Skip to main content

Workload Alerts

Overview​

The Workload Alerts feature in DataDios enables you to configure conditions that proactively monitor workload metrics. When thresholds are crossed or anomalies are detected, notifications are triggered. This ensures timely awareness of performance issues and helps prevent failures before they escalate.

Alerts are tied directly to workload graphs, so you can create them based on the exact metrics you are monitoring in dashboards.

Prerequisites​

Before creating and using alerts, make sure:

  • Datasource is created
    Navigate to Data Source → Sources and create a datasource (PostgreSQL, SQL Server, etc.).

  • Workload extraction is enabled
    Go to Workload → Extract → Extract Workload and start the extraction. Without metric data, alerts cannot be evaluated.

note

Without metric data, alerts cannot be evaluated.

Creating Alerts​

Step 1: From a Graph​

  1. Open the Workload Dashboard.
  2. Select the relevant Datasource where your workload metrics are displayed.
  3. Open the graph’s context menu (⋮ three dots).
  4. Select “Create Alert”.

Step 2: Configure Alert Parameters​

In the alert creation form, configure:

  • Alert Name → A descriptive name.
  • Alert Metric → The metric from the graph (e.g., Load By Waits (Database)).
  • Condition → The logic (e.g., CPU > 2, or query_duration > 2000ms).
  • Evaluation Interval → How often the condition should be checked (e.g., every 1 or 5 minutes).
  • Notification Medium → Choose Email, Webhook, or UI Notifications.
  • Recipients → Provide one or more email addresses (comma-separated).

Example Alert​

{
  "Alert Name": "High CPU on Prod DB",
  "Metric": "CPU",
  "Condition": ">",
  "Threshold": 2,
  "Interval": 1,
  "Medium": "Email",
  "Recipients": "dba-team@datadios.com"
}

Click Save to activate the alert.

info

Once the alert is created, it will be available in the Scheduler → Alerts section.

Demo Video​

Watch the walkthrough video below for setting up and managing alerts:

Example Use Case​

Scenario: You want to be notified when your production database CPU usage goes above 80%.

  1. Open the DB Load graph in the workload dashboard.
  2. Click ⋮ and select Create Alert.
  3. Configure:
    • Alert Name: CPU Alert - Prod DB
    • Metric: CPU
    • Condition: >
    • Threshold: 0.8
    • Interval: 30 minute
    • Medium: Email
    • Recipients: ops@datadios.com
  4. Save and activate.

Now, whenever CPU exceeds 80%, the alert will trigger, log into history, and notify your ops team.

tip

Always configure alerts with a reasonable evaluation interval to avoid unnecessary notifications.

Best Practices​

  • Keep thresholds actionable (avoid too many false positives).
  • Use short intervals (1–5 min) for critical production metrics.
  • Use notifications + email for high-severity alerts.
  • Regularly review and update alerts as workloads evolve.

Troubleshooting​

  • Alert not triggering?

    • Verify workload extraction is active.
    • Check that the evaluation interval is reasonable.
    • Confirm recipients’ email addresses are valid.

  • Too many alerts firing?

    • Adjust thresholds to reduce sensitivity.
    • Use filters (like specific databases or metrics).