Security

Overview
Custom domain and automated encryption (SSL/TLS)
Nginx advanced configuration
Using Cloudflare
Network Firewall
IP rate limiter
Output Cache
Termination protection & grace period
Manage SSH Keys
How to connect to an Elestio service with SSH keys
SMTP service
Multi-factor authentication
How to Enable or Disable IPv6 on a Linux Virtual Machine
How to configure NGINX to Listen on IPv6

Overview

We automatically setup all services with several layers of security activated by default.

This includes:

Each of these features are accessible from your Dashboard via the Security tab.

Custom domain and automated encryption (SSL/TLS)

If you have activated the firewall on your service ensure port 80 is open or else certificate creation/renewal will fail

Access the "Manage SSL Domains" option within the "Custom Domain Names" row located in the overview tab of the service dashboard.

From there, you can manage allowed domains for SSL. If you want to add a new domain, type it and press enter to add it to the list of authorized domains. You will also need to create a DNS entry to point your domain to the IP address of your service.

You can create an A or CNAME record to point to your service. CNAME is preferred as it won't change even if your IP changes (e.g., this can happen if you clone/migrate your service to another provider).

Once added, you can verify if your DNS entry is propagated with a tool like https://dnschecker.org/

Once propagated, SSL should work instantly on your service. The certificate will be automatically generated and renewed.

Important

For some software, you must also update an env var to indicate the domain to be used. To do that in the overview screen of your service click on the UPDATE CONFIG button > Env tab > there update the domain env var with your domain

Cloudflare users

If you are using Cloudflare reverse proxy (orange cloud icon), please check detailed instructions about Cloudflare & Elestio here:

https://docs.elest.io/books/security/page/using-cloudflare

Troubleshooting SSL not generated

You can display the nginx log with this command in a terminal:

cd /opt/elestio/nginx;
docker-compose logs -f;

Press Ctrl+C to stop displaying the live logs

Reset SSL_DATA folder

In some cases, the /opt/elestio/nginx/ssl_data folder can become corrupt. If this happens, connect to a terminal and try this:

cd /opt/elestio/nginx/;
docker-compose down;
mv ./ssl_data/ ./ssl_data_old/;
mkdir ./ssl_data/;
chmod 777 ./ssl_data/;
docker-compose up -d

Once executed, just open your custom website URL again, and your certificate should be generated and your site served over SSL/TLS.

Nginx advanced configuration

Edit the website nginx configuration

From the service overview screen it's possible to edit the nginx configuration for the elestio default subdomain domain & custom domain, to do that go to the Security tab, then click on Nginx configuration then click on Config button.

There you will see the site configuration in nginx, you are able from there to add or remove headers, to configure the reverse proxy and much more. Nginx is very powerful and provide tons of customizations and control.

Edit the global Nginx configuration

It's also possible to edit the global nginx configuration, to do that follow those steps:

1) create a file named nginx.conf in /opt/elestio/nginx/

2) Paste exactly this content in the file, then you can customize if for your needs:


worker_processes auto;

worker_rlimit_nofile 100000;

events {
  worker_connections 4000;
}

http {
  include mime.types;
  default_type application/octet-stream;
  sendfile on;
  tcp_nopush on;
  tcp_nodelay on;
  server_tokens off;

  # reasonable defaults
  client_header_timeout 10m;
  client_body_timeout 10m;
  send_timeout 10m;
  client_max_body_size 5120M;

  connection_pool_size 256;
  client_header_buffer_size 32k;
  large_client_header_buffers 4 64k;
  request_pool_size 4k;
  
  server_names_hash_bucket_size 4096;

  output_buffers 1 32k;
  postpone_output 1460;

  proxy_buffer_size   128k;
  proxy_buffers   4 256k;
  proxy_busy_buffers_size   256k;

  ignore_invalid_headers on;

  # enable gzip support
  include server-gzip.conf;

  # auto-ssl lua magic for automatic generation of certs
  include resty-http.conf;

  server {
    listen 80 default_server;

    include resty-server-http.conf;
  }

  include /etc/nginx/conf.d/*.conf;
}

stream {
 include /etc/nginx/streams/*.conf;
}

3) edit this file: /opt/elestio/nginx/docker-compose.yml and add there a new line under volumes

