Cisco Unified Communications Manager deployment on virtualized servers
Cisco supports running Cisco Unified Communications Manager (Unified CM) under ESXi. For more information about running Unified CM under ESXi, see Unified Communications VMware Requirements on http://www.cisco.com/go/uc-virtualized.
Note
Before you begin an OS installation procedure, see the licensing procedures for your VMware software at http://www.vmware.com/support/licensing/. VMware requires you to combine the licenses for multiple processors.
Cisco Unified Communications Manager VMware Tools and ESXi
VMware Tools are specialized drivers for virtual hardware that is installed in the UC applications when they are running virtualized. It is very important that the VMware tools version running in the UC application be in sync with the version of ESXi being used. For information on how to upgrade the tools, see:
Cisco Unified CM uses Automatic Update Statistics, an intelligent statistics update feature that monitors the changes made in the database tables and updates only tables that need statistic updates. This feature saves considerable bandwidth, especially on VMware deployments of Cisco Unified CM. Automatic Update Statistics is the default indexing method.
For more information about database services, see the Cisco Unified Serviceability Administration Guide.
New identity
Cisco supports the New Identity process for use with Cisco Unified Communications Manager (Unified CM). The New Identity process is designed to start with a Unified CM application that is fully installed and configured with common settings. Often, the initial VM is saved as a VMware template and cloned as new Unified CM publisher nodes come online.
The New Identity process copies the VMware template and changes a set of primary settings, such as the IP address and hostname, to give a new VM a unique identity in the network.
When prompted for the floppy/USB drive on the Pre-existing Configuration Information window, power down the VM.
Step 3
Clone or convert the VM into a VM template.
Step 4
For a new subscriber, deploy the template and mount a virtual floppy drive that contains the configuration file from the AFG tool.
New identity caveats
When you run the New Identity process, note the following:
Although you can provide a new OS administrator user ID in the XML file, you cannot change the OS administrator user ID during the New Identity process.
Each cloned VM has the same network configuration as the VMware template. The network must be functional during the New Identity process. If you run the cloned VMs on the same LAN there can be duplicate IP addresses. Ensure that you do not run the VMware template, or multiple VMs from the initial template, at the same time on the same LAN.
The NTP server must be accessible before you can configure it on the Unified CM application. Ensure that the VM has access to the new NTP server.
If DNS is used, DNS servers must be accessible when you run the New Identity process.
For Cisco Unity Connection, you must set the SMTP domain address after you run the New Identity process.
For Cisco Unified Presence, you must set the post-installation steps that configure the Unified CM system with which Cisco Unified Presence communicates after you run the New Identity process.
Installation, upgrade, and migration options
Note
Be aware that when you mount an ISO file via the VM console, VMware does not eject the disc at the end of the install process.
Tip
Always mount your DVD ISO file from the Edit Settings menu in VMware.
There are two main approaches to installing, upgrading, and migrating servers:
Specs-based
Tested reference configuration
For information relating to specs-based configuration, refer to the topics related to VMware specs-based support.
See topics related to VM feature support, system requirements, and migration for a description of any changes for installation, upgrade, and migration of VMware Specs-based support.
Migrating from an existing server to a VMware Specs-Based configuration follows a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
Migrating from a Media Convergence Server (MCS server) to a Cisco UCS B-Series Blade Server follows a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
The following procedure provides the tasks for the migration process. For more information, see the related topics, as well as the following documentation:
Cisco Unified Communications Operating System Administration Guide
Disaster Recovery System Administration Guide
Installing Cisco Unified Communications Manager
Replacing a Single Server or Cluster for Cisco Unified Communications Manager
Refer to the topic "Changing the Cluster IP Addresses for Publisher Servers That Are Defined by Host Name" in the Changing the IP Address and Host Name for Cisco Unified Communications Manager guide.
Procedure
Step 1
Review the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
This document describes how to replace server hardware, which is very similar to migrating from an MCS server to a Cisco UCS B-Series Blade Server server. You should perform the document's pre-replacement and post-replacement tasks, and review the procedures for installing Cisco Unified Communications Manager and migrating data.
Step 2
Upgrade the MCS server to Cisco Unified Communications Manager Release 8.0(2) or later.
Step 3
If the UCS VM will use a different IP address or hostname than the MCS server, change the IP address and hostname of the MCS server to the values that the UCS VM will use.
This is required for DRS backup and restore to work.
Step 4
Perform a DRS backup on the MCS server.
Note
Cisco UCS B-Series Blade Servers do not support tape drive as the backup media.
Step 5
Use the Answer File Generator to generate a license MAC for the Cisco UCS B-Series Blade Server server.
The license MAC is required to obtain licenses for the server. After you obtain the license MAC, you can rehost the licenses for your new server.
Step 6
Create the virtual machine (VM) on the Cisco UCS B-Series Blade Server server that will be used as the replacement for the MCS node.
Use the Cisco-provided OVA template to create VMs.
Step 7
Install Cisco Unified Communications Manager Release 8.0(2c) or later on the Cisco UCS B-Series Blade Server.
Note
Cisco UCS B-Series Blade Servers do not support installation from a DVD.
Note
Cisco UCS B-Series Blade Servers do not support a hardware clock; you must use NTP. The option to configure a hardware clock is not available in the installation program.
Step 8
Install Cisco Unified Communications Manager Release 8.0(2c) or later on the Cisco UCS B-Series Blade Server.
Step 9
Perform a DRS restore to restore the data backed up from the MCS server to the Cisco UCS B-Series Blade Server server.
Step 10
Upload the new licenses to the Cisco UCS B-Series Blade Server server.
If you did not obtain licenses for the new server already, you must request the licenses first.
Note
The previous license will no longer be valid. However, you have 30 additional days in which to use your previous license.
Cisco UCS B-Series Blade Servers use "soft media" such as ISO or FLP (virtual floppy) for procedures that require external media (such as installation and upgrade). Physical external devices such as USB drives are not supported.
Note
Backup and restore are not supported on soft media.
The virtual USB interface is not supported on VMware. The following are examples of differences in external media support between MCS servers and Cisco UCS B-Series Blade Servers:
Install logs cannot get dumped to a USB key. These logs get dumped to a file through the serial port of the VM.
The answer file generated by the Answer File Generator (platformConfig.xml) cannot get read from a USB key to perform an unattended installation. Instead, you must put the answer file into a FLP image to be mounted in the floppy drive.
USB tape drive backup is not supported. Use SFTP instead.
Music On Hold through a USB connection is not supported. Use a Cisco 7800 Series Media Convergence Server instead.
Cisco Messaging Interface (CMI) for Message Waiting Indication (MWI) is not supported over the serial port. Use a Cisco 7800 Series Media Convergence Server instead.
UCS C260 rack-mount server TRC1
Cisco supports running Cisco Unified Communications Manager on the Cisco UCS C260 M2 Rack-Mount Server Rack-Mount Server Tested Reference Configuration 1 (TRC1) with a specific configuration of direct attached storage (DAS).
The following sections describe the changes for installation, upgrade, and migration in Cisco UCS C260 M2 Rack-Mount Server:
To operate Cisco UCS C260 M2 Rack-Mount Server successfully, you should have the experience and skills to manage a host server running VMware ESXi. If you do not have this experience and want to obtain the required information quickly, consider using VMware GO, a Web-based application that facilitates VMware installations. For more information, refer to the VMware GO documentation.
This following sections describe how to perform a fresh installation of Cisco UCS C260 M2 Rack-Mount Server:
Configuration checklist for installing and configuring the server
Prepare for installation
Configuration checklist for installing and configuring the server
The following table provides a checklist of the major steps required to install and configure Cisco UCS C260 M2 Rack-Mount Server. The Related Documentation column contains references to documentation that is related to the step.
Configure the RAID settings to the following specifications: The first 8 drives are configured as a 7 + 1 RAID 5 array. The remaining 8 drives are configured as a 7 + 1 RAID 5 array.
This section describes how to prepare to install a Cisco UCS C260 M2 Rack-Mount Server in a standalone configuration (it is not in a datacenter).
It is suggested that you allocate the following resources before installation:
Space in a rack to receive a 2 RU Cisco UCS C260 M2 Rack-Mount Server. This rack needs to accommodate the "square mount" rails shipped with the Cisco UCS C260 M2 Rack-Mount Servers.
5 or 6 Ethernet ports on a switch close to the Cisco UCS C260 M2 Rack-Mount Server:
One standard Ethernet port for the dedicated CIMC management port, if desired
Four 802.1q trunked ports for the LOM NICs
An IP address for CIMC management. If the dedicated port is used, it should be attached to the appropriate LAN
A VLAN ID and IP address for the host. This is the Cisco UCS C260 M2 Rack-Mount Server's ESXi management address
A hostname and configure DNS, if desired, for the hosts' hostname
VLAN IDs and IP addresses for the VMs
Upon receipt of the Cisco UCS C260 M2 Rack-Mount Server you will:
Install the C-260 M2 in the rack
Attach the CIMC management port to the designated switch port
Attach the LOM NICs to their designated, trunked switch ports
Attach a VGA console, or a KVM to the VGA and keyboard ports. This is necessary until CIMC has been configured.
Set up CIMC for Cisco C260 rack-mount server TRC1
Configuring the CIMC allows you to perform all subsequent configuration and installation using the CIMC console. In addition, the CIMC provides a measure of hardware monitoring.
Follow this procedure to configure CIMC:
Procedure
Step 1
Attach a VGA console and keyboard directly to the server using
A dedicated Ethernet management port on the Cisco UCS C260 M2 Rack-Mount Server. This requires no VLAN and is the simplest to configure. This is a single NIC and there is no redundancy for the CIMC using this configuration.
The LOM NICs. Although this option is available, it is not recommended.
Step 2
During boot, hit F8 to enter CIMC configuration.
Step 3
In the CIMC configuration screen, under IPV4 (Basic):
Clear the DHCP enabled check box using the spacebar
Enter values for the CIMC IP, Subnet mask, and Gateway.
Step 4
Leave VLAN (Advanced) clear.
Step 5
Under Default User (Basic), enter the default CIMC user, admin and a password.
Note
The CIMC username is not configurable and is set to admin.
Step 6
Once configured, the CIMC is accessible via http. Point a browser to the IP address configured above and login as admin, using the password configured above.
Note
Cisco does not support or restrict customers to any specific version of BIOS. The current version is assumed to be compatible with latest release of ESXi. The same is true for the BIOS configuration. The default BIOS settings as shipped from the factory require no modification.
Set up RAID for Cisco UCS C260 rack-mount server TRC1
Configure the RAID settings to the following specifications:
The 16 drives present on the system are set up as two separate 8-drive RAID-5 logical volumes.
With eight 300GB drives, this corresponds to a total storage capacity of 1.93 TB each, as seen by ESXi.
Follow this procedure to perform this task:
Procedure
Step 1
Check your current RAID configuration:
During boot, the system asks you to enter either Ctrl-H or Ctrl-Y
To use the GUI to configure RAID, enter Ctrl - H.
To use the Preboot CLI to configure RAID, enter CTRL-Y.
Type the following command:
-ldinfo -l0 -a0
-ldinfo -l0 -a1
This command displays the number of drives, RAID level, and so forth for the specified logical drive. There should be 8 drives each in two separate RAID 5 arrays for logical drive 0 and 1.
Step 2
Use the following sequence of commands to set the recommended RAID configuration; two RAID 5 volumes of 8 drives each:
Enter CTRL-Y to enter the Preboot CLI when prompted during boot
Enter the following Preboot CLI command to clear configuration:
-cfgclr -a0
-cfgclr -a1
Step 3
To determine the enclosure ID and drive numbering, which is required before you can configure RAID, run the following:
-encinfo -a0 -page 20
-encinfo -a1 -page 20
Note
This command can generate more than one page of output, so enter -page 20 to look at 20 lines at a time. Look for the Device ID of the enclosure that has a non-zero Number of Physical Drives. Use this Device ID (also called Enclosure ID) in the following commands.
Step 4
There is a single enclosure for Cisco UCS C260 M2 Rack-Mount Server. The enclosure ID is not predictable, so you need to substitute the Device ID acquired, above, for <encl> in the commands, below. When all drives are in a single enclosure, the slot numbering starts at zero. This may not be true in all cases, so verify the slot numbering with the following command:
-pdinfo -physdrv [<encl>:0] -a0
If this command generates meaningful output, the drives start at zero. If it generates an error, the drives start at one.
Note
Please substitute your enclosure ID for <encl> in the commands, below. If you have determined your drives start at one, above, you need to start with drive one instead of zero
Step 5
Use the following command to setup RAID 5 on the existing 8 drives on each RAID controller:
To clear data on previously used drives and initialize a new array, use the -ldinit -start -full -l0 -a0 command. Allow command to finish before exiting the Preboot CLI.
Step 6
After you configure the logical volume, exit the Preboot CLI by entering q.
Step 7
During the boot process, make sure that Quiet Boot is disabled, and press Ctrl - H at the LSI screen when prompted. This brings you to the MegaRaid BIOS Configuration utility. At this screen, you will see 2 LSI MegaRAID SAS adapters. Select Adapter 0 to begin and click Start.
