217 lines
6.7 KiB
Markdown
217 lines
6.7 KiB
Markdown
# 🛡️ Backup Monitor
|
||
|
||
A self-hosted backup monitoring dashboard with MongoDB backend, designed for **Borgmatic** (but works with any tool that can send HTTP requests).
|
||
|
||
   
|
||
|
||
## Features
|
||
|
||
- **Dashboard** – Real-time overview of all backup hosts with status cards
|
||
- **Host Management** – Add, edit, disable, delete hosts via Web UI (no config files)
|
||
- **History** – 90-day retention with per-day calendar heatmap and size charts
|
||
- **Detailed Stats** – Duration, original/deduplicated/compressed size, file counts
|
||
- **Uptime Kuma Integration** – Automatic push per host after each backup
|
||
- **Stale Detection** – Configurable threshold (default: 26h) marks missed backups
|
||
- **Auto-Refresh** – Dashboard updates every 30 seconds
|
||
- **Dark Theme** – Clean, modern UI with status-colored indicators
|
||
- **Zero Config** – Hosts auto-register on first push, or add manually via UI
|
||
|
||
## Quick Start
|
||
|
||
```bash
|
||
# Clone
|
||
git clone https://github.com/feldjaeger/backup-monitor.git
|
||
cd backup-monitor
|
||
|
||
# Start
|
||
docker compose up -d
|
||
|
||
# Open
|
||
open http://localhost:9999
|
||
```
|
||
|
||
## Docker Compose
|
||
|
||
```yaml
|
||
services:
|
||
backup-monitor:
|
||
build: .
|
||
container_name: backup-monitor
|
||
restart: always
|
||
ports:
|
||
- "9999:9999"
|
||
environment:
|
||
- MONGO_URI=mongodb://mongo:27017
|
||
- STALE_HOURS=26 # Hours before a host is marked "stale"
|
||
depends_on:
|
||
- mongo
|
||
|
||
mongo:
|
||
image: mongo:4.4 # Use 7+ if your CPU supports AVX
|
||
container_name: backup-mongo
|
||
restart: always
|
||
volumes:
|
||
- mongo_data:/data/db
|
||
|
||
volumes:
|
||
mongo_data:
|
||
```
|
||
|
||
## Push API
|
||
|
||
After each backup, send a POST request:
|
||
|
||
```bash
|
||
# Minimal push (just hostname + status)
|
||
curl -X POST "http://localhost:9999/api/push?host=myserver&status=ok"
|
||
|
||
# Full push with stats (JSON)
|
||
curl -X POST -H "Content-Type: application/json" \
|
||
-d '{
|
||
"host": "myserver",
|
||
"status": "ok",
|
||
"duration_sec": 342,
|
||
"original_size": 5368709120,
|
||
"deduplicated_size": 104857600,
|
||
"compressed_size": 83886080,
|
||
"nfiles_new": 47,
|
||
"nfiles_changed": 12,
|
||
"message": "Backup completed successfully"
|
||
}' \
|
||
http://localhost:9999/api/push
|
||
```
|
||
|
||
### Borgmatic Integration
|
||
|
||
Add to your `borgmatic.yml`:
|
||
|
||
```yaml
|
||
after_backup:
|
||
- >-
|
||
bash -c '
|
||
STATS=$(borgmatic info --archive latest --json 2>/dev/null | python3 -c "
|
||
import sys,json
|
||
d=json.load(sys.stdin)[0][\"archives\"][-1]
|
||
s=d.get(\"stats\",{})
|
||
print(json.dumps({
|
||
\"host\":\"$(hostname)\",
|
||
\"status\":\"ok\",
|
||
\"duration_sec\":int(s.get(\"duration\",0)),
|
||
\"original_size\":s.get(\"original_size\",0),
|
||
\"deduplicated_size\":s.get(\"deduplicated_size\",0),
|
||
\"compressed_size\":s.get(\"compressed_size\",0),
|
||
\"nfiles_new\":s.get(\"nfiles\",0)
|
||
}))" 2>/dev/null || echo "{\"host\":\"$(hostname)\",\"status\":\"ok\"}");
|
||
curl -fsS -m 10 -X POST -H "Content-Type: application/json" -d "$STATS" "http://YOUR_SERVER:9999/api/push" || true
|
||
'
|
||
|
||
on_error:
|
||
- >-
|
||
curl -fsS -m 10 -X POST -H "Content-Type: application/json"
|
||
-d '{"host":"'$(hostname)'","status":"error","message":"Backup failed"}'
|
||
"http://YOUR_SERVER:9999/api/push" || true
|
||
```
|
||
|
||
## API Reference
|
||
|
||
| Method | Endpoint | Description |
|
||
|--------|----------|-------------|
|
||
| `GET` | `/` | Web UI |
|
||
| `GET/POST` | `/api/push` | Push backup status (query params or JSON) |
|
||
| `GET` | `/api/hosts` | List all hosts with current status |
|
||
| `POST` | `/api/hosts` | Add a host `{"name": "...", "kuma_push_url": "..."}` |
|
||
| `PUT` | `/api/hosts/<name>` | Update host `{"enabled": bool, "kuma_push_url": "..."}` |
|
||
| `DELETE` | `/api/hosts/<name>` | Delete host and all history |
|
||
| `GET` | `/api/history/<host>?days=30` | Backup history for a host |
|
||
| `GET` | `/api/calendar/<host>?days=30` | Calendar heatmap data (aggregated by day) |
|
||
| `GET` | `/api/summary` | Dashboard summary (counts, today stats) |
|
||
|
||
## Uptime Kuma Integration
|
||
|
||
1. Create a **Push** monitor in Uptime Kuma for each host
|
||
2. Copy the push URL (e.g. `https://status.example.com/api/push/borg-myserver?status=up&msg=OK`)
|
||
3. In Backup Monitor: Click on a host → Edit → Paste the Kuma Push URL
|
||
4. After each backup push, Backup Monitor automatically forwards the status to Uptime Kuma
|
||
|
||
## Environment Variables
|
||
|
||
| Variable | Default | Description |
|
||
|----------|---------|-------------|
|
||
| `MONGO_URI` | `mongodb://mongo:27017` | MongoDB connection string |
|
||
| `STALE_HOURS` | `26` | Hours without backup before host is marked stale |
|
||
| `WEBHOOK_URLS` | _(empty)_ | Comma-separated webhook URLs for notifications |
|
||
| `WEBHOOK_EVENTS` | `error,stale` | Events that trigger webhooks |
|
||
|
||
## Prometheus Integration
|
||
|
||
The `/metrics` endpoint exposes backup metrics in Prometheus format:
|
||
|
||
```
|
||
backup_hosts_total 21
|
||
backup_host_status{host="myserver"} 1 # 1=ok, 0=error, -1=stale
|
||
backup_host_last_seconds{host="myserver"} 3600 # seconds since last backup
|
||
backup_host_duration_seconds{host="myserver"} 342
|
||
backup_host_size_bytes{host="myserver"} 5368709120
|
||
backup_host_dedup_bytes{host="myserver"} 104857600
|
||
backup_host_files_new{host="myserver"} 47
|
||
backup_today_total 22
|
||
backup_today_bytes 47280909120
|
||
```
|
||
|
||
Add to your `prometheus.yml`:
|
||
|
||
```yaml
|
||
scrape_configs:
|
||
- job_name: 'backup-monitor'
|
||
static_configs:
|
||
- targets: ['backup-monitor:9999']
|
||
scrape_interval: 60s
|
||
```
|
||
|
||
## Webhook Notifications
|
||
|
||
Set `WEBHOOK_URLS` to receive notifications on backup errors or stale hosts:
|
||
|
||
```yaml
|
||
environment:
|
||
- WEBHOOK_URLS=https://n8n.example.com/webhook/backup-alert,https://other.webhook/endpoint
|
||
- WEBHOOK_EVENTS=error,stale # which events trigger webhooks
|
||
```
|
||
|
||
Webhook payload:
|
||
|
||
```json
|
||
{
|
||
"event": "error",
|
||
"host": "myserver",
|
||
"message": "Backup failed",
|
||
"timestamp": "2026-04-05T06:00:00Z"
|
||
}
|
||
```
|
||
|
||
Events:
|
||
- `error` – Fired immediately when a backup reports status "error"
|
||
- `stale` – Fired when a host exceeds `STALE_HOURS` without a backup (once per host, resets on next successful backup)
|
||
|
||
## Data Retention
|
||
|
||
- History entries are automatically deleted after **90 days** (MongoDB TTL index)
|
||
- Hosts are never auto-deleted – remove them manually via UI or API
|
||
|
||
## Screenshots
|
||
|
||
### Dashboard
|
||
Dark-themed overview with summary cards, host grid with status badges, and 14-day minibar charts per host.
|
||
|
||
### Host Detail
|
||
Slide-out drawer with 30-day calendar heatmap, data volume chart, and detailed backup history table.
|
||
|
||
## Tech Stack
|
||
|
||
- **Backend:** Python 3.12, Flask, Gunicorn
|
||
- **Database:** MongoDB 4.4+
|
||
- **Frontend:** Vanilla JS, CSS (no framework dependencies)
|
||
|
||
## License
|
||
|
||
MIT
|