Hosting: VPS

A virtual private server, mostly known as VPS, is a virtual machine running in the cloud with you, the developer, having full control over the system.

Server Rental

To be able to follow this guide, you first need to rent a VPS. This section will explain how to do that. If you already have a VPS to work on, skip to the next section.

In this guide, we will use the services of Hostinger.

You are free to choose the provider of your choice. All providers provide the same services, so you won’t have any problems with the technical part of this article. You can take this part as an overview of how the server rental operation works. If you are a beginner, you can use this guide to rent your first ever server!

Analog of a Server

If you cannot or do not want to rent a server but you still want to play around with running a bot on a VPS, you can follow this tutorial on a virtual machine instead. To do this, use an application such as VirtualBox. Create a virtual machine with the desired Linux distribution to simulate a Linux server.

Go to the VPS Hosting page. We will use the “KVM 1” plan. The resources of “KVM 1” are enough for bots with a large audience, and even more so for our test bot.

Click the “Add to cart” button. You will be automatically redirected to the checkout page, where you will also immediately register with Hostinger.

Change the Rental Term!

The typical lease term is 1-2 year (a marketing ploy), and it costs a lot of money. You probably don’t need it, so you can start by renting a server for a month, which is much cheaper.

In any case, Hostinger provides a 30-day money-back guarantee.

After making your payment, you will be able to set up your server:

Location. We recommend that you choose the location closest to Amsterdam. The main Bot API server is located in Amsterdam. If you use your own Bot API server, choose the closest location to it instead.
Server type. Choose the option “Clean OS.”
Operating system. We will use Ubuntu 22.04. If you choose a different system, some steps may be different, so be careful.
Server name. Pick any name you like.
Root password. Make a strong password and keep it in a safe place!
SSH key. Skip this step. We will set up SSH keys later.

After the server is created, you can connect to it using SSH:

SSH (Secure Shell) is a network protocol that can be used to remotely control a computer.

ssh root@<ip-address>

Replace <ip-address> with the IP address of your server which you can find on the server management page.

Configuring SSH

It can be difficult and tedious to remember which IP address and whose name you need to connect to a server. To eliminate these routine steps and improve your server experience, you can configure SSH by creating a ~/.ssh/config (https://linuxhandbook.com/ssh-config-file) file on your computer that stores all the data you need to connect to the server under certain arbitrary identifiers. This is beyond the scope of this article, so you’ll have to configure it yourself.

Separate User for Each Application

In this guide, all actions with the server will be performed as the root user. This is done on purpose to simplify this guide. However, in reality, the root user should only be responsible for general services (web server, database, etc.), and applications should be run by individual non-root users. This approach ensures the security of confidential data and prevents hacking of the entire system. At the same time, it imposes some inconveniences. Describing all these points unnecessarily increases the complexity of the article, which we try to avoid.

Starting the Bot

We now have a server at our disposal where we can run the bot to keep it running around the clock.

To simplify the beginning of the article, we skipped the step of delivering the code to the server automatically each time after pushing your code, but it is described below.

For now, you can copy local files to a remote server using the following command. Note that -r copies recursively, so you only need to specify the root directory of your project:

scp -r <path-to-local-project-root> root@<ip-address>:<path-to-remote-directory>

Replace <path-to-local-project-root> with the path to the project directory on your local disk, <ip-address> with the IP address of your server, and <path-to-remote-directory> with the path to the directory where the bot’s source code should be stored on the server.

As mentioned above, you should now be able to open a remote terminal on your VPS by starting an SSH session.

ssh root@<ip-address>

Note how your command prompt changes. This indicates that you are now connected to the remote machine. Every command you type will be run on your VPS. Try running ls to confirm that you have successfully copied over your source files.

The remainder of this page will assume that you are able to connect to your VPS. All following commands need to be run in an SSH session.

Don’t forget to install the runtime!

To run the bot, you need to install Node.js or Deno on the server, depending on the runtime in which the bot will run. This is beyond the scope of this article, so you will need to do it yourself. You probably already did this when getting started, so you should be familiar with the steps. 😉

Below are two ways you can keep your bot running smoothly: using systemd or PM2.

systemd

systemd is a powerful service manager which is pre-installed on many Linux distributions, mainly Debian-based ones such as Ubuntu.

