Grafana
Grafana is the visualization and alert management layer of BizFirst Observe. It connects to Loki, Prometheus, and Tempo simultaneously — giving engineers a single interface for logs, metrics, and traces without switching tools.
Grafana's Role in the Stack
Grafana is the only component in BizFirst Observe that is user-facing. It serves three primary functions:
Dashboards
Pre-built and custom dashboards combining all three signal types. Variables make dashboards multi-tenant and multi-environment.
Explore
Ad-hoc query interface. Split view lets you query Loki and Tempo side-by-side. Essential for incident debugging.
Alerting
Define alert rules against any data source. Contact points: Slack, email, PagerDuty, webhook. Notification policies control routing and deduplication.
Pre-Built BizFirstGO Dashboard Variables
All pre-built dashboards use template variables that filter data to the selected tenant and environment:
# Dashboard variable: $tenant
# Type: Query — populated from Loki labels
label_values({job="processengine"}, tenant_id)
# Dashboard variable: $environment
# Type: Custom
production, staging, development
# Dashboard variable: $node_type
# Type: Query — from Prometheus
label_values(bizfirst_node_execution_duration_seconds_count, node_type)Pre-Built Dashboard Summary
| Dashboard | Key Panels | Primary Audience |
|---|---|---|
| Flow Studio Overview | Execution rate, error rate, p50/p95/p99 latency | Operations, engineering leads |
| Node Performance | Per-node-type heatmap, error rates by node type | Platform engineers |
| HIL Analytics | Backlog gauge, suspension duration histogram, approval rates | Process owners, operations |
| EdgeStream Throughput | Messages/sec per topic, delivery latency p99 | Integration engineers |
| Error Analysis | Error log volume, top error message patterns | Developers debugging issues |
For the complete Grafana reference — data source configuration, building custom dashboards, alert rule creation, and access control — see Guide6: Grafana.