- /opt/elestio/nginx/nginx.conf:/usr/local/openresty/nginx/conf/nginx.conf

It should look like this:

version: '2'
services:
  nginx:
    image: elestio/nginx-auto-ssl:latest
    container_name: elestio-nginx
    restart: always
    network_mode: "host"
    volumes:
      - /opt/elestio/nginx/nginx.conf:/usr/local/openresty/nginx/conf/nginx.conf
      - /opt/elestio/nginx/ssl_data:/etc/resty-auto-ssl
      - /opt/elestio/nginx/conf.d:/etc/nginx/conf.d
      - /opt/elestio/nginx/streams:/etc/nginx/streams
      - /opt/elestio/nginx/logs:/var/log/nginx/
      - /root/.acme.sh/vm.elestio.app/fullchain.cer:/etc/nginx/certs/cert.pem
      - /root/.acme.sh/vm.elestio.app/vm.elestio.app.key:/etc/nginx/certs/key.pem
    env_file:
      - .env

4) Restart nginx with this command:

cd /opt/elestio/nginx/;
docker-compose down;
docker-compose up -d;

Now nginx reverse proxy will use your custom global configuration

Using Cloudflare

Cloudflare DNS:

If you wish to use Cloudflare for DNS ONLY, you can configure it just like any other DNS provider, and simply follow the steps for adding a custom domain as usual.

WARNING: Your domain DNS entry must have a GRAY cloud, not an ORANGE (proxied) cloud next to the entry.

Using Cloudflare's proxy for your domain without additional configuration will cause all incoming connections to fail!

This is the correct image shown for DNS-only entries.

Cloudflare Proxy

Even though Elest.io automatically provides SSL and has a firewall, there can be advantages to using Cloudflare for Proxying traffic, notably DDoS attacks and automatic filtering of scripted attacks.

Note: Cloudflare only proxies traffic on certain ports. If you want to use this hostname for SSH, FTP, or other services whose ports are not listed in the above link, you must configure Cloudflare to provide DNS only or use Cloudflare's Spectrum offer.

Because Elest.io already creates an SSL certificate for your website trusted by a root CA, the recommended configuration is to set Cloudflare to use Strict SSL verification when connecting to your server.

Before continuing, ensure you have already configured the domains as per the instructions on the previous page.

Option 1: To set up strict SSL verification for your whole domain:

Navigate to the SSL/TLS section of your domain's dashboard.
Select the "Full (strict)" option.
Your changes will be saved automatically. You're done!

Option 2: To set up strict SSL verification for a specific subdomain:

In your domain's dashboard, navigate to Rules > Configuration Rules and click Create Rule
Name your rule, and configure the incoming request filters.

3. Configure the SSL to Strict

4. Click Save

Option 3: Manual configuration (Advanced)

If you need a custom implementation, you can disable the creation of an SSL certificate with the following steps.

Create a CNAME record for your Cloudflare entry and point to the CNAME provided for that service in the Elestio dashboard.

These changes can be overwritten in the future if you modify the list of domains via the Elest.io dashboard.

1) Connect to the VM with SSH and type this:
nano /opt/elestio/nginx/.env

there remove your domain from the first line and save with CTRL+X

then type this command:
cd /opt/elestio/nginx;
docker-compose down;
docker-compose up -d;

After that, nginx won't try again to obtain an SSL certificate for your domain.

Network Firewall

By default, we only open the ports necessary for the application you have deployed.

How can I restrict access to my service by IP address?

From the Dashboard, select "Security", then "Show Settings" on the Firewall row

From there you can modify, remove, or add new rules to open a port from your service to the internet (or just to a specific target IP).

All services come preconfigured with firewall rules that match the software you are deploying.

You have to keep port 80 open to any ipv4/ipv6 or else Letsencrypt won't be able to generate an SSL certificate.

Here is a compilation of the ports necessary for Elestio Automation:

Mandatory	Application	Protocol	Port	Usage
✔️	Input	TCP	22	Automation SSH
❌	Input	UDP	4242	Nebula/ Global IP
❌	Input	TCP	18345	VS Code
❌	Input	TCP	18374	Open Terminal
❌	Input	TCP	18346	File Explorer
❌	Input	TCP	18445	Tail Logs
❌	Input	TCP	18344	Terminal

❌ => Ports are necessary only if you are utilizing specific tools and activating global private IP functionality.

IP rate limiter

From the Dashboard, click on Security then Rate Limiter > Show Options.

From here, you can easily modify and adjust your service's rate limiter configurations, by amount and per minute or second, per IP address.

By default, all services are preconfigured with a rate limiter of 150 requests per minute and per IP address.

