Alarms

In the Fabric Services System, alarms arise when a managed object enters an undesirable state; for example, if a managed node goes out of service, an associated alarm is raised.

The Fabric Services System supports the following alarm types, each with its own subtypes:
  • Communication
  • Configuration
  • Environment
  • Equipment
  • Operational
The Fabric Services System includes the following tools that can help you manage alarms:
  • the Alarms panel on the Dashboard, which summarizes current alarms; for more information about this panel see The dashboard
  • the Alarm List, which you can use to view and manage individual alarms
  • the Policy manager, which you can use to customize the severity level for specific types of alarm, or to suppress alarms of a specific type entirely.

For a list of individual alarms currently supported by the Fabric Services System, see Appendix: Supported alarms.

Alarm states

In the Fabric Services System, an alarm can adopt the following states:

  • Acknowledged: An acknowledged alarm still displays in the Alarm List; but when viewing details for the individual alarm its state displays as Acknowledged and any note you added to the alarm while acknowledging it is displayed as well. You can use the Acknowledge state as the basis for filtering or sorting the alarm list.
  • Closed: A Closed alarm still displays in the Alarm List, but this state can be the basis for filtering the alarms included in the list. Closing an alarm does not resolve the condition that caused the alarm to be raised in the first place.
  • Cleared: An alarm is Cleared when the condition that raised the alarm has been resolved. Unlike Acknowledged and Closed, the Cleared state cannot be assigned manually by a Fabric Services System operator. Only the device or devices that raised the original alarm can determine and communicate its closure.

Alarm list

The Alarm List displays a list of current alarms known to the Fabric Services System. From this page you can view details about each alarm's state, and also acknowledge any alarms.

To view and manage alarms with the Alarm List:

  1. From the main menu, select Alarm List.
    The alarm list displays, showing all active alarms (where "active" refers to alarms that have not been cleared).
    Note: Cleared alarms are not included in this list because the "Cleared" filter is set to "False" by default. To view cleared alarms in this list, clear that filter. For more information about managing lists and filters, see Lists.
    Note: A set of default columns display in the alarms list:
    • Severity
    • Alarm type
    • Node name
    • Resource name
    • Cleared
    • Occurrence
    • Last Raised

    There are other columns available to show more information about each alarm. For information about adding or removing columns from any list, see Lists.

  2. To view details about an alarm and its state:
    1. Select an alarm in the list.
    2. At the right edge of the row, click and select State Details from the displayed action list.
    3. Click the ALARM STATE tab to view details about the alarm's severity, a description of the alarm, and the time it was raised.
    4. Click the OPERATOR STATE tab to view the state assigned by the operator to address the alarm (either Acknowledged or Closed).
    5. When you are finished, click CLOSE to return to the Alarm List page.
  3. To acknowledge an alarm:
    Acknowledging an alarm marks it as received, but does not clear the alarm from the Alarm List.
    1. Select an alarm in the list.
    2. At the right edge of the row, click and select Acknowledge from the displayed action list.
    3. Optionally, enter any comments about the acknowledgement in the Additional Info field.
    4. Click SAVE.
    The Alarm is marked as Acknowledged (but not Closed).
  4. To Close an alarm:
    Closing an alarm prevents the alarm from appearing in the Alarm List, but does not resolve the condition that raised the alarm in the first place.
    1. Select an alarm in the list.
    2. At the right edge of the row, click and select Close from the displayed action list.
    3. Optionally, enter any comments about the acknowledgement in the Additional Info field.
      Note: The text you enter here is displayed in the Additional Info column of the OPERATOR STATE tab in the Alarm Details overlay.
    4. Click SAVE.

Policies

Policies allow you to customize the severity level associated with individual supported alarms.

A policy affects the alarm types of all future alarms raised; it does not retroactively modify any existng alarms of the same type.

Each policy can include a start time and an end time; these are boundaries on the time of day during which the policy applies. An alarm raised outside these boundaries has its default severity instead of the severity level defined by the policy. If no start and end time are defined, the policy is always active.

You can also use a policy to suppress an alarm entirely while the policy is in effect.

The Fabric Services System supports the definition of a policy's scope in two mutually exclusive ways:
  • by key value, which allows you to trigger a policy based on the name of the object (node, fabric, intent, or region) affected by the alarm
  • by alarm category and type, to apply the policy regardless of the object affected.

To customize an alarm's severity:

  1. From the main menu, select Policies.
  2. Click + CREATE A POLICY.
  3. Enter a Name and Description for the policy.
  4. In the Policy Definition panel, enter a Start Time and End Time for the policy.
    An alarm that would be affected by this policy uses the customized severity level only if it is raised during this period. If it is raised outside this period, it uses the default severity level.
  5. Do one of the following:
    • To configure a policy based on the object it affects, click the Key Value toggle to enable it and go to step 6.
    • To configure a policy based on alarm category and type, leave the Key Value toggle disabled and go to step 7.
  6. With the Key Value toggle enabled, do the following:
    1. Click on the Key Value Objects drop-down and select one or more of the displayed Key Value candidates:
      • Node Name
      • Fabric Name
      • Intent Name
      • Region Name
      For each selected item, a field displays.
    2. Use the displayed field or fields to provide the unique name of each object type you selected.
      This name identifies the unique object of that type for which the system changes the alarm severity or suppresses alarms, depending on how you configure the remainder of the policy.
    3. Go to step 8.
  7. In the Policy Definition panel, do the following:
    1. Select an Alarm Category from among the following:
      • Communication
      • Equipment
      • Operational
      • FSS
    2. Click the Alarm Type drop-down and check the box beside one or more alarms types in the displayed list.
    3. Select an Alarm Severity:
      • Major
      • Minor
      • Critical
      • Warning
      • Default
  8. Configure the way this policy modifies alarms:
    1. Select a Priority from 0 to 9, with 0 being the highest priority.
    2. Optionally enable Suppress Alarms toggle.
      Enabling this option means that alarms of this type are disabled and are not triggered while the policy is in effect.
    3. Optionally enable the Deployed Intent Alarms Only toggle.
      This option applies to "Communication – Interface Down" alarms. If this option is enabled, alarms are raised only on interfaces that are part of the intent configuration.
  9. Click CREATE.