Analyzing triggered alerts

View and analyze triggered alerts to understand when observed conditions are met, investigate issues, and update alert status.

Roles and permissions

For information about required roles and permissions for viewing alerts, see Roles and permissions for alerts.

Before you start viewing and analyzing alerts:

Alerts must be defined and active (see Creating alerts).
Make sure that you understand alert types and trigger conditions (see Data Observability alerts).

To view triggered alerts:

On the Data Observability home page, select the Triggered alerts tab.
Locate the alert by name, or click the Filter icon to narrow down the results.

For alerts based on issue detection, open the filter window, select Alert type > Job run state, and click Apply.
For other alerts, you can use the alert type that is based on the alert condition or any other filter.
Click an alert to open its details.

From the alert details, you can:

For failed DataStage job runs, the system also provides:

In the Triggered alerts tab, you can also inform other users about the status of the triggered alert:

Mark the alert as Acknowledged to indicate that someone is working on the issue.
When the problem is resolved, mark the alert as Resolved.

Follow this workflow when you analyze triggered alerts:

Identify alert type: Understand what condition triggered the alert.
Review alert details: Check job name, run time, and trigger conditions.
Investigate job run: Click job run link to view logs and metrics.
Determine root cause:
- For failed jobs: Use AI troubleshooting or review error logs.
- For threshold alerts: Compare actual versus expected values.
- For state alerts: Verify job completion status.
Take corrective action:
- Fix DataStage flow issues.
- Adjust job parameters.
- Update alert thresholds if needed.
Update alert status: Mark as Acknowledged (investigating) or Resolved (fixed).
Monitor recurrence: Check whether the same alert triggers again.