Getting the Start Command

Get the absolute path to your runtime:
DenoNode.js
sh
```
which deno
```
1
sh
```
which node
```
1
You should have the absolute path to your bot’s directory.

Your start command should look like the following:

<runtime_path> <options> <entry_file_relative_path>

# Path to the bot directory: /home/user/bot1/

# Deno example:
# /home/user/.deno/bin/deno --allow-all run mod.ts

# Node.js example:
# /home/user/.nvm/versions/node/v16.9.1/bin/node index.js

Creating the Service

Go to the services directory:
sh
```
cd /etc/systemd/system
```
1
Open your new service file with an editor:
sh
```
nano <app-name>.service
```
1
Replace <app-name> with any identifier. <app-name>.service will be the name of your service.
Add the following content:
text
```
[Unit]
After=network.target

[Service]
WorkingDirectory=<bot-directory-path>
ExecStart=<start-command>
Restart=on-failure

[Install]
WantedBy=multi-user.target
```
1
2
3
4
5
6
7
8
9
10
Replace <bot-directory-path> with the absolute path to your bot’s directory and <start-command> with the command you received above.
Here is a brief explanation of the service configuration:
- After=network.target — indicates that the application should be launched after the Internet module is loaded.
- WorkingDirectory=<bot-directory-path> — sets the current working directory of the process. This allows you to use relative assets, such as the .env file, which contains all the necessary environment variables.
- ExecStart=<start-command> — sets the startup command.
- Restart=on-failure — indicates that the application should restart after a crash.
- WantedBy=multi-user.target — defines the system state in which the service should be launched. multi-user.target — is a typical value for servers.
For more information on the unit files, read this.
Reload systemd whenever you edit the service:
sh
```
systemctl daemon-reload
```
1

Managing the Service

# Replace `<service-name>` with the file name of the service you created.

# To start the service
systemctl start <service-name>

# To view the service logs
journalctl -u <service-name>

# To restart the service
systemctl restart <service-name>

# To stop the service
systemctl stop <service-name>

# To enable the service to start when the server boots
systemctl enable <service-name>

# To disable service startup when the server boots
systemctl disable <service-name>

Starting the service should now launch your bot!

PM2

PM2 is a daemon process manager for Node.js that will help you manage and keep your app online 24/7.

PM2 is designed specifically to manage applications written in Node.js. However, it can also be used to manage applications written in other languages or runtimes.

Installing

NPMYarnpnpm

npm install -g pm2

yarn global add pm2

pnpm add -g pm2

Creating an Application

PM2 offers two ways to create an application:

Use the command line interface.
Use the configuration file.

The first method is convenient when getting to know PM2. However, during deployment, you should use the second method, which is what we did in our case.

Create a ecosystem.config.js file on the server in the directory where the bot build is stored with the following content:

module.exports = {
  apps: [{
    name: "<app-name>",
    script: "<start-command>",
  }],
};

Replace <app-name> with any identifier and <start command> with the command to start the bot.

Managing the Application

Below are the commands you can use to control the application.

# If the `ecosystem.config.js` file is in the current directory,
# you can specify nothing to start the application.
# If the application is already running, this command will restart it.
pm2 start

# All of the following commands require you to specify the name of the application
# or the `ecosystem.config.js` file.
# To apply the action to all applications, specify `all`.

# To restart the app
pm2 restart <app-name>

# To reload the app
pm2 reload <app-name>

# To stop the app
pm2 stop <app-name>

# To delete an app
pm2 delete <app-name>

Saving the Application Operation

If the server reboots, your bot will not resume working. For the bot to resume work, you need to prepare PM2 for this.

On the server in the terminal, run the following command:

pm2 startup

You will be provided with a command that you must execute to make PM2 automatically start after the server reboots.

Then run one more command:

pm2 save

This command will save the list of current applications so that they can be launched after a server reboot.

If you have created a new application and want to save it as well, simply run pm2 save again.

Running the Bot on Webhooks

To run a bot on webhooks, you will need to use a web framework and NOT call bot.start().