Step 8
Select New Configuration and click Next.
Step 9
Select Manual Configuration.
Step 10
On the next screen, you need to add drives to a Drive Group. Select one drive and then select all others by holding down on Shift and the down arrow key. Click Add to Array.
Step 11
Click Accept DG.
Step 12
The drive group must be added to a span. Select DG0 and click Add to Span.
Step 13
Once the drive group is part of a span, you can configure RAID on it.
Select RAID 5 from the list of available options.
(Optional) Cisco recommends that you select 128KB from the Strip Size drop down list.
Step 14
Set Read Policy to read ahead = always.
Step 15
Set Write Policy to one of the following:
- write back with bbu- if using a RAID card with SuperCap (e.g. RAID-9266CV)
- always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies.
Note
Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Step 16
Click Update Size to finalize the RAID 5 volume and to determine the size of the resulting volume. It will work out to 1.903 TB. You are presented with a warning relating to BBU, as the BBU is used whenever possible, but with the understanding that if the BBU is discharged or undergoing a learning cycle, performance will be degraded. Click Yes if this is acceptable.
Step 17
Click Next on the next screen to accept the Virtual Drive you just created (VD 0).
Step 18
Click Next. At the next screen, you are presented with the option to initialize the array. Click:
Fast Initialize
Go
Set Boot Drive
Go
Step 19
The RAID configuration is now complete for the first RAID controller. Go back to the controller selection by clicking on Controller selection. This time, select Adapter 1.
Step 20
Repeat all steps performed for Adapter 0 with this new Adapter to set up the second RAID 5 array.
Set up BIOS boot order for Cisco C260 rack-mount server TRC1
The hard drive is not selectable in the BIOS Boot Order menu until the first logical volume has been defined. Once RAID has been configured, you need to make the second boot device the hard drive, as described in the following procedure:
Procedure
Step 1
Press F2 during boot to enter BIOS setup.
Step 2
Move the cursor over to Boot Options.
Step 3
Verify that the CD ROM is selected for Boot Option #1.
Step 4
Verify that the hard drive (the RAID 5 Array) is selected for Boot Option #2.
Step 5
Verify that VT is enabled in advanced CPU options.
Step 6
Verify that VT I/O Redirection is disabled in the CPU options.
The server will now try to boot the CD ROM drive first and the hard drive second.
Disk management for Cisco UCS C260 M2 rack-mount server TRC1
The Cisco UCS C260 M2 Rack-Mount Server comes with 16 hard drives. These drives were configured into two logical volumes and require no further management at this point.
Disks are hot-swappable. This does not mean that you will be able to swap drives ad-hoc on failures. A process exists to swap drives. When a drive fails, you need to:
Reboot and enter the Preboot CLI
Mark the defective drive for removal using -PdPrpRmv -physdrv [<encl>:<slot>] -a0
Replace the drive
The RAID array is rebuilt automatically when the replacement disk is inserted.
Note
Although Preboot CLI is recommended, it is also possible to perform this task through the LSI MegaRaid GUI, where you can swap drives out on the fly without having to power-cycle the server to get into the preboot CLI. However, this requires you to procure a separate machine (Windows or Linux) on the same subnet as the ESXi host, installed with the LSI MegaRaid utility.
ESXi installation and setup for C260 M2 rack-mount server TRC1
The following sections provide a sequence of steps for bringing ESXi into service at the customer site.
Preparation for ESXi Installation
Prior to installing ESXi, it is assumed that:
The IPL order in the BIOS is configured to boot the CDROM first
For the Cisco UCS C260 M2 Rack-Mount Server, the 16 drives are configured in two separate RAID-5 arrays.
These steps should have been performed by the factory prior shipping.
Install ESXi on one of the RAID-5 arrays. Both RAID arrays are acceptable.
ESXi Installation takes less than 5 minutes. Once installation is complete, remove the install CD and reboot the machine.
Following a reboot, a grey and yellow ESXi console is displayed with 2 options:
F2 to customize the system
F12 to restart or halt the system
At this point, press F2 and configure the system in accordance with your network.
Installing vSphere Client
Once the host is on the network, you can browse to its IP address to bring up a web-based interface. The vSphere client is Windows-based, so the download and install must be performed from a Windows PC.
This install proceeds like any other Windows application install, and takes only a few minutes to perform. Once the vSphere client is installed, you can bring it up and log into the host using the host name or IP address, the root login ID, and the password configured, above.
The host may also be joined to a vCenter if available and if you wish to manage the host through vCenter.
Configuring LOM NICs and Virtual Switches
The following options may be configured:
Simple vSwitch0 (default VMware virtual switch)
For larger datacenters using vCenter, you can configure distributed virtual switches (ex: distributed vSwitch or the Nexus 1000V distributed virtual switch)
For all options, you must define a port group for each VLAN running on the virtual switch. These port groups are selected when configuring a Virtual Machine’s network adapter, to place the virtual machine on a given LAN.
VM installation and configuration C260 M2 rack-mount server TRC1
ISOs and VM Templates
ISOs are available on the media kit you receive with the Cisco UCS C260 M2 Rack-Mount Server, and the templates are posted to Cisco.com.
Using the vSphere Client to Create the VM
Cisco provides templates on a URL to download and transfer to a host. Use the following procedure to create the VM:
Procedure
Step 1
Deploy the appropriate OVA template for their application using the CCO URL as the source.
Step 2
Make the CDROM drive available to the newly deployed VM.
Step 3
Click on Options > Boot Options > The next time the virtual machine boots, force entry into BIOS Setup Screen.
Step 4
Insert the installation media (from the media kit) in the system CDROM drive.
Step 5
Power on the VM, select Boot and promote CD ROM to boot before the hard drive.
Step 6
Save BIOS settings and boot.
You should be in the normal install screens for your application at this point.
Migrate to Cisco UCS C260 M2 Rack-Mount Server TRC1
Migrating from a Media Convergence Server (MCS server) to a Cisco UCS C260 M2 Rack-Mount Server server follows a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
The following table provides an overview of the migration process and references to other pertinent documentation.
Configuration steps
Related procedures and topics
Step 1
Review the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager. This document describes how to replace server hardware, which is very similar to migrating from an MCS server to a Cisco UCS C260 M2 Rack-Mount Server. You should perform the document’s pre-replacement and post-replacement tasks, and review the procedures for installing Cisco Unified Communications Manager and migrating data.
Replacing a Single Server or Cluster for Cisco Unified Communications Manager
Step 2
Upgrade the MCS server to Cisco Unified Communications Manager Release 8.0(2) or later.
Cisco Unified Communications Operating System Administration Guide
Step 3
If the UCS VM will use a different IP address or hostname than the MCS server, change the IP address and hostname of the MCS server to the values that the UCS VM will use.
This is required for DRS backup and restore to work.
Refer to the topic “Changing the Cluster IP Addresses for Publisher Servers That Are Defined by Host Name” in the Changing the IP Address and Host Name for Cisco Unified Communications Manager guide.
Step 4
Perform a DRS backup on the MCS server.
Disaster Recovery System Administration Guide
Step 5
Use the Answer File Generator to generate a license MAC for the Cisco UCS C260 M2 Rack-Mount Server. The license MAC is required to obtain licenses for the server.
After you obtain the license MAC, you can rehost the licenses for your new server.
Perform a DRS restore to restore the data backed up from the MCS server to the Cisco UCS C260 M2 Rack-Mount Server.
Disaster Recovery System Administration Guide
Step 9
Upload the new licenses to the Cisco UCS C260 M2 Rack-Mount Server.
If you did not obtain licenses for the new server already, you must request the licenses first.
Note
The previous license will no longer be valid. However, you have 30 additional days in which to use your previous license. Refer toNew licensing procedure customer impact
Cisco Unified Communications Operating System Administration Guide
Cisco UCS C260 M2 rack-mount server TRC1 daily operations
At this point the application is installed and in operation. Daily operations from the application's viewpoint are no different from installation on a physical server, including:
Application configuration and integration with other applications
RTMT performance monitoring
SNMP monitoring and alarms
DRS backup and restore
CDR collection
Device, trunk, gateway configuration and monitoring
The following sections describe how to perform these tasks.
Monitoring Hardware from the VM
Applications running in a VM have no ability to monitor the physical hardware. Any hardware monitoring must be done from the CIMC, ESXi plugins, vCenter or by physical inspection (for flashing LEDs, and so on).
Monitoring of hardware is the customer's responsibility. It is assumed the customer is familiar with virtualized environments and has the know how to manage hardware in these environments.
Monitoring from CIMC
The CIMC provides the following hardware monitoring:
An overview of CPU, memory and power supply health
An overview of hardware inventory, including CPUs, Memory, Power Supplies and Storage
Monitoring of sensors for Power Supplies, Fans, Temperature,Voltage and Current
A system event log that contains BIOS and Sensor entries
LSI MegaRAID controller information, which includes physical and virtual drive layout and Battery Backup Unit information from the Inventory -> Storage tab. This information was usually accessible for earlier UCS servers only by installing the MegaRAID plugin from ESXi.
Monitoring from ESXi
Monitoring server health from ESXi is possible by:
Logging into the ESXi console and inspecting system /var/log/messages for telltale entries
Monitoring from vSphereClient and vCenter
The vSphere Client provides the following monitoring:
Hardware and system alarms defined under the Alarms tab in the vSphere Client when logged into vCenter.
VM resource usage under the Virtual Machines tab in the vSphere Client, as well as under the Performance tab for each VM
Host performance and resource usage under the Performance tab for the Host.
Cisco UCS C240 M3 rack-mount server TRC1
Cisco supports running Cisco Unified Communications Manager on the Cisco UCS C240 M3 Rack-Mount server tested reference configuration 1 (TRC1). For more information about tested reference configurations for specific server models, see http://docwiki.cisco.com/wiki/Tested_Reference_Configurations_%28TRC%29.
For information on installation, upgrade, and migration to the Cisco UCS C240 M3 Rack-Mount server TRC1, see the related topics.
To operate Cisco UCS C240 M3 Rack-Mount Server successfully, you should have the experience and skills to manage a host server running VMware ESXi. If you do not have this experience and want to obtain the required information quickly, consider using VMware GO, a Web-based application that facilitates VMware installations. For more information, refer to the VMware GO documentation.
Install and set up Cisco C240 M3 rack-mount server TRC1
Before You Begin
To prepare to install a Cisco UCS C240 M3 Rack-Mount Server TRC1 in a standalone configuration (it is not in a datacenter), Cisco recommends that you allocate the following resources before installation:
Space in a rack to receive a Cisco UCS C240 M3 Rack-Mount Server. This rack needs to accommodate the square mount rails shipped with Cisco UCS C240 M3 Rack-Mount Servers.
8 to 11 Ethernet ports on a switch close to the Cisco UCS C240 M3 Rack-Mount Server:
One standard Ethernet port for the dedicated CIMC management port, if desired.
Eight ports for the quad-port Intel NICs.
Two ports for the LOM NICs, if required.
An IP address for CIMC management. If the dedicated port is used, it should be attached to the appropriate LAN.
A VLAN ID and IP address for the host. This is the Cisco UCS C240 M3 Rack-Mount Server ESXi management address.
A hostname and configure the DNS, if desired, for the hostname.
VLAN IDs and IP addresses for the VMs.
Upon receipt of the Cisco UCS C240 M3 Rack-Mount Server:
Install the C-240 M3 Rack-Mount Server in the rack.
Attach the CIMC management port to the designated switch port.
Attach the LOM NICs to their designated, trunked switch ports.
Attach a VGA console, or a KVM to the VGA and keyboard ports. This is necessary until CIMC has been configured.
Configuration checklist for installing and configuring the server
The following table provides a checklist of the major steps required to install and configure a Cisco UCS C240 M3 Rack-Mount Server TRC1. The Purpose column contains references to documentation that is related to the step.
Set up CIMC for Cisco C240 M3 rack-mount server TRC1
Configuring the CIMC allows you to perform all subsequent configuration and installation using the CIMC console. In addition, the CIMC provides a measure of hardware monitoring.
Follow this procedure to configure CIMC:
Procedure
Step 1
Attach a VGA console and keyboard directly to the server using
A dedicated Ethernet management port on the Cisco UCS C240 M3 Rack-Mount Server. This does not require a VLAN and is the simplest to configure. This is a single NIC and there is no redundancy for the CIMC using this configuration.
The LOM NICs. Although this option is available, Cisco does not recommend it.
Step 2
During boot, press F8 to enter the CIMC configuration.
Step 3
In the CIMC configuration screen, under IPV4 (Basic):
Clear the DHCP enabled check box using the spacebar.
Enter values for the CIMC IP address, subnet mask, and gateway.
Step 4
Leave VLAN (Advanced) clear.
Step 5
Under Default User (Basic), enter the default CIMC user, admin, and a password.
Note
The CIMC username is not configurable and is set to admin.
Step 6
Once configured, the CIMC is accessible through http. Point a browser to the IP address that you configured and login as admin, using the password that you configured.
Set up RAID for Cisco C240 M3 rack-mount server TRC1
Configure the RAID settings to the following specifications:
Set up the 16 drives present on the system as two 8-drive RAID-5 logical volumes.
With eight 300GB drives, this corresponds to a total storage capacity of 1.93 TB each, as seen by ESXi.
During boot, the system asks you to enter either Ctrl-H or Ctrl-Y.
Set up RAID with Preboot CLI for Cisco C240 M3 rack-mount server TRC1
Use the following sequence of commands to set the recommended RAID configuration; two 8-drive RAID-5 logical volumes:
Important:
If required, use the following settings for the Read and Write policies:
Set Read Policy to read ahead = always.
Set Write Policy to one of the following:
write back with bbu – if using a RAID card with SuperCap (for example: RAID-9266CV)
always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies. Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Procedure
Step 1
Check your current RAID configuration:
During boot, the system asks you to enter either Ctrl-H or Ctrl-Y
To use the Preboot CLI to configure RAID, enter CTRL-Y.
Type the following commands:
-ldinfo -l0 -a0
-ldinfo -l1 -a0
These commands display the number of drives, RAID level, and so forth for the specified logical drive. There should be 8 drives each in two separate RAID-5 arrays for logical drive 0 and 1.
Step 2
Use the following sequence of commands to set the recommended RAID configuration; two 8-drive RAID-5 logical volumes:
Enter CTRL-Y to enter the Preboot CLI when prompted during boot
Enter the Preboot CLI command -cfgclr -a0 to clear the configuration.
Step 3
To determine the enclosure ID and drive numbering, which is required before you can configure RAID, run the following command:
-encinfo -a0 -page 20
Note
This command can generate more than one page of output, so enter -page 20 to look at 20 lines at a time. Look for the Device ID of the enclosure that has a non-zero Number of Physical Drives. Use this Device ID (also called Enclosure ID) in the following commands.
Step 4
There are two enclosures for a Cisco UCS C240 M3 Rack-Mount Server TRC1. The enclosure ID is not predictable, so you must substitute the Device ID acquired above for <encl> in the commands below. When all drives are in a single enclosure, the slot numbering starts at zero. This may not be true in all cases, so verify the slot numbering with the following command:
-pdinfo -physdrv [<encl>:0] -a0
If this command generates meaningful output, the drives start at zero. If it generates an error, the drives start at one.
Note
Substitute your enclosure ID for <encl> in the commands below.
Step 5
Use the following commands to setup RAID 5 on the existing 8 drives on each RAID controller:
Note
The following example assumes that the drive numbering starts at 1.
The following commands are not necessary for new drives that have not been used.
Use the -ldinit -start -l0 -a0 and –ldinit –start –l1 –a0 commands to perform a fast initialize.
To clear data on previously used drives and to slow (or full) initialize a new array, use the -ldinit -start -full -l0 -a0 and -ldinit -start -full -l1 -a0 commands. Allow the commands to finish before exiting the Preboot CLI. When both commands –ldinit –showprog –l0 –a0 and –ldinit –showprog –l1 –a0 show that initialization is not running, it is safe to exit the Preboot CLI.
Note
Cisco has noticed that a slow initialize can take up to 95 minutes or more to fully complete for array sizes of 1 TB+ that are used in a UCS C240 M3 Rack-Mount Server TRC1 deployment.
Step 7
After you configure the logical volume, exit the Preboot CLI by entering q.
Note
The LSI adapter has factory default values for the drive rebuild rate, patrol read rate, and other settings. Cisco recommends leaving the default values unchanged.
Set up RAID from the GUI for C240 M3 rack-mount server TRC1
Procedure
Step 1
During the boot process, make sure that Quiet Boot is disabled.
Press Ctrl - H at the LSI screen when prompted.
This brings you to the MegaRaid BIOS Configuration utility. Click Start.
Step 2
On the next screen
Select New Configuration.
Click Next.
Step 3
When prompted on the next screen to clear the configuration, click Yes.
Step 4
On the next screen
Select Manual Configuration.
Click Next.
Step 5
On the next screen, you add the first eight drives to a Drive Group.
Click on one drive and then select the other drives by holding down on Shift and the down arrow key.
Click Add to Array.
Click Accept DG.
On the same screen, you add the remaining eight drives to another Drive Group. Click on Drive9 and then select all other drives by holding down on Shift and the down arrow key.
Click Add to Array.
Click Accept DG.
Click Next to accept the Drive Group.
Step 6
The drive group must be added to a span.
Select Drive Group0.
Click Add to Span.
Click Next to accept the span definition.
Step 7
Once the drive group is part of a span, you can configure RAID on it.
Select RAID 5 from the list of available options.
Select 128KB from the Strip Size drop down list.
Set Read Policy to read ahead = always.
Set Write Policy to one of the following:
- write back with bbu- if using a RAID card with SuperCap (e.g. RAID-9266CV)
- always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies.
Note
Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Click Update Size to finalize the RAID 5 volume and to determine the size of the resulting volume. It will work out to 1.903 TB.
Click Accept to accept the virtual drive definition.
Click Next.
Step 8
You are presented with a warning relating to BBU, as the BBU is used whenever possible, but with the understanding that if the BBU is discharged or undergoing a learning cycle, performance will be degraded. Click Yes if this is acceptable.
Step 9
Click Back on the next screen, to go back and add the second RAID-5 array (VD 1).
Step 10
Select Drive Group 1 from the drop down list.
Click Add to Span.
Click Next.
Step 11
At the Raid selection screen
Select RAID 5 from the list of available options.
Select 128KB from the Strip Size drop down list.
Set Write Policy to one of the following:
- write back with bbu- if using a RAID card with SuperCap (e.g. RAID-9266CV)
- always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies.
Note
Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Click Update Size.
Click Accept to accept the new virtual VD1.
Step 12
Click Yes at the BBU warning screen.
Step 13
Click Next at the Virtual Drive Definition screen, to indicate that you are finished defining virtual drives.
Step 14
Click Accept at the Configuration Preview screen to accept the RAID configuration.
Step 15
Click Yes to save the configuration.
Step 16
Click Yes to start drive initialization.
Step 17
Click Home to exit the configuration wizard when both drives report their status as Optimal.
Step 18
Click Exit to exit the GUI.
Note
After RAID configuration is complete on the drives, the system may try to initialize (format) the new RAID array. When this happens, the current initialization progress can be seen from the Web BIOS screen. Wait for the background initialization to complete before proceeding with any of the subsequent server configuration steps such as installing ESXi. Cisco has noticed that a slow initialize can take up to 95 minutes or more to fully complete for array sizes of 1 TB+ used in a UCS C240 M3 Rack-Mount Server TRC1 deployment.
Set up BIOS boot order for Cisco C240 M3 rack-mount server TRC1
Note
Cisco does not support or restrict customers to any specific version of BIOS. The current version is assumed to be compatible with the latest release of ESXi. The same is true for the BIOS configuration. The default BIOS settings as shipped from the factory require no modification.
The hard drive is not selectable in the BIOS Boot Order menu until the first logical volume has been defined. Once RAID has been configured, make the second boot device the hard drive, as described in the following procedure:
Procedure
Step 1
Press F2 during boot to enter BIOS setup.
Step 2
Move the cursor over to Boot Options.
Step 3
Verify that the Cisco Virtual CD/DVD (Virtual CD/DVD drive) is selected for Boot Option #1.
Step 4
Verify that the hard drive (one of the RAID-5 arrays) is selected for Boot Option #2.
Step 5
Verify that VT is enabled in advanced CPU options.
Step 6
Verify that VT I/O Redirection is disabled in the CPU options.
The server will now try to boot the virtual CD drive first and the hard drive second.
Disk management for Cisco UCS C240 M3 rack-mount server TRC1
The Cisco UCS C240 M3 Rack-Mount Server TRC1 comes with 16 hard drives. These drives were configured as two 8-drive RAID-5 logical volumes and require no further management at this point.
Disks are hot-swappable. This does not mean that you can swap drives ad-hoc on failures. A process exists to swap drives. When a drive fails, you must:
Reboot and enter the Preboot CLI
Mark the defective drive for removal using -PdPrpRmv -physdrv [<encl>:<slot>] -a0
Replace the drive
The RAID array is rebuilt automatically when the replacement disk is inserted.
Note
Although Preboot CLI is recommended, it is also possible to perform this task through the LSI MegaRaid GUI, where you can swap drives out without having to power-cycle the server to get into the Preboot CLI. However, this requires you to procure a separate machine (Windows or Linux) on the same subnet as the ESXi host, installed with the LSI MegaRaid utility.
ESXi installation and setup for C240 M3 rack-mount server TRC1
The following sections provide a sequence of steps for bringing ESXi into service at the customer site.
Preparation for ESXi Installation
Prior to installing ESXi, it is assumed that:
The IPL order in the BIOS is configured to boot the virtual CD/DVD first.
For the Cisco UCS C240 M3 Rack-Mount Server TRC1, the 16 drives are configured as two separate RAID-5 arrays.
These steps should have been performed by the factory prior to shipping.
Install ESXi on one of the RAID-5 arrays. You can install ESXi on either of the two arrays.
ESXi installation takes less than five minutes. Once the installation is complete, remove the install CD and reboot the machine.
Following the reboot, a grey and yellow ESXi console is displayed with two options:
F2 to customize the system.
F12 to restart or halt the system.
At this point, press F2 and configure the system in accordance with your network.
Installing vSphere Client
Once the host is on the network, you can browse to its IP address to bring up a web-based interface. The vSphere client is Windows-based, so you must perform the download and installation from a Windows PC.
The installation proceeds like any other Windows application installation, and completes in a few minutes. Once the vSphere client is installed, you can bring it up and log into the host using the host name or IP address, the root login ID, and the password that you configured.
The host may also be joined to a vCenter if available and if you wish to manage the host through vCenter.
Configuring LOM NICs and Virtual Switches
You can configure the following options:
Simple vSwitch0 (the default VMware virtual switch).
For larger datacenters using vCenter, you can configure distributed virtual switches such as the distributed vSwitch or the Nexus 1000V distributed virtual switch.
For all options, you must define a port group for each VLAN running on the virtual switch. These port groups are selected when configuring a virtual machine network adapter to place the virtual machine on a given LAN.
VM installation and configuration Cisco C240 M3 rack-mount server TRC1
ISOs and VM templates
DVDs are available in the media kit that you receive with the Cisco UCS C240 M3 Rack-Mount Server TRC1, and the OVA templates are posted to Cisco.com.
Using the vSphere client to create the VM
Cisco provides templates on a URL to download and transfer to a host. Use the following procedure to create the VM:
Procedure
Step 1
Deploy the appropriate OVA template for your application using the Cisco.com URL as the source.
Step 2
Make the virtual CD drive available to the newly deployed VM.
Step 3
Refer to the Release Notes for the OVA that you are deploying for instructions on setting up the BIOS boot order.
Step 4
Map the installation media (from the media kit) to the virtual CD/DVD drive.
Step 5
Save the BIOS settings and boot.
You should be in the normal install screens for your application at this point.
Migrate to Cisco C240 M3 rack-mount server TRC1
Migrating from a Media Convergence Server (MCS server) to a Cisco UCS C240 M3 Rack-Mount Server follows a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
The following table provides an overview of the migration process and references to other pertinent documentation.
Configuration steps
Related procedures and topics
Step 1
Review the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager. This document describes how to replace server hardware, which is very similar to migrating from an MCS server to a Cisco UCS C240 M3 Rack-Mount Server. You should perform the documented pre-replacement and post-replacement tasks, and review the procedures for installing Cisco Unified Communications Manager and migrating data.
Replacing a Single Server or Cluster for Cisco Unified Communications Manager
Step 2
Upgrade the MCS server to Cisco Unified Communications Manager Release 8.0(2) or later.
Cisco Unified Communications Operating System Administration Guide
Step 3
If the UCS VM will use a different IP address or hostname than the MCS server, change the IP address and hostname of the MCS server to the values that the UCS VM will use.
This is required for DRS backup and restore to work.
Refer to the topic “Changing the Cluster IP Addresses for Publisher Servers That Are Defined by Host Name” in the Changing the IP Address and Host Name for Cisco Unified Communications Manager guide.
Step 4
Perform a DRS backup on the MCS server.
Disaster Recovery System Administration Guide
Step 5
Use the Answer File Generator to generate a license MAC for the Cisco UCS C240 M3 Rack-Mount Server. The license MAC is required to obtain licenses for the server.
After you obtain the license MAC, you can rehost the licenses for your new server.
Perform a DRS restore to restore the data backed up from the MCS server to the Cisco UCS C220 M3 Rack-Mount Server.
Disaster Recovery System Administration Guide
Step 9
Upload the new licenses to the Cisco UCS C240 M3 Rack-Mount Server.
If you did not obtain licenses for the new server already, you must request the licenses first.