Rate limiter is used only for web traffic.

Output Cache

From the Dashboard, navigate to the Security tab and select Show Options under Output Cache.

From here, you can modify your output cache configurations.

All GET requests are cached for 3 seconds, which is useful in preventing Denial of Service (DOS) attacks.

All services come preconfigured with Output cache by default.

Output cache is used for web traffic only.

Termination protection & grace period

You can enable or disable the Termination protection option from your Dashboard Overview, using the toggle on the right-hand side. This setting is disabled by default.

It's not possible to change software versions, delete, shut down, power off, reset or reboot your service when Termination protection is enabled. To make these changes, you must first disable Termination protection.

Our grace period for storing backups after the deletion of service is 7 days, making it easier for you to restore your service in this window of time for any reason.

Manage SSH Keys

From the Dashboard, navigate to the Security tab and select the Show Options button to Manage SSH Keys.
From here, you can add or remove SSH keys allowed on the server.

Add an SSH key
Click on the Add key. Simply give your key a title and save!

Deleting an SSH key
Select the 'trash' icon to the right of the key you wish to delete. We'll always double-check with you before making a deletion, just to be sure!

How to connect to an Elestio service with SSH keys

Connecting to an Elestio service Using Existing SSH Keys from a Mac/Linux Machine

Step 1: Check for Existing SSH Key Pair

First, verify if you have an existing SSH key pair by checking the .ssh directory. Open a terminal and execute the following command:

ls ~/.ssh

The file with the .pub extension is your public key.

If you don't have an existing SSH key pair, you can generate a new one using the following command:

ssh-keygen -b 4096 -t rsa

Step 2: Copy your SSH Public Key to your Elestio service

Copy the Public Key to Clipboard:
Navigate to the .ssh directory by typing

cd .ssh

Display the contents of your public key using cat. For example, if your public key is id_ed255518.pub, you can view it with:

cat ~/.ssh/id_ed255518.pub

Copy the displayed public key and add it to your Elestio service.

Add the Public Key to Elestio:
- Open your Elestio service and navigate to the 'Security' tab.
- Go to 'Manage SSH Keys' and click on the 'Add key' button.
- Paste your public key into the provided field.

Step 3: Confirm and Test Connection

To test the SSH connection to your VM, use the following command from your local machine:

ssh root@IPV4_TARGET_VM

Replace "IPV4_TARGET_VM" with the actual IPv4 address of your target VM, which you can find in the 'Overview' tab of your service.

For example:

ssh root@128.144.84.22

If the connection is successful, you'll be logged into your VM 🚀

Connecting to an Elestio service Using Existing SSH Keys from a Windows Machine with WSL (Windows Subsystem for Linux)

Step 1: Set up WSL (Windows Subsystem for Linux)

Before proceeding, you need to ensure that WSL is installed and configured on your Windows machine. Here's how you can set up WSL:

Enable WSL: Open PowerShell as Administrator and run the following command:
```
wsl --install
```
This command will automatically enable the necessary features and install the latest version of WSL on your system.
Install a Linux Distribution: After WSL is installed, visit the Microsoft Store and search for your preferred Linux distribution (e.g., Ubuntu). Click "Install" to download and install it.
Initialize WSL: Once the distribution is installed, launch it from the Start menu. This will initialize the distribution and prompt you to create a user account and set up a password.
Update and Upgrade: After the setup is complete, it's a good idea to update and upgrade the packages within your Linux distribution to ensure you have the latest versions. You can do this by running the following commands:
```
sudo apt update sudo apt upgrade
```
Follow the same tutorial to connect to an Elestio service using existing SSH keys from a Mac/Linux machine.

SMTP service

All deployed services include a basic preconfigured SMTP service, useful for sending alerts and notifications from your service.

This is free, but comes with a few limitations:

You can only send transactional emails. Marketing emails are not permitted.

Any violation of this will lead to a suspension or termination of your service.
You can send up to 300 transactional emails per hour. That's up to 7200 emails per day!
All emails must be sent from [domain]@vm.elestio.app where [domain] is the URL of your service.

Attempts at sending from any other email address will be rejected.
The SMTP service is only available from the global private network IP of your VM.

Of course in several cases you will want to change the smtp configuration in the web UI of your software to use another smtp service. It's useful if you want to be able to configure another sender address or to overcome any limitations stated above.

