Skip to main content

Notifications & Alerts

Stay on top of your data pipeline without watching it. Get instant alerts when runs fail, tests break, or schemas change unexpectedly. Delivered where your team already works.

#Slack
Email
🔔PagerDuty
{ }Webhook
Slack#data-alerts: Run failed on dim_customers
PagerDutyIncident triggered: prod pipeline stalled
EmailDaily digest: 14/14 jobs passed
WebhookPOST /hooks/pipeline-status 200 OK

Never Miss a Failure

Data pipelines run around the clock. You should not have to. dbdeux monitors every run and proactively notifies you when something needs attention:

  • Run failures with error details and affected models
  • Test failures with specific test results and failure reasons
  • Schema changes when unexpected modifications are detected
  • Long-running jobs when a run exceeds expected duration
  • Pipeline health daily/weekly digest of overall pipeline status

Delivery Channels

Notifications reach you through the channels your team already uses:

ChannelHow it works
SlackMessages posted to channels or DMs. Connect via OAuth in Settings
EmailRich HTML emails with run summaries. Enabled by default for all users
PagerDutyIncident alerts via Events API v2. Configure a routing key per scheduled job
WebhooksJSON payloads to any URL. Integrate with Opsgenie or custom systems
In-appNotification center in the dbdeux UI with full history

Slack Integration

The Slack integration sends rich, actionable messages:

  • Run status with model counts and duration
  • Direct links to failed models in the IDE
  • Thread replies for related events (e.g., retry success after initial failure)
  • Channel-based routing so the right team sees the right alerts

Webhook Payloads

For custom integrations, webhooks deliver structured JSON payloads containing:

  • Event type (run_completed, test_failed, schema_changed)
  • Full run metadata (duration, model count, user who triggered)
  • Failure details with error messages
  • Links back to the dbdeux UI for investigation

Alert Configuration

Fine-tune exactly what you get notified about:

Per-Event Rules

Choose which events trigger notifications:

  • Run completed (success, failure, or both)
  • Test failed (all failures, or only for specific models/tags)
  • Schema change detected (any change, or only removals/type changes)
  • Git activity (PR merged, branch updated)

Routing Rules

Send different events to different channels:

  • Critical production failures → Slack #data-alerts + PagerDuty webhook
  • Test failures in staging → Email to the model owner
  • Schema changes → Slack #data-schema-changes
  • Daily pipeline digest → Email to the analytics team

Quiet Hours

Set schedules when non-critical notifications are held and delivered as a digest later. Critical alerts (production failures) always come through immediately.

Organization Defaults

Set notification defaults at the organization level so new projects inherit a consistent baseline:

  • Configure default channels (Slack, email, webhook) for the entire org
  • New projects automatically inherit org defaults
  • Projects can override defaults with their own rules when needed
  • A banner in the project wizard shows when inherited defaults are active

Per-Project Overrides

Each project can have its own notification rules. A critical production project might alert the whole team on any failure, while a sandbox project only notifies the individual developer.

Notification History

The in-app notification center keeps a full searchable history:

  • Filter by event type, project, or date range
  • See which notifications were acknowledged vs. unread
  • Link directly to the run, model, or schema diff from any notification

PagerDuty Integration

For on-call alerting, dbdeux integrates directly with PagerDuty:

  • Configure a PagerDuty routing key per scheduled job
  • When a job fails, an incident is triggered via the PagerDuty Events API v2
  • The incident payload includes run metadata, failure details, and a link back to dbdeux
  • PagerDuty then routes the alert through your team's escalation policy

Set the routing key in the Notify step of the schedule wizard, or edit it on an existing job's detail page.