Foglight for Infrastructure 5.9.3 - User Guide

Creating agent instances

After a successful agent deployment, you can proceed to create agent instances using the agent types included in the newly-deployed infrastructure agent package: UnixAgentPlus, UnixAgent, WindowsAgent, or MultiHostProcessMonitorAgent.

NOTE: UnixAgentPlus includes the Linux and Oracle Solaris agent types.

NOTE: UnixAgent includes the AIX, HP-UX, Linux, and Oracle Solaris agent types.

NOTE: MultiHostProcessMonitorAgent includes both the Windows and Linux platforms.

The recommended method for creating Foglight for Infrastructure agents for the hosts that you want to monitor is by using the Add Monitored Host wizard or the Add Monitored Host - List wizard. For detailed information about these wizards, see sections Adding a monitored host and Adding multiple monitored hosts.

For generic information about creating agent instances and editing properties using the Agent Status dashboard, see “Create agent instances on monitored hosts” in the Administration and Configuration Help. Edit the agent instance properties, as necessary for each monitored host. For generic information about editing agent properties, see “Edit instance-specific agent properties” in the Administration and Configuration Help.

For detailed information about Foglight for Infrastructure agents, see About the WindowsAgent, About the UnixAgentPlus, About the UnixAgent, and About the MultiHostProcessMonitorAgent.

Defining credentials

Credentials are security data that provide the Foglight for Infrastructure agents with the permission to monitor system resources, such as a host or a range of hosts.

NOTE: Credentials are needed for any remote agent. They are not needed when the Foglight Agent Manager is physically located on the host being monitored.

If you create agent instances using the Add Monitored Host wizard or the Add Monitored Host - List wizard, the credentials setup is part of the wizard’s flow. For details, see sections Adding a monitored host and Adding multiple monitored hosts.

If you create agent instances using the Agent Status dashboard, you need to set up the credentials required for your Foglight for Infrastructure Agent in order to start collecting data. The credentials provided for monitoring should be those of a local user account on the remote system. The local user account should also have a local home directory. If there is a failure in the remote storage or remote user services, a local account is accessible to the agent, allowing it to detect and report the problem. If a network account, or an account with a remotely mounted home directory is used, this may prevent the agent from accessing the remote system when a problem occurs. For detailed information about setting up credentials in Foglight, see the “Managing Credentials” topic in the Foglight Administration and Configuration Help.

Activating the agent

After you create the agent instance and define the proper credentials, you can enable the agent to start the data collection:

TIP: The activate operation fails if the credentials are not properly defined. Click Get Log to see the list of errors and warnings.

NOTE: You can also create/activate an agent when adding a monitored host. For more information, see Adding a monitored host.

Monitoring the infrastructure

After activating the agent, you can start monitoring your infrastructure environment. If your agent is configured correctly, the Hosts dashboard should now display the data being collected.

NOTE: The Add Host icon and the Monitored column are available only if Foglight for Infrastructure is installed.

The Monitored column shows whether the host is already configured for monitoring , or is not monitored yet (in which case, you can click the icon to add a monitored host).

Once sufficient data has been collected, the CPU (Utilization, Current), Memory Current, Disk (I/O, Usage), and Network Current columns and the Utilization sparklines display the resource utilization for each monitored host. Hover over each of these columns to see the resource utilization graph for a host.