Alerts
Missed schedule alerts, robot offline alerts, severity levels, and alert management
RPA Watch can automatically detect when scheduled processes don't run on time or when monitored robots go offline — and notify the right people via email. Alerts provide proactive monitoring so you can catch issues before they impact your operations.
Alert Types
| Type | Trigger | Description |
|---|---|---|
| Missed Schedule | A scheduled process didn't run at its expected time | Detected by comparing schedule cron expressions against actual job executions |
| Robot Offline | A monitored robot lost connectivity | Detected when a robot's session is no longer available in the RPA provider |
Enabling Alerts
Alerts are configured per account in Account Settings:
1. Navigate to Account Settings in the sidebar
2. Go to the Alerts tab
3. Toggle Enable to turn on alert monitoring
4. Configure the Grace Period and Cooldown Period
5. Click Save
| Setting | Range | Default | Description |
|---|---|---|---|
| Grace Period | 5–120 minutes | 15 minutes | How long to wait after the expected time before triggering an alert |
| Cooldown Period | 30–1440 minutes | 120 minutes | Minimum time between alerts for the same schedule |
Requires Account Admin or Tenant Admin role.
How Detection Works
Missed Schedule Detection
1. Every 15 minutes, RPA Watch evaluates all active schedules in accounts with alerts enabled
2. For each schedule, it calculates when the process should have run based on the cron expression and timezone
3. If no matching job is found within the grace period, a missed schedule alert is created
4. Schedules running more frequently than every 5 minutes are excluded to avoid noise
5. The cooldown period prevents repeated alerts for the same schedule
Robot Offline Detection
1. When [robot monitoring](/docs/robots) is enabled for a robot, RPA Watch tracks its connectivity
2. If a monitored robot's status changes to Disconnected, a robot offline alert is generated
3. The alert resolves automatically when the robot comes back online
## Alert Severity
Each alert has a severity level:
| Severity | Description |
|---|---|
| Low | Minor — the issue is unlikely to have immediate impact |
| Medium | Moderate — attention needed but not urgent |
| High | Important — should be addressed soon |
| Critical | Urgent — immediate action required |
Email Notifications
When an alert is triggered, email notifications are sent to:
- The process owner (set in the process metadata)
- All process members (users assigned to the process)
The alert record stores the list of notified email addresses for audit purposes.
Viewing Alerts
Navigate to Alerts in the sidebar to see all alerts. The alerts page shows:
| Column | Description |
|---|---|
| Account | Which account the alert belongs to (visible for Tenant Admins) |
| Source | The robot name (for offline alerts) or process name (for missed schedules) |
| Details | Schedule name or "Machine disconnected" |
| Expected At | When the process was expected to run |
| Severity | Color-coded severity badge |
| Status | Active or Resolved |
| Notified | Number of people notified (hover to see email addresses) |
| Alerted | How long ago the alert was triggered |
Filtering Alerts
- Status filter: View Active, Resolved, or All alerts
- Account filter: Tenant admins can filter by specific account
Resolving Alerts
Alerts can be resolved in two ways:
- Automatic: When the missed run eventually completes, the alert auto-resolves
- Manual: Click the Resolve button on any active alert to mark it as resolved
Resolved alerts are kept in history for audit and review purposes.
Visibility by Role
| Role | What You See |
|---|---|
| Tenant Admin | All alerts across all accounts (with account filter) |
| Consultant | Alerts for assigned processes only |
| Account Admin | All alerts in their account |
| Account User | Alerts for processes they own or are a member of |