Alarm Mode

By default, alert rules in vuSmartMaps work as alarms. In alarm mode, the system continuously monitors the alert conditions. It triggers an active alert notification when the conditions become true and a clear notification when they become false.

For example, let's say you have an alert rule to monitor host CPU usage. When the CPU usage exceeds the set threshold, vuSmartMaps sends a notification. As long as the CPU usage remains above the threshold, no additional notifications are sent. Once the CPU usage falls below the threshold, vuSmartMaps sends a clear notification.

Figure: The points at which Alarm Active Notification and Clear Notifications are sent

The clear notification for alarms will include details about how long the alarm was active and provide metrics and information for the entire duration of the alarm.

Disabling Alarm Mode

You can configure an alert rule to work in non-alarm mode if you need general information notifications. When alarm mode is disabled, vuSmartMaps will generate notifications whenever the alert rule conditions are met. It will also continue to send alert notifications at regular intervals as long as the conditions are met.

However, in this mode, vuSmartMaps won't track the alert state, and it won't send clear notifications when the conditions are no longer met.

Figure: The points at which Notifications are sent in non-alarm mode

Non-alarm mode-based alerts are particularly useful for one-time event cases. For instance, when monitoring Syslog messages and you only want to be notified when a specific event like 'System rebooted' is detected, you can configure a non-alarm mode to check for this condition. In such cases, vuSmartMaps will send notifications each time the event occurs without tracking it as an ongoing alarm.

Changing from one mode to another

Switching between alarm mode and non-alarm mode is a flexible process for users. You can change an alert rule from alarm mode to non-alarm mode or vice versa by adjusting the configuration and saving it.

It's important to remember that when you switch from alarm mode to non-alarm mode, the system stops tracking the alarm state for the rule. This means that no clear notification will be sent for alarm conditions that were active at the time of the configuration change.

The same principle applies when an alert rule in alarm mode is disabled in the configuration. In this case as well, the system won't send an explicit clear notification.

Behind the Scenes - vuSmartMaps Alert Rule Engine

Internally, the vuSmartMaps NG Alert Rule Engine continuously evaluates all configured alert rules. During each evaluation cycle, it checks whether configured conditions evaluate to true and generates notifications based on the alert state and configured channels.

Frequency of Rule Evaluation

You can control how often vuSmartMaps evaluates alert rules.

• In the user interface, navigate to Alert Controls
• Select the desired rule evaluation frequency

note

The time window over which a metric is evaluated (for example, last 5 minutes) is independent of the rule execution frequency (for example, every 1 minute). These two settings serve different purposes and operate independently.

Alarm Update Notifications

For an active alarm, update notifications may be generated in the following cases:

• When the severity of an active alarm changes
• When the state of a metric or condition contributing to the alert changes

Alarm State and Channel Behavior

Alarm Activation

When an alert condition transitions from false to true:

• Notifications are sent through all configured channels
• A new alarm record is created in the datastore

Alarm Active State

While the alarm condition remains true:

• The alert continues to be updated internally at the configured evaluation frequency
• Alert Dashboards reflect:
  • Current metric values
  • Alarm duration
  • Updated condition evaluations

However:

• No repeated notifications are sent via Email, SMS, or WhatsApp
• Ticketing systems receive updates to already opened tickets (ticketing update support is available in newer releases)

Alarm Clear

When the alert condition transitions from true to false:

• Clear notifications are sent through all configured channels except the Report channel
• A clear event is recorded in the vuSmartMaps datastore

Channel Behavior Summary

Channel	New Alarm	Alarm Periodic Update	Alarm Change in severity and affected metrics	Alarm Clear
Data Store	✅	✅	✅	✅
Email	✅		✅	✅
Report	✅
ITSM	✅		✅	✅
Runbook	✅
WhatsApp	✅		✅	✅
SMS	✅		✅	✅

Non-Alarm Mode and Channels

When alarm mode is disabled for an alert:

• Notifications are generated at regular intervals (default: every 5 minutes)
• Notifications continue as long as the condition remains true
• No alarm state tracking or clear notifications are generated
• When the condition clears, notifications simply stop

Enabling Debug Logs for Alerts

The Alert module can generate detailed debug logs. Since this can be verbose in environments with many alert rules, vuSmartMaps allows selective debug logging per alert rule.

This can be configured in the auxiliary configuration file:

configs/alert_rule.yml

Example

# Add alert rule names for which debug logs should be enabled
alert-debug-list:
  - test123

Logs generated through this mechanism are recorded at the WARNING level.

Controlling Alert Email Content

By default, alert emails include the following sections:

• Additional fields (hashtags)
• Historical alarm data
• Contextual metrics table
• Information metrics table

If certain sections are not required, they can be enabled or disabled at the system level using the auxiliary configuration file:

configs/alert_rule.yml

Example Configuration

# Control additional field tags in email
email-tags: true

# Control historical data in emails
history-in-emails: true

# Control contextual metrics table
EmailAlerterContextualMetrics: false

# Control information metrics table
information-table: true

Managing Multiple Alert Nodes

In a containerized or Docker-based deployment, multiple Alert nodes can be run simultaneously.

• Alert rules are automatically distributed across nodes for load balancing
• Each alert rule is consistently assigned to the same node based on the rule name
• This ensures stable execution and predictable state handling

Users can also manually assign specific alert rules to a particular Alert node if required.