Installing in a cluster environment


This topic provides information about installing and configuring TrueSight Middleware and Transaction Monitor (TMTM) Services in a clustered environment.

There are many clustered environments from different vendors across the Windows and Unix/Linux platforms and the example given in this topic, focuses on a Microsoft hardware cluster environment, using an active/passive configuration. You can apply these guidelines for other hardware and software environments such as Linux RHEL. For these other environments you need to consult the respective software documentation for installation and configuration details.

TMTM services are not dependent on, or aware of any specific clustered environment, and as such, active/active environments are not supported. TMTM can be considered like any other application you run in a clustered environment and the same setup installation and configuration considerations should apply.    

Clusters provide the following benefits:

  • Availability/failover. When one application or node in a cluster fails, the failure can be detected and the application or node moves its processing to a secondary node with limited interruption to users.
  • Manageability. Clusters allow administrators to inspect the status of all cluster resources, and move workload around onto different servers within the cluster. This is useful for manual load balancing, and to perform "rolling updates" on the servers without taking important data and applications offline.

This topic describes how to install and configure the core components in a Microsoft hardware cluster environment, using an active/passive configuration. You can apply these guidelines during the installation of other hardware and software environments.Consult the respective software documentation for installation and configuration details.

An active/active environment is not supported. 

Related topics

Installing-the-core-components-of-the-product 

Working-with-the-subscriber-services

Before you begin

  • Configure your cluster hardware and software correctly. Ensure that your cluster is functioning properly before installing TrueSight Middleware and Transaction Monitor (TMTM) Services. For example, test that node failover is working correctly.
  • Decide if you are going to place your TMTM Services in the "Cluster Group", or if you are going to host them on a virtual server. The rest of this document assumes that you are placing the services in the default "Cluster Group". If this is not the case, ensure that the group containing the services has its own virtual IP address and shared disk resources.

Note

BMC recommends that the shared storage and virtual IP resources should be dedicated to the group where the TMTM Services are installed.

To install the prerequisite software

  1. Install database connectivity software on each node of the cluster, and ensure that each node can connect to your database.

    Note

    It is possible to host the database on the cluster as well, but that would require a configuration that is beyond the scope of this document. If you select this route, ensure that each node of your cluster is powerful enough to run the database and the Services, and amend the following instructions to ensure that the Services are dependent upon the database service.

  2. Execute the appropriate TMTM database creation and initialization scripts against your database. The scripts are available on the installation media or files that you downloaded from EPD.
  3. Create the cluster group or virtual server hosting the TMTM Services and ensure that it contains all the other (non-TMTM) resources, including shared storage and virtual IP resources.

Note

If you wish to have TMTM event actions that write messages to IBM WebSphere MQ queues and you are using a local queue manager (as opposed to a client connection to a remote queue manager), the local queue manager must be a resource in the same group as the TMTM Services. You should make the TMTM Event Service dependent upon the queue manager resource.

To install the TMTM software and configure the cluster group

  1. Log on to the active side of the cluster.
  2. Install the TMTM software as described in Installing, but select a destination directory that is on the shared disk resource in the group where you place the TMTM Services. If the installer prompts you to reboot after installation, answer no, and reboot at the end of these steps.
  3. In the Windows Service Control Manager, set all the TMTM Services to manual start (they are set for an automatic start by default).
  4. Using Microsoft Cluster Administrator, failover the group where you are installing the TMTM Services. The resources move to the other cluster node so that you can install the software there.
  5. Repeat 1 through 3 on the now-active node.
  6. In the Microsoft Cluster Administrator, create a Generic Service resource for each of the Services (TMTM Security, TMTM Topic, TMTM History, TMTM Event, TMTM Client Gateway, TMTM ProactiveNet, and TMTM Application).
    • The TMTM Application Service should depend on the group's virtual IP address and the shared disk resources.
    • The TMTM should depend on the TMTM Application Service.
    • The TMTM History Service should depend on the TMTM Topic Service.
    • The TMTM Event Service should depend on the TMTM Topic Service and the TMTM Application Service. If you are writing actions that send WebSphere MQ messages, and have a local queue manager, then the TMTM Event Service should also depend on the group's local queue manager resource.
    • The TMTM ProactiveNet Service should depend on the TMTM Topic Service.
    • The TMTM Client Gateway Service should depend on the TMTM Application Service.
  7. If the file ss_state.bin exists in the TMTM directory, delete it.
  8. If, in step 2, the installer prompted you to reboot one or more cluster nodes, reboot those nodes now.

To verify the cluster configuration

The following is a list of tests you should perform to verify the functionality of your cluster. This is a basic list only and you should consult with your cluster administrator for additional test scenarios as appropriate.

  • Simulate a node failover and ensure that the TMTM services come up and function correctly.
  • Simulate a group failover and ensure that the TMTM services come up and function correctly.
  • Simulate failover of a single TMTM service and ensure that the TMTM services come up and function correctly.

Where to go from here

To deploy agents to monitored systems, see Installing-the-Agent-and-monitoring-extensions.

To configure policies to deploy agents, see Getting-started-with-policies.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*