Note
The previous license will no longer be valid. However, you have 30 additional days in which to use your previous license. Refer toNew licensing procedure customer impact
Cisco Unified Communications Operating System Administration Guide
Cisco C240 M3 rack-mount server TRC1 daily operations
At this point the application is installed and in operation. Daily operations from the viewpoint of the application do not differ from an installation on a physical server, including:
Application configuration and integration with other applications
RTMT performance monitoring
SNMP monitoring and alarms
DRS backup and restore
CDR collection
Device, trunk, and gateway configuration and monitoring
Monitoring Hardware from the VM
Applications running in a VM can not monitor the physical hardware. You must monitor the hardware from the CIMC, ESXi plugins, vCenter or by physical inspection (for flashing LEDs, and so on).
The customer is responsible for hardware monitoring. It is assumed the customer is familiar with virtualized environments and knows how to manage hardware in these environments.
Monitoring from CIMC
The CIMC provides the following hardware monitoring:
An overview of CPU, memory and power supply health
An overview of hardware inventory, including CPUs, memory, power supplies and storage
Monitoring of sensors for power supplies, fans, temperature, voltage and current
A system event log that contains BIOS and sensor entries
LSI MegaRAID controller information, which includes physical and virtual drive layout and Battery Backup Unit information from the Inventory > Storage tab. This information was usually accessible for earlier UCS servers only by installing the MegaRAID plugin from ESXi.
Monitoring from ESXi
To monitor server health from ESXi:
Log into the ESXi console and inspect system /var/log/messages for telltale entries
Monitoring from vSphere client and vCenter
The vSphere client provides the following monitoring:
Hardware and system alarms defined under the Alarms tab in the vSphere client when logged into vCenter.
VM resource usage under the Virtual Machines tab in the vSphere client, as well as under the Performance tab for each VM.
Host performance and resource usage under the Performance tab for the host.
Cisco UCS C220 M3 rack-mount server TRC1
Cisco supports running Cisco Unified Communications Manager on the Cisco UCS C220 M3 Rack-Mount server tested reference configuration 1 (TRC1). For more information about tested reference configurations for specific server models, see http://docwiki.cisco.com/wiki/Tested_Reference_Configurations_%28TRC%29.
Note
Cisco UCS C220 M3 Rack-Mount server TRC1 is not supported for use with Cisco Business Edition 6000. Deploy Cisco Business Edition 6000 on Cisco UCS C220 M3 Rack-Mount Server TRC2.
For information on installation, upgrade, and migration to the Cisco UCS C220 M3 Rack-Mount server TRC1, see the related topics.
To operate Cisco UCS C220 M3 Rack-Mount Server successfully, you should have the experience and skills to manage a host server running VMware ESXi. If you do not have this experience and want to obtain the required information quickly, consider using VMware GO, a Web-based application that facilitates VMware installations. For more information, refer to the VMware GO documentation.
Install and set up Cisco C220 M3 rack-mount server TRC1
Before You Begin
To prepare to install a Cisco UCS C220 M3 Rack-Mount Server TRC1 in a standalone configuration (it is not in a datacenter), Cisco recommends that you allocate the following resources before installation:
Space in a rack to receive a Cisco UCS C220 M3 Rack-Mount Server. This rack needs to accommodate the square mount rails shipped with Cisco UCS C220 M3 Rack-Mount Servers.
5 to 7 Ethernet ports on a switch close to the Cisco UCS C220 M3 Rack-Mount Server:
One standard Ethernet port for the dedicated CIMC management port, if desired.
Four ports for the quad-port Intel NIC.
Two ports for the LOM NICs, if required.
An IP address for CIMC management. If the dedicated port is used, it should be attached to the appropriate LAN.
A VLAN ID and IP address for the host. This is the Cisco UCS C220 M3 Rack-Mount Server ESXi management address.
A hostname and configure the DNS, if desired, for the hostname.
VLAN IDs and IP addresses for the VMs.
Upon receipt of the Cisco UCS C220 M3 Rack-Mount Server:
Install the C-220 M3 Rack-Mount Server in the rack.
Attach the CIMC management port to the designated switch port.
Attach the LOM NICs to their designated, trunked switch ports.
Attach a VGA console, or a KVM to the VGA and keyboard ports. This is necessary until CIMC has been configured.
Configuration checklist for installing and configuring the server
The following table provides a checklist of the major steps required to install and configure a Cisco UCS C220 M3 Rack-Mount Server TRC1. The Purpose column contains references to documentation that is related to the step.
Set up CIMC for Cisco C220 M3 rack-mount server TRC1
Configuring the CIMC allows you to perform all subsequent configuration and installation using the CIMC console. In addition, the CIMC provides a measure of hardware monitoring.
Follow this procedure to configure CIMC:
Procedure
Step 1
Attach a VGA console and keyboard directly to the server using
A dedicated Ethernet management port on the Cisco UCS C220 M3 Rack-Mount Server. This does not require a VLAN and is the simplest to configure. This is a single NIC and there is no redundancy for the CIMC using this configuration.
The LOM NICs. Although this option is available, Cisco does not recommend it.
Step 2
During boot, press F8 to enter CIMC configuration.
Step 3
In the CIMC configuration screen, under IPV4 (Basic):
Clear the DHCP enabled check box using the spacebar
Enter values for the CIMC IP address, subnet mask, and gateway.
Step 4
Leave VLAN (Advanced) clear.
Step 5
Under Default User (Basic), enter the default CIMC user, admin, and a password.
Note
The CIMC username is not configurable and is set to admin.
Step 6
Once configured, the CIMC is accessible through http. Point a browser to the IP address that you configured and login as admin, using the password that you configured.
Set up RAID for Cisco C220 M3 rack-mount server TRC1
Configure the RAID settings to the following specifications:
The 8 drives present on the system are set up as an 8-drive RAID-5 logical volume.
With eight 300GB drives, this corresponds to a total storage capacity of 1.93 TB, as seen by ESXi.
During boot, the system asks you to enter either Ctrl-H or Ctrl-Y.
Set up RAID with Preboot CLI for Cisco C220 M3 rack-mount server TRC1
Use the following sequence of commands to set the recommended RAID configuration; one 8-drive RAID-5 logical volume.
Important:
If required, use the following settings for the Read and Write policies:
Set Read Policy to read ahead = always.
Set Write Policy to one of the following:
write back with bbu – if using a RAID card with SuperCap (for example: RAID-9266CV)
always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies. Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Procedure
Step 1
Check your current RAID configuration:
During boot, the system asks you to enter either Ctrl-H or Ctrl-Y
To use the Preboot CLI to configure RAID, enter CTRL-Y.
Type the following command:
-ldinfo -l0 -a0
This command displays the number of drives, RAID level, and so forth for the specified logical drive. There should be 8 drives in a RAID-5 array for logical drive 0.
Step 2
Use the following sequence of commands to set the recommended RAID configuration; one 8-drive RAID-5 logical volume:
Enter CTRL-Y to enter the Preboot CLI when prompted during boot.
Enter the Preboot CLI command -cfgclr -a0 to clear the configuration.
Step 3
To determine the enclosure ID and drive numbering, which is required before you can configure RAID, run the following command:
-encinfo -a0 -page 20
Note
This command can generate more than one page of output, so enter -page 20 to look at 20 lines at a time. Look for the Device ID of the enclosure that has a non-zero Number of Physical Drives. Use this Device ID (also called Enclosure ID) in the following commands.
Step 4
There is a single enclosure for a Cisco UCS C220 M3 Rack-Mount Server. The enclosure ID is not predictable, so you must substitute the Device ID acquired above for <encl> in the commands below. When all drives are in a single enclosure, the slot numbering starts at zero. This may not be true in all cases, so verify the slot numbering with the following command:
-pdinfo -physdrv [<encl>:0] -a0
If this command generates meaningful output, the drives start at zero. If it generates an error, the drives start at one.
Note
Substitute your enclosure ID for <encl> in the commands, below. If you have determined your drives start at one, start with drive one instead of zero
Step 5
Use the following command to setup RAID 5 on the existing 8 drives:
The following commands are not necessary for new drives that have not been used.
Use the -ldinit -start -l0 -a0 command to perform a fast initialize.
To clear data on previously used drives and to slow (or full) initialize a new array, use the -ldinit -start -full -l0 -a0 command. Allow the command to finish before exiting the Preboot CLI. When the command –ldinit –showprog –l0 –a0 shows that initialization is not running, it is safe to exit the Preboot CLI.
Note
Cisco has noticed that a slow initialize can take up to 95 minutes or more to fully complete for array sizes of 1 TB+ that are used in a UCS C220 M3 Rack-Mount Server TRC1 deployment.
Step 7
After you configure the logical volume, exit the Preboot CLI by entering q.
Note
The LSI adapter has factory default values for the drive rebuild rate, patrol read rate, and other settings. Cisco recommends leaving the default values unchanged.
Set up RAID from the GUI for C220 M3 rack-mount server TRC1
Procedure
Step 1
During the boot process, make sure that Quiet Boot is disabled.
Press Ctrl - H at the LSI screen when prompted.
This brings you to the MegaRaid BIOS Configuration utility. Click Start.
Step 2
On the next screen
Select New Configuration.
Click Next.
Step 3
When prompted on the next screen to clear the configuration, click Yes.
Step 4
On the next screen
Select Manual Configuration.
Click Next.
Step 5
On the next screen, you add drives to a Drive Group.
Click on one drive and then select all other drives by holding down on Shift and the down arrow key.
Click Add to Array.
Click Accept DG.
Click Next to accept the Drive Group.
Step 6
The drive group must be added to a span.
Select Drive Group0.
Click Add to Span.
Click Next to accept the span definition.
Step 7
Once the drive group is part of a span, you can configure RAID on it.
Select RAID 5 from the list of available options.
(Optional) Cisco recommends that you select 128KB from the Strip Size drop down list.
Set Read Policy to read ahead = always.
Set Write Policy to one of the following:
- write back with bbu- if using a RAID card with SuperCap (e.g. RAID-9266CV)
- always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies.
Note
Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Click Update Size to finalize the RAID 5 volume and to determine the size of the resulting volume. It will work out to 1.903 TB.
Click Accept to accept the virtual drive definition.
Click Next.
Step 8
You are presented with a warning relating to BBU, as the BBU is used whenever possible, but with the understanding that if the BBU is discharged or undergoing a learning cycle, performance will be degraded. Click Yes if this is acceptable.
Step 9
Click Next on the next screen to accept the Virtual Drive you created (VD 0).
Step 10
Click Accept on the next screen to accept the Virtual Drive you created (VD 0).
Step 11
Click Yes to save the configuration.
Step 12
Click Yes to start drive initialization.
Step 13
At the next screen, you are presented with the options to initialize the array.
Click Fast Initialize
Click Go
Click Set Boot Drive
Click Go
Step 14
The RAID configuration is now complete.
Click Home to return to the main page.
Click Exit to exit the GUI.
Note
After RAID configuration is complete on the drives, the system may try to initialize (format) the new RAID array. When this happens, the current initialization progress can be seen from the Web BIOS screen. Wait for the background initialization to complete before proceeding with any of the subsequent server configuration steps such as installing ESXi. Cisco has noticed that a slow initialize can take up to 95 minutes or more to fully complete for array sizes of 1 TB+ used in a UCS C220 M3 Rack-Mount Server TRC1 deployment.
Set up BIOS boot order for Cisco C220 M3 rack-mount server TRC1
Note
Cisco does not support or restrict customers to any specific version of BIOS. The current version is assumed to be compatible with the latest release of ESXi. The same is true for the BIOS configuration. The default BIOS settings as shipped from the factory require no modification.
The hard drive is not selectable in the BIOS Boot Order menu until the first logical volume has been defined. Once RAID has been configured, make the second boot device the hard drive, as described in the following procedure:
Procedure
Step 1
Press F2 during boot to enter BIOS setup.
Step 2
Move the cursor over to Boot Options.
Step 3
Verify that the Cisco Virtual CD/DVD (Virtual CD/DVD drive) is selected for Boot Option #1.
Step 4
Verify that the hard drive (the RAID-5 array) is selected for Boot Option #2.
Step 5
Verify that VT is enabled in advanced CPU options.
Step 6
Verify that VT I/O Redirection is disabled in the CPU options.
The server will now try to boot the virtual CD drive first and the hard drive second.
Disk management for Cisco UCS C220 M3 rack-mount server TRC1
The Cisco UCS C220 M3 Rack-Mount Server TRC1 comes with 8 hard drives. These drives were configured as an 8-drive RAID-5 logical volume and require no further management at this point.
Disks are hot-swappable. This does not mean that you can swap drives ad-hoc on failures. A process exists to swap drives. When a drive fails, you must:
Reboot and enter the Preboot CLI
Mark the defective drive for removal using -PdPrpRmv -physdrv [<encl>:<slot>] -a0
Replace the drive
The RAID array is rebuilt automatically when the replacement disk is inserted.
Note
Although Preboot CLI is recommended, it is also possible to perform this task through the LSI MegaRaid GUI, where you can swap drives out without having to power-cycle the server to get into the Preboot CLI. However, this requires you to procure a separate machine (Windows or Linux) on the same subnet as the ESXi host, installed with the LSI MegaRaid utility.
ESXi installation and setup for C220 M3 rack-mount server TRC1
The following sections provide a sequence of steps for bringing ESXi into service at the customer site.
Preparation for ESXi Installation
Prior to installing ESXi, it is assumed that:
The IPL order in the BIOS is configured to boot the virtual CD/DVD first.
