Skip to main content

Daemon mode

Daemon mode runs Reqon as a long-running service, executing scheduled missions automatically.

Starting the daemon​

reqon ./missions/ --daemon

How it works​

  1. Daemon loads all missions from the specified path
  2. Parses schedule configurations
  3. Runs missions according to their schedules
  4. Continues until stopped

Command options​

reqon ./missions/ --daemon [options]

Options:
  --auth <file>        Credentials file
  --verbose            Enable detailed logging
  --check-interval     How often to check schedules (default: 1000ms)

Example setup​

Mission files​

missions/
├── sync-customers.vague
├── sync-orders.vague
└── daily-report.vague

sync-customers.vague​

mission SyncCustomers {
  schedule: every 15 minutes

  source API { auth: bearer, base: "https://api.example.com" }
  store customers: file("customers")

  action Sync {
    get "/customers" { since: lastSync }
    store response -> customers { key: .id, upsert: true }
  }

  run Sync
}

Running​

reqon ./missions/ --daemon --auth ./credentials.json --verbose

Process management​

Foreground​

reqon ./missions/ --daemon

Press Ctrl+C to stop.

Background (Linux)​

nohup reqon ./missions/ --daemon > reqon.log 2>&1 &

Systemd service​

Create /etc/systemd/system/reqon.service:

[Unit]
Description=Reqon Data Sync Daemon
After=network.target

[Service]
Type=simple
User=reqon
WorkingDirectory=/opt/reqon
ExecStart=/usr/bin/npx reqon ./missions/ --daemon --auth ./credentials.json
Restart=on-failure
RestartSec=10

[Install]
WantedBy=multi-user.target

Enable and start:

sudo systemctl enable reqon
sudo systemctl start reqon

Docker​

Dockerfile:

FROM node:20-alpine

WORKDIR /app
COPY package*.json ./
RUN npm install

COPY . .

CMD ["npx", "reqon", "./missions/", "--daemon"]

docker-compose.yml:

version: '3'
services:
  reqon:
    build: .
    volumes:
      - ./missions:/app/missions
      - ./credentials.json:/app/credentials.json
      - reqon-data:/app/.vague-data
    environment:
      - NODE_ENV=production
    restart: unless-stopped

volumes:
  reqon-data:

Graceful shutdown​

Reqon handles SIGINT and SIGTERM:

  1. Stops accepting new runs
  2. Waits for in-progress missions to complete
  3. Saves state
  4. Exits cleanly
# Graceful stop
kill -TERM $(pgrep -f "reqon.*daemon")

# Or with systemd
systemctl stop reqon

Health checks​

Status endpoint​

Enable HTTP health endpoint:

reqon ./missions/ --daemon --health-port 8080

Check health:

curl http://localhost:8080/health
# {"status":"ok","uptime":3600,"missionsLoaded":3}

File-based health​

mission HealthCheck {
  schedule: every 1 minute

  action Check {
    store { status: "ok", timestamp: now() } -> health
  }
}

Monitor the health store file.

Logging​

Log levels​

# Default logging
reqon ./missions/ --daemon

# Verbose logging
reqon ./missions/ --daemon --verbose

# Environment variable
REQON_LOG_LEVEL=debug reqon ./missions/ --daemon

Log output​

[2024-01-20 09:00:00] [INFO] Starting Reqon daemon
[2024-01-20 09:00:00] [INFO] Loaded 3 missions
[2024-01-20 09:00:00] [INFO] SyncCustomers: Next run at 09:15:00
[2024-01-20 09:00:00] [INFO] SyncOrders: Next run at 09:05:00
[2024-01-20 09:00:00] [INFO] DailyReport: Next run at 2024-01-21 00:00:00
[2024-01-20 09:05:00] [INFO] SyncOrders: Starting run
[2024-01-20 09:05:02] [INFO] SyncOrders: Completed (2.1s)

Structured logging​

REQON_LOG_FORMAT=json reqon ./missions/ --daemon
{"timestamp":"2024-01-20T09:00:00Z","level":"info","message":"Starting run","mission":"SyncCustomers"}

Monitoring​

Metrics​

Export metrics with --metrics-port:

reqon ./missions/ --daemon --metrics-port 9090

Prometheus format:

# HELP reqon_mission_runs_total Total mission runs
reqon_mission_runs_total{mission="SyncCustomers",status="success"} 142
reqon_mission_runs_total{mission="SyncCustomers",status="failure"} 3

# HELP reqon_mission_duration_seconds Mission run duration
reqon_mission_duration_seconds{mission="SyncCustomers",quantile="0.5"} 2.1

Alerting​

Create alerting rules:

groups:
  - name: reqon
    rules:
      - alert: ReqonMissionFailing
        expr: rate(reqon_mission_runs_total{status="failure"}[5m]) > 0.1
        for: 10m
        labels:
          severity: warning
        annotations:
          summary: "Mission {{ $labels.mission }} is failing"

Best practices​

Use separate credentials​

# Production credentials
reqon ./missions/ --daemon --auth ./prod-credentials.json

Run as non-root​

# Create dedicated user
useradd -r -s /bin/false reqon

# Run as that user
sudo -u reqon reqon ./missions/ --daemon

Persistent storage​

Ensure .vague-data is on persistent storage:

volumes:
  - /var/lib/reqon:/app/.vague-data

Health monitoring​

Always enable health checks:

reqon ./missions/ --daemon --health-port 8080

Log rotation​

# Use logrotate
/var/log/reqon/*.log {
    daily
    rotate 7
    compress
    missingok
    notifempty
}