Multi-factor authentication

By default, Elestio uses Email-based MFA, each time you log in to Elestio we will send you an email with a one-time code to enter into our UI to be able to connect. This protection is in place to enforce security and avoid account hacking.

We also have TOTP-based MFA, this is more secure because it's based on an app installed on your phone to generate TOTP codes instead of us sending them by email. So even if your mailbox is compromised your Elestio account will still be safe.

The process to activate TOTP MFA on your account

Open the account security tab here: https://dash.elest.io/account/security
Click on Configure button in Manage Two-Factor Authentication.
Select the Authenticator App tab.
Download an authenticator app: Authy (recommended) or Google Authenticator or Microsoft Authenticator
Open your authenticator app then scan the QR code on the screen
Generate a code with your app and enter it on the Elestio screen
Click on Validate

Done, Strong MFA is now enabled on your account, and will be required to login into your Elestio account

You should keep the text version somewhere safe (in orange in the screenshot), this will allow you to recover in case you lose your phone or authenticator app.

If you have lost both your authenticator app and text secret, you can contact our support team via email with proof of identity to get MFA removed from your account.

The process to Activate Email Based MFA on your account

Open the account security tab here: https://dash.elest.io/account/security
Click on Configure button in Manage Two-Factor Authentication.
Select the Email Based tab.
Click on Enable button to activate it.

The process to Deactivate MFA on your account

Open the account security tab here: https://dash.elest.io/account/security
Click on Configure button in Manage Two-Factor Authentication.
Select the Disabled tab.
To confirm the action, type disable in the confirmation input field on the confirmation modal.
Click on Disable button to disable it.

How to Enable or Disable IPv6 on a Linux Virtual Machine

IPv6 is the latest version of the Internet Protocol, designed to address the limitations of IPv4. In some scenarios, such as managing Docker Hub rate limits, disabling IPv6 may be necessary to ensure smooth operation. This guide provides step-by-step instructions for enabling or disabling IPv6 on a Linux VM.

If you are encountering issues with the Docker Hub rate limit, it is recommended to disable IPv6 on your VM by following the steps outlined in the "Disabling IPv6" section below.

First, open the terminal by selecting the "Open Terminal" option located at the top-right corner of the service details overview. Then, proceed with the steps for either enabling or disabling IPv6, depending on your requirements.

Enabling IPv6

If IPv6 is currently disabled on your VM and you want to enable it, follow these steps:

1. Temporarily Enable IPv6: To enable IPv6 without rebooting:

sudo sysctl -w net.ipv6.conf.all.disable_ipv6=0
sudo sysctl -w net.ipv6.conf.default.disable_ipv6=0

This change will revert after a system reboot.

2. Permanently Enable IPv6: To enable IPv6 persistently:

Edit the `sysctl.conf` file:

sudo nano /etc/sysctl.conf

Locate and comment out the following lines (if present):

# net.ipv6.conf.all.disable_ipv6 = 1
# net.ipv6.conf.default.disable_ipv6 = 1

Apply the changes:

sudo sysctl -p

IPv6 will now remain enabled even after a reboot.

Disabling IPv6

If IPv6 is causing issues, such as with Docker Hub rate limits, follow these steps to disable it.

1. Temporarily Disable IPv6: To disable IPv6 without rebooting:

sudo sysctl -w net.ipv6.conf.all.disable_ipv6=1
sudo sysctl -w net.ipv6.conf.default.disable_ipv6=1

This change is not permanent and will be reset after a reboot.

2. Permanently Disable IPv6: To disable IPv6 persistently:

Edit the `sysctl.conf` file:

sudo nano /etc/sysctl.conf

Add the following lines to the file:

net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1

Apply the changes:

sudo sysctl -p

IPv6 will now remain disabled even after a reboot.

How to configure NGINX to Listen on IPv6

IPv6 is quickly becoming standard. If your VM already supports it, you just need to tell NGINX to listen on an IPv6 socket.

(If IPv6 is disabled on the VM itself, start with the “Enable IPv6 on your VM” article and come back here.)

1. Open the dashboard

Go to your project → Services → Security tab.
Click NGINX Configuration › Show config.

2 . Edit the server block for your domain

Choose your domain and find edit the NGINX config then find

listen 443 ssl http2;

Add the IPv6 line directly beneath:

listen 443 ssl http2; listen [::]:443 ssl http2;

(Keep the original IPv4 line—now NGINX will listen on both stacks.)

Click Apply & Restart. NGINX reloads and your server is now ready to serve HTTPS traffic over IPv6.