For the Cisco UCS C220 M3 Rack-Mount Server TRC1, the 8 drives are configured in a RAID-5 array.
These steps should have been performed by the factory prior to shipping.
ESXi installation takes less than five minutes. Once the installation is complete, remove the install CD and reboot the machine.
Following the reboot, a grey and yellow ESXi console is displayed with two options:
F2 to customize the system.
F12 to restart or halt the system.
At this point, press F2 and configure the system in accordance with your network.
Installing vSphere Client
Once the host is on the network, you can browse to its IP address to bring up a web-based interface. The vSphere client is Windows-based, so you must perform the download and installation from a Windows PC.
The installation proceeds like any other Windows application installation, and completes in a few minutes. Once the vSphere client is installed, you can bring it up and log into the host using the host name or IP address, the root login ID, and the password that you configured.
The host may also be joined to a vCenter if available and if you wish to manage the host through vCenter.
Configuring LOM NICs and Virtual Switches
You can configure the following options:
Simple vSwitch0 (the default VMware virtual switch).
For larger datacenters using vCenter, you can configure distributed virtual switches such as the distributed vSwitch or the Nexus 1000V distributed virtual switch.
For all options, you must define a port group for each VLAN running on the virtual switch. These port groups are selected when configuring a virtual machine network adapter to place the virtual machine on a given LAN.
VM installation and configuration Cisco C220 M3 rack-mount server TRC1
ISOs and VM templates
DVDs are available in the media kit that you receive with the Cisco UCS C220 M3 Rack-Mount Server TRC1, and the OVA templates are posted to Cisco.com.
Using the vSphere client to create the VM
Cisco provides templates on a URL to download and transfer to a host. Use the following procedure to create the VM:
Procedure
Step 1
Deploy the appropriate OVA template for your application using the Cisco.com URL as the source.
Step 2
Make the virtual CD drive available to the newly deployed VM.
Step 3
Refer to the Release Notes for the OVA that you are deploying for instructions on setting up the BIOS boot order.
Step 4
Map the installation media (from the media kit) to the virtual CD/DVD drive.
Step 5
Save the BIOS settings and boot.
You should be in the normal install screens for your application at this point.
Migrate to Cisco C220 M3 rack-mount server TRC1
Migrating from a Media Convergence Server (MCS server) to a Cisco UCS C220 M3 Rack-Mount Server follows a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
The following table provides an overview of the migration process and references to other pertinent documentation.
Configuration steps
Related procedures and topics
Step 1
Review the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager. This document describes how to replace server hardware, which is very similar to migrating from an MCS server to a Cisco UCS C220 M3 Rack-Mount Server. You should perform the documented pre-replacement and post-replacement tasks, and review the procedures for installing Cisco Unified Communications Manager and migrating data.
Replacing a Single Server or Cluster for Cisco Unified Communications Manager
Step 2
Upgrade the MCS server to Cisco Unified Communications Manager Release 8.0(2) or later.
Cisco Unified Communications Operating System Administration Guide
Step 3
If the UCS VM will use a different IP address or hostname than the MCS server, change the IP address and hostname of the MCS server to the values that the UCS VM will use.
This is required for DRS backup and restore to work.
Refer to the topic “Changing the Cluster IP Addresses for Publisher Servers That Are Defined by Host Name” in the Changing the IP Address and Host Name for Cisco Unified Communications Manager guide.
Step 4
Perform a DRS backup on the MCS server.
Disaster Recovery System Administration Guide
Step 5
Use the Answer File Generator to generate a license MAC for the Cisco UCS C220 M3 Rack-Mount Server. The license MAC is required to obtain licenses for the server.
After you obtain the license MAC, you can rehost the licenses for your new server.
Perform a DRS restore to restore the data backed up from the MCS server to the Cisco UCS C220 M3 Rack-Mount Server.
Disaster Recovery System Administration Guide
Step 9
Upload the new licenses to the Cisco UCS C220 M3 Rack-Mount Server.
If you did not obtain licenses for the new server already, you must request the licenses first.
Note
The previous license will no longer be valid. However, you have 30 additional days in which to use your previous license. Refer toNew licensing procedure customer impact
Cisco Unified Communications Operating System Administration Guide
Cisco C220 M3 rack-mount server TRC1 daily operations
At this point the application is installed and in operation. Daily operations from the viewpoint of the application do not differ from an installation on a physical server, including:
Application configuration and integration with other applications
RTMT performance monitoring
SNMP monitoring and alarms
DRS backup and restore
CDR collection
Device, trunk, and gateway configuration and monitoring
Monitoring Hardware from the VM
Applications running in a VM can not monitor the physical hardware. You must monitor the hardware from the CIMC, ESXi plugins, vCenter or by physical inspection (for flashing LEDs, and so on).
Monitoring of hardware is the responsibility of the customer. It is assumed the customer is familiar with virtualized environments and knows how to manage hardware in these environments.
Monitoring from CIMC
The CIMC provides the following hardware monitoring:
An overview of CPU, memory and power supply health
An overview of hardware inventory, including CPUs, memory, power supplies and storage
Monitoring of sensors for power supplies, fans, temperature, voltage and current
A system event log that contains BIOS and sensor entries
LSI MegaRAID controller information, which includes physical and virtual drive layout and Battery Backup Unit information from the Inventory > Storage tab. This information was usually accessible for earlier UCS servers only by installing the MegaRAID plugin from ESXi.
Monitoring from ESXi
To monitor server health from ESXi:
Log into the ESXi console and inspect system /var/log/messages for telltale entries
Monitoring from vSphere client and vCenter
The vSphere client provides the following monitoring:
Hardware and system alarms defined under the Alarms tab in the vSphere client when logged into vCenter.
VM resource usage under the Virtual Machines tab in the vSphere client, as well as under the Performance tab for each VM.
Host performance and resource usage under the Performance tab for the host.
Cisco UCS C220 M3 Rack-Mount Server TRC2
Cisco supports running Cisco Unified Communications Manager on the Cisco UCS C220 M3 Rack-Mount Server tested reference configuration 2 (TRC2). For more information about tested reference configurations for specific server models, see http://docwiki.cisco.com/wiki/Tested_Reference_Configurations_%28TRC%29.
Note
Cisco UCS C220 M3 Rack-Mount Server TRC2 is supported for use as part of Cisco Business Edition 6000 or with other applications.
Raid capacity of a Cisco UCS C220 M3 Rack-Mount Server TRC2
With four 500 GB drives, the total storage capacity of a Cisco UCS C220 M3 Rack-Mount server TRC2 as seen by ESXi is 929.46 GB.
Installation, upgrade, and migration to a Cisco UCS C220 M3 Rack-Mount Server TRC2
The procedures for the installation, upgrade, and migration to a Cisco UCS C220 M3 Rack-Mount Server TRC2 are the same as the procedures for the installation, upgrade, and migration to Cisco UCS C200 Rack-Mount Servers. For more information, see the related topics, as well as the following documentation:
Cisco supports running Cisco Unified Communications Manager on the Cisco UCS C220 M3 Rack-Mount Server tested reference configuration 3 (TRC3). For more information about tested reference configurations for specific server models, see http://docwiki.cisco.com/wiki/Tested_Reference_Configurations_%28TRC%29.
For information about installation, upgrade, and migration to a Cisco UCS C220 M3 Rack-Mount Server TRC3, see the related topics.
To operate a Cisco UCS C220 M3 Rack-Mount Server successfully, you should have the experience and skills that are required to manage a host server running VMware ESXi. If you do not have this experience and want to obtain the required information quickly, consider using VMware GO, a web-based application that facilitates VMware installations. For more information, refer to the VMware GO documentation.
Install and set up Cisco C220 M3 Rack-Mount Server TRC3
Before You Begin
To prepare to install a Cisco UCS C220 M3 Rack-Mount Server TRC3 in a standalone configuration (not in a data center), Cisco recommends that you allocate the following resources before installation:
Space in a rack to receive a Cisco UCS C220 M3 Rack-Mount Server. This rack needs to accommodate the square mount rails that are shipped with a Cisco UCS C220 M3 Rack-Mount Server.
Two or three Ethernet ports on a switch that is close to the Cisco UCS C220 M3 Rack-Mount Server:
One standard Ethernet port for the dedicated Cisco Integrated Management Controller (CIMC) port, if desired.
Two 802.1q trunked ports for the LAN on Motherboard (LOM) Network Interface Cards (NIC).
An IP address for CIMC. If you use the dedicated port, you should attach it to the appropriate LAN.
A VLAN ID and IP address for the host. This is the Cisco UCS C220 M3 Rack-Mount Server ESXi management address.
A hostname. If desired, configure a DNS for the hostname.
VLAN IDs and IP addresses for the VMs.
Upon receipt of the Cisco UCS C220 M3 Rack-Mount Server, perform the following tasks:
Install the C220 M3 Rack-Mount Server in the rack.
Attach the CIMC port to the designated switch port.
Attach the LOM NICs to their designated trunked switch ports.
Attach a VGA console, or a KVM, to the VGA and keyboard ports. This is necessary until you have configured the CIMC.
Install and set up server
The following procedure summarizes the major tasks that are required to install and set up a Cisco UCS C220 M3 Rack-Mount Server TRC3. For more information, see the related topics and the following documentation:
Set up CIMC for Cisco C220 M3 Rack-Mount Server TRC3
After you configure the Cisco Integrated Management Controller (CIMC), you can perform all subsequent configuration and installation tasks using the CIMC console. In addition, the CIMC provides a measure of hardware monitoring.
Follow this procedure to configure the CIMC.
Procedure
Step 1
Attach a VGA console and keyboard directly to the server using one of the following options:
A dedicated Ethernet management port on the Cisco UCS C220 M3 Rack-Mount Server. This option does not require a VLAN and is the simplest to configure. This is a single NIC and there is no redundancy for the CIMC using this configuration.
The LOM NICs. Although this option is available, Cisco does not recommend it.
Step 2
During boot, press F8 to enter CIMC configuration.
Step 3
In the CIMC configuration window, under IPV4 (Basic):
Use the spacebar to uncheck the DHCP enabled check box.
Enter values for the CIMC IP address, subnet mask, and gateway.
Step 4
Leave VLAN (Advanced) unchecked.
Step 5
Under Default User (Basic), enter the default CIMC user, admin, and a password.
Note
The CIMC username is not configurable and is set to admin.
Step 6
After you configure CIMC, you can access it through an HTTP connection. Point a browser to the CIMC IP address that you configured and log in as admin, using the password that you configured.
What to Do Next
To use the CIMC to check the hardware that is installed on your C220 M3 Rack-Mount Server:
Enter the default CIMC user, admin, and the password that you configured.
Choose Server > Inventory.
There are tabs for CPUs, Memory, Power Supplies, and Storage. You can inspect the number and type of CPUs, the number and size of DIMMs, and the disk drives to verify that your hardware configuration is correct.
Set up RAID for Cisco C220 M3 Rack-Mount Server TRC3
Check RAID Configuration
During boot, the system asks you to press either Ctrl-H or Ctrl-Y:
To use the Preboot CLI to check the RAID configuration, press Ctrl-Y.
Enter the command -ldinfo -l0 -a0.
This command displays the number of drives, RAID level, and so forth for the specified logical volume. There should be an 8-drive RAID-5 array for logical volume 0. If the RAID configuration is wrong, use the Preboot CLI or the GUI to configure RAID.
Set up RAID
Configure the eight drives on the system as a single 8-drive RAID-5 logical volume. With eight 300 GB drives, the total storage capacity of a Cisco C220 M3 Rack-Mount Server TRC3 as seen by ESXi is 1.903 TB.
During boot, the system asks you to press either Ctrl-H or Ctrl-Y:
To use the Preboot CLI to configure RAID, press Ctr-Y.
Set up RAID with Preboot CLI for Cisco C220 M3 Rack-Mount Server TRC3
Important:
If required, use the following settings for the Read and Write policies:
Set Read Policy to read ahead = always.
Set Write Policy to one of the following options:
write back with bbu - Choose this setting if you are using a RAID card with SuperCap (for example: RAID-9266CV).
always write back - Choose this setting if you are using a RAID card with legacy Battery Backup (BBU) instead of SuperCap (for example, RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies. Whenever possible, use the new RAID cards with SuperCap and make sure that the UCS is attached to an Uninterruptible Power Supply (UPS).
Procedure
Step 1
During boot, the system asks you to press either Ctrl-H or Ctrl-Y. To use the Preboot CLI to configure RAID, press Ctrl-Y.
Step 2
To clear the configuration, enter the command -cfgclr -a0.
Step 3
To determine the enclosure ID, enter the command -encinfo -a0 -page 20.
Note
This command can generate more than one page of output. Enter -page 20 to look at 20 lines at a time. Look for the Device ID of the enclosure that has a nonzero Number of Physical Drives. Use this Device ID (also called Enclosure ID) for <encl> in the following commands in this procedure.
Step 4
To verify the slot numbering, enter the command -pdinfo -physdrv [<encl>:0] -a0.
