6.1.2 SLM system configuration in a cluster system
To run SLM in a cluster system, the system configuration must satisfy the following conditions:
The logical hosts require a shared disk and logical IP addresses that can be inherited from the active server to the standby server. The shared disk and logical IP addresses must satisfy the conditions described in 6.1.1 Prerequisites for cluster system operations.
The same OS must be used throughout the cluster system. Failover between different OSs is not supported.
The cluster system must have an active-standby configuration.
This following subsections explain the system configuration for running SLM - Manager and SLM - UR in a cluster system.
- Organization of this subsection
(1) System configuration for SLM - Manager
This subsection explains a system configuration example for SLM - Manager in a cluster system. For details about the system configuration for JP1/Base in a cluster system, see the SLM/Base User's Guide. If you run JP1/Base with a cluster configuration, you must specify the logical host name of JP1/Base in the jbsHostName property in SLM - Manager's system definition file (jp1itslm.properties).
|
![[Figure]](GRAPHICS/F0000236.GIF)
SLM - Manager places the information that is to be inherited from the active server to the standby server in the event of a failover in shared folders on the shared disk. The files to be placed in the shared folders are created when an environment is set up for SLM - Manager on the logical host. If there are no shared folders when an environment is set up for SLM - Manager on the logical host, the setup command creates them.
A file system area for the database is placed on the shared disk. The name that is used for the folders is shown below.
- Folder name for the database file system area:
shared-folder#\JP1SLM\database
- #
A different shared folder must be specified at setup for each logical host.
Only the SLM system information and performance data for monitored services are stored in the database file system area on the shared disk. The other file system areas used by SLM are created on local disks of the individual logical hosts.
(2) System configuration for SLM - UR
The following figure shows an example of a system configuration for SLM - UR in a cluster system.
|
![[Figure]](GRAPHICS/F0000231.GIF)
Note that SLM - UR uses the shared disk so that any access logs will be switched over to the standby server in the event of a failover. To determine whether a shared disk is required in the system configuration, check the cluster software specifications, as well as the destination path where the files than make up the access logs are to be recorded.
When SLM - UR is run in a cluster system, a network tap that copies HTTP packets passing through the switch's mirror ports is required between SLM - UR and the switch. The network tap copies the HTTP packets that pass the switch and sends them to the individual SLM - URs that make up the cluster system.