3.3.1 Before installation and setup

(a) Cluster system

Make sure that the following conditions are satisfied:

• The cluster system is being controlled by cluster software.

• The cluster software is set to control startup and termination of any PFM - RM for Virtual Machine instances that run on a logical host.

  Important note

  • In Windows Server 2003, if the application error message box is displayed in Dr. Watson, it may prevent a failover, and therefore you must suppress error notification that uses the message box. For details about how to suppress error notification, see the OS documentation. Note that if you suppress error notification, the collection of information when an application error occurs may be affected.

  • When an application error occurs, a dialog box for reporting the error to Microsoft is displayed. If this dialog box is displayed, it may prevent a failover, and therefore you must suppress error notification. For details about how to suppress error notification, see the OS documentation.

(b) Shared disk

Make sure that the following conditions are satisfied:

• Each logical host must have a shared disk, and it must be possible to switch from the executing node to a standby node.

• The shared disk must be physically connected to each node via Fibre Channel or SCSI. Performance Management does not support a configuration in which a network drive or a disk replicated via a network is used as a shared disk.

• Even if a problem causes a process that is using the shared disk to remain when a failover occurs, it must be possible to forcibly take the shared disk offline and enable a failover using a means such as cluster software.

• If multiple PFM products are used on a single logical host, the folder name of the shared disks must be the same. You can change the storage destination of the Store database and store it in a different folder on the shared disk.

(c) Logical host names and logical IP addresses

Make sure that the following conditions are satisfied:

• Each logical host must have a logical host name, and each logical host name must have a corresponding logical IP address, and it must be possible to switch from the executing node to a standby node.

• The logical host names and logical IP addresses must be set in the hosts file and name server.

• If DNS is used, a host name excluding the domain name, not the FQDN name, must be used as the logical host name.

• Physical host names and logical host names must be unique within the system.

  Important note

  • Do not specify a physical host name (host name displayed by the hostname command) for the logical host name. Otherwise, communication may not be processed correctly.

  • A logical host name can use from 1 through 32 bytes of alphanumeric characters. Spaces and the following symbols cannot be specified:

    / \ : ; * ? ' " < > | & = , .

  • You cannot specify a logical host name that begins with the character string localhost, an IP address, or a hyphen (-).

(d) Settings for using IPv6

Performance Management supports network configurations in IPv4 and IPv6 environments. Performance Management is able to operate with a network configuration in which IPv4 and IPv6 environments coexist.

PFM - RM for Virtual Machine can communicate with PFM - Manager via IPv6.

Note that for Performance Management to operate with that network configuration, the OS of the host on which PFM - RM for Virtual Machine is installed must be Windows Server 2008 R2 or Windows Server 2012. In addition, the OS of the host on which PFM - Manager is installed must be Windows Server 2008 R2, Windows Server 2012 or Linux.

For information about the applicable scope of communications in IPv4 and IPv6 environments, see K. Communications in IPv4 and IPv6 Environments.

To make it possible to communicate via IPv6, the setting for using IPv6 must be enabled on both the PFM - Manager host and the PFM - RM host. This setting can be enabled by using the jpcconf ipv6 enable command. However, there are cases in which execution of this command is necessary and others where it is unnecessary.

Cases in which the jpcconf ipv6 enable command needs to be executed:

• When switching from an IPv4 environment to an IPv6 environment on each host

• When switching the PFM - Manager environment from IPv4 to IPv6 in an environment in which both IPv4 and IPv6 are used

Cases in which the jpcconf ipv6 enable command does not need to be executed:

• When each host already operates in an IPv6 environment

• When PFM - Manager operates in an IPv6 environment in an environment in which both IPv4 and IPv6 are used

The following shows an example of executing the jpcconf ipv6 enable command:

jpcconf ipv6 enable

Make sure that you execute the jpcconf ipv6 enable command on both the executing node and the standby node.

For details about the jpcconf ipv6 enable command, see the chapter that describes commands in the manual Job Management Partner 1/Performance Management Reference. For details about the conditions or occasions for executing the jpcconf ipv6 enable command, see the chapter that describes network configuration examples in an environment that includes IPv6 in the Job Management Partner 1/Performance Management Planning and Configuration Guide.

When you use IPv6 for communication between a monitored host and PFM - RM for Virtual Machine, specify the name of a monitored host where name resolution can be performed.

Communication between PFM - RM for Virtual Machine and a monitoring target is performed with an IP address that can be resolved. Also, if an IPv4 environment and an IPv6 environment are both used, and communication between PFM - RM for Virtual Machine and the monitoring target fails with an IP address that can be resolved, the communication is not retried by using another IP address.

For example, if communication fails when IPv4 is used, IPv6 is not used to retry communication. Similarly, if communication fails when IPv6 is used, IPv4 is not used to retry communication. Make sure that a connection can be established beforehand.

(e) WMI connection (when the virtual environment uses Hyper-V)

Confirm that the following condition is satisfied:

• The same user account that can connect to the monitored host via WMI is created on both the executing node and the standby node.

For details about the WMI connection settings, see 2.5.2(1) Setting up WMI connection.

(f) SSH connection (when the virtual environment uses KVM)

Confirm that the following conditions are satisfied:

• A private key is stored at the same path on both the executing node and the standby node.

• The private key can be used to connect to the monitored host.

• PuTTY is installed at the same path on both the executing node and the standby node.

  Important note

  Use either of the following methods to register the private key and public key:

  • After creating a private key on the active server and copying the private key to the standby server, pair the private key with the public key that is distributed from the active server to the monitored host.

  • After creating public keys separately on the active and standby servers, register both of these public keys on the monitored host so that either of the public keys can be paired with the private key.

For details about the SSH connection settings, see 2.5.3 For KVM.