Note
When all drives are in a single enclosure, the slot numbering starts at 0. This may not be true in all cases. If this command generates meaningful output, the drives start at 0. If it generates an error, the drives start at 1.
Step 5
To set up RAID-5 on the eight drives, enter the command –CfgSpanAdd –r10 –Array0[<encl>:0,<encl>:1] –Array1[<encl>:2,<encl>:3] –a0.
Note
If you have determined in the previous step that your drives start at 1, start with drive 1 instead of 0.
Step 6
To clear data on previously used drives and to initialize the new array, perform one of the following options:
To perform a fast initialize, enter the command -ldinit -start -l0 -a0.
To perform a slow (or full) initialize, enter the command -ldinit -start -full -l0 -a0.
Note
A slow initialize can take up to 95 minutes or more to fully complete for the array sizes that are used in a UCS C220 M3 Rack-Mount Server TRC3 deployment.
Note
Allow the initialization to finish before exiting the Preboot CLI. When the command –ldinit –showprog –l0 –a0 shows that initialization is not running, it is safe to exit the Preboot CLI.
Note
The previous commands are not necessary for new drives that have not been used.
Step 7
After you configure the logical volume, enter q to exit the Preboot CLI.
Note
The LSI adapter has factory default values for the drive rebuild rate, patrol read rate, and other settings. Cisco recommends that you do not change the default values.
Set up RAID from the GUI for C220 M3 Rack-Mount Server TRC3
Procedure
Step 1
During the boot process, make sure that Quiet Boot is disabled.
To launch the MegaRAID BIOS Configuration utility, press Ctrl-H at the LSI screen when prompted.
Step 2
On the Adapter Selection screen, choose Adapter 0 and click Start.
Step 3
On the MegaRAID BIOS Config Utility Virtual Configuration screen, click Configuration Wizard.
Step 4
On the MegaRAID BIOS Config Utility Configuration Wizard screen, click the New Configuration radio button and click Next.
Step 5
On the MegaRAID BIOS Config Utility Confirm Page screen, click Yes.
Step 6
On the MegaRAID BIOS Config Utility Configuration Wizard screen, click the Manual Configuration radio button and click Next.
Step 7
On the MegaRAID BIOS Config Utility Config Wizard - Drive Group Definition screen, you add drives to a drive group.
From the drop-down list in the Drives pane, select Slot 0 through Slot 7.
Click Add to Array, to add the drives to Drive Group 0.
Click Accept DG, to accept the drive group.
Click Next.
Step 8
On the MegaRAID BIOS Config Utility Config Wizard - Span Definition screen, you add the drive group to the span.
In the Array With Free Space pane, select Drive Group0 from the drop-down list.
Click Add to Span.
Click Next, to accept the span definition.
Step 9
On the MegaRAID BIOS Config Utility Config Wizard - Virtual Drive Definition screen, you select the RAID configuration.
Choose RAID 5 from the RAID level drop-down list.
(Optional) Cisco recommends that you choose 128 KB from the Strip Size drop-down list.
Choose Always Read Ahead from the Read Policy drop-down list.
Choose one of the following options from the Write Policy drop-down list:
Write Back with BBU- Choose this setting if you are using a RAID card with SuperCap (for example, RAID-9266CV)
Always Write Back - Choose this setting if you are using a RAID card with the legacy Battery Backup (BBU) instead of SuperCap (for example, RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies.
Note
Whenever possible, use the new RAID cards with SuperCap and make sure that the UCS is attached to an Uninterruptible Power Supply (UPS).
Click Update Size to finalize the RAID-5 configuration and to determine the size of the resulting volume. The volume size should be 1.903 TB.
Click Accept, to accept the virtual drive definition.
Step 10
On the MegaRAID BIOS Config Utility Confirm Page screen, you are presented with a warning relating to BBU. The BBU is used whenever possible, but with the understanding that if the BBU is discharged or undergoing a learning cycle, performance will be degraded. Click Yes, if this is acceptable.
Step 11
On the MegaRAID BIOS Config Utility Config Wizard - Virtual Drive Definition screen, click Next to accept the Virtual Drive that you created (VD 0).
Step 12
On the MegaRAID BIOS Config Utility Config Wizard - Preview screen, click Accept to accept the configuration that you have defined.
Step 13
On the MegaRAID BIOS Config Utility Confirm Page screen, click Yes to save the configuration.
The MegaRAID BIOS Config Utility Confirm Page screen refreshes.
Step 14
On the MegaRAID BIOS Config Utility Confirm Page screen, click Yes to start the drive initialization.
Step 15
On the MegaRAID BIOS Config Utility Virtual Drives screen, click the Set Boot Drive radio button and click Go.
Step 16
The RAID configuration is now complete.
Click Home to return to the main MegaRAID BIOS Config Utility Virtual Configuration screen.
Click Exit to exit the GUI.
Note
After RAID configuration is complete on the drives, the system may try to initialize (format) the new RAID array. When this happens, the current initialization progress can be seen from the MegaRAID BIOS Config Utility Virtual Drives screen. Wait for the background initialization to complete before you proceed with any of the subsequent server configuration steps such as installing ESXi. A slow initialize can take up to 95 minutes or more to fully complete for the array sizes that are used in a UCS C220 M3 Rack-Mount Server TRC3 deployment.
Set up BIOS boot order for Cisco C220 M3 Rack-Mount Server TRC3
Note
Cisco does not support or restrict customers to a specific version of BIOS. The current version is assumed to be compatible with the latest release of ESXi. The same is true for the BIOS configuration. You do not need to modify the default BIOS settings.
You cannot select the hard drive in the BIOS Boot Order menu before you have defined the first logical volume. After you configure RAID, make the second boot device the hard drive, as described in the following procedure.
Procedure
Step 1
Press F2 during boot to enter the BIOS setup.
Step 2
Move the cursor to Boot Options.
Step 3
Verify that the CD ROM is selected for Boot Option #1.
Step 4
Verify that the hard drive (the RAID-5 array) is selected for Boot Option #2.
Step 5
Verify that VT is enabled in the advanced CPU options.
Step 6
Verify that VT I/O Redirection is disabled in the CPU options.
The server will now try to boot the virtual CD ROM drive first and the hard drive second.
Disk management for Cisco UCS C220 M3 Rack-Mount Server TRC3
The Cisco UCS C220 M3 Rack-Mount Server TRC3 comes with eight hard drives. These drives were configured as an 8-drive RAID-5 logical volume and do not require further management.
The drives are hot-swappable. This does not mean that you can swap drives ad-hoc when a drive fails. A process exists to swap drives. When a drive fails, you must:
Reboot and enter the Preboot CLI.
To enter the Preboot CLI, enter Ctrl-Y when prompted.
To mark the defective drive for removal, enter the command -PdPrpRmv -physdrv [<encl>:<slot>] -a0.
Replace the drive.
The RAID array is rebuilt automatically when the replacement drive is inserted.
Note
Although Cisco recommends that you use the Preboot CLI, you can also perform this task through the LSI MegaRaid GUI, where you can swap drives without having to power-cycle the server to get into the Preboot CLI. However, this method requires that you procure a separate machine (Windows or Linux), which is installed with the LSI MegaRaid utility and deployed on the same subnet as the ESXi host.
Install and set up ESXi for C220 M3 Rack-Mount Server TRC3
The following sections provide a sequence of steps to bring ESXi into service at your site.
Preparation for ESXi Installation
Prior to installing ESXi, Cisco assumes that:
The Initial Program Load (IPL) order in the BIOS is configured to boot the CD ROM first.
For the Cisco UCS C220 M3 Rack-Mount Server TRC3, the eight drives are configured in a RAID-5 array.
These configuration steps are performed at the factory prior to shipping the server.
ESXi installation takes less than five minutes. After the installation is complete, remove the installation CD and reboot the server.
After reboot, a gray-and-yellow ESXi console displays with two options:
Press F2 to customize the system.
Press F12 to restart or halt the system.
At this point, press F2 and configure the system in accordance with your network.
Install vSphere Client
After the host is on the network, you can browse to the host IP address to open a web-based interface. The vSphere client is Windows-based, so you must perform the download and installation from a Windows PC.
The vSphere client installation proceeds like the installation of any other Windows application, and completes in a few minutes. After you install the vSphere client, you can open it and log in to the host using the hostname or IP address, the root login ID, and the password that you configured.
You also can join the host to a vCenter, if available, and if you want to manage the host through vCenter.
Configure LAN on Motherboard Network Interface Cards and Virtual Switches
You can configure the following options:
Simple vSwitch0 (the default VMware virtual switch).
For larger data centers that use vCenter, you can configure distributed virtual switches such as the distributed vSwitch or the Nexus 1000V distributed virtual switch.
For all options, you must define a port group for each VLAN that runs on the virtual switch. You choose these port groups when you configure a virtual machine network adapter to place the virtual machine on a given LAN.
Install and set up VM for Cisco C220 M3 Rack-Mount Server TRC3
ISOs and VM templates
DVDs are available in the media kit that you receive with the Cisco UCS C220 M3 Rack-Mount Server TRC3, and the OVA templates are posted to Cisco.com.
Cisco provides OVA templates on a URL to download and transfer to a host. Use the following procedure to create the VM.
Procedure
Step 1
Deploy the appropriate OVA template for your application using the Cisco.com URL as the source.
Step 2
Make the CD ROM drive available to the newly deployed VM.
Step 3
Choose Options > Boot Options.
The next time the VM boots, force entry into the BIOS Setup screen.
Step 4
Insert the installation media, from the media kit, in the system CD ROM drive.
Step 5
Power on the VM, choose Boot, and promote the CD ROM to boot before the hard drive.
Step 6
Save the BIOS settings and boot.
The window for your application opens.
Migrate to Cisco C220 M3 Rack-Mount Server TRC3
To migrate from a Media Convergence Server (MCS server) to a Cisco UCS C220 M3 Rack-Mount Server, you follow a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
The following table provides an overview of the migration process and references to other pertinent documentation.
Configuration steps
Related documentation
Step 1
Review the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager. This document describes how to replace server hardware. Replacing server hardware is very similar to migrating from an MCS server to a Cisco UCS C220 M3 Rack-Mount Server. Perform the documented pre-replacement and post-replacement tasks, and review the procedures for installing Cisco Unified Communications Manager and migrating data.
Replacing a Single Server or Cluster for Cisco Unified Communications Manager
Step 2
Upgrade the MCS server to Cisco Unified Communications Manager Release 8.0(2) or a later release.
Cisco Unified Communications Operating System Administration Guide
Step 3
If the UCS virtual machine (VM) will use a different IP address or hostname than the MCS server, change the IP address and hostname of the MCS server to the values that the UCS VM will use.
This change is required for DRS backup and restore to work.
See the topic “Changing the Cluster IP Addresses for Publisher Servers That Are Defined by Host Name” in the Changing the IP Address and Host Name for Cisco Unified Communications Manager guide.
Step 4
Perform a DRS backup on the MCS server.
Disaster Recovery System Administration Guide
Step 5
Use the Answer File Generator to generate a license MAC for the Cisco UCS C220 M3 Rack-Mount Server. The license MAC is required to obtain licenses for the server.
After you obtain the license MAC, you can rehost the licenses for your new server.
Installing Cisco Unified Communications Manager
Step 6
Create the VM on the Cisco UCS C220 M3 Rack-Mount Server that will be used as the replacement for the MCS node.
Step 7
Install the same release of Cisco Unified Communications Manager on the Cisco UCS C220 M3 Rack-Mount Server that you installed on the MCS server.
Installing Cisco Unified Communications Manager
Step 8
Perform a DRS restore operation to restore the data that you backed up from the MCS server to the Cisco UCS C220 M3 Rack-Mount Server.
Disaster Recovery System Administration Guide
Step 9
Upload the new licenses to the Cisco UCS C220 M3 Rack-Mount Server.
If you did not obtain licenses for the new server already, you must request the licenses first.
Note
The previous licenses are not valid. However, you can still use the licenses for 30 days.
Cisco Unified Communications Operating System Administration Guide
Cisco C220 M3 Rack-Mount Server TRC3 daily operations
At this point the application is installed and in operation. Daily operations do not differ from an installation on a physical server, including:
Application configuration and integration with other applications
RTMT performance monitoring
SNMP monitoring and alarms
DRS backup and restore
CDR collection
Device, trunk, and gateway configuration and monitoring
Hardware monitoring from the VM
Applications that run in a VM cannot monitor the physical hardware. You must monitor the hardware from the CIMC, the ESXi plug-ins, or vCenter or by physical inspection (for flashing LEDs, and so on).
Monitoring of hardware is the responsibility of the customer. Cisco assumes that the customer is familiar with virtualized environments and knows how to manage hardware in these environments.
Monitoring from CIMC
The CIMC provides the following hardware monitoring:
An overview of CPU, memory, and power supply health
An overview of hardware inventory, including CPUs, memory, power supplies, and storage
Monitoring of sensors for power supplies, fans, temperature, voltage, and current
A system event log that contains BIOS and sensor entries
Monitoring from ESXi
To monitor server health from ESXi, log in to the ESXi console and inspect system /var/log/messages for relevant entries.
