Skip to content

Alerting support to the Debezium Platform #2314

Description

@mfvitale

Which use case/requirement will be addressed by the proposed feature?

The platform exposes pipeline monitoring through 14 predefined panels (DDD-38), but monitoring is passive: users must manually watch dashboards to detect problems.

Alerting closes this gap by letting users define threshold-based rules against existing monitoring panels and receive notifications when conditions are breached.

Key capabilities:

  • Alert rules: users define rules by selecting a monitoring panel, a comparison operator, a numeric threshold, and an optional "for" duration to suppress transient
    spikes
  • Global evaluation: rules fire for any pipeline that breaches the threshold, the alert event identifies which pipeline triggered it
  • Alert lifecycle: three-state model (Pending, Firing, Resolved) with incident tracking (one record per fire/resolve cycle)
  • Severity levels: Critical, Warning, Info
  • Notification channels: Email (SMTP) and Webhook (HTTP POST), plus always-on in-app visibility (history page and alert badge)
  • Alert history: queryable, filterable, paginated incident history with configurable retention (default 30 days)
  • UI: new Alerts section with Rules, Channels, and History sub-pages, plus a global alert badge in the header

Out of scope for initial implementation: alert grouping, silencing, escalation policies, repeat notifications, raw PromQL mode, per-pipeline rule scoping, real-time push
(SSE/WebSocket).


Implementation ideas (optional)

DDD-57: Alerting for Debezium Platform

Metadata

Metadata

Assignees

Type

No type

Projects

Status
In review

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions