The documentation set for this product strives to use bias-free language. For the purposes of this documentation set, bias-free is defined as language that does not imply discrimination based on age, disability, gender, racial identity, ethnic identity, sexual orientation, socioeconomic status, and intersectionality. Exceptions may be present in the documentation due to language that is hardcoded in the user interfaces of the product software, language used based on RFP documentation, or language that is used by a referenced third-party product. Learn more about how Cisco is using Inclusive Language.
This chapter includes the following sections:
You can install Cisco UCS Central using either one of the following:
Cisco UCS Central, release 1.1 and newer provides you the option to install in a standalone or cluster set up. You must obtain the software from Cisco.com and save it in your local drive before installation.
Make sure that you have your Cisco.com username and password ready to be able to successfully download the Cisco UCS Central software.
Step 1 | In a web browser, navigate to Cisco.com. |
Step 2 | Under Support, click All Downloads. |
Step 3 | In the center pane, click Unified Computing and Servers. |
Step 4 | If prompted, enter your Cisco.com username and password to log in. |
Step 5 | In the right
pane, click the link for the
Cisco UCS Central
software in the format that you want.
You can download the Cisco UCS Central software in the following formats:
You can also download the admin password reset ISO image from this location. |
Step 6 | On the page from which you download the software, click the Release Notes link to download the latest version of the Release Notes. |
Step 7 | Click the link for the release of the Cisco UCS Central software that you want to download. |
Step 8 | Click one of the following buttons and follow the instructions provided: |
Step 9 | Follow the prompts to complete your download of the software. |
Step 10 | Read the Release Notes before deploying the Cisco UCS Central VM. |
You can install Cisco UCS Central through either the OVA file or ISO Image in standalone mode.
Note | After the Cisco UCS Central VM starts up for the first time, it performs a one-time post-installation configuration. Allow the installation to complete before you log in. |
Note | For VMware, we recommend that you install from the OVA file. Installing from the ISO file works with limitations. See https://tools.cisco.com/bugsearch/bug/CSCuv32055 for more information. |
Step 1 | Create a VM with
the following settings:
| ||||||||||||||
Step 2 | In the settings for the VM, do the following: | ||||||||||||||
Step 3 | Under the same IDE controller as the first virtual drive, create a second virtual drive for the VM with no less than 40GB of available disk space. | ||||||||||||||
Step 4 | In the Time synchronization to disable it. , uncheck | ||||||||||||||
Step 5 | Mount the Cisco UCS Central ISO image to the CD/DVD drive. | ||||||||||||||
Step 6 | Start the VM and connect to the console. | ||||||||||||||
Step 7 | From the
Cisco
UCS Central Installation menu on the ISO image, choose
Install
Cisco UCS Central.
The Cisco UCS Central installer checks that the VM has the required RAM and disk space (two disks, both with 40GBs). If the VM meets the requirements, the installer formats the disks, transfers the files, and installs Cisco UCS Central. | ||||||||||||||
Step 8 | When the Cisco
UCS Central VM has completed the initial part of the installation process,
answer the following questions in the VM console window:
| ||||||||||||||
Step 9 | Unmount the Cisco UCS Central ISO image from the virtual CD/DVD drive. | ||||||||||||||
Step 10 | Reboot the Cisco UCS Central VM. |
Note | When you install Cisco UCS Central on KVM Hypervisor, the set up runs in graphics mode. Installing in text mode is not supported. |
Step 1 | Create a VM with
the following settings:
| ||||||||||||
Step 2 | Under the same IDE controller as the first virtual drive, create a second virtual drive for the VM with no less than 40GB of available disk space. | ||||||||||||
Step 3 | Mount the Cisco UCS Central ISO image to the CD/DVD drive. | ||||||||||||
Step 4 | Start the VM and connect to the console. | ||||||||||||
Step 5 | From the
Cisco
UCS Central Installation menu on the ISO image, choose
Install
Cisco UCS Central.
The Cisco UCS Central installer checks that the VM has the required RAM and disk space (two disks, both with 40GBs). If the VM meets the requirements, the installer formats the disks, transfers the files, and installs Cisco UCS Central. | ||||||||||||
Step 6 | When the Cisco
UCS Central VM has completed the initial part of the installation process,
answer the following questions in the VM console window:
| ||||||||||||
Step 7 | Unmount the Cisco UCS Central ISO image from the virtual CD/DVD drive. | ||||||||||||
Step 8 | Reboot the Cisco UCS Central VM. |
You can install Cisco UCS Central in two virtual machines in a highly available configuration. In cluster mode, one VM acts as the primary node and the other as the secondary node. This cluster configuration provides redundancy and high availability in case of a VM failure.
