IncidentRelay 2.5.1 adds configurable directory synchronization, richer alert analytics and native GitLab deployment webhooks. It also strengthens API authorization and improves notification delivery, Docker builds and everyday navigation.
- Directory synchronization. Import teams, users and on-call schedules from external APIs using configurable endpoints, authentication headers and field mappings. Preview changes before applying them, schedule periodic updates and control how existing resources are updated, edited or removed. Users can be matched by email for sign-in through an existing OIDC configuration. Certificate verification can be disabled explicitly for individual sources using self-signed certificates.
- Alert analytics. Compare activity with the previous period, explore a day/hour heatmap and identify alert patterns with the largest increase in noise. Open the corresponding alert groups directly from analytics results. Translations have been completed across all supported languages.
- Native GitLab deployment webhooks. Send GitLab Deployment events directly to Change Intelligence without custom webhook payloads. Map projects and environments to services within the integration’s team. Duplicate and out-of-order events are handled consistently. Change Intelligence intake and automatic correlation remain opt-in.
- API automation. Manage Change integrations and system settings using appropriately scoped personal API tokens, making these endpoints easier to use from scripts and Terraform workflows.
- Persistent alert filters. Filter selections are preserved, and common service and involvement filters stay synchronized between Dashboard and Alerts.
- Performance improvements. Reduced repeated database work during alert intake, added batched trace persistence and improved PostgreSQL connection pooling. Independent alert groups can deliver notifications concurrently on PostgreSQL; SQLite retains sequential delivery.
- Security fixes. Centralized token authorization strengthens scope, Group and Team restrictions across API endpoints, alert intake, maintenance, orchestration and CalDAV. Child tokens cannot exceed their parent’s permissions. Alert deduplication and pending-event handling respect route boundaries. Outbound SAML metadata and browser-push requests have additional protection against server-side request forgery.
- Reliability fixes. Fixed voice ACK and resolve callbacks arriving during notification delivery. Database-backed encryption-key validation detects configuration changes before services start, including deployments with ephemeral runtime directories.
- Interface improvements. Teams, Silences and Escalation Policies display more information directly in their tables. Incident rows use status colors consistent with alerts. Team managers can create resources permitted by their roles. The Grafana dashboard supports datasource selection across its panels and dependent variables.
- Docker improvements. Multi-stage builds keep compilation tools out of the runtime image. Readiness checks validate database connectivity and migration state before dependent services start.