Ace Observability Platform

A unified observability platform for metrics, logs, and traces. Built with Vue.js, Go, and the VictoriaMetrics ecosystem. Enterprise auth (RBAC, SSO, audit logging) included free.

Features

• Multi-datasource support — VictoriaMetrics, Prometheus, Loki, VictoriaLogs, Tempo, VictoriaTraces, ClickHouse, Elasticsearch, CloudWatch
• Visual query builder — Builder/Code mode toggle with metric search, label filters, aggregation, and live query preview (MetricsQL/PromQL compatible)
• AI-powered query assist — inline AI suggestions with tool-calling for metric discovery. Configurable AI providers or GitHub Copilot integration
• Export to dashboard — save any Explore query directly as a dashboard panel
• Grafana dashboard import — connect to Grafana, browse dashboards, and import with fidelity report
• AI chat-to-dashboard — describe what you want in natural language, get a complete dashboard
• 16+ panel types — line, stat, gauge, bar gauge, bar chart, pie, table, heatmap, histogram, scatter, candlestick, state timeline, status history, flame graph, node graph, canvas, logs, trace list, trace heatmap
• Drag-and-drop layout — 12-column grid with resize handles
• Live log streaming — real-time log tailing for Loki and VictoriaLogs
• Log-to-trace correlation — click a trace ID in logs to jump to the trace view
• Enterprise auth — RBAC (Admin/Editor/Viewer/Auditor), Google/Microsoft/Okta SSO, group-to-role mapping, audit logging
• Dark/light mode with the "Kinetic" design system (burnished brass palette, Space Grotesk typography)

Tech Stack

• Frontend: Vue 3 (Composition API + TypeScript), Vite, Tailwind CSS v4, ECharts
• Backend: Go 1.25, net/http
• Database: PostgreSQL 16 (metadata), Valkey (cache)
• Supported datasources: VictoriaMetrics, Prometheus, Loki, VictoriaLogs, Tempo, VictoriaTraces, ClickHouse, Elasticsearch, CloudWatch

Quick Start

Option A: Tilt + Kubernetes (recommended for development)

Requires: Go 1.25+, Docker, a local k8s cluster (Colima, kind, or Docker Desktop), kubectl, helm, tilt

# Start local k8s cluster (Colima example)
colima start --kubernetes --cpu 4 --memory 8

# Start the full stack
tilt up -- victoria-metrics victoria-logs victoria-traces tempo loki prometheus \
  otel-collector telemetrygen elasticsearch clickhouse alertmanager vmalert grafana

# Seed admin user and datasources with k8s service URLs
make seed-tilt

# Seed demo dashboards
cd backend && go run ./cmd/seed-dashboards -org victoria

# Open the app
open http://localhost:5173

All datasource services are optional. Enable only what you need:

# Victoria stack only
tilt up -- victoria-metrics victoria-logs victoria-traces otel-collector telemetrygen

# LGTM stack only
tilt up -- prometheus loki tempo otel-collector telemetrygen

# Everything
tilt up -- victoria-metrics victoria-logs victoria-traces prometheus loki tempo \
  otel-collector telemetrygen elasticsearch clickhouse alertmanager vmalert grafana

The otel-collector config is generated dynamically from enabled services. Telemetrygen produces realistic web app telemetry (HTTP requests, database queries, distributed traces) via OpenTelemetry.

Option B: Docker Compose

# Start infrastructure (pick a datasource profile)
make compose-up PROFILES=victoria

# Start telemetry generators
make telemetrygen PROFILES=victoria

# Start backend (hot reload with air)
make backend

# Start frontend (Vite dev server)
make frontend

# Seed admin user and datasources
make seed

Available profiles: victoria, lgtm, elk, clickhouse. Combine with commas: PROFILES=victoria,lgtm.

Default Credentials

Field	Value
Email	admin@admin.com
Password	Admin1234

Local Ports

Service	Port
Frontend	http://localhost:5173
Backend API	http://localhost:8080
VictoriaMetrics	http://localhost:8428
VictoriaLogs	http://localhost:9428
VictoriaTraces	http://localhost:10428
Prometheus	http://localhost:9090
Loki	http://localhost:3100
Tempo	http://localhost:3200
Elasticsearch	http://localhost:9200
ClickHouse	http://localhost:8123
AlertManager	http://localhost:9093
VMAlert	http://localhost:8880
Grafana	http://localhost:3000
Tilt Dashboard	http://localhost:10350

Seeded Organizations

The seed command creates four organizations, each with datasources for its stack:

Organization	Datasources
Victoria	VictoriaMetrics, VictoriaLogs, VictoriaTraces, VMAlert, AlertManager
LGTM	Prometheus, Loki, Tempo
Elastic	Elasticsearch
ClickHouse	ClickHouse

Demo Dashboards

Six pre-built dashboards are available for the Victoria organization:

1. HTTP Service Overview — request rate, error rate, latency percentiles (P50/P95/P99), status code breakdown
2. Application Performance — avg request duration, latency trends, resource usage
3. Database Health — connection pool stats, query duration, PostgreSQL vs Redis breakdown
4. Infrastructure Overview — service uptime, CPU utilization, scrape health, memory trends
5. Log Intelligence — service-filtered log views (API gateway, orders, payments), error analysis, slow query detection
6. Trace Explorer — latency heatmap, recent distributed traces

Seed them with:

cd backend && go run ./cmd/seed-dashboards -org victoria

Container Images

Published to GHCR on every release (multi-arch: amd64 + arm64):

docker pull ghcr.io/aceobservability/ace-backend:latest
docker pull ghcr.io/aceobservability/ace-frontend:latest

Tags: vX.Y.Z, X.Y.Z, X.Y, latest, sha-<commit>

Development

Running Tests

# Frontend (Vitest + happy-dom)
cd frontend && bun run test

# Backend (Go testing)
cd backend && go test ./...

Linting

make lint          # all linters
make backend-lint  # golangci-lint
make frontend-lint # Biome

Building

# Frontend
cd frontend && bun run build

# Backend
cd backend && go build ./cmd/api

Project Structure

ace/
├── frontend/               # Vue 3 application
│   ├── src/
│   │   ├── components/     # UI components (QueryBuilder, LogViewer, panels/)
│   │   ├── composables/    # Shared state (useDatasource, useOrganization, useAIProvider, useToast)
│   │   ├── views/          # Page views (MetricsExploreTab, LogsExploreTab, TracesExploreTab)
│   │   └── utils/          # Chart theme, dashboard helpers
│   └── package.json
├── backend/                # Go API
│   ├── cmd/
│   │   ├── api/            # Main API server
│   │   ├── seed/           # Admin user + org + datasource seeder
│   │   ├── seed-dashboards/# Demo dashboard seeder
│   │   └── seed-correlated/# Telemetry generator (realistic web app traces/logs/metrics)
│   └── internal/
│       ├── handlers/       # HTTP handlers (datasource proxy, AI chat, dashboards, auth, SSO)
│       ├── datasource/     # Datasource clients (Prometheus, VictoriaMetrics, Loki, Tempo, ClickHouse, ES)
│       ├── models/         # Data models
│       ├── db/             # Database connection, migrations
│       ├── auth/           # JWT, password hashing, middleware
│       ├── authz/          # RBAC authorization
│       └── audit/          # Append-only audit logging
├── deploy/
│   ├── charts/ace/              # Production Helm chart
│   ├── charts/ace-local-infra/  # Local dev Helm chart (all datasource services)
│   └── docker/                  # Docker Compose configs per stack
├── Tiltfile                # Local dev orchestration
├── Makefile                # Common commands
├── DESIGN.md               # Kinetic v2 design system
└── RELEASE.md              # Release process

Versioning and Releases

• Semantic Versioning (vMAJOR.MINOR.PATCH) with Conventional Commits
• release-please opens release PRs from changes on main
• Merge creates a GitHub Release with notes + CHANGELOG update
• Auto-published: backend binaries, frontend tarball, container images, SBOMs, checksums
• See RELEASE.md for maintainer workflow