With a cluster configuration, the VMs use shared storage on a LUN disk for the database and the images repository. As a result, you download firmware images to the shared storage. The statistics used for reporting are also collected and stored in the shared storage.
When you install Cisco UCS Central in a cluster configuration, be aware of the following guidelines:
The two VMs in the cluster are not on the same server. If both VMs are in the same server, a single host failure would bring down the cluster.
Both hosts must have the same versions of ESX, HyperV or KVM installed on them.
Both hosts must share same lun to configure shared storage.
Both VMs must be on the same sub-net.
You must install the same release version of Cisco UCS Central in both VMs.
You must install the first node completely and then install the second node. Parallel installation may corrupt or overwrite the partition table and shared storage might lose all deployed data.
In a cluster configuration, Cisco UCS Central supports RDM only with a single path, but not with multi-path.
For Cisco UCS Central to perform well in cluster mode, make sure to configure and connect shared storage according to the industry best practices. Be aware of the following guidelines:
Configure high speed SAN connections to enable quick access to shared storage.
Select a better performing RAID type to configure the shared LUN.
Make sure the storage is write cache enabled with enough space, proper page size and watermark settings. For example, if you have an EMC storage array, you should have the following cache configuration:
You must have only one path to the shared storage. Multi-path is not supported. If there is more than one path when you add the shared storage to Cisco UCS Central VM, you must disable all additional paths. To disable additional paths,
When you add shared storage to Node B, make sure to directly connect to the ESX host. You must add the disk as an RDM.
If you are installing Cisco UCS Central in cluster mode, you must use only NFS shared storage based high availability. RDM shared storage is not supported for KVM Hypervisor.
The NFS server stores database and images used by various applications on Cisco UCS Central. The initial configuration of NFS server might take sometime. Make sure to wait until the Timeout message ends.
The NFS server must be located as close to Cisco UCS Central VM as possible. We recommend that the NFS server is located in the same subnet as Cisco UCS Central.
Step 1 | Create a partition or a volume with 40Gb or more. |
Step 2 | Export the NFS
Directory.
Use an editor to open file /etc/exports and export the NFS directory. Example: /nfs *(rw,sync,no_root_squash) Where /nfs is the directory which needs to be exported. You must export the directory with the options rw and no_root_squash |
Step 3 | Restart the NFS
service.
Example:/sbin/service nfs restart |
Step 4 | Delete any firewall rules on the NFS server that may block the mount of NFS server directories from Cisco UCS Central VM. |
You can edit and change an existing NFS server or the NFS shared storage directory using the following commands:
You can change the shared storage from RDM to NFS using Cisco UCS Central CLI. Make sure to change to NFS server in the primary node. Changing the secondary node separately is not required.
You can log into the system after this massage and enter showstorage-device to confirm the storage device is changed.
Note | The Cisco UCS Central VM performs a one-time installation process the first time it starts up. Allow the installation to complete before you log in. |
Make sure you have the following information:
Network data such as hostname, IP address, default gateway, DNS server, and DNS domain name
Whether you are setting up a new cluster
Admin username and password
Shared secret for communications between the cluster nodes and with Cisco UCS Manager
IP address of the peer Cisco UCS Central node
Virtual IP address
Step 1 | Save the Cisco UCS Central OVA or ISO file to a folder that you can access from the hypervisor. |
Step 2 | Open or import
the
Cisco UCS Central
OVA file into a supported hypervisor, as required by the hypervisor.
Do not continue with the next step until the VM has finished booting. |
Step 3 | Add shared storage. See Adding and Setting up an RDM Shared Storage on VMware or Adding and Setting up an RDM Shared Storage on Hyper-V. |
Step 4 | Power up the Cisco UCS Central VM. |
Step 5 | Open up a console window to the Cisco UCS Central VM. |
Step 6 | When the
Cisco UCS Central
VM has completed the initial part of the installation process, answer the
following questions in the VM console window:
|
Make sure you have the following information:
Step 1 | Save the Cisco UCS Central OVA or ISO file to a folder that you can access from the hypervisor. |
Step 2 | Open or import
the
Cisco UCS Central
OVA file into a supported hypervisor, as required by the hypervisor.
Do not continue with the next step until the VM has finished booting. |
Step 3 | Setup the shared storage. See Adding and Setting up an RDM Shared Storage on VMware or Adding and Setting up an RDM Shared Storage on Hyper-V. |
Step 4 | Power up the Cisco UCS Central VM. |
Step 5 | Open up a console window to the Cisco UCS Central VM. |
Step 6 | When the
Cisco UCS Central
VM has completed the initial part of the installation process, answer the
following questions in the VM console window:
|
You must have only one path to the shared storage. Multi-path is not supported.
If there is more than one path when you add the shared storage to Cisco UCS Central VM, you must disable all additional paths. To disable additional paths,
Step 1 | Create a LUN
which is no less that 40 GB on the storage array and map it to the Hyper-V host
on which you have installed Node A.
The disk must be in offline mode. |
Step 2 | Add the disk to the VM as a 'Physical hard disk' using a new SCSI Controller. |
Step 3 | In Powershell command windows, run the Set-ExecutionPolicy unrestricted command. |
Step 4 | Disabling SCSI
filtering is required for SCSI-3 PGR to work. To disable SCSI filtering for
this disk, save and run the following script in both Hyper-V machines, with the
virtual machine's name as the parameter:
Script: $HyperVGuest = $args[0] $VMManagementService = gwmi Msvm_VirtualSystemManagementService -namespace "root\virtualization" foreach ($Vm in gwmi Msvm_ComputerSystem -namespace "root\virtualization" -Filter "elementName='$HyperVGuest'") { $SettingData = gwmi -Namespace "root\virtualization" -Query "Associators of {$Vm} Where ResultClass=Msvm_VirtualSystemGlobalSettingData AssocClass=Msvm_ElementSettingData" $SettingData.AllowFullSCSICommandSet = $true $VMManagementService.ModifyVirtualSystem($Vm,$SettingData.PSBase.GetText(1)) | out-null } Example:If you have stored the script in Hyper-V host, for example in c:\, if Node A's name is UCSC-Node-1 and the file name is DisableSCSIFiltering.ps1, then open the Powershell window to run the script: C:\> .\DisableSCSIFiltering.ps1 UCSC-Node-1. You must run this script on both Hyper-V machines with the right VM names. |
Step 5 | Map the LUN you
added to Node A to the Hyper-V host on which you have installed node B.
This enables both Hyper-V hosts to view the same LUN. |
Step 6 | Add this LUN to Node B. |
Note | In a cluster set up, when the RDM link goes down on the primary node, DMEs cannot write to the database. This causes a crash on the primary node and failover to the subordinate node. The subordinate node takes over as the primary node. The database is then mounted in read-write mode on the new primary node. Because the RDM link is down, umount fails on the old primary node. When the RDM link comes up, the database is mounted on the old primary (current subordinate) node in read-only mode. As a workaround, you can restart pmon services on the current subordinate node or restart the node itself. Either of these processes will unmount the read-only partition and enable proper cleanup. |
You must have only one path to the shared storage. Multi-path is not supported.
If there is more than one path when you add the shared storage to Cisco UCS Central VM, you must disable all additional paths. To disable additional paths,
Step 1 | Create a LUN which is no less that 40 GB on the storage array and map it to the ESXi host on which you have installed Node A. |
Step 2 | Add the storage array to the VM as Raw Device Mapping in physical compatibility mode. Make sure to select all default options. |
Step 3 | Change the path selection policy for raw Device mapping hard disk to Fixed (VMware). You have now added a shared storage to VMware. Do the following to setup the shared storage. |
Step 4 | Map the LUN you
added to Node A to the ESXi host on which you have installed node B.
This enables both ESXi hosts to view the same LUN. |
Step 5 | Open a separate
VSphere Client session for this ESXi Host.
You should not add the VM using the VCenter Server. If you do, that will disallow conflicting LUN mapping. |
Step 6 | Add it to the VM as Raw Device Mapping in physical compatibility mode. Make sure to select all default options. |
Step 7 | Change the path selection policy for Raw Device Mapping to Fixed VMware. |
Note | In a cluster set up, when the RDM link goes down on the primary node, DMEs cannot write to the database. This causes a crash on the primary node and failover to the subordinate node. The subordinate node takes over as the primary node. The database is then mounted in read-write mode on the new primary node. Because the RDM link is down, umount fails on the old primary node. When the RDM link comes up, the database is mounted on the old primary (current subordinate) node in read-only mode. As a workaround, you can restart pmon services on the current subordinate node or restart the node itself. Either of these processes will unmount the read-only partition and enable proper cleanup. |
If you answer yes to enable Statistics Collection question during install , then you must specify the database details during Cisco UCS Central installation.
