[RUBY] Launch Rails on EC2


The procedure for "using the AWS server" is written in 5 parts.

Please check the articles below.

Initial setting of EC2Create a database in production environmentStart Rails on EC2 ← Imakoko ④ Set up the Web serverAutomate deployment

Settings for launching Rails on EC2

Run in terminal (in EC2)

[[email protected]***-**-**-*** ~]$ ssh-keygen -t rsa -b 4096
Generating public/private rsa key pair.
Enter file in which to save the key (/home/ec2-user/.ssh/id_rsa):
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/ec2-user/.ssh/id_rsa.
Your public key has been saved in /home/ec2-user/.ssh/id_rsa.pub.
The key fingerprint is:
3a:8c:1d:d1:a9:22:c7:6e:6b:43:22:31:0f:ca:63:fa [email protected]
The key's randomart image is:
+--[ RSA 4096]----+
|    +            |
| . . =           |
|  = . o .        |
| * o . o         |
|= *     S        |
|.* +     .       |
|  * +            |
| .E+ .           |
| .o              |

Next, copy the value of the ssh public key displayed by the "cat command".

By executing the "cat command", you can check the specific contents of the file on the terminal.

[[email protected]***-**-**-*** ~]$ cat ~/.ssh/id_rsa.pub
ssh-rsa AAAAB3NzaC1yc2E......

Next, register the public key displayed by the cat command in GitHub. Access the following URL


Fill in "Title" Paste the public key into "Key" Press "ADD SSH Key" to save

After registering the key on GitHub, check if you can connect with ssh with the following command

[[email protected]***-**-**-*** ~]$ ssh -T [email protected]
Hi <Github username>! You've successfully authenticated, but GitHub does not provide shell access.

If "Permission denied (public key)." Is displayed, The ssh key setting is incorrect, so check the work again.

Application server settings

"Application server" accepts "requests" from browsers Software that actually runs Rails applications

You also need to run the application server on the EC2 server. The necessary tools for that are ** `ʻUnicorn`` **

"Unicorn" is an application server that is often used on servers that are open to the world. It can be started by "unicorn_rails command" instead of rails s command.

Install Unicorn

Add the following contents to the bottom of the Gemfile and bundle install

group :production do
  gem 'unicorn', '5.4.1'

Create and edit "unicorn.rb"

Create "config / unicorn.rb" and edit the contents as follows


#Put the directory where the application code on the server is installed in a variable
app_path = File.expand_path('../../', __FILE__)

#Determine application server performance
#Worker that accepts requests and generates responses(worker)Determine the number of
worker_processes 1

#Specify the directory where the application is installed
#When Unicorn runs Rails code, it specifies a directory to look for files that it actually references, such as routing.
working_directory app_path

#Specify the location of the files required to start Unicorn
#Unicorn will generate a file with the process id when it starts. Specify the location.
pid "#{app_path}/tmp/pids/unicorn.pid"

#Specify the port number
#Decide which port number you want to accept requests. This time, specify port 3000.
listen 3000

#Specify a file to log errors
stderr_path "#{app_path}/log/unicorn.stderr.log"

#Specify the file to record the normal log
stdout_path "#{app_path}/log/unicorn.stdout.log"

#Set maximum time to wait for Rails application response
timeout 60

#The following is an applied setting, so the explanation is omitted.

preload_app true
GC.respond_to?(:copy_on_write_friendly=) && GC.copy_on_write_friendly = true

check_client_connection false

run_once = true

before_fork do |server, worker|
  defined?(ActiveRecord::Base) &&

  if run_once
    run_once = false # prevent from firing again

  old_pid = "#{server.config[:pid]}.oldbin"
  if File.exist?(old_pid) && server.pid != old_pid
      sig = (worker.nr + 1) >= server.worker_processes ? :QUIT : :TTOU
      Process.kill(sig, File.read(old_pid).to_i)
    rescue Errno::ENOENT, Errno::ESRCH => e
      logger.error e

after_fork do |_server, _worker|
  defined?(ActiveRecord::Base) && ActiveRecord::Base.establish_connection

Reflect your edits in the remote repository

"Commit → push" to the remote repository If it can be reflected in the remote repository of GitHub, reflect all the local contents so far in EC2.

#Create a new directory with the mkdir command
[[email protected]***-**-**-*** ~]$ sudo mkdir /var/www/

#Ec2 permissions on the created www directory-Change to user
[[email protected]***-**-**-*** ~]$ sudo chown ec2-user /var/www/

Next, get the "repository URL" from GitHub and clone it. You can copy the URL by pressing the green "Code". Image from Gyazo

Once you have copied the URL, clone the code

[[email protected]***-**-**-*** ~]$ cd /var/www/
[[email protected]***-**-**-*** www]$git clone paste the copied URL

Success if there is no error in the above input.