To monitor the RAID controller, install plug-ins such as LSI MegaRAID Release 3.6.
Monitoring from vSphere client and vCenter
The vSphere client provides the following monitoring:
Hardware and system alarms that are defined under the Alarms tab in the vSphere client when you are logged in to vCenter.
VM resource usage under the Virtual Machines tab in the vSphere client, as well as under the Performance tab for each VM.
Host performance and resource usage under the Performance tab for the host.
C220 M3 Rack-Mount Server is a single point of failure
For a cluster of up to four nodes that is capable of handling hundreds of thousands of Busy Hour Call Completions (BHCC), the C220 M3 Rack-Mount Server is a single point of failure.
The entire cluster can be brought out of service if the C220 M3 Rack-Mount Server is powered off, rebooted, or suffers a critical hardware outage. While Cisco provides power supply, network, and hard drive redundancy, recovery is not possible from a hardware failure that involves the C220 M3 Rack-Mount Server CPU, memory, motherboard, disk controller, or other critical hardware.
Cisco UCS C210 rack-mount servers
Cisco supports running Cisco Unified Communications Manager on the Cisco UCS C-Series Rack-Mount Server.
To operate Cisco UCS C210 Rack-Mount Servers successfully, you should have the experience and skills to manage a host server running VMware ESXi. If you do not have this experience and want to obtain the required information quickly, consider using VMware GO, a Web-based application that facilitates VMware installations. For more information, refer to the VMware GO documentation.
Install and set up Cisco UCS C210 Rack-Mount Server
The following procedure provides the tasks required to install and configure Cisco UCS C210 Rack-Mount Servers. For more information, see the related topics, as well as the following documentation:
Standalone Cisco UCS C210 Rack-Mount Server installation preparation
This section describes how to prepare to install a Cisco UCS C210 Rack-Mount Servers server in a standalone configuration, meaning that it is not in a datacenter.
You should allocate the following resources before installation:
Space in a rack to receive a 2 RU Cisco UCS C-Series Rack-Mount Server
3 Ethernet ports on a switch close to the Cisco UCS C-Series Rack-Mount Server:
One port for the CIMC
Two ports for the LAN on motherboard (LOM) NICs
Optionally, up to four IP addresses for the Broadcom NIC, if your server has it
An IP address for the CIMC management port
An IP address for the virtual host. This is the Cisco UCS C-Series Rack-Mount Server's IP address and is used by ESXi.
A maximum of four IP addresses for the LAN on motherboard (LOM) NICs
A hostname, and optionally configure DNS for the virtual hosts' hostname
IP addresses for the VMs
Set up RAID for Cisco UCS C210 Rack-Mount Server
Configure the RAID settings to the following specifications:
The first 2 drives are configured as a RAID 1 (mirrored) drive. This drive is for ESXi installation.
The remaining drives, if any exist, are configured as a RAID 5 drive. This drive is for Cisco Unified Communications Manager application VMs.
Important:
If required, use the following settings for the Read and Write policies:
Set Read Policy to read ahead = always.
Set Write Policy to one of the following:
write back with bbu – if using a RAID card with SuperCap (for example: RAID-9266CV)
always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies. Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Follow this procedure to perform this task:
Procedure
Step 1
Make sure that Quiet Boot is disabled by following these steps:
Boot the server and watch for the F2 prompt during bootup.
Press F2 when prompted to enter the BIOS Setup utility.
On the Main page of the BIOS Setup utility, set Quiet Boot to Disabled. This allows non-default messages, prompts, and POST messages to display during bootup instead of the Cisco logo screen.
Press F10 to save your changes and exit the utility.
Step 2
During server bootup, press Ctrl+Y to enter the preboot CLI.
Step 3
Enter the following commands to determine the current RAID configuration:-ldinfo-l0-a0
The required configuration is two drives in a RAID 1 array for logical drive 0, and for servers with more drives, the remaining drives in a RAID 5 array for Logical drive 1. If the RAID configuration is wrong, continue with this procedure.
Note
Do not continue with this procedure if RAID is configured correctly.
Step 4
Enter the command --cfgclr-a0 to clear the RAID configuration.
Caution
Clearing the RAID configuration deletes all data on the hard drives.
Step 5
Determine the Device ID of the enclosure that contains the disk drives by entering the command -encinfo--a0-page-20.
If necessary, page through the output to find the Device ID of the enclosure that has a non-zero entry for Number of Physical Drives. Substitute this Device ID value where deviceID appears in the commands in this procedure.
Step 6
Determine the starting slot number in the enclosure that you identified by entering the command -pdinfo-physdrv[deviceID:0]-a0.
If this command generates meaningful output, the drives start at slot zero. If it generates an error, the drives start at slot one.
Step 7
Configure the first RAID array by entering one of the following commands, depending on the starting slot number:
If your drives start at slot zero, run this command:
-cfgldadd-r1[deviceID:0,deviceID:1]-a0
If your drives start at slot one, run this command:
-cfgldadd-r1[deviceID:1,deviceID:2]-a0
Step 8
If your server contains 6 total disk drives, enter the following command to configure the second RAID array:
If your server contains 10 total disk drives, configure the second RAID array by entering one of the following commands, depending on the starting slot number:
If your drives start at slot zero, run this command:
-cfgldadd-r5[deviceID:2,deviceID:3,deviceID:4,deviceID:5,deviceID:6,deviceID:7,deviceID:8,deviceID:9]-a0
If your drives start at slot one, run this command:
-cfgldadd-r5[ deviceID:3,deviceID:4,deviceID:5,deviceID:6,deviceID:7,deviceID:8,deviceID:9,deviceID:10]-a0
Step 10
If the hard drives did not have a RAID configuration previously, you are done configuring RAID. If the hard drives had a RAID configuration before, continue with this procedure.
Step 11
Enter the following commands to initialize the logical volumes.
-ldinit-start-full-l0-a0 (l0 is the letter l and the number 0, not the number 10)
-ldinit-start-full-l1-a0 (l1 is the letter l and the number 1, not the number 11)
This clears data on the drives and initializes the new array.
Step 12
Allow these commands to finish running before exiting the Preboot CLI. Enter the following commands to display the progress of the commands:
-ldinit-showprog-l0-a0
-ldinit-showprog-l1-a0
When both commands report that no initialization is running, it is safe to quit the Preboot CLI.
Step 13
After configuring the two logical volumes, you can exit the Preboot CLI by entering q.
vSphere Client installation
When the virtual host is available on the network, you can browse to its IP address to bring up a web-based interface. The vSphere Client is Windows-based, so the download and install must be performed from a Windows PC.
Once the vSphere Client is installed, you can run it and log into the virtual host using the virtual host's name or IP address, the root login ID, and the password you configured.
You can join the host to a vCenter if you want to manage it through vCenter.
Datastore used for VM alignment
Note
This section does not apply to servers that contain only 2 disk drives. There is only one logical volume on such servers.
In VMWare ESXi 4.x and earlier versions, a second logical volume is automatically imported unaligned during installation. To ensure that the ESXi partition used for VMs will be aligned, you should delete the unaligned datastore (the larger disk partition), then recreate the datastore using vSphere client.
When you install VMWare ESXi, a single datastore is created. This datastore has the EXSi is installed. VMs have better disk performance when all partitions (physical, ESXi and VM) start on the same boundary. This prevents disk blocks being fragmented across the different boundaries.
Create virtual machines for Cisco UCS C210 Rack-Mount Server installation
Cisco provides a VM template for you to download and transfer to your virtual host. Use this template to create the VMs for Cisco UCS C210 Rack-Mount Servers installation.
Before you deploy the template and create VMs, you should have a hostname and IP address allocated for each new VM.
Follow these steps to create a VM and prepare to install Cisco UCS C210 Rack-Mount Servers on it:
Procedure
Step 1
Download the VM template for your application.
Step 2
Upload the template to a datastore on the Cisco UCS C-Series Rack-Mount Server.
It is recommended to use the smaller datastore (with ESXi installed on it) for this.
Step 3
Make this template available to the Cisco UCS C-Series Rack-Mount Server.
Step 4
Deploy the template file using vSphere Client. Enter the following information for the new VM:
hostname
datastore—Select the larger datastore
Step 5
Complete creating the VM.
At this point a new VM has been created with the correct amount of RAM, number of CPUs, size and number of disks for the intended application.
Step 6
Install Cisco Unified Communications Manager on the VM.
Install Cisco Unified Communications Manager on VM
Follow this procedure to install Cisco Unified Communications Manager on a new VM:
Procedure
Step 1
In vSphere Client, edit the VM to force entry into BIOS setup the next time the VM reboots.
Step 2
Make the Cisco Unified Communications Manager installation media available to the VM DVD-ROM drive.
Step 3
Power on the VM, then in BIOS setup, promote CD ROM to boot before the hard drive.
Step 4
Complete booting the VM.
The Cisco Unified Communications Manager installation program starts. During the installation, the VM displays a message to indicate that the guest operating system has locked the CD-ROM door, and prompts you to disconnect and override the lock. This message displays twice during the installation process. Click No on each occurrence and continue the installation.
For information about performing the installation, see the document Installing Cisco Unified Communications Manager.
Note
Cisco UCS C210 Rack-Mount Servers does not support a hardware clock; you must use NTP. The option to configure a hardware clock is not available in the installation program.
Migrate to Cisco UCS C210 Rack-Mount Servers
Migrating from a Media Convergence Server (MCS server) to a Cisco UCS C210 Rack-Mount Servers server follows a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
The following procedure provides the tasks for the migration process. For more information, see the related topics, as well as the following documentation:
Cisco Unified Communications Operating System Administration Guide
Disaster Recovery System Administration Guide
Installing Cisco Unified Communications Manager
Replacing a Single Server or Cluster for Cisco Unified Communications Manager
Refer to the topic "Changing the Cluster IP Addresses for Publisher Servers That Are Defined by Host Name" in the Changing the IP Address and Host Name for Cisco Unified Communications Manager guide.
Procedure
Step 1
Review the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
This document describes how to replace server hardware, which is very similar to migrating from an MCS server to a Cisco UCS C210 Rack-Mount Servers server. You should perform the document's pre-replacement and post-replacement tasks, and review the procedures for installing Cisco Unified Communications Manager and migrating data.
Step 2
Upgrade the MCS server to Cisco Unified Communications Manager Release 8.0(2) or later.
Step 3
If the UCS VM will use a different IP address or hostname than the MCS server, change the IP address and hostname of the MCS server to the values that the UCS VM will use.
This is required for DRS backup and restore to work.
Step 4
Perform a DRS backup on the MCS server.
Step 5
Use the Answer File Generator to generate a license MAC for the Cisco UCS C210 Rack-Mount Servers server.
VMware support for Cisco UCS C210 Rack-Mount Servers
Consider the following, when using Cisco UCS C210 Rack-Mount Servers:
Install, upgrade, and recovery procedures now use "soft media" such as ISO or FLP (virtual floppy) if the server does not have a DVD drive.
The answer file generated by the Answer File Generator (platformConfig.xml) cannot get read from a USB key to perform an unattended installation. Instead, you must put the answer file into a FLP image to be mounted in the floppy drive.
Music On Hold external source and Cisco Messaging Interface requires a mixed MCS/UCS cluster.
USB tape backup is not supported.
NIC teaming is configured at the VMware virtual switch.
Hardware SNMP and syslog move to VMware and UCS Manager.
Install logs are written only to the virtual serial port.
Unattended installs use virtual floppy instead of USB.
Basic UPS Integration, as used with a Cisco 7800 Series Media Convergence Server, is not supported.
Boot order is controlled by the BIOS of the VMware VM.
Hardware BIOS, firmware, and drivers must be the required level and configured for compatibility with Cisco Unified Communications Manager-supported VMware product and version.
Hardware MIBs are not supported.
Hardware Failure alert and Hardware failure syslog messages are not available.
CLI does not support hardware information.
SNMP Hardware agent does not run on VMware.
Real–Time Management Tool Client—Hardware alerts are not generated. The scope is limited to Virtual Machine and not to Physical Machine.
CDP reports as a Virtual Machine.
Certain kinds of Cisco UCS B-Series Blade Server and Cisco UCS C-Series Rack-Mount Server hardware alerts are only available via CIM alerting, and must be viewed in VMware vCenter or an equivalent CIM-compliant console.
For more information on the Cisco UCS C-Series Rack-Mount Server, go to the following URL:
Applications running in a VM have no ability to monitor the physical hardware. Any hardware monitoring must be done from the CIMC, ESXi plugins, vCenter, or by physical inspection (for flashing LEDs, etc.).
Hardware monitoring from CIMC
The CIMC provides the following hardware monitoring:
An overview of CPU, memory and power supply health
An overview of hardware inventory, including CPUs, memory, power supplies and storage
Monitoring of sensors for power supplies, fans, temperature and voltage
A system event log that contains BIOS and sensor entries
Hardware and system monitoring from vSphere Client and vCenter
The vSphere Client provides the following monitoring features:
When you are logged into vCenter, the vSphere Client displays hardware and system alarms defined on the Alarms tab.
VM resource usage displays on the Virtual Machines tab, and on the Performance tab for each VM.
