:B5:40:00:31
FabricInterconnect-A /org/mac-pool/block* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /org/mac-pool # show
MAC Pool:
Name Size Assigned
-------------------- ---------- --------
MyMACPool 50 1
FabricInterconnect-A /org/mac-pool #
WWNN and WWPN Pools ease the administrative burden of assigning individual WWNN and WWPN addresses to each Service Profile. In a typical UCS deployment, you would create fairly large WWN pools. After they are assigned to a Service Profile, which is then assigned to a blade server, you would note the WWN assignment and configure your SAN accordingly.
To configure WWN Pools through the UCS Manager GUI:
To configure WWN Pools through the UCS Manager CLI:
FabricInterconnect-A# scope org
FabricInterconnect-A /org # create wwn-pool WWNTEST node-wwn-assignment
FabricInterconnect-A /org/wwn-pool* # create block 20:00:00:25:b5:00:00:00 20:00:00:25:b5:00:00:31
FabricInterconnect-A /org/wwn-pool/block* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /org/wwn-pool # show
WWN Pool:
Name Purpose Size Assigned
-------------------- ------------------- ---------- --------
MyWWNNPool Node Wwn Assignment 50 0
FabricInterconnect-A /org/wwn-pool #
The utilization of UUID Suffix Pools within a Service Profile eliminates the need to manually configure the UUID of the server associated with the Service Profile.
To configure UUID Suffix Pools through the UCS Manager GUI:
To configure UUID Suffix Pools through the UCS Manager CLI:
FabricInterconnect-A# scope org
FabricInterconnect-A /org # create uuid-suffix-pool MyUUIDPool
FabricInterconnect-A /org/uuid-suffix-pool* # create block 0000-000300000001 0000-000300000032
FabricInterconnect-A /org/uuid-suffix-pool/block* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /org/uuid-suffix-pool # show
UUID Suffix Pool:
Name UUID Prefix Size Assigned
-------------------- ------------------ -------- --------
MyUUIDPool 8C1D5178-8C02-11D9 50 1
FabricInterconnect-A /org/uuid-suffix-pool #
The use of Policies is not mandatory during the initial UCS configuration. However, it is important to be aware that there are various Policies that can be created and applied to Service Profiles. For detailed information about this feature, please see the CCO TechNote entitled “Configure Chassis and Server Discovery Policies for Cisco UCS”.
You should be somewhat familiar with the concept of Service Profiles from the previous sections. The Service Profile, or logical server, is the fundamental backbone of the stateless capabilities within UCS. The Service Profile represents a logical view of a single blade server, without the detailed knowledge of the underlying hardware. The profile object contains the server personality, for example, the values contained in the pools that were created in the previous steps.
Service Profiles can be created manually, cloned from an existing Service Profile, or created in batch using a Service Profile Template. This guide will focus on the manual creation of a Service Profile. For more detailed information on Service Profiles, please see the following CCO TechNotes:
This document will guide you through the creation and association of a single Service Profile to a blade server. Please note that this process should be repeated for each blade server in the chassis.
While a Service Profile can exist that inherits the identity of the blade server, we will focus on creating a Service Profile that overrides the server’s identity, which is required for stateless computing.
In this procedure, we will use the Expert Wizard to create a new Service Profile:
Create exactly one vNIC in the profile for each port you would like to make accessible to your blade server’s operating system.
Use Table 1 in the Configure Network Connectivity section of this document to determine the maximum number of vNICs that are supported with your exact hardware configuration.
Once you click Add, the “Create vNICs” pop-up window appears
vNIC/vHBA Placement specifies the mezzanine card on which the vNICs and vHBAs are placed. This step of the setup wizard is only applicable on the B250 full width blade server, which has multiple mezzanine cards.
A 1:1 mapping of Service Profiles to blade servers can exist. You cannot assign a Service Profile to more than one blade server at any given time.
To bind a Service Profile through the UCS Manager GUI:
To bind a Service Profile through the UCS Manager CLI:
FabricInterconnect-A# scope org
FabricInterconnect-A /org # scope service-profile MyServiceProfile
FabricInterconnect-A /org/service-profile # associate server 1/1
FabricInterconnect-A /org/service-profile* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /org/service-profile # show
Service Profile:
Service Profile Name Type Server Assignment Association
-------------------- ----------------- ------- ---------- -----------
MyServiceProfile Instance 1/1 Assigned Associating
FabricInterconnect-A /org/service-profile #
The items listed below are optional, but common configuration settings that would normally be configured as part of an initial UCS deployment.