Production settings

The EC2 instance that is currently running does not have enough computer power, An error may occur when installing Gem. Specifically, there is not enough memory related to the processing power of the computer. This is the limit of instances that can be run for free, so it can't be helped ...

Therefore, before making future settings, a process called "Swap file" is performed to increase the memory.

A "Swap file" is a file prepared to temporarily increase the amount of memory. EC2 does not prepare a Swap file by default, so preparing it will prevent memory shortage processing errors.

Increase the area of Swap files

Make an ssh connection to EC2 and move to your home directory Run in terminal (in EC2)

[[email protected]***-**-**-*** ~]$ cd 

Execute the following commands in order

[[email protected]***-**-**-*** ~]$ sudo dd if=/dev/zero of=/swapfile1 bs=1M count=512

#Wait for a while and succeed if the following is displayed
512+0 record input
512+0 record output
536870912 bytes(537 MB)Copied, 7.35077 seconds, 73.0 MB/Seconds
[[email protected]***-**-**-*** ~]$ sudo chmod 600 /swapfile1
[[email protected]***-**-**-*** ~]$ sudo mkswap /swapfile1

#Success if the following is displayed
Set swap space version 1, size= 524284 KiB
No label, UUID=74a961ba-7a33-4c18-b1cd-9779bcda8ab1
[[email protected]***-**-**-*** ~]$ sudo swapon /swapfile1
[[email protected]***-**-**-*** ~]$ sudo sh -c 'echo "/swapfile1  none        swap    sw              0   0" >> /etc/fstab'

This completes allocating the Swap file area.

Install gem in EC2

Install the gem required to launch the application cloned on EC2 with the following command

Run in terminal (in EC2)

#Move to the cloned directory
[[email protected]***-**-**-*** www]$ cd  /var/www/Application under development

#Check the version of ruby
[[email protected]***-**-**-*** <Repository name>]$ ruby -v
ruby 2.6.5p114 (2019-10-01 revision 67812) [x86_64-linux]

Next, install bundler to manage gems in production environment. First, check which version of bundler was used in the locally developed application.

Run in terminal (local)

#Run in the directory of the application under development

% bundler -v
Bundler version 2.1.4 ← Example

Introduced the same version of bundler as local on the EC2 side. In the above case, install the version of "bundler 2.1.4" and execute bundle install.

Run in terminal (in EC2)

# 「2.1.In the part of "4", the version of bundler confirmed locally is introduced.
[[email protected]***-**-**-*** <Repository name>]$ gem install bundler -v 2.1.4

#The following commands may take several minutes or more to process
[ec2[email protected]***-**-**-*** <Repository name>]$ bundle install

Set environment variables

The value of the environment variable can be used from Rails by writing "ENV ['']". This time, specify the environment variable "SECRET_KEY_BASE".

"Secret_key_base" is a character string used to encrypt cookies. You need to prepare it when you run a Rails application. Also, since it is a value that should not be leaked to the outside, this is also referred to from the environment variable.

First, create secret_key_base with the following command

Run in terminal (in EC2)

# 「2.1.In the part of "4", the version of bundler confirmed locally is introduced.
[[email protected]***-**-**-*** <Repository name>]$ rake secret

Next, set environment variables in EC2. Environment variables are saved in a file called "/ etc / environment". Use the vim command to write environment variables. At that time, use the option "sudo".

"Sudo" is an option with the role of "execute a command with all privileges". Since "files under / etc" cannot be written or saved without strong authority, add "sudo" to the beginning of the command.

Run in terminal (in EC2)

# 「2.1.In the part of "4", the version of bundler confirmed locally is introduced.
[[email protected]***-**-**-*** ~]$ sudo vim /etc/environment

Type "i" to enter input mode. After that, it is described as follows.

#Enter the password for the database root user (see the article from the link below)

DATABASE_PASSWORD='Database root user password'
SECRET_KEY_BASE='The secret created earlier_key_base'

After completing the input, execute in the order of "esc key" → ": wq" and save. Log out of the production environment once to reflect the set environment variables.

Run in terminal (in EC2)

# 「2.1.In the part of "4", the version of bundler confirmed locally is introduced.
[[email protected]***-**-**-*** ~]$ exit
Connection to 52.xx.xx.xx closed.

Next, check if the environment variables you have just set are reflected properly. Connect to EC2 again with the following command.

Terminal (local)

$ ssh -i [The name of the downloaded key].pem ec2-user@[Elastic IP associated with the created EC2 instance]

After reconnecting, combine the "env command" and "grep command" to Make sure that the environment variables you set earlier are applied.

env command: Command to display the set environment variables grep command: Command used to search environment variables, etc. The usage is "grep (search target)".

Run in terminal (in EC2)

# 「2.1.In the part of "4", the version of bundler confirmed locally is introduced.
[[email protected]***-**-**-*** ~]$ env | grep SECRET_KEY_BASE

[[email protected]***-**-**-*** ~]$ env | grep DATABASE_PASSWORD
DATABASE_PASSWORD='Database root user password'

Release port

The EC2 instance you just launched can be accessed with ssh, It is not connected at all by other communication methods such as HTTP. Therefore, the EC2 instance used as a server needs to open the "port" so that HTTP can be connected in advance. Earlier, I wrote "listen 3000" in config / unicorn.rb, This means "start the Rails server on port 3000".

In order to set the port, it is necessary to change the setting called "security group" of EC2.

Set security group port

Select the target instance from the EC2 instance list screen and select it. Click the "Security Group" link ("launch-wizard-2" in the figure) at the bottom right of the figure Image from Gyazo](https://gyazo.com/cef10db71b1cb250b5fcc8851bc46d1b)

Then click "Edit Inbound Rule" in the "Inbound" tab Image from Gyazo

After the screen transition, click "Add rule" and edit as follows Type: Custom TCP Protocol: TCP Port range: 3000 Source: Custom / Image from Gyazo

After setting, click "Save Rule". That's all for opening the port.

Launch Rails in production

Use the "unicorn_rails command" to start Rails in a production environment.

First, edit the production environment settings of "database.yml" with VS Code. Edit the local database.yml as follows to match the mysql settings in the production environment.

config/database.yml (local)

  <<: *default
  database:(* Do not edit here)
  username: root
  password: <%= ENV['DATABASE_PASSWORD'] %>
  socket: /var/lib/mysql/mysql.sock

Next, edit "commit → push"

Next, reflect the contents of GitHub on EC2 Execute the following command in the terminal (in EC2)

[[email protected]***-**-**-*** <Repository name>]$ git pull origin master

Next, create a database in EC2. There is an option "RAILS_ENV = production". RAILS_ENV = production: Options that are attached when executing commands in the production environment.

Run in terminal (in EC2)

[[email protected]***-**-**-*** <Repository name>]$ rails db:create RAILS_ENV=production
Created database '<Database name>'
[[email protected]***-**-**-*** <Repository name>]$ rails db:migrate RAILS_ENV=production

Once you've done that, start Rails.

Run in terminal (in EC2)

[[email protected]***-**-**-*** <Repository name>]$ cd /var/www/[Repository]
[[email protected]***-**-**-*** <Repository name>]$ bundle exec unicorn_rails -c config/unicorn.rb -E production -D

Go to http: // : 3000 / with your browser and check if the site is displayed.

When I check it with a browser, the view remains broken. An error message (We're sorry, but something went wrong.) May be displayed.

This is not good, so organize the "asset files".

Asset file: A file that manages images, CSS, JavaScript, etc. The process of compressing this asset file and transferring the data is called "compilation". If you do not do this, the CSS will not be reflected in the production environment and the view will collapse, Problems such as the browser not being displayed due to an error occur.

Compile asset files

Run in terminal (in EC2)

[[email protected]***-**-**-*** <Repository name>]$ rails assets:precompile RAILS_ENV=production

When I'm done so far, I started Rails again, but since the server is already up, I need to restart Rails.

Follow the steps below to restart Rails in production

First, check "Unicorn process". Use the "ps command" to check the process from the terminal. Then kill the process with the "kill command".

ps command: A command to check the currently running process. kill command: A command to stop the currently running process.

Run in terminal (in EC2)

[[email protected]***-**-**-*** <Repository name>]$ ps aux | grep unicorn

Then you should see the process as below.

#### **`Terminal`**

ec2-user 17877  0.4 18.1 588472 182840 ?       Sl   01:55   0:02 unicorn_rails master -c config/unicorn.rb -E production -D
ec2-user 17881  0.0 17.3 589088 175164 ?       Sl   01:55   0:00 unicorn_rails worker[0] -c config/unicorn.rb -E production -D
ec2-user 17911  0.0  0.2 110532  2180 pts/0    S+   02:05   0:00 grep --color=auto unicorn

Above "ec2-The number to the right of "user" is the process id (PID) 「unicorn_The process labeled "rails master" is the Unicorn process body The process id at this time is "17877"

Terminal (in EC2))Run on

 [ec2-user @ ip-***-**-**-*** <repository name>] $ kill <confirmed unicorn rails master process id> ← This time 17878

View the executed process again and confirm that it has been terminated.

 [ec2-user @ ip-***-**-**-*** <repository name>] $ ps aux | grep unicorn
ec2-user 17911  0.0  0.2 110532  2180 pts/0    S+   02:05   0:00 grep --color=auto unicorn

If the execution result is as above, the stop of Unicorn is completed.