Here is a sample code to run the bot on webhooks that should be added to the main bot file:

Node.jsDeno

import { webhookCallback } from "grammy";
import { fastify } from "fastify";

const server = fastify();

server.post(`/${bot.token}`, webhookCallback(bot, "fastify"));

server.listen();

import { webhookCallback } from "https://deno.land/x/grammy@v1.32.0/mod.ts";

const handleUpdate = webhookCallback(bot, "std/http");

Deno.serve(async (req) => {
  if (req.method === "POST") {
    const url = new URL(req.url);
    if (url.pathname.slice(1) === bot.token) {
      try {
        return await handleUpdate(req);
      } catch (err) {
        console.error(err);
      }
    }
  }
  return new Response();
});

Domain Rental

To connect a bot running on webhooks to the outside world, you need to purchase a domain. We are going to explain this with Hostinger again, but there are many other services, too, and they all work similarly.

Go to the domain name search page. In the text input field, enter a domain name of the form <name>.<zone>. For example, example.com.

If the desired domain is free, click the Add button next to it. You will be automatically redirected to the checkout page, where you will also immediately register with Hostinger if you are not already registered. Pay for the domain.

Domain Pointing to VPS

Before your domain can work with your VPS, you need to point the domain to your server. To do this, in the Hostinger Control Panel, click the “Manage” button next to your domain. Next, go to the DNS record management page by clicking on the “DNS / Name Servers” button in the menu on the left.

First, find out the IP address of your VPS.

In the list of DNS records, find the record of type A with the name @. Edit this record by changing the IP address in the “Points to” field to the IP address of your VPS, and set the TTL to 3600.

Next, find and delete the record of type CNAME with the name www. Instead, create a new record of type A with the name www, pointing to the IP address of your VPS, and set the TTL to 3600.

If you run into problems, use the other method described in the knowledge base.

Setting up a Web Server

For the website to work and the bot to start receiving updates from Telegram, you need to set up a web server. We will use Caddy.

Caddy is a powerful open source web server with automatic HTTPS.

Web Server

We use Caddy because, unlike mainstream web servers like Nginx or Apache, it automatically configures SSL certificates. This makes the article much easier. However, you are free to choose any web server you want.

Installation

The following five commands will download and automatically start Caddy as a systemd service called caddy.

apt install -y debian-keyring debian-archive-keyring apt-transport-https curl
curl -1sLf 'https://dl.cloudsmith.io/public/caddy/stable/gpg.key' | gpg --dearmor -o /usr/share/keyrings/caddy-stable-archive-keyring.gpg
curl -1sLf 'https://dl.cloudsmith.io/public/caddy/stable/debian.deb.txt' | tee /etc/apt/sources.list.d/caddy-stable.list
apt update
apt install caddy

See the Caddy installation guide for other installation options.

Check the status of Caddy:

systemctl status caddy

Troubleshooting

Some hosting providers provide VPS with a pre-installed web server, for example, Apache. Multiple web servers cannot run on the same machine at the same time. For Caddy to work, you need to stop and shut down another web server:

systemctl stop <service-name>
systemctl disable <service-name>

Replace service-name with the name of the web server service that is interfering with Caddy.

Now, if you open your server’s IP address in a browser, you will see a typical page with instructions on how to set up Caddy.

Configuring

In order for Caddy to process requests coming to our domain, we need to change the Caddy configuration.

Run the following command to open the Caddy configuration file:

nano /etc/caddy/Caddyfile

You will see the following default configuration:

text

# The Caddyfile is an easy way to configure your Caddy web server.
#
# Unless the file starts with a global options block, the first
# uncommented line is always the address of your site.
#
# To use your own domain name (with automatic HTTPS), first make
# sure your domain's A/AAAA DNS records are properly pointed to
# this machine's public IP, then replace ":80" below with your
# domain name.

:80 {
  # Set this path to your site's directory.
  root * /usr/share/caddy

  # Enable the static file server.
  file_server

  # Another common task is to set up a reverse proxy:
  # reverse_proxy localhost:8080

  # Or serve a PHP site through php-fpm:
  # php_fastcgi localhost:9000
}

# Refer to the Caddy docs for more information:
# https://caddyserver.com/docs/caddyfile

For the bot to work, make the configuration look like this:

text

<domain> {
  reverse_proxy /<token> localhost:<port>
}

Replace <domain> with your domain, <token> with your bot token, and <port> with the port on which you want to run your bot.

Reload Caddy every time you change the site’s configuration file using the following command:

systemctl reload caddy

Now all requests to the address https://<domain>/<token> will be redirected to the address http://localhost:<port>/<token>, where the bot’s webhook is running.