D : Default (internal Postgres db). The internal database is not recommended if you have more than 5 Cisco UCS domains registered with Cisco UCS Central.
P : Postgre
O: Oracle
M: Microsoft SQL Server
If you select either P or O for either of the external database options, make sure you have the following Database information ready:
Type : Oracle, PostgreSQL and MSSQL are the supported options.
Server Name or IP Address: This must be accessible from Cisco UCS Central.
Port: You can configure any custom DB port for accessing the Database server. You must have enabled this port in the firewall settings to enable Cisco UCS Central access the database server through this port.
Name : Name of the database where statistics data would be stored.
Username: The user with administrative privileges to the database to create, delete, read and write.
Password: We recommend to have password expiry set to never or 1 year, so the statistics collection is not interrupted due to expired DB password.
You cannot use a Cisco UCS Central, release 1.1 OVA file to restore a full-state backup from Cisco UCS Central, release 1.0.
Note | This procedure describes the process to restore using an OVA file. |
You must have a backup file with extension .tgz from a Cisco UCS Central system that you want to use to restore the configuration of the Cisco UCS Central VM. For information on how to back up a Cisco UCS Central system, see Managing Back up and Restore in Cisco UCS Central User Manual and CLI Reference Manual.
Step 1 | Save the Cisco UCS Central OVA file to a folder that you can access from the hypervisor. |
Step 2 | Open or import
the
Cisco UCS Central OVA file into a supported hypervisor,
as required by the hypervisor.
Do not continue with the next step until the VM has finished booting. |
Step 3 | If you have not already done so as part of importing the OVA file, power up the Cisco UCS Central VM. |
Step 4 | Open up a console window to the Cisco UCS Central VM. |
Step 5 | When the
Cisco UCS Central VM has completed the initial part of
the installation process, answer the following questions in the VM console
window:
After you confirm that you want to proceed with the configuration, the network interface reinitializes with your settings and Cisco UCS Central becomes accessible via the IP address. |
After Cisco UCS Central is restored, login to Cisco UCS Central and download the firmware images into Image library. If any firmware images are referenced in the service profiles, then you must make sure that the images are downloaded and available in the image library before you re-acknowledge a Cisco UCS domain from the suspended state.
The restored VM is configured to Node A by default. If this is a new cluster, you must install Node B and add it to the cluster mode.
Note | We recommend that you use a different NFS shared directory, or completely clean up the previously used NFS shared directory before starting the full-state restore. |
You must have a backup file with extension .tgz from a Cisco UCS Central system that you want to use to restore the configuration of the Cisco UCS Central VM. For information on how to back up a Cisco UCS Central system, see Managing Back up and Restore in Cisco UCS Central User Manual and CLI Reference Manual.
When you are restoring a cluster setup, you must map shared storage before initiating restore.
Step 1 | Save the Cisco UCS Central OVA or ISO file to a folder that you can access from the hypervisor. |
Step 2 | Open or import
the
Cisco UCS Central
OVA file into a supported hypervisor, as required by the hypervisor.
Do not continue with the next step until the VM has finished booting. |
Step 3 | Add shared storage. See Adding and Setting up an RDM Shared Storage on VMware or Adding and Setting up an RDM Shared Storage on Hyper-V. |
Step 4 | Power up the Cisco UCS Central VM. |
Step 5 | Open up a console window to the Cisco UCS Central VM. |
Step 6 | When the
Cisco UCS Central
VM has completed the initial part of the installation process, answer the
following questions in the VM console window:
|
Map the same shared storage device to node B, configure Node B for the cluster. See Installing Cisco UCS Central on Node B
You can migrate a Cisco UCS Central instance to support Data Center migration or Disaster Recovery use cases. The following pre-requisites must be met to migrate a Cisco UCS Central instance:
The Fully Qualified Domain Name (FQDN) and Hostname must remain the same across sites. The IP address of the Cisco UCS Central instance can be changed during the migration.
All the connected Cisco UCS Manager instances must continue to be reachable from the Cisco UCS Central instance in the same way they were registered (domain hostname or IP address).
Note | Migrating Cisco UCS Central is not supported when the Cisco UCS Central's IP address changes. However, there are no restrictions on Cisco UCS Manager hostname or IP address changes during the migration. You can migrate a Cisco UCS Central instance and install in a standalone or cluster mode. |