Configure a DNS server that will be used to resolve hostnames to IP addresses within UCS.
More detailed information on configuring DNS can be found in the following CCO TechNote: Configure the DNS Server for Cisco UCS.
To configure DNS through the UCS Manager GUI:
To configure your UCS to utilize a DNS Server with address 192.168.1.1:
FabricInterconnect-A# scope system
FabricInterconnect-A /system # scope services
FabricInterconnect-A /system/services # create dns 192.168.1.1
FabricInterconnect-A /system/services* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /system/services # show dns
Domain Name Servers:
IP Address: 192.168.1.1
Use this section to help you configure SNMP for both polling and sending Traps. UCS supports SNMP Version 2c and Version 3, but this document will guide you through configuring Version 2c.
For detailed information on the information that can be gathered via SNMP from UCS, please see the following link on CCO: MIB Quick Reference for Cisco UCS Manager, Release 1.x.
To configure SNMP through the UCS Manager GUI:
To set the SNMP community to ‘public’ (Note: It is recommended that you choose a different community other than ‘public’ for both SNMP and SNMP Traps), execute the following:
FabricInterconnect-A# scope monitoring
FabricInterconnect-A /monitoring # set snmp community public
To verify after the change has been made:
FabricInterconnect-A /monitoring # show snmp
Name: snmp
Admin State: Disabled
Port: 161
Community: public
To set the SNMP-Trap community to ‘public’, execute the following
FabricInterconnect-A# scope monitoring
FabricInterconnect-A /monitoring # create snmp-trap 192.168.1.1
FabricInterconnect-A /monitoring/snmp-trap* # set community public
FabricInterconnect-A /monitoring/snmp-trap* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /monitoring/snmp-trap # show detail
SNMP Trap:
Hostname: 192.168.1.1
Port: 162
Community: public
Version: V2c
V3 Privilege: Noauth
To enable SNMP after the settings have been configured:
FabricInterconnect-A /monitoring # enable snmp
To verify after the change has been made:
FabricInterconnect-A /monitoring # show snmp
Name: snmp
Admin State: Enabled
Port: 161
Community: public
UCS has standard syslogging capabilities and can write to a local and remote facility. More detailed information on configuring Syslog can be found in the following CCO TechNote:
To configure Syslog through the UCS Manager GUI:
To configure Syslog through the UCS Manager GUI:
FabricInterconnect-A# scope monitoring
FabricInterconnect-A /monitoring # set syslog remote-destination server-1 hostname syslog1.cisco.com state enabled
FabricInterconnect-A /monitoring* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /monitoring # show syslog
console
state: Disabled
level: Critical
monitor
state: Disabled
level: Critical
file
state: Enabled
level: Critical
name: messages
size: 4194304
remote destinations
Name Hostname State Level Facility
-------- -------------------- -------- ------------- --------
Server 1 syslog1.cisco.com Enabled Critical Local7
Server 2 none Disabled Critical Local7
Server 3 none Disabled Critical Local7
FabricInterconnect-A /monitoring #
UCS can synchronize its clock with an NTP server on your network. Use the following procedure to configure this feature.
To configure NTP through the UCS Manager GUI:
To configure your UCS to utilize an NTP Server with address 192.168.1.1:
FabricInterconnect-A# scope system
FabricInterconnect-A /system # scope services
FabricInterconnect-A /system/services # create ntp 192.168.1.1
FabricInterconnect-A /system/services* # commit-buffer
To verify after the change has been made:
FabricInterconnect-A /system/services # show ntp
NTP Servers:
IP Address: 192.168.1.1
Before configuring UCS for TACACS support, please ensure that your TACACS Server has been pre-configured with the appropriate key and that the individual IP addresses of each Fabric Interconnect have been added accordingly. Note that the TACACS request will be sourced from the individual IP address of each Fabric Interconnect.
For more details on configuring UCS for TACACS authentication, as well as the corresponding configuration required on the CiscoSecure ACS Server (if applicable), see the CCO TechNote entitled Setup TACACS Authentication for Cisco UCS.
