Configure elastic connectors for MPN alarm collection

  • Release version: Australia
  • Updated June 16, 2026
  • 2 minutes to read
  • Configure a connector instance to collect fault management alarm data from a Mobile Private Network (MPN) Elastic index and forward events to Event Management.

    Before you begin

    Role required: tsom_assurance_admin

    About this task

    The MPN connector collects alarm and event data from managed network infrastructure by querying Elastic fault management indices at a configured polling interval. Each connector instance collects from one or more of these indices, each corresponding to a specific network domain. The MPN pull connector definition supports three connector instances: one for alarm collection and two for metrics collection. This procedure covers configuring the alarm collection instance.

    You can create multiple connector instances, one per index, and assign each to a dedicated MID Server to distribute the polling workload across your deployment. A default index value is provided for each connector instance.

    The alarm collection instance also clears stale alarms: alarms that were cleared at the source but were missed during a normal collection cycle and would otherwise remain open in Event Management. At a regular interval, the connector rechecks the source for active alarms that have not been updated within a set time window and clears any that are no longer active. Default values are provided, so no action is required unless you want to tune this behavior.

    Procedure

    1. Navigate to All > Event Management > Integrations > Connector Definitions.
    2. Select Nokia MPN Pull Connector.
    3. In the Default schedule field, enter the polling time (in seconds) to be used for all connector instances.
      The default value is 120 seconds.
    4. In the Connector Instances related list, select Nokia MPN Alarms.
    5. On the Connector Instances page, in the Event collection schedule field, enter how long each event collection should run, in seconds.
      This value overrides the default schedule set on the connector definition. The MPN Alarms instance defaults to 60 seconds.
    6. In the Index field, provide the index connector instance values.

      Enter the index pattern for the network domain this connector instance collects from. To collect from more than one domain, enter the patterns as a comma-separated list. The following table lists the standard index patterns:

      Index connector instance values configuration screen.
      Table 1. Index connector instance values
      Description Name
      Radio network elements radio-fm*
      Core network elements core-fm*
      Application-layer network elements application-fm*
      IXR (Interconnect Router) devices ixr-fm*
      DAC (Digital Access Controller) devices dac-fm*
    7. Add the MID Server.
      1. In the MID Servers for Connectors section, select the plus icon next to Insert a new row.
      2. Enter the name of the MID Server in the MID Server field.
      3. Select the green check mark to save your selection.
    8. Enable the connector instance by selecting the Active check box.
    9. Optional: Tune how stale alarms are detected and cleared.
      Set the following connector instance values on the alarm collection instance. If you don't set them, the default values apply.
      Table 2. Stale alarm detection parameters
      Parameter Default Description
      stale_threshold_minutes 60 How long, in minutes, an active alarm can go without an update at the source before the connector treats it as stale and clears it in Event Management.
      stale_check_interval_minutes 30 How often, in minutes, the connector runs the stale alarm check. This check runs independently of the event collection schedule.
    10. Select Update.
    11. Optional: View your configured alarms by navigating to em_event.LIST and searching for API Alarm Notification.