---
title: "NMS Policy Settings"
slug: "nms-policy-settings"
description: "Manage NMS Alert Policies in the Kentik portal with step-by-step guidance on settings, types, and notification channels for effective monitoring."
updated: 2026-07-17T00:46:41Z
published: 2026-07-17T00:46:41Z
canonical: "kb.kentik.com/nms-policy-settings"
stale: true
---

> ## Documentation Index
> Fetch the complete documentation index at: https://kb.kentik.com/llms.txt
> Use this file to discover all available pages before exploring further.

# NMS Policy Settings

The settings for NMS policy types in Kentik's **Alerting** system are detailed here.

> [!NOTE]
> **Notes**:
> 
> - For information about Flow policy type settings, see [**Flow Policy Settings**](/v1/docs/flow-policy-settings).
> - For background on Kentik’s Alerting system, first read [**Alerting**](/v1/docs/alerting-4), [**Alerting Page**](/v1/docs/alerting-page), [**Notifications**](/v1/docs/notification-channel), and [**Mitigations**](/v1/docs/mitigations-page).

Including the **General** tab (see [**Flow Policy Settings**](/v1/docs/flow-policy-settings#general-settings)), for [**NMS policy types**](/v1/docs/nms-policy-settings#nms-policy-types) the following alert policy settings tabs are available:

- [**Dataset**](/v1/docs/nms-policy-settings#dataset-settings-nms)
- [**Activate & Clear**](/v1/docs/nms-policy-settings#activate-clear-settings-nms)
- [**Baseline**](/v1/docs/nms-policy-settings#baseline-settings-nms)
- [**Dashboards & Labels**](/v1/docs/nms-policy-settings#dashboards-labels-settings-nms) (Classic Threshold policy type only).

## Dataset Settings (NMS)

The **Dataset** settings tab of an NMS alert policy is where you define the scope of NMS telemetry data that the policy will evaluate.

- **Measurements**: Click the field to select one or more NMS metrics from the dropdown menu (e.g., `/kentik/agent/cpu`, `/kentik/agent/memory`). Selected metrics appear as lozenges; click the **X** on any lozenge to remove it.
- **Devices**: Specifies which infrastructure components or agents this policy will monitor.
  - **All Devices**: Monitors all available infrastructure by default.
  - **Edit Devices**: Opens the [**Data Sources**](/v1/docs/nms-policy-settings#data-sources-dialog) dialog to filter by specific devices, Sites, or Labels.
- **Dimension Filters**: Narrow the scope of monitored data based on specific dimension criteria (see [**Filters Settings**](/v1/docs/filtering-settings)). Click **Add Dimension Filters** to open the filter configuration workflow (see [**Group By Dimensions**](/v1/docs/dimension-settings#choosing-query-dimensions)).

### Data Sources Dialog

The Data Sources dialog allows you to add devices to the alert policy, including:

- **Search**: Enter terms to filter the Device Selector list.
- **Cancel**: Click the **X** at the upper right, or **Cancel** at the lower right, to exit the dialog without saving.
- **Device Selector:** Left pane listing methods of selecting devices:
  - **All Devices** (default): Include all devices.
  - **Labels**: Add devices by selecting one or more labels (see [**Labels**](/v1/docs/labels)).
  - **Sites**: Add devices by selecting one or more sites (see [**Sites**](/v1/docs/sites)).
- **View Selections**: Right pane summarizing device selections:
  - Lozenges for each selected label/site
  - Click **X** to remove the selection
  - Click **>** on a site lozenge to show assigned devices
- **Save**: Click **Save** to confirm the device selections and exit.

## Activate & Clear Settings (NMS)

The **Activate & Clear** settings tab of an NMS alert policy is where you configure the technical rules and thresholds that must evaluate to true to trigger an alert.

### Severity

- **Severity Dropdown**: Assign a severity level (Minor, Warning, Major, Severe, Critical) to the alert state generated by this policy.

### Alert Condition Groups

Build conditional logic rules using nested groups and operators.

- **Logical Operators**: Choose all or any from the dropdown to determine how multiple condition groups or individual conditions are evaluated together.
- **Condition Criteria**: For each rule line, specify:
  - **Measurement**: The base NMS metric path (e.g., `/kentik/agent/cpu`).
  - **Metric**: The specific type of evaluation (e.g., `Utilization` or `Times`).
  - **Operator**: Choose the relational operator (e.g., `equals`, `greater than`).
  - **Value**: Enter the threshold value (the UI will indicate expected formats, such as percent as float or float).
- **Add Actions**: Click **+ Add Condition** to add a row inside the current group, or **+ Add Nested Condition Group** to create a deeper logical subset.

