Collect data from your system devices

  • Release version: Yokohama
  • Updated January 30, 2025
  • 2 minutes to read
    • Summarize
    Summarized using AI
    This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.

    Summary of Collect data from your system devices

    This guide explains how to collect various types of data—such as monitoring data, visibility data, log data, or user metrics—from your system devices using the Agent Client Collector in the Yokohama release (updated January 30, 2025). Data collection is performed by running check commands on your servers or databases.

    Show full answer Show less

    Policies and Checks

    A policy defines which Configuration Items (CIs) are monitored and which checks run on those CIs. Each policy contains:

    • The CIs monitored by the Agent Client Collector
    • The set of checks to run, including when and on which CIs

    Each check within a policy includes these attributes:

    • Command to execute
    • Check type
    • Command timeout
    • Plugins to use (securely downloaded tar.gz files)

    Scheduling Checks

    Checks run based on one of two scheduling models:

    • Interval-based scheduling: Runs checks after a minimum number of seconds between runs (e.g., once per hour). If the agent is down during a scheduled run, it executes the check within one hour after recovery. For longer intervals, the execution time adjusts randomly to maintain the defined cadence.
    • Cron-based scheduling: Runs checks at fixed times or intervals, such as daily at 6:00 PM or every two hours.

    Checks complete their run when results are sent to the MID Server (for MID deployments) or to ITOM Cloud Services Pod (for MID-less deployments).

    Determining Where Checks Run

    Policies monitor CIs mapped to agent hosts. You can create filters to specify which CI types to monitor. This allows targeted data collection on the appropriate devices.

    Updating Policy Mappings

    When a policy is activated, the Agent Client Collector Framework maps monitored CIs to that policy. This mapping refreshes every seven days after the initial mapping.

    Additionally, every 15 minutes (default), the Agent Client Collector checks the CMDB for changes matching filter conditions and updates mappings accordingly. You can adjust this frequency by modifying the snagentsyncfiltersintervalmin property in the System Properties.

    Run a check command on your server or database to gather data from those devices. Depending on the check that is invoked, collected data may be monitoring data, visibility data, log data, or user metrics.

    Policies

    A policy consists of the Configuration Items (CIs) monitored by the Agent Client Collector, and the checks that run on those CIs. Policies define which checks to run, as well as when and on which CIs to run them.

    Checks

    A policy runs a set of checks. Each check is defined with the following attributes:
    • Command to run
    • Check type
    • Command timeout
    • Plugins to be used (tar.gz files that are securely and automatically downloaded to the agent)

    Determining when checks run

    Policy checks run based on one of the following scheduling models:
    • Interval-based scheduling: The minimum number of seconds between each check run. For example:
      • An interval value is at least once per hour (at least once every 3,600 seconds).
      • The agent was down during the scheduled interval when the check was to run.
      When the agent is back up, it runs the check within one hour.

      If a check interval is less than one hour, the check is executed based on the indicated amount of time since the check's most recent run. For example, for a check scheduled to run once every eight hours:

      • The check runs at 00:00 (midnight)
      • The check runs again at 08:00 (8 AM)

      However, if the agent is down at 8:00 and is back up at 09:03, the check runs within an hour; that is, between 09:03 and 10:03. If the check runs at 09:57, the next run is at 17:57 (5:57 PM). In these cases, the check picks the running time at random to ensure running of once every eight hours.

      A check completes its run when the results are sent to the MID Server (in a MID deployment) or ITOM Cloud Services (ICS) Pod (in a MID-less deployment).

    • Cron-based scheduling: Runs at fixed times or intervals (for example, every day at 6:00 PM or every two hours).

    Determining where checks run

    Policies monitor the configuration items (CIs) that are mapped to the agent hosts. To determine the specific CI types to be monitored, you can create filters. For details on configuring policy filters, see Create a service filter for a policy.

    Updating policy mappings

    When activating a policy, Agent Client Collector Framework maps monitored CIs to a policy. This policy mapping repeats every seven days after the initial mapping. Every 15 minutes (by default), the Agent Client Collector monitors the CMDB for any changes matching the filter conditions, and updates the mapping accordingly. To modify the frequency by which mapping between the CMDB and CI is updated, modify the sn_agent_sync_filters_interval_min property on the System Properties page (All > System Properties > All Properties).