Connecting a Webhook to Telegram

All you have to do is tell Telegram where to send the updates. To do this, open your browser and visit the page at the following link:

text

https://api.telegram.org/bot<token>/setWebhook?url=https://<domain>/<token>

Replace <token> with your bot token and <domain> with your domain.

CI/CD

CI/CD is an important part of the modern software development process. This guide covers almost all of the CI/CD pipeline.

We will focus on writing scripts for GitHub and GitLab. You can easily adapt the examples below to your CI/CD service of choice, such as Jenkins, Buddy, etc., if needed.

SSH Keys

To deliver files to the server, you need to set up passwordless authentication, which is implemented using SSH keys.

The following commands should be run on your personal computer.

Change to the directory with the SSH keys:

cd ~/.ssh

Generate a new key pair:

GitHubGitLab

ssh-keygen -t rsa -m PEM

ssh-keygen -t ed25519

This command will generate a public and private key of the type and format you want for GitHub and GitLab. You can also specify a custom key name if you wish.

Next, send the public key to the server:

ssh-copy-id -i <key-name>.pub root@<ip-address>

Replace <key-name> with the name of the generated key and <ip-address> with the IP address of your server.

Note that the public key can be located on many servers, and the private key should be only with you and GitHub or GitLab.

You can now connect to the server without having to enter a password.

Example Workflows

Node.js (GitHub)

Use

yml

name: Main

on:
  push:
    branches:
      - main

jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
        with:
          fetch-depth: 0
      - uses: actions/setup-node@v3
        with:
          node-version: "latest"
      - run: npm ci
      - name: Build
        run: npm run build
      - uses: actions/upload-artifact@v3
        with:
          name: source
          path: |
            dist/*.js
            package.json
            package-lock.json
  deploy:
    needs: build
    runs-on: ubuntu-latest
    steps:
      - uses: actions/download-artifact@v3
        with:
          name: source
          path: dist/
      - name: Deploy
        uses: easingthemes/ssh-deploy@v4
        env:
          SOURCE: "dist package.json package-lock.json"
          ARGS: "--delete -az"
          SSH_PRIVATE_KEY: ${{ secrets.SSH_PRIVATE_KEY }}
          REMOTE_HOST: ${{ secrets.REMOTE_HOST }}
          REMOTE_USER: ${{ secrets.REMOTE_USER }}
          TARGET: "<target-directory>"
          SCRIPT_AFTER: |
            cd <target-directory>
            npm i --omit=dev
            <start-command>

where <target-directory> is replaced with the name of the directory where the bot build is stored on the server, and <start-command> with the command to start your bot, which can for example be a call to pm2 or systemctl.

This script sequentially performs two tasks: build and deploy. After build is executed, the artifact of this task, namely the dist directory containing the bot build, is passed to the deploy task.

Files are delivered to the server using the rsync utility, which is implemented by easingthemes/ssh-deploy. After the files are delivered to the server, the command described in the SCRIPT_AFTER environment variable is executed. In our case, after the files are delivered, we go to the bot directory, where we install all the dependencies except devDependencies, and restart the bot.

Note that you need to add three secret environment variables:

SSH_PRIVATE_KEY—this is where the private SSH key you created in the previous step should be stored.
REMOTE_HOST—the IP address of your server should be stored here.
REMOTE_USER—the name of the user on whose behalf the bot is launched should be stored here.

Node.js (GitLab)

Use

yml

image: node:latest

stages:
  - build
  - deploy

Build:
  stage: build
  before_script: npm ci
  script: npm run build
  artifacts:
    paths:
      - dist/

Deploy:
  stage: deploy
  before_script:
    - "command -v ssh-agent >/dev/null || ( apt-get update -y && apt-get install openssh-client -y )"
    - "command -v rsync >/dev/null || ( apt-get update -y && apt-get install rsync -y )"
    - eval $(ssh-agent -s)
    - echo "$SSH_PRIVATE_KEY" | tr -d '\r' | ssh-add -
    - mkdir -p ~/.ssh
    - chmod 700 ~/.ssh
    - ssh-keyscan "$REMOTE_HOST" >> ~/.ssh/known_hosts
    - chmod 644 ~/.ssh/known_hosts
  script:
    - rsync --delete -az dist package.json package-lock.json $REMOTE_USER@$REMOTE_HOST:<target-directory>
    - ssh $REMOTE_USER@$REMOTE_HOST "cd <target-directory> && npm i --omit=dev && <start-command>"

The files are delivered to the server using the rsync utility, which we must install before executing the main script. After the files are delivered, we connect to the server using SSH to run a command to install all dependencies except devDependencies and restart the application.