###Finally, run the command to restart Rails 「RAILS_SERVE_STATIC_FILES="1" has the role of specifying Rails to find the compiled asset.

Terminal (in EC2))Run on

 [ec2-user @ ip-***-**-**-*** <repository name>] $ RAILS_SERVE_STATIC_FILES = 1 unicorn_rails -c config / unicorn.rb -E production -D

Http in browser://<Elastic IP>:3000/Go to and check if the site is displayed.

##When Rails doesn't start well unicorn_When running rails 「master failed to start, check stderr log for details」 If you get, you need to check the error log of unicorn.

Use something called "less command" when viewing log files

less command: A command that allows you to check the contents of a file. The "cat command" has a similar role.

Terminal (in EC2))Run on

 [[email protected]***-**-**-*** <repository name>] $ less log / unicorn.stderr.log
I, [2016-12-21T04:01:19.135154 #18813]  INFO -- : Refreshing Gem list
I, [2016-12-21T04:01:20.732521 #18813]  INFO -- : listening on addr= fd=10
E, [2016-12-21T04:01:20.734067 #18813] Mysql2::Error::ConnectionError: Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (2)
/var/www/furima/shared/bundle/ruby/2.6.0/gems/mysql2-0.5.3/lib/mysql2/client.rb:90:in `connect'
/var/www/furima/shared/bundle/ruby/2.6.0/gems/mysql2-0.5.3/lib/mysql2/client.rb:90:in `initialize'
/var/www/furima/shared/bundle/ruby/2.6.0/gems/activerecord- `new'
/var/www/furima/shared/bundle/ruby/2.6.0/gems/activerecord- `mysql2_connection'
/var/www/furima/shared/bundle/ruby/2.6.0/gems/activerecord- `new_connection'

##「We’re sorry,but ~ "is displayed First of all, "production.Check "log".

production.log: A place to record server logs, "a place to record events in EC2"

Terminal (in EC2))Run on

 [[email protected]***-**-**-*** <repository name>] $ less log / production.log

 (Display production.log)

####tail -use the f command tail -f command: A command that displays the latest log for only 10 lines. It is a tool that allows you to check the log easily, so use it as much as possible.

that's all. Thank you for your hard work.

Continue →Here

Recommended Posts

Launch Rails on EC2
Launch Rails on EC2 (manual deployment)
Deploy RAILS on EC2
Tailwind on Rails
EC2 on Docker-compose
Launch docker container on EC2 (personal memorandum)
Test run on rails
Ruby on Rails Elementary
Ruby on Rails basics
Yay! I'm on Rails!
Ruby On Rails Association
Recipe for deploying Rails apps on AWS EC2
[Rails] Launch a new project
Resolve ActiveRecord :: NoDatabaseError on rails6
rails new application launch command
Launch a terminal on VScode
Ruby on rails learning record -2020.10.03
Portfolio creation Ruby on Rails
Install Sidekiq pro on Rails
Ruby on rails learning record -2020.10.04
[Ruby on Rails] Debug (binding.pry)
Ruby on rails learning record -2020.10.05
Ruby on rails learning record -2020.10.09
Ruby on Rails config configuration
Ruby on Rails basic learning ①
[Ruby on Rails] about has_secure_password
Ruby on rails learning record-2020.10.07 ②
Yay! I'm on Rails! Repetition
Install Rails on macOS Catalina
Commentary on partial! --Ruby on Rails
Post a video on rails
Install docker on AWS EC2
[Rails] EC site cart function
Ruby on rails learning record-2020.10.07 ①
Cancel Ruby on Rails migration
Ruby on rails learning record -2020.10.06
Ruby on Rails validation summary
Ruby on Rails Basic Memorandum
Launch Nuxt.js + Rails API on Docker and try CRUD operation
How to deploy jQuery on Rails
[Ruby on Rails] Read try (: [],: key)
[Ruby on Rails] yarn install --check-files
Ruby on Rails variable, constant summary
Installing Ruby + Rails on Ubuntu 18.04 (rbenv)
[Ruby on Rails] Introduced paging function
Basic knowledge of Ruby on Rails
Progate Ruby on Rails5 Looking Back
[Rails] DB design for EC site
React.js CSS in Rails app deployed on EC2 cannot be loaded
How to use Ruby on Rails
How to deploy Bootstrap on Rails
[Ruby on Rails] Add / Remove Columns
Ruby on Rails Japanese-English support i18n
Error when deploying EC2 on CircleCI
(Ruby on Rails6) "Erase" posted content
[Ruby on Rails] CSV output function
Ruby on Rails 6.0 environment construction memo
[Ruby on Rails] What is Bcrypt?
Rails on Tiles (how to write)
Rails on Docker environment construction procedure
[Ruby on Rails] Confirmation page creation