Deploy Twenty CRM on a VPS

Twenty is an open-source CRM (a modern Salesforce/HubSpot alternative). The self-hosted stack is a Postgres database, Redis, and the Twenty server + worker containers, served through a single reverse-proxied entrypoint. It's a comparatively light stack and runs comfortably on a small VPS.

1. Prerequisites

• RamNode VPS on Ubuntu 24.04 LTS
• Minimum 4 GB RAM, 2 vCPUs, 25 GB disk for a small team; bump RAM if you'll have dozens of concurrent users
• DNS A record for the subdomain, e.g. crm.yourdomain.com
• Root/sudo SSH access

2. Base server setup

apt update && apt -y upgrade
apt -y install curl git ufw

ufw allow OpenSSH
ufw allow 80/tcp
ufw allow 443/tcp
ufw enable

3. Install Docker

curl -fsSL https://get.docker.com | sh
apt -y install docker-compose-plugin
systemctl enable --now docker

4. Set up Twenty

mkdir -p /opt/twenty && cd /opt/twenty
curl -fsSL https://raw.githubusercontent.com/twentyhq/twenty/main/packages/twenty-docker/docker-compose.yml -o docker-compose.yml
curl -fsSL https://raw.githubusercontent.com/twentyhq/twenty/main/packages/twenty-docker/.env.example -o .env

Edit .env:

• PG_DATABASE_PASSWORD — strong unique password
• APP_SECRET — generate with openssl rand -base64 32
• SERVER_URL — set to https://crm.yourdomain.com (this must match the public URL exactly — Twenty uses it for auth callbacks, invite links, and file URLs)
• STORAGE_TYPE — local is fine for a single VPS; switch to s3 and fill in the corresponding vars if you want uploads on external object storage
• Mail settings (EMAIL_FROM_ADDRESS, EMAIL_DRIVER, SMTP credentials) — required for invite emails and password resets to actually send

5. Start the stack

docker compose up -d

This starts Postgres, Redis, the server container (API + frontend), and a worker container (background jobs — emails, imports, workflow automation). Watch logs:

docker compose logs -f server

The server listens on port 3000 by default.

6. nginx reverse proxy + TLS

apt -y install nginx certbot python3-certbot-nginx

/etc/nginx/sites-available/twenty:

server {
    listen 80;
    server_name crm.yourdomain.com;

    client_max_body_size 50M;   # file/attachment uploads

    location / {
        proxy_pass http://127.0.0.1:3000;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection "upgrade";   # GraphQL subscriptions / real-time updates
    }
}

ln -s /etc/nginx/sites-available/twenty /etc/nginx/sites-enabled/
nginx -t && systemctl reload nginx
certbot --nginx -d crm.yourdomain.com

7. First login

Visit https://crm.yourdomain.com and complete the workspace creation wizard — this creates the first admin user. From there, invite teammates via Settings → Members (requires working SMTP from step 4).

8. Backups

Only Postgres and, if using local storage, the uploads volume need regular backup:

docker exec twenty-db-1 pg_dump -U postgres default > /root/backups/twenty-pg-$(date +%F).sql

docker run --rm -v twenty_server-local-data:/data -v /root/backups:/backup alpine \
  tar czf /backup/twenty-uploads-$(date +%F).tar.gz -C /data .

Cron this nightly and copy backups off the VPS.

9. Updating

cd /opt/twenty
docker compose down
docker compose pull
docker compose up -d

Twenty runs its own DB migrations automatically on server container startup — check the changelog before major version jumps, since CRM schema migrations can take a few minutes on larger datasets.

Troubleshooting notes

• Invite/reset emails never arrive → almost always an SMTP config issue in .env; test with a known-good SMTP relay before assuming the app is broken.
• Login redirects loop or fail after TLS setup → check SERVER_URL matches the actual public HTTPS URL exactly (scheme + host, no trailing slash mismatch).
• Background jobs (imports, automations) not running → confirm the worker container is up alongside server; they're separate containers and only server being healthy doesn't mean jobs are processing.