Model your architecture
Define services, map dependencies, link monitors. When something goes down, see exactly what's affected and who owns it.
Blast radius at a glance
When a service goes down, the catalog shows every downstream service that depends on it. Instantly.
Service-level monitoring
Link monitors to services so 'API Gateway is degraded' reflects real telemetry. Not a stale wiki page.
Clear ownership
Every service has an owner. When an incident fires, you know who to ping. No guessing, no stale on-call docs.
Define services with structure
Each service is a first-class entity with metadata that matters. Not just a name in a spreadsheet. Tier, owner, status, and a description that responders actually read.
- Service tier (Tier 1, Tier 2, Tier 3) for prioritizing incident response
- Owner field. Single point of accountability per service
- Status that reflects health (operational, degraded, partial outage, major outage, maintenance)
- Free-text description for context. What the service does, who it serves
- Available on Pro and Business plans
API Gateway
OperationalMap dependencies
Services don't exist in isolation. Link upstream and downstream dependencies so when something fails, the graph tells you what else is at risk.
- Define upstream and downstream dependency edges between services
- Visualize the dependency graph from any service detail page
- Cycle detection. Signalog warns if you accidentally create a circular dependency
- Filter the graph by tier so Tier 1 dependencies stand out
Dependency graph · Auth Service
Upstream (depends on)
Downstream (depended on by)
Wire monitors to services
A service's status should reflect what's actually happening. Link any monitor to any service so 'API Gateway' shows green when its health checks are passing. And red when they're not.
- Many-to-many. One service can have multiple monitors, one monitor multiple services
- Service status auto-updates from linked monitor state
- Status page components can mirror service status for public visibility
- Per-monitor 'service impact'. See the full set of affected services
API Gateway
OperationalLinked monitors · 3
Link incidents to services
When an incident is declared, attach the affected services. The catalog now knows what's broken and the incident page shows which services were impacted. Fueling postmortems and trend analysis.
- Attach one or more services to any incident with a multi-select
- Service detail page shows incident history per service
- Postmortems pull affected service names into the impact section
- Retrospective analytics break down incidents by service
Payment processing degradation
Affected services · 3
Postmortem · impact section
This incident affected Stripe Integration, Billing Worker, and Subscription Renewal for 12 minutes. Estimated 14 customers experienced delayed charges.