Configuring the Agent

Terminology

Depending on the version of the PATROL Agent, the terminology varies slightly:

• For PATROL Agent versions up to 11.3.02, you can specify Integration Services to integrate the agent with BMC TrueSight.
• For PATROL Agent versions from 20.02 and higher, you can specify Streaming Endpoints to integrate the agent with either BMC TrueSight or BMC Helix.

While the interface reflects the terminology applicable to the agent, the actual configuration variable is the same in both cases: /AgentSetup/integration/integrationServices.

tcp:<is1Hostname>:3183,tcp:<is2Hostname>:3183,...

https:<endpointHostname>:443

Agent Tags

Tags are used to specify which Infrastructure Policies set in TrueSight or Helix applies to the agent. Infrastructure Policies can be associated to specific Agent Tags.

You can assign Tags to a PATROL Agent to decide which Infrastructure Policies will be applied to it. For each Tag, you can add an optional description.

Events Routing

Several options are available to define how PATROL events are routed to TrueSight Event Management:

Advanced Settings

When you configure Monitoring Studio X from the Web interface, you may want to block any updates from CMA policies that could interfere or supersede your settings. To prevent changes from Infrastructure Policies on the PATROL Agent, simply set the Disable Configuration Updates from TrueSight CMA Policies to YES.

Additionally, you can fine-tune the integration with BMC TrueSight with the following configuration variables:

• Publish Hostname
• Integration Service Connection Timeout
• Retries
• Retries per Integration Services
• Wait Before Retries
• Text Parameter Maximum Length

SSL Certificate of the HTTP Server

Out of the box, Monitoring Studio X uses a self-signed SSL certificate to operate its HTTPS server. Your browser will warn you that the connection to this HTTPS server is unsecure, because it cannot verify the identity of the server.

To avoid giant security warnings in your browser when you connect to Monitoring Studio X Web Interface, you can configure the HTTP server to use a proper SSL certificate corresponding to its hostname and domain name with the below procedure:

1. Obtain a certificate from a Certificate Authority (external or in your organization) and the associated private key. The certificate file is typically a *.crt, *.cer or *.pem file.

2. Store the certificate and the associated private key in a PKCS#12 keystore file. This can be done with several tools like OpenSSL or keytool.

   Example with OpenSSL:

   $ openssl pkcs12 -export -inkey yourdomain.key -in yourdomain.crt -out mystore.p12 -name "My Certificate" -password p4ssw0rd
   

3. Save the *.p12 file on the host running the PATROL Agent, typically in the $PATROL_HOME directory.

4. Configure Monitoring Studio X HTTP Server to use this *.p12 file as the SSL Certificate PKCS#12 Keystore and specify the password to open, decrypt and verify the integrity of the keystore.

5. Restart the JVM. The next screen should show you that the HTTP server is responding. However, due to security measures in browsers, the Web application will probably not be able to properly report when the HTTP server starts responding with the new certificate. After a few seconds, simply refresh the page of the browser and make sure the certificate of the HTTP server is the one you configured.

Note about times

The Agent Information displays various information about time, notably the current time of the system where the PATROL Agent is running. As this system clock may not be synchronized with your own system (where your browser is running), this page highlights the clock skew between the agent and your browser.

For consistency, all times about events and data on the agent are adjusted to your local time, taking into account time zones and clock skew.

Example:

• Agent time is 9:15:00
• Browser time is 9:20:07
• There is a 307-second (5 minutes and 7 seconds) clock skew between the agent and the browser (agent is 307 seconds behind your browser).
• An alarm was triggered at 9:12:00 on the agent. At this time, it was already 9:17:07 on your browser.
• This event will be displayed in your browser as happening at 9:17:07 (2 minutes ago) (which is more logical than displaying 9:12:00 (2 minutes ago) while your clock says it's currently 9:20:07).

Persistent Thresholds

These alert thresholds are stored in the PATROL Agent's configuration under /___tuning___.

These thresholds definitions may apply to a parameter of a specific instance, or to a parameter of an entire class (i.e. all of its instances). In the latter case, the table shows the number of instances that the threshold definition apply to.

Thresholds can be edited by clicking on the corresponding thresholds definitions. Depending on the available information and type of thresholds, the editing dialog may display different information.

CMA

Thresholds displayed in the CMA tab all come from Infrastructure Policies defined in TrueSight CMA, and applied to this PATROL Agent.

These thresholds definitions cannot be modified or deleted. But because the table is a practical summary of all thresholds defined in all applicable policies for this agent, it allows you to quickly troubleshoot policy issues.

Current

The thresholds displayed in the Current tab summarizes all alert thresholds on all parameters for all instances of all classes monitored on this PATROL Agent. Only the parameters listed in this table will trigger alerts and events.

This table will differ from the Persistent table as some KMs programmatically set alerts thresholds using the set_alarm_ranges() PSL function. These supersede thresholds definitions in the /___tuning___ configuration tree, and even those defined through CMA policies.

You can edit these thresholds right from the table by clicking on the thresholds definitions.

If you enable the Persistent Thresholds in this dialog, your settings will also be saved in the /___tuning___ configuration tree. You may need to manually refresh the Persistent tab to get the new thresholds definitions.