Guidelines for load balancers

Use a load balancer to ensure seamless failover when TrueSight Operations Management is configured for high availability. The following topic provides information on how to prepare and configure a load balancer.

Terminologies

  • Node — A node is a logical object on the load balancer System that identifies the IP address of a physical resource on the network. You can explicitly create a node, or you can instruct Local Traffic Manager to automatically create one when you add a pool member to a load balancing pool.
  • Load balancing pool — A load balancing pool is a logical set of devices, such as web servers, that you group together to receive and process traffic. Instead of sending client traffic to the destination IP address specified in the client request, Local Traffic Manager sends the request to any or available servers that are members of that pool.
  • Health monitors — The purpose of health monitors is check the health, performance, and load. If a monitored device, link, or service does not respond within a specified timeout period, or the status indicates that the service is down or performance is degraded or that the load is excessive, the load balancer system redirects the traffic to another resource. The load balancer identifies the active node using health monitors.
  • Virtual server — A virtual server is a traffic management object on the load balancer system that is represented by an IP address and a service. Clients on an external network can send application traffic to a virtual server, which then directs the traffic according to your configuration instructions. The main purpose of a virtual server is often to balance traffic load across a pool of servers on an internal network.

High-availability of TrueSight Operations Management components

  • In a Presentation Server HA pair, only one node will be active and the other node will be in standby .
  • In an Infrastructure Management HA pair, only one node will be active and the other node will be in standby at any point in time.
  • In an Integration Service cluster, both the nodes can be active at any point in time.
  • In a Cell HA pair, both the nodes can be active at any point in time.

Load balancer placement

The following illustration identifies the recommended placement of load balancers with respect to the TrueSight Operations Management components.


Use case 1: Load balancer for the TrueSight Presentation Server

See #1 in the preceding illustration.

To configure the load balancer for Presentation Server, the health monitors should be as follows:

Step 1: Create health monitor based on HTTP Queries

TrueSight version

Protocol

Health check URL

Healthy State

Comments

11.x

HTTP

HTTP://<tsps server node name>/LogoutUser.jsp 

HTTP/1.1 200

The health check monitor should be created against both the Presentation Server nodes.

10.x

HTTP

HTTP://<tsps server node name>

HTTP/1.1 200

The health check monitor should be created against both the Presentation Server nodes.

Step 2: Add ‘Virtual server of TSPS’ as referrer host to the Presentation Server

Run the following commands from the Presentation Sever command prompt (installationDirectory\truesightpserver\bin) to add or append the 'Virtual server of TSPS' as referrer host name:

  • tssh properties list
    Check tspsProxyHosts if any host(s) are listed.
  • Add 'Virtual server of TSPS' to the existing value:
    • tssh properties set tspsProxyHosts <Virtual server of TSPS (FQDN format preferred)>
    • tssh properties reload

Use case 2: Load balancer for the TrueSight Infrastructure Management server

See #2 in the preceding illustration.

To configure the load balancer for Infrastructure Management, the health monitors should be as follows:

Step 1: Create health monitor based on HTTP Queries

TrueSight version

Protocol

Health check URL

Healthy State

Comments

11.x

HTTP

HTTP://<tsim server node name>/jsp/ShowError.jsp

HTTP/1.1 200

The health check monitor should be created against both the Infrastructure Management nodes.

10.x

HTTP

HTTP://<tsim server node name>/jsp/login.jsp

HTTP/1.1 200

The health check monitor should be created against both the Infrastructure Management nodes.

Step 2: Add ‘ Virtual server of TSIM’ referrer host to the Infrastructure Management Server

  • Open the pronet.conf file located at the installationDirectory\pw\custom\conf directory.
    • Add or append the 'Virtual server of TSIM (FQDN format preferred)' to the following pronet.conf property:
      • pronet.tsim.proxy.hosts=<existing host(s), if any>,<Virtual server of TSIM (FQDN format preferred)>
      • Save the file changes.
      • Run the following command to reload the properties:
      • pw jproperties reload

Use case 3: Load balancer for the TrueSight Integration server and PATROL Agents

See #3 in the preceding illustration.

To configure the load balancer for Integration Service and PATROL Agents, the health monitors should be as follows:

Step 1: Create health monitor based on TCP port queries

PATROL Agent Version

Protocol

Check

Healthy State

Comments

9.x - 11.x

TCP

Port check Integration server at 3183

Port Connected

Assumption that the Integration Service is running on port 3183.

Step 2: Add ‘ Virtual server of TSIS’ referrer host to the PATROL Agents

In the PATROL Agent, modify the following setting:

/AgentSetup/integration/integrationServices" = {REPLACE="<Virtual server of TSIS (FQDN format preferred)>:3183"},)

Step 3: Set the algorithm for load balancer

The load balancer should follow the algorithm of round-robin with least connections.

Use case 4: Load balancer for the Cell

See #4 in the preceding illustration.

To configure the load balancer for the Cell, the health monitors should be as follows:

Step 1: Create health monitor based on TCP port queries

Protocol

Check

Healthy State

Comments

TCP

Port check at 1828

Port Connected

Assumption that the Cell is running on port 1828.

Step 2: Add ‘ Virtual server of Cell’ referrer host to the event sources

In the event sources, change the gateway to the 'Virtual server of Cell'.

Step 3: Set the protocol for load balancer

The load balancer should route the traffic to the active Cell.

Was this page helpful? Yes No Submitting... Thank you

Comments

  1. Cassie Derrick

    Application Visibility use case would be very helpful here.

    Sep 30, 2019 06:02
    1. Harihara Subramanian

      Hello Cassie Derrick,

      Working with the SMEs to get some use case info.

      May 07, 2020 08:54
      1. Harihara Subramanian

        As per my following discussions with the SMEs closing this.

        Hi Hari,

        In our TrueSight Application Management 11.3 document, we have a Planning web page where it have a couple of links for High Availability App Visibility and EUEM.  Below is the link to the Planning web page:

         

        https://docs.bmc.com/docs/applicationmanagement/113/planning-772581442.html

         

        In this Planning web page, there is a link that discuss different high availability planning for the various TSOM components.  See that link below:

         

        https://docs.bmc.com/docs/TSOperations/113/high-availability-deployment-for-truesight-operations-management-843619585.html

         

        and then below is a link for high availability planning for EUEM-Software Edition:

         

        https://docs.bmc.com/docs/TSOperations/113/high-availability-recommendations-for-real-end-user-experience-monitoring-software-edition-843619794.html

         

        and below is a link for high availability planning for App Visibility Manager:

         

        https://docs.bmc.com/docs/TSOperations/113/app-visibility-manager-high-availability-deployment-843619753.html

         

        App Visibility does not have a web page similar to the web page in your email, but the last link I provided above is close to it.  Also in the last link is a link to another web page where we provide a sample configuration of a F5 load balancer.  You can probably give the customer all of the above links or just the last link.

        May 26, 2020 02:41
  2. Kumar Vanka

    Is there any information that can be included regarding the load balancer 'Persistence' setting? Specifically, the ISN load balancer's 'Persistence Type' setting to handle Patrol Agents that restart quite often.

    Jun 23, 2020 10:15
    1. Sanjay Prahlad

      Hello Kumar,

      This page contains information on what has been tested in the BMC labs. If additional tests are performed, we will update this page.

      Nov 05, 2020 10:48
  3. Nikhil Shetty

    /AgentSetup/integration/integrationServices" = {REPLACE="<Virtual server of TSIS (FQDN format preferred)>:3183"}, -

    Do we need to add tcp: before Virtual server of TSIS ? tcp:Virtual server of TSIS :3183?

    Oct 14, 2020 06:52
  4. Javier Hernandez

    Hi Nikhil Shetty


    Yes, correct, you need to set tcp: before the virtual server of TSIS.


    Thanks

    Javier

    Oct 14, 2020 07:11
    1. Nikhil Shetty

      Also the Load balancer algorithm for Integration service says to use Round robin with least connection. But in F5 we can configure either Least connection or Round robin, not both . Can you please update on this

      Oct 14, 2020 04:15
      1. Sanjay Prahlad

        Hello Nikhil, our internal testing has been with round-robin.

        Nov 05, 2020 10:47