Host performance and resource usage display on the Performance tab for the Host.
When ESXi is used standalone (without vCenter), hardware status and resource usage are available, but alarming is not possible.
Ensure that you use ESXi, rather than ESX, to run Cisco UCS C200 Rack-Mount Servers. However, the rack server can be part of a VMware vCenter that includes ESX hosts.
To operate Cisco UCS C200 Rack-Mount Servers successfully, you should have the experience and skills to manage a host server running VMware ESXi. If you do not have this experience and want to obtain the required information quickly, consider using VMware GO, a Web-based application that facilitates VMware installations. For more information, refer to the VMware GO documentation.
Cisco UCS C200 Rack-Mount Servers uses "soft media" such as ISO or FLP (virtual floppy) for procedures that require external media (such as installation and upgrade). Physical external devices such as USB drives are not supported.
Note
Backup and restore are not supported on soft media.
The virtual USB interface is not supported on VMware. The following are examples of differences in external media support between MCS servers and Cisco UCS C200 Rack-Mount Servers:
Install logs cannot get dumped to a USB key. These logs get dumped to a file through the serial port of the VM.
The answer file generated by the Answer File Generator (platformConfig.xml) cannot get read from a USB key to perform an unattended installation. Instead, you must put the answer file into a FLP image to be mounted in the floppy drive.
USB tape drive backup is not supported. Use SFTP instead.
Music On Hold through a USB connection is not supported. Use a Cisco 7800 Series Media Convergence Server instead.
Cisco Messaging Interface (CMI) for Message Waiting Indication (MWI) is not supported over the serial port. Use a Cisco 7800 Series Media Convergence Server instead.
Cisco UCS C200 Rack-Mount Server installation
This section describes how to perform a fresh installation of Cisco UCS C200 Rack-Mount Servers.
Install and set up Cisco UCS C200 Rack-Mount Server
The following procedure provides the tasks required to install and configure Cisco UCS C200 Rack-Mount Servers. For more information, see the related topics, as well as the following documentation:
Standalone Cisco UCS C200 Rack-Mount Server installation preparation
This section describes how to prepare to install a Cisco UCS C200 Rack-Mount Servers server in a standalone configuration, meaning that it is not in a datacenter.
You should allocate the following resources before installation:
Space in a rack to receive a 2 RU Cisco UCS C-Series Rack-Mount Server
3 Ethernet ports on a switch close to the Cisco UCS C-Series Rack-Mount Server:
One port for the CIMC
Two ports for the LAN on motherboard (LOM) NICs
Optionally, up to four IP addresses for the Broadcom NIC, if your server has it
An IP address for the CIMC management port
An IP address for the virtual host. This is the Cisco UCS C-Series Rack-Mount Server's IP address and is used by ESXi.
A maximum of four IP addresses for the LAN on motherboard (LOM) NICs
A hostname, and optionally configure DNS for the virtual hosts' hostname
IP addresses for the VMs
Set up RAID for Cisco UCS C200 Rack-Mount Servers
Configure the RAID settings to the following specifications:
RAID 10 array for logical drive 0
RAID 10 volume of 4 drives
Important:
If required, use the following settings for the Read and Write policies:
Set Read Policy to read ahead = always.
Set Write Policy to one of the following:
write back with bbu – if using a RAID card with SuperCap (for example: RAID-9266CV)
always write back – if using a RAID card with legacy Battery Back-Up (BBU) instead of SuperCap (e.g. RAID-9266). This option helps prevent a UC application performance impact in the event that the BBU goes into learning mode or the battery dies. Whenever possible, use the new RAID cards with SuperCap and make sure the UCS is attached to an Uninterruptible Power Supply (UPS).
Follow this procedure to perform this task:
Procedure
Step 1
Boot the server and watch for the F2 prompt during bootup.
Step 2
Press F2 when prompted to enter the BIOS Setup utility.
Step 3
On the Main page of the BIOS Setup utility, verify or set Quiet Boot to Disabled. This allows non-default messages, prompts, and POST messages to display during bootup instead of the Cisco logo screen.
Step 4
Press F10 to save your changes and exit the utility.
Step 5
During server bootup, press Ctrl+Y to enter the preboot CLI.
Step 6
Enter the following commands to determine the current RAID configuration:-ldinfo-l0-a0
The required configuration is four drives in a RAID 10 array for logical drive 0. If the RAID configuration is wrong, continue with this procedure.
Note
Do not continue with this procedure if RAID is configured correctly.
Step 7
Enter the command --cfgclr-a0 to clear the RAID configuration.
Caution
Clearing the RAID configuration deletes all data on the hard drives.
Step 8
Determine the Device ID of the enclosure that contains the disk drives by entering the command -encinfo--a0-page-20.
Note
For UCS C200 M2 Rack-Mount Servers with 4 drives, the Device ID is typically 252. If this is not the case, please use the Enclosure ID obtained from the output of encinfo.
Step 9
Determine the starting slot number in the enclosure that you identified by entering the command -pdinfo-physdrv[deviceID:0]-a0.
If this command generates meaningful output, the drives start at slot zero. If it generates an error, the drives start at slot one.
Note
UCS C200 M2 Rack-Mount Servers with 4 drives, the drives start at slot zero.
Step 10
Configure the RAID array by entering the following command:
If the hard drives did not have a RAID configuration previously, you are done configuring RAID. If the hard drives had a RAID configuration before, continue with this procedure.
Step 12
Enter the following command to initialize the logical volumes.
-ldinit-start-full-l0-a0 (l0 is the letter l and the number 0, not the number 10)
This clears data on the drives and initializes the new array. Allow this command to finish running before exiting the Preboot CLI.
Step 13
If you want to do so, you can enter the following command to display the progress of the command you entered in Step 12:
-ldinit-showprog-l0-a0
When the display command in Step 13 reports that no initialization is running, it is safe to quit the Preboot CLI.
Step 14
After configuring the two logical volumes, you can exit the Preboot CLI by entering q.
Tip
If you can't get to the Preboot CLI even after Quiet Boot is disabled (in other words, if you have no Ctrl+Y option and that key sequence isn't working), you can configure RAID 10 by using the WebBios (Ctrl+H). To use the Web Bios, you must have a USB Keyboard and a USB mouse, rather than a PS2 keyboard and mouse. Use 2 drive groups (DGs) with DG0 containing Disks 0,1 and DG1 containing Disks 2,3.
Install vSphere Client
When the virtual host is available on the network, you can browse to its IP address to bring up a web-based interface. The vSphere Client is Windows-based, so the download and install must be performed from a Windows PC.
Once the vSphere Client is installed, you can run it and log into the virtual host using the virtual host's name or IP address, the root login ID, and the password you configured.
You can join the host to a vCenter if you want to manage it through vCenter.
Create virtual machines for Cisco UCS C200 Rack-Mount Server installation
Cisco provides a VM template for you to download and transfer to your virtual host. Use this template to create the VMs for Cisco UCS C200 Rack-Mount Servers installation.
Before you deploy the template and create VMs, you should have a hostname and IP address allocated for each new VM.
Follow these steps to create a VM and prepare to install Cisco UCS C200 Rack-Mount Servers on it:
Procedure
Step 1
Download the VM template for your application.
Step 2
Upload the template to a datastore on the Cisco UCS C-Series Rack-Mount Server.
Step 3
Make this template available to the Cisco UCS C-Series Rack-Mount Server.
Step 4
Deploy the template file using vSphere Client. Enter the following information for the new VM:
hostname
datastore—Select datastore
Step 5
Complete creating the VM.
At this point a new VM has been created with the correct amount of RAM, number of CPUs, size and number of disks for the intended application.
Step 6
Install Cisco Unified Communications Manager on the VM.
Migrating from a Media Convergence Server (MCS server) to a Cisco UCS C200 Rack-Mount Servers server follows a procedure that is very similar to replacing server hardware, which is described in the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
The following procedure provides the tasks for the migration process. For more information, see the related topics, as well as the following documentation:
Cisco Unified Communications Operating System Administration Guide
Disaster Recovery System Administration Guide
Installing Cisco Unified Communications Manager
Replacing a Single Server or Cluster for Cisco Unified Communications Manager
Refer to the topic "Changing the Cluster IP Addresses for Publisher Servers That Are Defined by Host Name" in the Changing the IP Address and Host Name for Cisco Unified Communications Manager guide.
Procedure
Step 1
Review the document Replacing a Single Server or Cluster for Cisco Unified Communications Manager.
This document describes how to replace server hardware, which is very similar to migrating from an MCS server to a Cisco UCS C200 Rack-Mount Servers server. You should perform the document's pre-replacement and post-replacement tasks, and review the procedures for installing Cisco Unified Communications Manager and migrating data.
Step 2
Upgrade the MCS server to Cisco Unified Communications Manager Release 8.0(2) or later.
Step 3
If the UCS VM will use a different IP address or hostname than the MCS server, change the IP address and hostname of the MCS server to the values that the UCS VM will use.
This is required for DRS backup and restore to work.
Step 4
Perform a DRS backup on the MCS server.
Step 5
Use the Answer File Generator to generate a license MAC for the Cisco UCS C200 Rack-Mount Servers server.
Install, upgrade, and recovery procedures now use "soft media" such as ISO or virtual floppy (FLP) if the server does not have a DVD drive.
NIC teaming is configured at the VMware virtual switch.
Hardware SNMP and syslog move to VMware and UCS Manager.
Install logs are written only to the virtual serial port.
Basic UPS Integration, as used with a Cisco 7800 Series Media Convergence Server, is not supported.
Boot order is controlled by the BIOS of the VMware VM.
Hardware BIOS, firmware, and drivers must be the required level and configured for compatibility with Cisco Unified Communications Manager-supported VMware product and version.
Hardware MIBs are not supported.
Hardware Failure alert and Hardware failure syslog messages are not available.
CLI does not support hardware information.
SNMP Hardware agent does not run on VMware.
Real–Time Management Tool Client—Hardware alerts are not generated. The scope is limited to Virtual Machine and not to Physical Machine.
CDP reports as a Virtual Machine.
Certain kinds of Cisco UCS C-Series Rack-Mount Server hardware alerts are only available via CIM alerting, and must be viewed in VMware vCenter or an equivalent CIM-compliant console.
For more information on the Cisco UCS C-Series Rack-Mount Server, go to the following URL:
Cisco Unified Communications Manager on Virtualized Servers uses a different licensing model than Cisco Unified Communications Manager on an MCS server. The MAC address of the NIC card is no longer used to associate the license to the server.
Instead, the license gets associated to a license MAC, which is a 12 digit HEX value created by hashing the following parameters that you configure on the server:
Time zone
NTP server 1 (or "none")
NIC speed (or "auto")
Hostname
IP Address (or "dhcp")
IP Mask (or "dhcp")
Gateway Address (or "dhcp")
Primary DNS (or "dhcp")
SMTP server (or "none")
Certificate Information (Organization, Unit, Location, State, Country)
The ways to obtain the license MAC are as follows:
If you use this method, ensure that you enter the identical parameter values in the Answer File Generator and the Cisco Unified Communications Manager installation program, or the license will be invalid.
After installation, navigate to Show > System in Cisco Unified Communications Manager Administration.
After installation, use the CLI command show status.
Obtaining New Licenses
The process to redeem a Product Activation Key (PAK) for licenses at www.cisco.com/go/license is changed for a license MAC. When redeeming a PAK for a license MAC at this URL, you get prompted to select the type of license that you want to obtain:
A physical MAC address — this is used when Cisco Unified Communications Manager will be installed on an MCS server.
A license MAC address — this is used when Cisco Unified Communications Manager will be installed on Cisco Unified Communications Manager on Virtualized Servers.
After you make this selection, the generation and installation of the license file follows the same process.
Obtaining Rehosted Licenses When You Change License MAC Parameters
When you change any of the parameters that create the license MAC, the license that you obtained with it becomes invalid. You must request a rehosting of the license to obtain a valid license. The old license continues to work for a 30-day grace period.
To rehost your licenses, you must open a case with the licensing team to obtain a license for your replacement server. Contact the licensing team at licensing@cisco.com.
During the grace period, you can change the settings back to the licensed values to make your original license valid again. If you need more than 30 days of grace period, change your settings back to the licensed values, then change them back to the new values that you want to use. You will get another 30- day grace period.
Virtual machine setup and licensing support
The virtual machine configuration for running Cisco Unified Communications Manager on Virtualized Servers must match the stated specifications to get support from Cisco.
While Cisco Unified Communications Manager can be installed and licensed in other virtual machine configurations, Cisco does not support these configurations.
TCP and UDP ports for vCenter Server, ESX hosts, and other network components' management access are listed in article 1012382 at the following URL:
http://kb.vmware.com
Migrate Cisco Unity Connection on a virtual machine
For information on migrating to
Cisco Unity Connection on a virtual machine, see the
"Migrating from a
Cisco Unity Connection Physical Server to a Connection 8.x Virtual Machine"
chapter in the applicable
Reconfiguration and Upgrade Guide for
Cisco Unity Connection at
http://www.cisco.com/en/US/products/ps6509/prod_installation_guides_list.html.