Production Deployment

Best practices and recommendations for running Rustmail in production.

Checklist

Before deploying to production:

Tested configuration locally
Verified bot permissions in Discord
Set up HTTPS for the panel
Configured backups
Set up monitoring
Documented your configuration

System Requirements

Minimum

1 CPU core
512 MB RAM
1 GB disk space

Security

HTTPS

Always use HTTPS for the panel in production. Options:

Reverse proxy with TLS termination (recommended)
- Nginx, Caddy, Traefik
- Automatic certificate renewal with Let’s Encrypt
Cloud load balancer
- AWS ALB, GCP Load Balancer, Cloudflare

See Configuration for proxy setup.

Firewall

Restrict access to necessary ports only:

# Allow SSH
ufw allow 22/tcp

# Allow HTTPS (reverse proxy)
ufw allow 443/tcp

# Block direct access to bot port (if using reverse proxy)
# ufw deny 3002/tcp

ufw enable

Secrets Management

Protect sensitive configuration:

Bot token
OAuth2 client secret
Database file

Options:

File permissions: chmod 600 config.toml
Docker secrets
Environment-based secret injection at deployment

Updates

Keep Rustmail updated for security fixes:

# Check current version
./rustmail --version

# Update (Docker)
docker pull ghcr.io/rustmail/rustmail:latest

High Availability

Rustmail is designed as a single-instance application. For high availability:

Database

Regular backups
Store backups off-server
Test restore procedures

Process Management

Use a process manager to ensure the bot restarts on failure:

Systemd (Linux):

[Unit]
Description=Rustmail Discord Bot
After=network.target

[Service]
Type=simple
User=rustmail
WorkingDirectory=/opt/rustmail
ExecStart=/opt/rustmail/rustmail
Restart=always
RestartSec=10

[Install]
WantedBy=multi-user.target

Docker:

services:
  rustmail:
    restart: unless-stopped

Monitoring

Monitor bot status:

# Simple health check
curl -f http://localhost:3002/api/bot/status

Integration with monitoring systems:

Prometheus metrics endpoint (future feature)
External uptime monitoring
Discord webhook alerts

Backup Strategy

What to Backup

Item	Location	Frequency
Database	`db.sqlite`	Daily
Configuration	`config.toml`	On change

Automated Backups

Example backup script:

#!/bin/bash
BACKUP_DIR="/backups/rustmail"
DATE=$(date +%Y%m%d_%H%M%S)

# Create backup directory
mkdir -p "$BACKUP_DIR"

# Backup database
cp /opt/rustmail/db.sqlite "$BACKUP_DIR/db_$DATE.sqlite"

# Backup config
cp /opt/rustmail/config.toml "$BACKUP_DIR/config_$DATE.toml"

# Keep last 30 days
find "$BACKUP_DIR" -type f -mtime +30 -delete

Add to crontab:

0 3 * * * /opt/rustmail/backup.sh

Off-site Backup

Sync backups to remote storage:

# rsync to remote server
rsync -avz /backups/rustmail/ backup-server:/backups/rustmail/

# AWS S3
aws s3 sync /backups/rustmail/ s3://your-bucket/rustmail-backups/

Logging

Log Levels

Rustmail outputs logs to stdout. In production:

# Redirect to file
./rustmail >> /var/log/rustmail/rustmail.log 2>&1

Log Rotation

With logrotate (/etc/logrotate.d/rustmail):

/var/log/rustmail/*.log {
    daily
    rotate 14
    compress
    delaycompress
    missingok
    notifempty
    create 0640 rustmail rustmail
}

Centralized Logging

For Docker deployments, use logging drivers:

services:
  rustmail:
    logging:
      driver: "syslog"
      options:
        syslog-address: "udp://logserver:514"
        tag: "rustmail"

Performance Tuning

SQLite Optimization

The database uses SQLite with default settings. For high-volume deployments:

-- Run these optimizations
PRAGMA journal_mode = WAL;
PRAGMA synchronous = NORMAL;
PRAGMA cache_size = -64000;  -- 64MB cache

Connection Pooling

Rustmail uses SQLx with connection pooling. Default settings are suitable for most deployments.

Maintenance

Planned Downtime

For updates requiring restart:

Close active tickets or notify staff
Stop the bot
Backup database
Apply update
Verify configuration
Start the bot
Verify functionality

Database Maintenance

Periodically optimize the database:

# Stop bot first
sqlite3 db.sqlite "VACUUM;"
sqlite3 db.sqlite "ANALYZE;"

Cleanup Old Data

If needed for storage or compliance:

-- Delete closed tickets older than 2 years
DELETE FROM thread_messages
WHERE thread_id IN (
  SELECT id FROM threads
  WHERE status = 0
  AND closed_at < datetime('now', '-2 years')
);

DELETE FROM threads
WHERE status = 0
AND closed_at < datetime('now', '-2 years');

VACUUM;

Troubleshooting

Bot Goes Offline

Check process status: systemctl status rustmail
Check logs for errors
Verify Discord API status
Check network connectivity
Verify token validity

Panel Inaccessible

Check bot process is running
Verify port 3002 is listening: netstat -tlnp | grep 3002
Check reverse proxy configuration
Verify SSL certificate validity
Check firewall rules

Performance Issues

Monitor CPU/memory usage
Check database size
Review ticket volume
Consider database maintenance

Data Recovery

From backup:

# Stop bot
systemctl stop rustmail

# Restore database
cp /backups/rustmail/db_20240115.sqlite /opt/rustmail/db.sqlite

# Verify
sqlite3 /opt/rustmail/db.sqlite "SELECT COUNT(*) FROM threads;"

# Start bot
systemctl start rustmail

Rustmail Documentation