Use the following procedure to collect cloning images:
Create a Physical L-Server for the Infrastructure Administrator
Refer to "Create an L-Server for the Infrastructure Administrator".
Manually Install an OS
Install an OS on the physical L-Server for the infrastructure administrator created in 1.
Refer to "Manual OS Installation".
Install a Multipath Driver (when using a Multipath for the Path to the SAN)
Install a multipath driver to the L-Server.
Refer to the manual of the multipath driver for details on installing multipath drivers.
Stop the L-Server
For details on stopping L-Servers, refer to "11.1.2 Stopping an L-Server" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
Change the SAN Path Status of the L-Server
Follow the procedure below to change the SAN path status of the L-Server.
Right-click the target L-Server in the orchestration tree, and select [Change Settings]-[Modify Specification] from the popup menu. The [Modify an L-Server] dialog is displayed.
Uncheck the "Single path mode" checkbox for FC path.
Click <OK>.
Install the Agent
Install the Resource Orchestrator agent to the physical L-Server for the infrastructure administrator.
For details on installing agents, refer to "2.2 Agent Installation" in the "Installation Guide CE".
Register the Agent
It is also necessary to install an agent after installing an OS, and then register the server used by an L-Server for an agent.
Use the following procedure to register agents.
Right-click the target L-Server in the orchestration tree, and select [Register]-[Agent] from the popup menu.
The [Register Agent] dialog is displayed.
Click <OK>.
Start the L-Server
For details on starting L-Servers, refer to "11.1.1 Starting an L-Server" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
Stop the L-Server
For details on stopping L-Servers, refer to "11.1.2 Stopping an L-Server" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
Collect a Cloning Image from the L-Server after the OS is Installed
Refer to "Collect a Cloning Image from the L-Server after the OS is Installed".
Note
This function is not available if ServerView Deployment Manager is used on the admin LAN. Use the cloning function of ServerView Deployment Manager. For details, refer to "Appendix L Co-Existence with ServerView Deployment Manager".
When cloning, only the content of the first disk recognized in the BIOS (the boot disk) of the managed server is collected.
Data disk content (second disk onwards) cannot be cloned. Use other backup software and the copy functions of storage units to collect and deploy the data.
If multiple storage areas (Windows drivers or Linux partitions) are set in the first disk, all the partitions are cloned.
Disk | Windows Drive | Cloning Target |
|---|---|---|
First disk | C: | Yes |
E: | Yes | |
Second disk | D: | No |
F: | No |
When collecting a cloning image, work must be stopped because the managed server is restarted.
The first partition must be a primary partition.
Cloning images can be collected with the following file systems. Note that LVM (Logical Volume Manager) partitions are not supported.
NTFS
EXT3
EXT4
LinuxSwap
The following conditions must be met for the managed server to collect the cloning image from and the managed server to deploy the cloning image to.
The model names are the same
The hardware configurations are the same, such as the optional cards and expansion cards, and their locations
The same BIOS settings have specified in accordance with "BIOS Settings of Managed Servers" in "4.1.2 Configuring the Server Environment" of the "ServerView Resource Orchestrator Setup Guide"
The LAN and SAN connections have the same redundancy methods and the same number of redundancy paths, and can access the same network devices and storage units
Note that LAN or fibre channel switches connected in a cascade configuration are viewed as a single device.
Some applications may require manual adjustments to function properly after cloning.
If necessary, manually perform such adjustments before or after the cloning process.
Up to four system image backup, restoration, and cloning image collection processes can be executed at the same time. If five or more processes are requested, the extra processes wait until the processes being executed are complete.
Restore operations executed by the backup/restore method during server switchover and failback also wait. When using auto-recovery and manual switchover operations with the backup/restore method, execute a maximum of three system image backup/restore or cloning image collection/deployment operations at the same time.
After collecting or deploying a cloning image, software required for connecting to external servers, etc. when the OS is started may not start correctly.
In this case, restart the OS after it is collected.
If the Watchdog (function that automatically resets or turns off the OS when a hang is detected because the OS does not respond for a certain period) function of the managed server is enabled, the OS may be automatically reset or turned off during cloning.
It is therefore highly recommended to disable the Watchdog function before a cloning operation.
For details, refer to the manual of the managed server.
When using MAK license authentication for the activation of Windows Server 2008, Sysprep can be executed a maximum of three times.
Since Sysprep is executed when deploying a cloning image, cloning image collection and deployment cannot be executed four or more times.
Therefore, it is recommended not to collect cloning images from managed servers that have had cloning images deployed, but to collect them from a dedicated master server.
As there is a chance that data will be damaged, do not perform collection or deployment of a cloning image while performing an iSCSI connection using a software initiator.
When using data disks, use the hardware initiator.
Create an L-Server for the Infrastructure Administrator
The L-Server for the infrastructure administrator is the L-Server to collect cloning images used for the entire system.
Create an L-Server for the infrastructure administrator using the L-Server template created in "8.5 Creating L-Server Templates".
In this case, perform the following configuration:
Select "None" for images.
Check the "Single path mode" checkbox for FC path.
For details on creating an L-Server for the infrastructure administrator using an L-Server template, refer to "10.1 Creation Using an L-Server Template" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
Manual OS Installation
Install the OS and applications required for deployment.