Note that you need to add three environment variables:

SSH_PRIVATE_KEY—this is where the private SSH key you created in the previous step should be stored.
REMOTE_HOST—the IP address of your server should be stored here.
REMOTE_USER—the name of the user on whose behalf the bot is launched should be stored here.

Deno (GitHub)

Use

yml

name: Main

on:
  push:
    branches:
      - main

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
        with:
          fetch-depth: 0
      - name: Deploy
        uses: easingthemes/ssh-deploy@v4
        env:
          SOURCE: "src deno.jsonc deno.lock"
          ARGS: "--delete -az"
          SSH_PRIVATE_KEY: ${{ secrets.SSH_PRIVATE_KEY }}
          REMOTE_HOST: ${{ secrets.REMOTE_HOST }}
          REMOTE_USER: ${{ secrets.REMOTE_USER }}
          TARGET: "<target-directory>"
          SCRIPT_AFTER: |
            cd <target-directory>
            <start-command>

This script sends files to the server using the rsync utility, which is implemented by easingthemes/ssh-deploy. After the files are delivered to the server, the command described in the SCRIPT_AFTER environment variable is executed. In our case, after the files are delivered, we go to the bot’s directory and restart the bot.

Note that you need to add three secret environment variables:

SSH_PRIVATE_KEY—this is where the private SSH key you created in the previous step should be stored.
REMOTE_HOST—the IP address of your server should be stored here.
REMOTE_USER—the name of the user on whose behalf the bot is launched should be stored here.

Deno (GitLab)

Use

yml

image: denoland/deno:latest

stages:
  - deploy

Deploy:
  stage: deploy
  before_script:
    - "command -v ssh-agent >/dev/null || ( apt-get update -y && apt-get install openssh-client -y )"
    - "command -v rsync >/dev/null || ( apt-get update -y && apt-get install rsync -y )"
    - eval $(ssh-agent -s)
    - echo "$SSH_PRIVATE_KEY" | tr -d '\r' | ssh-add -
    - mkdir -p ~/.ssh
    - chmod 700 ~/.ssh
    - ssh-keyscan "$REMOTE_HOST" >> ~/.ssh/known_hosts
    - chmod 644 ~/.ssh/known_hosts
  script:
    - rsync --delete -az src deno.jsonc deno.lock $REMOTE_USER@$REMOTE_HOST:<target-directory>
    - ssh $REMOTE_USER@$REMOTE_HOST "cd <target-directory> && npm i --omit=dev && <start-command>"

This script sends files to the server using rsync, which has to be installed previously. After the files are copied, we connect to the server using SSH to restart the bot.

Note that you need to add three environment variables:

SSH_PRIVATE_KEY—this is where the private SSH key you created in the previous step should be stored.
REMOTE_HOST—the IP address of your server should be stored here.
REMOTE_USER—the name of the user on whose behalf the bot is launched should be stored here.

You should now see how every code push to the main branch will automatically be deployed to your VPS. Development go brrrrr 🚀

Hosting: VPS ​

Server Rental ​

Starting the Bot ​

systemd ​

Getting the Start Command ​

Creating the Service ​

Managing the Service ​

PM2 ​

Installing ​

Creating an Application ​

Managing the Application ​

Saving the Application Operation ​

Running the Bot on Webhooks ​

Domain Rental ​

Domain Pointing to VPS ​

Setting up a Web Server ​

Installation ​

Configuring ​

Connecting a Webhook to Telegram ​

CI/CD ​

SSH Keys ​

Example Workflows ​

Node.js (GitHub) ​

Node.js (GitLab) ​

Deno (GitHub) ​

Deno (GitLab) ​

Hosting: VPS

Server Rental

Starting the Bot

systemd

Getting the Start Command

Creating the Service

Managing the Service

PM2

Installing

Creating an Application

Managing the Application

Saving the Application Operation

Running the Bot on Webhooks

Domain Rental

Domain Pointing to VPS

Setting up a Web Server

Installation

Configuring

Connecting a Webhook to Telegram

CI/CD

SSH Keys

Example Workflows

Node.js (GitHub)

Node.js (GitLab)

Deno (GitHub)

Deno (GitLab)