### Operational Behaviors

- **Acknowledgement Required**: Turn **On** to require an operator to manually acknowledge the alert before it can progress or clear (see [**Alerts List**](/v1/docs/alerting-page#alerts-list)).
- **Manual Clear Required**: Turn **On** to require an operator to explicitly close out the alert once metrics return to normal (see [**Alerts List**](/v1/docs/alerting-page#alerts-list)).
- **Notification Channels**: Select or search for routing destinations (e.g., Slack, Email) where alerts should be sent. Click **Add New Channel** to build an integration on the fly (see [**About Notification Channels**](/v1/docs/notification-channel#about-notification-channels)).
- **Test Notification Channels**: Click to send test notifications to all assigned channels. Active only when notification channels are selected.
- **Activation Delay**: Specify the number of minutes to wait before officially firing an active alert. If the condition clears during this window, no alert is sent.
- **Clearance Delay**: Specify the number of minutes to wait before officially clearing a resolved alert. If the condition triggers again during this window, the alert remains active.

## Baseline Settings (NMS)

The **Baseline** settings tab of an NMS alert policy is where you configure the [**historical data buckets**](/v1/docs/nms-policy-settings#baseline-presets) and [**aggregation methods**](/v1/docs/nms-policy-settings#building-using-the-baseline) used to establish normal (i.e. baseline) traffic patterns for comparison.

> [!NOTE]
> **Note**: NMS policies use the same underlying baseline configuration engine as [**Flow policies**](/v1/docs/flow-policy-settings).

### Baseline Presets

Select one of the cards at the top of the tab to instantly apply a baseline configuration:

- **DEFAULT**: Collects 25 keys an hour, going back 4 weeks. Recommended for most general uses.
- **PRECISION**: Collects 300 keys an hour, going back 4 weeks. Recommended for highly accurate sampling.
- **EXPRESS**: Collects 25 keys an hour, going back 1 week. Recommended for a quick setup.
- **CUSTOM**: Automatically selected if you manually alter any of the preset values below.

### About Your Baseline

The **About Your Baseline** pane is a dynamically generated text block confirming the metric being measured (derived from your selected paths in the **Dataset** tab) and the time interval for data points (e.g., 60 Seconds).

### Building & Using the Baseline

- **Window**: How far back the history goes (e.g., 4 weeks from 1 day ago) to determine the bucket count.
- **Bucket depth**: The number of top keys collected in every hour of the window.
- **Rollup aggregation**: The mathematical method (e.g., 98th Percentile) used to handle the hourly data points.
- **Compare to**: Determines which historical buckets are evaluated against current traffic (e.g., the current hour of every day of the week).
- **Bucket width**: Derives the value from just that hour or by aggregating surrounding hours.
- **Final aggregation**: The math method used to combine values into a single comparison metric.
- **Use separate patterns for weekdays and weekends**: A checkbox to calculate UTC weekends independently from weekdays to account for regular business-cycle variations.

## Dashboards & Labels Settings (NMS)

The **Dashboards & Labels** settings tab of an NMS alert policy allows you to organize your policy and associate it with relevant visualization tools within the Kentik platform.

> [!NOTE]
> **Note**: There are no dashboard settings available for NMS Policy Types. See [**Dashboards & Labels Settings**](/v1/docs/flow-policy-settings#dashboards-labels-settings) for information about using labels.

## NMS Policy Types

The following table describes the policy types for NMS alert policies, listing the related metrics for each:

| Policy Type | Description | Metrics |
| --- | --- | --- |
| **Agent** | Agent health (resource usage, status) | - `/kentik/agent/cpu` - `/kentik/agent/disk` - `/kentik/agent/health` - `/kentik/agent/memory` - `/kentik/agent/status` - `/kentik/agent/status/alert` |
| **Agent Capability** | Capability health (enablement, status) | - `/kentik/agent/capability/health` - `/kentik/agent/capability/status` |
| **BGP Neighbor** | BGP peering session states and metrics (session state changes, advertised prefixes) | - `/protocols/bgp/neighbors` - `/protocols/bgp/neighbors/prefixes` |
| **Component** | Device component states and metrics (CPU, power supply) | - `/components` - `/components/cpu/utilization` - `/components/fan` - `/components/memory` - `/components/power-supply` - `/components/temperature` |
| **Device** | System-level device states and metrics (up/down status, CPU/memory) | - `/devices/apc/ups/battery` - `/devices/apc/ups/input` - `/devices/apc/ups/output` - `/devices/bluecat/dns` - `/devices/bluecat/replication/node` - `/devices/bluecat/replication/peer` - `/devices/cisco/cellular` - `/devices/cisco/cfw/hardware` - `/devices/cisco/cfw/statistics/buffer` - `/devices/cisco/cfw/statistics/connection` - `/devices/digi/modem` - `/lldp/global` - `/mpls/entity/status` - `/mpls/neighbor` - `/protocols/bgp/global` - `/system` - `/system/cpus/summary` |
| **Event** | SNMP traps or syslog messages via regex filters | - Syslog - SNMP Trap > [!NOTE] > **Note**: Click **Edit Dimension** to add dimensions other than the defaults. |
| **Interface** | Interface states and counter metrics (oper status, traffic spikes) | - `/interfaces/counters` - `/lldp/interfaces` - `/lldp/interfaces/neighbors` - `/protocols/isis/interfaces` |
| **Classic Threshold** | Rolling window aggregations and classic baselines (average over time thresholds) | All |
| **Custom** | Any NMS measurement including cloud and custom metrics from integrations | All |