For manual OS installation, installation by remote console is recommended after starting the MB (Management Blade) or iRMC (Remote Management Controller) screen.
Additionally, apply any required patches and other necessary settings.
Make sure that the source server operates properly after those steps.
Check whether the DHCP client is enabled on the managed server to collect the cloning image from.
Cloning images with the same name can be saved up until the maximum number of image versions.
When collecting a new cloning image when the maximum number of image versions has already been reached, select the cloning image to delete.
The maximum number of image versions is three by default.
For details of how to change the maximum number of image versions, refer to "3.1.4 Changing the Maximum Number of Cloning Image Versions (Physical Servers)" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
When the target of operation is a PRIMEQUEST server, confirm that boot option settings on the target server and the boot option set in BIOS are set to Legacy boot.
If either setting is UEFI, change the settings to Legacy boot.
For details on changing the boot options, refer to "3.2.10 Changing Boot Options" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
[Windows]
Enable NetBIOS over TCP/IP
A volume license is required for cloning, and must be entered during the installation of Resource Orchestrator Agent.
For details, refer to "2.2.1.2 Collecting and Checking Required Information" and "2.2.2 Installation [Windows/Hyper-V]" in the "Installation Guide CE".
If no volume license is entered during the agent installation, or if this information is changed after installation, edit the following license configuration file on the source server. Note that the structure and location of this file depend on the version of Windows that is being used.
For Windows Server 2003
Installation_folder\Agent\scw\SeparateSetting\sysprep\sysprep.inf
Edit the following line to enter a valid product ID.
ProductID=Windows product key (*1) |
*1: 5-digit values separated by hyphens
Example
ProductID=11111-22222-33333-44444-55555 |
Note
If there is a mistake in the product ID value or format, an error occurs when the collected cloning image is deployed. Make sure to enter a valid product ID when editing the definition file.
For Windows Server 2008
Installation_folder\Agent\scw\SeparateSetting\ipadj\activation.dat
Specify the parameters in the format "parameter=value" in each line of the [ActivationInfo] section of the definition file.
Refer to the following table for details on each parameter.
Format | Parameter | Value |
|---|---|---|
KMS | .cmd.remotescript.1.params.kmscheck (Mandatory) | KMS host search type. Select one of the following:
|
.cmd.remotescript.1.params.kmsname | Host name (FQDN), computer name or IP address of the KMS host. | |
.cmd.remotescript.1.params.kmsport | KMS host port number. If omitted, "1688" will be set. | |
MAK | .cmd.remotescript.1.params.makkey (Mandatory) | MAK key. |
Common | .cmd.remotescript.1.params.ieproxy | Host name (FQDN) and the port number of the proxy server. The host name and port number are separated by a colon (":"). |
.cmd.remotescript.1.params.password | Administrator password. An existing password setting will be displayed as an encrypted character string. | |
encrypted | The encryption status of the Administrator password. |
Example
With KMS (Automatic Search)
[ActivationInfo] |
With KMS (Manual Settings)
[ActivationInfo] |
With MAK
[ActivationInfo] |
If the Administrator password has been changed, execute the following command. The password specified in the .cmd.remotescript.1.params.password parameter of the definition file is changed to the encrypted string, and the line "encrypted=yes" is added to indicate that the password is encrypted.
For details, refer to "5.4 rcxadm deployctl" in the "Command Reference".
>"Installation_folder\Agent\bin\rcxadm" deployctl passwd -encrypt <RETURN> |
With MAK (Already Encrypted Password)
[ActivationInfo] |
[Windows]
For physical L-Servers, specify the NIC number defined for the rack mount or tower servers decided in "4.2.1.1 Admin LAN Network Design" and "4.2.1.3 Physical Network Design for the Public LAN and iSCSI LAN" as the subscript.
Example
If the NIC number of the leftmost NIC on the back of a rack mount server is defined as "1", specify "Local area connection 1" here.
[Linux]
For physical L-Servers, specify the number with 1 subtracted from NIC number defined for the rack mount or tower servers decided in "4.2.1.1 Admin LAN Network Design" and "4.2.1.3 Physical Network Design for the Public LAN and iSCSI LAN" as the subscript.
Example
If the NIC number of the leftmost NIC on the back of a rack mount server is defined as "1", specify "eth0" here.
Information
The following are examples of the methods for aligning the subscript of a network interface name and NIC number of the NIC at back of the rack mount server.
For details, refer to the OS manual.
Red Hat Enterprise Linux
Configure the MAC address of the NIC at the back of the rack mount server for HWADDR in the following file: /etc/sysconfig/network-scripts/ifcfg-ethX
SLES
Use udev.
Collect a Cloning Image from the L-Server after the OS is Installed
Cloning images can be collected from the L-Server after an OS is installed.
In the orchestration tree on the [Resource] tab of the ROR console, right-click the L-Server created in "1. Create a physical L-Server for the infrastructure administrator", and select [Cloning] - [Collect] from the popup menu.
The [Collect a Cloning Image] dialog is displayed.
Configure the necessary items.
The cloning image is stored in the specified image pool.
For details on collecting cloning images, refer to "11.5.1 Collecting and Registering Cloning Images" in the "User's Guide for Infrastructure Administrators (Resource Management) CE".