To configure TACACS through the UCS Manager GUI:
Once a TACACS+ server has been configured, enable TACACS authentication using the following steps:
Set the TACACS Server IP address to 192.168.1.1:
FabricInterconnect-A# scope security
FabricInterconnect-A /security #
FabricInterconnect-A /security/tacacs # create server 192.168.1.1
FabricInterconnect-A /security/tacacs/server* # set key
Enter the key: <enter the key here – it will not display on the screen>
Confirm the key: <confirm>
FabricInterconnect-A /security/tacacs/server* #
FabricInterconnect-A /security/tacacs/server* # set port 49
FabricInterconnect-A /security/tacacs/server* # commit-buffer
FabricInterconnect-A /security/tacacs/server #
To verify after the change has been made:
FabricInterconnect-A /security/tacacs/server # show detail
AAA server:
Hostname or IP address: 192.168.1.1
Order: 1
Port: 49
Key: ****
FabricInterconnect-A /security/tacacs/server #
Verify and change Authentication settings:
To verify current settings:
FabricInterconnect-A# scope security
FabricInterconnect-A /security #
FabricInterconnect-A /security # show authentication
Console authentication: Local
Default authentication: Local
Role Policy For Remote Users: Assign Default Role
Set the default authentication method to TACACS for remote connections (SSH, Telnet, HTTP, HTTPS):
FabricInterconnect-A /security # set authentication default tacacs
FabricInterconnect-A /security* #commit-buffer
To verify after the change has been made:
FabricInterconnect-A /security # show authentication
Console authentication: Local
Default authentication: Tacacs
Role Policy For Remote Users: Assign Default Role
Before configuring UCS for RADIUS support, please ensure that your RADIUS Server has been pre-configured with the appropriate key and that the individual IP addresses of each Fabric Interconnect have been added accordingly. Note that the RADIUS request will be sourced from the individual IP address of each Fabric Interconnect.
For more details on configuring UCS for RADIUS authentication, as well as the corresponding configuration required on the CiscoSecure ACS Server (if applicable), see the CCO TechNote entitled Set up RADIUS Authentication for Cisco UCS.
To configure RADIUS through the UCS Manager GUI:
Once a RADIUS server has been configured, enable RADIUS authentication using the following steps:
To configure RADIUS through the UCS Manager CLI:
FabricInterconnect-A# scope security
FabricInterconnect-A /security # scope radius
FabricInterconnect-A /security/radius # create server 192.168.1.1
FabricInterconnect-A /security/radius/server* # set key
Enter the key:
Confirm the key:
FabricInterconnect-A /security/radius/server* #
FabricInterconnect-A /security/radius/server* # commit-buffer
FabricInterconnect-A /security/radius/server#
To verify after the change has been made:
FabricInterconnect-A /security/radius/server # show detail
AAA server:
Hostname or IP address: 192.168.1.1
Order: 1
Port: 1812
Key: ****
FabricInterconnect-A /security/radius/server #
Verify and change Authentication settings:
To verify current settings:
FabricInterconnect-A# scope security
FabricInterconnect-A /security #
FabricInterconnect-A /security # show authentication
Console authentication: Local
Default authentication: Local
Role Policy For Remote Users: Assign Default Role
Set the default authentication method to RADIUS for remote connections (SSH, Telnet, HTTP, HTTPS):
FabricInterconnect-A /security # set authentication default radius
FabricInterconnect-A /security* #commit-buffer
To verify after the change has been made:
FabricInterconnect-A /security # show authentication
Console authentication: Local
Default authentication: Radius
Role Policy For Remote Users: Assign Default Role
A UCS configuration backup takes a snapshot of all or part of the system configuration and exports the backup file to a location on your network. The backup can be performed while the system is up and running. The backup operation only saves information from the management plane. It does not have any impact on the server or network traffic.
To create a backup operation through the UCS Manager GUI:
To create a backup operation through the UCS Manager CLI:
cae-ucs-rtp-B# scope system
cae-ucs-rtp-B /system # create backup sftp://admin@backup.cisco.com///admin@backup.cisco.com/ all-configuration enabled
cae-ucs-rtp-B /system/backup* # set remote-file MyUCSBackup
cae-ucs-rtp-B /system/backup* # set password
Password:
cae-ucs-rtp-B /system/backup* # commit-buffer
To verify after the change has been made:
cae-ucs-rtp-B /system/backup # show
Backup:
Hostname Type User Protocol Administrative State Description
---------- --------------------- ---------- -------- -------------------- -----------
backup.cisco.com
All Configuration admin Sftp Enabled
cae-ucs-rtp-B /system/backup #
Installing an operating system is the last step in the initial configuration of UCS. This step is not covered in this document, but great detail on this topic can be found in the following in the chapter entitled “Installing an OS on a Server” within the Cisco UCS Manager GUI Configuration Guide, Release 1.x on CCO.
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: