ivadragonfly/statuspage

Statuspage

Statuspage is a self-hosted status page with automatic service health checking, incident management, and multi-channel notifications.

• Backend: Go 1.24 · Chi router · GORM · SQLite (WAL mode)
• Frontend: PHP 8.3 (server-side rendered, no JS framework)
• Auth: JWT + Cookie sessions + OIDC (Keycloak)
• TLS: Let's Encrypt (ACME HTTP-01, auto-renewal)
• API: REST (OpenAPI 3) + GraphQL + Prometheus metrics
• Notifications: Email, Slack, ***, ***, Twilio, Mattermost, LINE, Webhook, Command

---

Quick Start

bash
cp .env.example .env
# edit .env — at minimum set HTTP_PORT
docker compose up -d

Open http://localhost/setup.php to run the first-time setup wizard.

With HTTPS (Let's Encrypt)

Port 80 must be publicly reachable for the ACME HTTP-01 challenge.

bash
cp .env.example .env
# set HTTP_PORT=80, HTTPS_PORT=443 in .env
docker compose up -d

Then in Admin → Settings → Let's Encrypt: enter your domain and email, save, and click Obtain / Renew Certificate. See docs/lets-encrypt.md for details.

---

Project Structure

statuspage/
├── backend/
│   ├── cmd/server/main.go          # Entry point
│   └── internal/
│       ├── api/
│       │   ├── router.go           # Chi routes
│       │   ├── middleware/auth.go  # JWT / session / RBAC
│       │   └── handlers/           # One file per resource
│       ├── certmanager/            # ACME HTTP-01 cert management
│       ├── checker/                # HTTP/TCP/UDP/ICMP/gRPC/SMTP/IMAP
│       ├── config/                 # Config loading (env / .env / yml)
│       ├── db/                     # SQLite init, AutoMigrate, helpers
│       ├── graphql/                # GraphQL schema + handler
│       ├── importer/               # services.yml bulk import
│       ├── metrics/                # Prometheus metrics
│       ├── models/                 # GORM models
│       ├── notifier/               # 9 notification channels
│       └── scheduler/              # Per-service timer scheduler
├── frontend/
│   ├── src/
│   │   ├── pages/admin/            # Admin dashboard pages
│   │   ├── pages/                  # Public status page, login, setup
│   │   ├── includes/               # header, footer, nav
│   │   ├── lang/                   # en.php, de.php
│   │   └── Lib/                    # ApiClient.php, I18n.php
│   └── public/                     # Web root (assets, index.php)
├── docker/
│   ├── nginx.conf                  # HTTP server block (port 80)
│   ├── supervisord.conf
│   └── entrypoint.sh
├── tests/backend/                  # Go tests
├── Dockerfile
├── docker-compose.yml
├── .env.example
├── config.yml.example
├── services.yml.example
├── openapi.json                    # Swagger/OpenAPI 3 spec
└── schema.graphql

---

Configuration

Priority order: environment variables > .env file > config.yml

See .env.example and config.yml.example for the full list.

---

Services YAML Import

Place a services.yml next to the ***ary or mount it at /app/services.yml in Docker. Services are imported automatically on startup; SHA256-based deduplication skips existing entries.

bash
cp services.yml.example services.yml
# edit, then restart the container
docker compose restart

See services.yml.example for all supported service types and fields.

---

API

Full reference: docs/api.md · Interactive UI: http://localhost/swagger

---

Prometheus Metrics

statuspage_service_up{service,type,group}        1 = online, 0 = offline
statuspage_service_latency_ms{service,type}      last check latency (ms)
statuspage_service_uptime_percent{service}       24-hour uptime %
statuspage_service_checks_total{service,result}  cumulative check counter
statuspage_open_incidents_total                  number of open incidents

Secure with API_SECRET as a Bearer token: Authorization: Bearer <secret>

---

Notifiers

Configure via Admin → Notifiers. See docs/notifiers.md for per-channel field mappings.

---

Scheduler

The scheduler runs a time.Timer per service. On each tick:

1. Acquires a semaphore slot (max SCHEDULER_CONCURRENCY parallel checks)
2. Runs the appropriate checker for the service type
3. Stores a Check record; updates online, last_latency, last_check
4. On first failure: creates a Failure record and marks the service offline
5. On recovery: resolves the open Failure and marks the service online
6. Sends notifications after notify_after consecutive failures
7. Reschedules for the next interval

---

Data Retention

Check records older than remove_after days (default 90) are purged automatically every hour.

---

Let's Encrypt

See docs/lets-encrypt.md.

---

Development

See docs/development.md.

---

License

MIT