Emmanuel Corrales

How to remove Git submodules

Fri, 11 Oct 2019 00:30:00 +0000

Here’s a simple guide on how to remove Git submodules. Lets say I want to remove a Git submodule called drafts.

Edit .gitmodules to remove lines that are similar to the example below.

[submodule "drafts"]
  path = drafts
  url = git://github.com/emcorrales/drafts.git

Stage the .gitmodules file.
```
git add .gitmodules
```
Remove the submodule from the git index. Ensure that there is no trailing slash. We’re deleting the reference to the submodule and not the directory.
```
git rm --cached drafts
```
Remove the untracked files left by the submodule.
```
rm -rf drafts
```

Commit changes.

git commit -m "Remove draft submodule.".

There are still references to the drafts submodule. Edit .git/config and remove lines that are similar to the example below.
```
[submodule "drafts"]
  url = git://github.com/emcorrales/drafts.git
```
Remove the copy of the submodule’s git repo.
```
rm -rf .git/modules/drafts
```
There you have it, you have succesfully removed the submodule from your local git repo.

How to use a self signed certificate with Gradle

Mon, 29 Jul 2019 00:30:00 +0000

Lately I’ve been getting a lot of issues regarding self-signed certificates with Java apps using Gradle as their build tool. The issues were mostly errors like “Exception in thread “main” javax.net.ssl.SSLHandshakeException” or packages that can’t be downloaded because of an invalid certificate. This was caused by Gradle not importing the corporate certificiate pre-installed on my client’s company issued laptop by default. This post is a guide about using a self signed certificate with Gradle.

First find out where the certificate is stored. The certificate can be in any directory but if using OpenSSL it is good practice to put the certificate on its default directory. Depending on the operating system, the default directory may vary.

$ openssl version -d
OPENSSLDIR: "/private/etc/ssl"

Export the self signed certificate to the Java keystore.

keytool -import -trustcacerts -alias root -file /private/etc/ssl/certs.pem -keystore $JAVA_HOME/jre/lib/security/cacerts -storepass changeit

The default trust store is located at $JAVA_HOME/jre/lib/security/cacerts. On the project’s root directory create or edit the gradle.properties file and add the line below.

org.gradle.jvmargs=-Djavax.net.ssl.keyStore="$JAVA_HOME/jre/lib/security/cacerts" -Djavax.net.ssl.keyStoreType=KeychainStore -Djavax.net.ssl.keyStorePassword=changeit

If you have properly done the steps above then there should no longer be any problem when running Gradle.

How to enable internet access of CentOS 7 on VirtualBox

Thu, 13 Jun 2019 15:30:33 +0000

I was setting up a CentOS 7 virtual machine using VirtualBox to replicate an existing setup when I discovered that I cannot access the internet even though my VM was setup to use a bridged adapter.

Diagnosing the problem

I wanted to know the status of network interfaces on my virtual machine so I execute the nmcli command.

$ nmcli
enp0s3: disconnected
        "Intel 82540EM"
        1 connection available
        ethernet (e1000), 08:00:27:A7:B4:52, autoconnect, hw, mtu 1500

lo: unmanaged
        "lo"
        loopback (unknown), 00:00:00:00:00:00, sw, mtu 65536

Use "nmcli device show" to get complete information about known devices and
"nmcli connection show" to get an overview on active connection profiles.

Consult nmcli(1) and nmcli-examples(5) manual pages for complete usage details.

The command shows that enp0s3 is diconnected so I checked its network configuration script.

$ cat /etc/sysconfig/network-scripts/ifcfg-enp0s3
TYPE=Ethernet
PROXY_METHOD=none
BROWSER_ONLY=no
BOOTPROTO=dhcp
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_FAILURE_FATAL=no
IPV6_ADDR_GEN_MODE=stable-privacy
NAME=enp0s3
UUID=1056c9af-dc47-429b-958c-d0e83b99d60f
DEVICE=enp0s3
ONBOOT=no

I found out that the reason I have no access to the internet is because it is not configured to be activated on boot. The last line ONBOOT=no means that this network interface won’t be activated on startup.

Configuring the network interface

To resolve this I opened a text editor to edit the network configuration and changed the last line to ONBOOT=yes.

sudo vi /etc/sysconfig/network-scripts/ifcfg-enp0s3

I have to restart the network and apply the changes.

sudo systemctl restart network.service

Now if I check the interface, it would show that it is connected.

$ nmcli
enp0s3: connected to enp0s3
        "Intel 82540EM"
        ethernet (e1000), 08:00:27:A7:B4:52, hw, mtu 1500
        ip4 default
        inet4 10.0.2.15/24
        route4 0.0.0.0/0
        route4 10.0.2.0/24
        inet6 fe80::f63f:8611:805b:579d/64
        route6 fe80::/64
        route6 ff00::/8

lo: unmanaged
        "lo"
        loopback (unknown), 00:00:00:00:00:00, sw, mtu 65536

DNS configuration:
        servers: 17.83.200.200 17.83.200.202
        domains: apple.com
        interface: enp0s3

Use "nmcli device show" to get complete information about known devices and
"nmcli connection show" to get an overview on active connection profiles.

Consult nmcli(1) and nmcli-examples(5) manual pages for complete usage details.

Testing access to the internet

Now to check if my virtual machine has access to the internet.

$ ping google.com
PING google.com (74.125.24.138) 56(84) bytes of data.
64 bytes from 74.125.24.138 (74.125.24.138): icmp_seq=1 ttl=41 time=71.7 ms
64 bytes from 74.125.24.138 (74.125.24.138): icmp_seq=2 ttl=41 time=73.8 ms
64 bytes from 74.125.24.138 (74.125.24.138): icmp_seq=3 ttl=41 time=73.1 ms

The command above shows that my virtual machine can now successfully connect to the internet.

NodeJS: Debugging and fixing HTTPS/SSL issues caused by changing the self signed certificate.

Tue, 30 Apr 2019 00:30:00 +0000

This past month I’ve been bombarded with lots of issues regarding HTTPS/SSL. The internal Certificate Authority of the company I’m servicing recently changed their root certicate. This caused the services that communicate with other services in their network through HTTPS/SSL to break and receive errors. I’ve been debugging a NodeJS app that is being plagued by these issues. In this post I’ll show you the steps I took to fix this.

The image above shows that the NodeJS app uses an API on their internal network. The NodeJS app is the client of the third party API and is having trouble getting a response through HTTPS even though the admin has installed the latest certificate on every server’s trust store and is ready for use.

Checking if the certificate is installed on the server.

The first thing I did is to check if the new certificate is installed on the server of the client app. I looked for the directory where the certificates are stored:

$ openssl version -d
OPENSSLDIR: "/usr/lib/ssl"

then I looked for the certifcate in that directory. Below is the command I used to look for a certificate called example.pem.

ls /usr/lib/ssl/certs | grep example.pem

Verifying the connection to the third party API.

I needed to figure out if there is a problem with the new certificate. The first thing that comes to my mind is to check if I can establish a HTTPS connection with the third party API so I executed this:

openssl s_client -connect api.example.com:443

Once I verified the HTTPS/SSL connection with the third party API, I proceeded to running the client app.

Running the client app.

The Node app I was debugging is too big and I just need to test the part that uses the API so I created a simple app that uses the same API. This will serve as proxy. Below are the contents of the file client.js. A simple NodeJS app that uses the third party API.

// client.js
const https = require('https');

https.get("https://api.example.com", (res) => {
  console.log(res.statusCode);
  if (res.statusCode !== 200) {
    throw new Error(`Expected 200, got ${res.statusCode}`);
  }

  res.on('data', data => process.stdout.write(data));
});

When I run the app I get error messages like these:

$ node client.js
Error: unable to verify the first certificate
    at TLSSocket.onConnectSecure (_tls_wrap.js:1036:34)
    at TLSSocket.emit (events.js:159:13)
    at TLSSocket._finishInit (_tls_wrap.js:637:8)

The app is having verification errors with the API’s certificate because Node doesn’t load the corresponding certificate that can verify it when creating a HTTPS request. The new certificate is installed on the server’s trust store but Node doesn’t use it because it already has a default list of CA’s built into its source. Thus, while it is true that it doesn’t do a lookup on the system hosted CA’s and that there is no “store” per se, there is a default list of CA’s that it accepts.

In order to use the new certificate, the app must be configured to load the CA upon deployment by passing the environment variable NODE_EXTRA_CA_CERTS like this:

NODE_EXTRA_CA_CERTS=/usr/lib/ssl/certs node client.js Hello, world

This time the app will no longer show an error because it knows where to find the new certificate.

How to properly debug shell scripts

Sun, 24 Feb 2019 22:30:00 +0000

I recently learned the proper way to debug shell scripts. It seems shell scripting has built-in features that can help developers find problems in their script. In this post I will be sharing how I debug scripts before and what is wrong with it while also showing the proper way to do it.

Printing commands upon execution.
Running the entire script on debug mode.
Terminating the entire script upon error.

Printing commands upon execution

Before, to debug a script I’m developing, I would do it like this.

echo "ls pictures"
ls pictures

This prints a message before the command is executed. This worked out well for me before however this means I have to run echo for every command that I want to debug. What if I am debugging multiple consecutive lines of a script? Suppose I have a script called foo.sh with contents like these:

#!/bin/bash
pwd
whoami
ls pictures
uname

If I’m just interested to see if the first three commands, pwd, whoami and ls are being executed then with my previous way of debugging I would edit its contents to look like this.

#!/bin/bash
echo "pwd"
pwd
echo "whoami"
whoami
echo "ls pictures"
ls pictures
uname

Thats a lot of echo and it looks tedious even though I’m just debugging the first three commands.

A better way to do this is to execute

set -x

before the commands I want to print. This command prints all the commands upon execution in the script until it is disabled by executing

set +x

The contents of foo.sh would look like this

#!/bin/bash
set -x # Start printing commands upon execution.
pwd
whoami
ls pictures
set +x # Stop printing commands upon execution.
uname

This is better because I only need to execute two commands to print all the commands being executed and I don’t have to execute echo which is another command that may potentially introduce some bugs upon execution of the script.

Running the entire script on Debug mode.

I can run the entire script on debug mode upon its execution from the command line if I pass the parameter “-x” to the script’s shebang which is always at the first line. Executing the command “set -x” is no longer necessary. The script foo.sh would look like this.

#!/bin/bash -x
pwd
whoami
ls pictures
uname

Terminating the script on error

There are times that if a command the script is executing fails, I want it to terminate the script immediately, preventing the execution of subsequent commands.

Suppose that I run the script foo.sh and the pictures directory doesn’t exist. The command “ls pictures” will not execute properly and will result an error but will still execute the uname command. If I don’t want to execute uname if an error occurs, I will insert the line below after the ls command.

[ ! $? -eq 0 ] && exit 1

and the script to to look like this

#!/bin/bash
set -x # Start printing commands upon execution.
pwd
whoami
ls pictures
[ ! $? -eq 0 ] && exit 1
uname
set +x # Stop printing commands upon execution.

This is how I used to do it but this also has the same problem from the previous section. Do I have to add this statement after every command that should terminate the script upon error? That is a tedious task, fortunately I can just add another parameter “-e” on the set command to achieve the same result and the script would look like this.

#!/bin/bash
set -xe # Start printing commands upon execution.
pwd
whoami
ls pictures
set +xe # Stop printing commands upon execution.
uname

I can also add “-e” as a parameter for the script’s shebang if I want the script to exit upon error of any command like this:

#!/bin/bash -xe
pwd
whoami
ls pictures
uname

That’s it! These are the better and proper ways I learned to debug my scripts.

Installing Oracle JDK for MacOS via Homebrew

Tue, 05 Feb 2019 00:30:00 +0000

MacOS has OpenJDK installed by default however I prefer to use Oracle’s version of JDK because its the official version. I don’t want to install it the same way Oracle instructs it on their docs as I find it very tedious. I’m a guy who loves automating stuff so I prefer to install it via Homebrew. I frequently do a clean install on my Mac every time there is a new version of OSX so I have to install JDK again and again. I’d rather just run a single installation script instead of heading over to Oracle’s website and following their instructions.

Steps to install and configure the Oracle JDK:

Install Homebrew and Cask
Install Oracle JDK
Setup the JAVA_HOME environment variable
Verifying installation

Homebrew and Cask

Homebrew is a package manager for Mac and has always been my preferred way to install my command line tools because I can integrate it with my setup scripts. To install it I’ll run

ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

then I’ll install Homebrew Cask which is an extension of Homebrew. It makes the installation of large binaries and graphical applications simpler.

brew tap caskroom/versions

JDK Installation

Before I install the JDK, I’ll check first which version it will install by default. I’m very picky about the version because most of the time I just use Java for Android development. I also prefer the older and more stable version of JDK so I run

brew cask info java

which will output

java: 11.0.2,9
https://jdk.java.net/
Not installed
From: https://github.com/Homebrew/homebrew-cask/blob/master/Casks/java.rb
==> Name
OpenJDK
==> Artifacts

This means that the latest version is JDK 11. I can install it now by running

brew cask install java

but I prefer to install JDK 8 over 11 so instead I’ll run

brew cask install java8

Setup Java_HOME environment variable

Once installed, I will set the JAVA_HOME environment variable by editing my .bash_profile

vim ~/.bash_profile

and inserting this line

export JAVA_HOME=$(/usr/libexec/java_home)

and applying these changes by running

source ~/.bash_profile

Verifying Installation

Now to confirm if the installation was sucessful I’ll run this command.

java -version

If the installation is successful, its output would be similiar to this

java version "1.8.0_202"
Java(TM) SE Runtime Environment (build 1.8.0_202-b08)
Java HotSpot(TM) 64-Bit Server VM (build 25.202-b08, mixed mode)

This tells me that I have installed the Oracle version of the JDK. However if the output is like this

java version "1.8.0_202"
OpenJDK Runtime Environment (build 1.8.0_202-b08)
OpenJDK 64-Bit Server VM (build 25.202-b08, mixed mode)

then I may have failed to install the JDK properly or the changes may not have been applied yet because I can see that OpenJDK is still being used. I’ll try to fix this by restarting my Mac then running “java -version” again.

Automating installation with a script

Below is a simple script to automate the installation of the latest Oracle JDK.

That’s it! Now you can automate your JDK installation on you Mac by running the script.

How I install and configure Docker for my Mac via Homebrew

Tue, 29 Jan 2019 02:30:00 +0000

This is the best way I can come up with to install Docker on my Mac. I never liked manually downloading the “.dmg” file and then moving it to the Applications directory. I will be sharing in this post how I installed it via Homebrew while also configuring bash completion for Docker.

Install Homebrew and Cask

Homebrew is a package manager for Mac and has always been my preferred way to install my tools because I can integrate it with my setup scripts. To install it execute

ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

then install Homebrew Cask which is an extension of Homebrew. It makes the installation of large binaries and graphical applications simpler.

brew tap caskroom/versions

Install Docker

To install Docker via Homebrew run

brew cask install docker

then after installation run

docker --version

to confirm its installation.

Setup Bash auto completion

Install bash completion via Homebrew.

brew install bash-completion

Create symlinks for Docker’s completion on Bash.

etc=/Applications/Docker.app/Contents/Resources/etc
ln -s $etc/docker.bash-completion $(brew --prefix)/etc/bash_completion.d/docker
ln -s $etc/docker-machine.bash-completion $(brew --prefix)/etc/bash_completion.d/docker-machine
ln -s $etc/docker-compose.bash-completion $(brew --prefix)/etc/bash_completion.d/docker-compose

Add this line to your ~/.bash_profile.

[ -f /usr/local/etc/bash_completion ] && . /usr/local/etc/bash_completion

Apply the changes by running

source ~/.bash_profile

Thats it! Now you have installed Docker while also enabling completion for Bash. If you know a better way to install it please don’t hesitate to comment below.

How I redirected all the web traffic for my old domain to the new one

Tue, 15 Jan 2019 00:30:00 +0000

I recently decided to change the domain name of my personal website from emmanuelcorrales.com to emcorrales.com because it is shorter and easier to remember. I don’t want visitors of emmanuelcorrales.com to receive a 404 Not Found or any other error so I redirected all the traffic to emcorrales.com. In this article I’ll be sharing the steps I took to accomplish this.

Steps to redirect web traffic from old domain to the new one:

Launch a server for my old domain.
Setup virtual hosts for the old domain with Apache.
Configure the virtual host to redirect to the new domain.
Enable HTTPS for Apache
Install Let’s Encrypt SSL Self Signed Certificate for Apache

Launch a server for my old domain

To assign a server to my old domain, I launched a new instance of an Ubuntu Server then configured the DNS A records of emmanuelcorrales.com which was hosted at GoDaddy to point to the server’s external IP address. This article will not cover how I configured my DNS records to point to my server since every registrar provides a unique way of changing their DNS records. You must already have knowledge on how to configure the DNS with your registrar.

Setup virtual hosts for the old domain with Apache

Next I created a virtual host for emmanuelcorrales.com using the same steps from a guide I recently wrote.

Configure the virtual host to redirect to the new domain.

I configured the virtual host for emmanuelcorrales.com to redirect to https://emcorrales.com by editing /etc/apache2/sites-available/emmanuelcorrales.com.conf

sudo vim /etc/apache2/sites-available/emmanuelcorrales.conf

to look like this

<VirtualHost *:80>
    ServerAdmin admin@emmanuelcorrales.com
    ServerName emmanuelcorrales.com
    ServerAlias emmanuelcorrales.com
    DocumentRoot /var/www/emmanuelcorrales.com
    ErrorLog ${APACHE_LOG_DIR}/error.log
    CustomLog ${APACHE_LOG_DIR}/access.log combined
    RedirectMatch permanent ^/(.*)$ https://emcorrales.com/$1
</VirtualHost>

The important part to note here is the line with the RedirectMatch. This line is the configuration I need to redirect traffic from emmanuelcorrales.com to https://emcorrales.com.

I restarted the Apache web server to apply the changes I’ve made to the configuration.

sudo service apache2 restart

After restarting, visiting http://emmanuelcorrales.com will redirect me to https://emcorrales.com.

Enable HTTPS for Apache

At this point HTTPS request is still not supported and visiting https://emmanuelcorrales.com causes an error on my browser. To support HTTPS request I configured Apache to enable SSL for my old domain by running the following commands:

sudo a2enmod ssl
sudo a2ensite default-ssl.conf
sudo service apache2 restart

After that I was able to access https://emmanuelcorrales.com through my browser. However, the browser warns me that my connection is not private because I have an invalid certificate. To remove this, I needed to install a certificate issued by a trusted Certificate Authority but unfortunately I don’t want to buy one so I installed the free certificate provided by Let’s Encrypt.

Install Let’s Encrypt SSL Self Signed Certificate for Apache

I cloned the repo of Let’s Encrypt to a directory on my server using Git.

sudo apt-get -y install git
cd /usr/local
sudo git clone https://github.com/letsencrypt/letsencrypt

Then I generated an SSL certifacte for my old domain by running the following commands below and entering all the required credentials.

cd /usr/local/letsencrypt
sudo ./letsencrypt-auto --apache -d emmanuelcorrales.com

After this the certificate was installed at /etc/letsencrypt/live and I confirmed the status of my SSL certificate by visiting the link below.

https://www.ssllabs.com/ssltest/analyze.html?d=emmanuelcorrales.com&latest

And that’s it. I have successfully redirected all urls from my old domain to the new one. If you know a better way to do this please don’t hesitate to approach me.

Apache: Setup a virtual host on Ubuntu 18.04 LTS

Wed, 09 Jan 2019 00:30:00 +0000

This guide is about setting up a virtual host on Ubuntu 18.04 LTS with Apache as the web server. We’ll assume that we want to host a simple site where the domain name will be example.com.

Install the Apache Web Server.

sudo apt-get update
sudo apt-get install apache2

Create a directory called example.com inside the /var/www and make sure that the current user can access it by changing its owner and permission.

sudo mkdir -p /var/www/example.com
sudo chown -R $USER:$USER /var/www/example.com
sudo chmod -R 755 /var/www

Create an index file which will be the entry point of the website. Run the command below to open an editor.

vim /var/www/example.com/index.html

Edit the contents to look like the code below.

<html>
  <body>
    <h1>The virtual host is working</h1>
  </body>
</html>

Make the site available for Apache by copying the default configuration to our site’s own configuration.

sudo cp /etc/apache2/sites-available/000-default.conf /etc/apache2/sites-available/example.com.conf

Open the copied configuration file with a text editor.

sudo vim /etc/apache2/sites-available/example.com.conf

Edit the contents to look like below.

<VirtualHost *:80>
    ServerAdmin webmaster@example.com
    DocumentRoot /var/www/example.com
    ErrorLog ${APACHE_LOG_DIR}/error.log
    CustomLog ${APACHE_LOG_DIR}/access.log combined
</VirtualHost>

We need to disable Apache’s default configuration so it won’t have any conflict with our configuration.

sudo a2dissite 000-default.conf

Enable our own config file by running the command below.

sudo a2ensite example.com.conf

Restart the Apache web server to apply the changes.

sudo systemctl restart apache2

Add example.com to the virtual hosts by editing /etc/hosts.

sudo vim /etc/hosts

and add this line

127.0.0.1 example.com

Test by using the curl command.

curl --header "hostname: example.com" example.com

It should respond with the contents of /var/www/example.com/index.html.

You may also type example.com in the browser and see the message “The virtual host is working.” if you set it up correctly.

Ruby on Rails: Capistrano for easy deployment and configuration to Ubuntu with Nginx and Puma

Wed, 26 Sep 2018 00:30:00 +0000

Capistrano is a tool written in ruby to automate the deployment of multiple versions of an application. In this post I’ll demonstrate how to use Capistrano to deploy and configure a Rails app to an Ubuntu Server with Nginx and Puma.

Setup the Ubuntu Server.
- Setup Ruby on Rails environment.
- Setup SSH.
Setup Capistrano.
Deploying the Rails app with Capistrano.

Setup the Ubuntu Server

ssh root@example.emmanuelcorrales.com

Create a super user called deploy.

useradd -m deploy -G sudo

su - deploy

Update Ubuntu then install Curl, Git and Nginx.

sudo apt-get update
sudo apt-get install curl git-core nginx -y

Setup Ruby on Rails environment

Install RVM.

gpg --keyserver hkp://keys.gnupg.net --recv-keys 409B6B1796C275462A1703113804BB82D39DC0E3
curl -sSL https://get.rvm.io | bash -s stable
source /home/deploy/.rvm/scripts/rvm
rvm requirements

Install Ruby via RVM.

rvm install 2.2.2
rvm use 2.2.2 --default
ruby --version

Install Rails and Bundler without docs to save space and nobody needs docs on the server.

deploy@server:~:$ gem install rails -v '5.2.0' -V --no-ri --no-rdoc
deploy@server:~:$ gem install bundler -V --no-ri --no-rdoc

Setup SSH

Create .ssh directory

deploy@server:~:$ mkdir -m 700 .ssh
deploy@server:~:$ touch ~/.ssh/authorized_keys
deploy@server:~:$ chmod 600 ~/.ssh/authorized_keys

Setup SSH access to the server. Create key pairs from your local machine.

ssh-keygen -t rsa

This will generate two files ~/.ssh/id_rsa and ~/.ssh/id_rsa.pub. Copy the public key ~/.ssh/id_rsa it to the server.

ssh-copy-id deploy@example.emmanuelcorrales.com

The server will also use the same public key to pull changes to the Rails from the repository. Add the newly created public key (~/.ssh/id_rsa.pub) to your repository’s deployment keys. If you are using Github you can find the instructions here.

Setup Capistrano

“Capify” the Rails app

Add this gem to your Gemfile.

group :development do
  gem 'capistrano', '~>3.10', require: false
end

Then install it.

bundle install

From your apps local root directory generate the necessary configuration files.

bundle exec cap install

This command will generate four files Capfile, config/deploy.rb config/deploy/staging.rb and config/deploy/production.rb.

We want to deploy the Rails app called example.emmanuelcorrales.com whose repository is hosted at git@github.com:EmmanuelCorrales/example.git. Different verisons of the app would be stored at the /home/default/example.emmanuelcorrales.com directory. Configure Capistrano to install it there by changing the contents of config/deploy.rb to look like the code below.

# config valid for current version and patch releases of Capistrano
lock "~> 3.11.0"

set :application, 'example.emmanuelcorrales.com'
set :repo_url, 'git@github.com:EmmanuelCorrales/example.git'

# Deploy configurations
set :deploy_via,      :remote_cache
set :deploy_to,       "/home/#{fetch(:user)}/apps/#{fetch(:application)}"

Add Ruby on Rails deployment tasks

Add this gem to your Gemfile under the capistrano gem.

gem 'capistrano-rails', '~>1.4', require: false

Then install it.

bundle install

Edit the Capfile to look like this:

# Load DSL and set up stages
require "capistrano/setup"

# Include default deployment tasks
require "capistrano/deploy"

# Load the git SCM plugin appropriate to your project:
require "capistrano/scm/git"
install_plugin Capistrano::SCM::Git

require "capistrano/rails"

# Load custom tasks from `lib/capistrano/tasks` if you have any defined
Dir.glob("lib/capistrano/tasks/*.rake").each { |r| import r }

Symlink Rails shared files and directories like log, tmp and public/uploads. Enable it by setting the linked_dirs and linked_files options at the deploy.rb.

## Linked Files & Directories (Default None):
append :linked_dirs, 'log', 'tmp/pids', 'tmp/cache', 'tmp/sockets', 'vendor/bundle', '.bundle', 'public/system', 'public/uploads'
append :linked_files, 'config/database.yml', 'config/secrets.yml', 'config/application.yml'

Configure Puma

Add this gem to your Gemfile under the capistrano gem.

gem 'capistrano3-puma', require: false

Then install it.

bundle install

Add these lines to your Capfile.

require 'capistrano/puma'
install_plugin Capistrano::Puma

Add these lines to the config/deploy.rb.

# Puma configurations
set :puma_threads, [4, 16]
set :puma_workers, 0
set :puma_bind, "unix://#{shared_path}/tmp/sockets/#{fetch(:application)}-puma.sock"
set :puma_state, "#{shared_path}/tmp/pids/puma.state"
set :puma_pid, "#{shared_path}/tmp/pids/puma.pid"
set :puma_access_log, "#{release_path}/log/puma.error.log"
set :puma_error_log, "#{release_path}/log/puma.access.log"
set :puma_preload_app, true
set :puma_worker_timeout, nil
set :puma_init_active_record, true

Configure Nginx

Add this line to your Capfile below the line install_plugin Capistrano::Puma.

install_plugin Capistrano::Puma

Add these lines to the config/deploy.rb. These are the parameters for the Nginx configurations that will be uploaded to the server.

# Nginx configurations
set :nginx_config_name, "#{fetch(:application)}_#{fetch(:stage)}"
set :nginx_flags, 'fail_timeout=0'
set :nginx_http_flags, fetch(:nginx_flags)
set :nginx_server_name, "localhost #{fetch(:application)}.local"
set :nginx_sites_available_path, '/etc/nginx/sites-available'
set :nginx_sites_enabled_path, '/etc/nginx/sites-enabled'
set :nginx_socket_flags, fetch(:nginx_flags)
set :nginx_ssl_certificate, "/etc/ssl/certs/#{fetch(:nginx_config_name)}.crt"
set :nginx_ssl_certificate_key, "/etc/ssl/private/#{fetch(:nginx_config_name)}.key"
set :nginx_use_ssl, false

Upload the nginx configuration to the server.

bundle exec cap production puma_nginx:config

Deploying the Rails app with Capistrano

Edit the contents of config/deploy/production.rb to look like this:

server "example.emmanuelcorrales.com", user: "deploy", roles: %w{web app db}

Deploy to the server.

bundle exec cap production deploy

Emmanuel Corrales

How to remove Git submodules

How to use a self signed certificate with Gradle

How to enable internet access of CentOS 7 on VirtualBox

Diagnosing the problem

Configuring the network interface

Testing access to the internet

NodeJS: Debugging and fixing HTTPS/SSL issues caused by changing the self signed certificate.

Checking if the certificate is installed on the server.

Verifying the connection to the third party API.

Running the client app.

How to properly debug shell scripts

Contents

Printing commands upon execution

Running the entire script on Debug mode.

Terminating the script on error

Installing Oracle JDK for MacOS via Homebrew

Steps to install and configure the Oracle JDK:

Homebrew and Cask

JDK Installation

Setup Java_HOME environment variable

Verifying Installation

Automating installation with a script

How I install and configure Docker for my Mac via Homebrew

Steps to install and configure Docker:

Install Homebrew and Cask

Install Docker

Setup Bash auto completion

How I redirected all the web traffic for my old domain to the new one

Steps to redirect web traffic from old domain to the new one:

Launch a server for my old domain

Setup virtual hosts for the old domain with Apache

Configure the virtual host to redirect to the new domain.

Enable HTTPS for Apache

Install Let’s Encrypt SSL Self Signed Certificate for Apache

Apache: Setup a virtual host on Ubuntu 18.04 LTS

Ruby on Rails: Capistrano for easy deployment and configuration to Ubuntu with Nginx and Puma

Table of Contents

Setup the Ubuntu Server

Setup Ruby on Rails environment

Setup SSH

Setup Capistrano

“Capify” the Rails app

Add Ruby on Rails deployment tasks

Configure Puma

Configure Nginx

Deploying the Rails app with Capistrano