Revision: Wed, 27 Nov 2024 11:25:20 GMT
v3.13 – outdated
This version of the documentation is outdated. Consider upgrading your project to Spiral Framework 3.14
Edit this page

Getting started — Deployment

Deploying a Spiral application can be a complex task, as it requires various steps to be performed in order to ensure that the application is properly configured and ready for production use. In this guide, we will go over several strategies for deploying an application, including basic file transfer, using source control, build scripts, and Docker builds.

Preparing the application

Deploying a Spiral application on a production server requires several key configurations to ensure proper operation and security.

Warning
Do not store .env file in your repository as it may contain sensitive information such as database credentials, API keys, and other secrets.

Here is a step-by-step guide on how to configure a Spiral application on a production server:

Disable debug mode

Make sure that the DEBUG variable is set to false in your application's environment configuration. This will prevent sensitive information from being displayed in case of an error.

dotenv
.env
DEBUG=false

Set the environment to production

Change the APP_ENV variable in your application's environment configuration to production. This will prevent accidental actions that should only be run on a development server.

dotenv
.env
APP_ENV=production

Enable tokenizer cache

Make sure that the TOKENIZER_CACHE_TARGETS variable is set to true in your application's environment configuration. This will disable scanning and analysis of the application's source code, which can improve performance on boot.

dotenv
.env
TOKENIZER_CACHE_TARGETS=true

Note
Read more about tokenizer in Component — Static analysis section.

Set the verbosity level

Set the VERBOSITY_LEVEL variable to a basic level to hide server errors from being displayed to the public.

dotenv
.env
VERBOSITY_LEVEL=basic

Configure the logger

Set the desired MONOLOG_DEFAULT_CHANNEL in the application's environment configuration. This will determine where the application's logs will be stored. Also set the MONOLOG_DEFAULT_LEVEL variable to error in your application's environment configuration. This will prevent debug and info errors from being logged, helping to keep your logs clean and easy to read.

dotenv
.env
MONOLOG_DEFAULT_CHANNEL=roadrunner
MONOLOG_DEFAULT_LEVEL=error

Cycle ORM

If you are using the Cycle ORM, make sure that the CYCLE_SCHEMA_CACHE and CYCLE_SCHEMA_WARMUP variables are set to true in your application's environment.

The CYCLE_SCHEMA_CACHE variable controls whether the ORM should cache the schema of the database tables. When set to true, the ORM will cache the schema, which can improve the performance of the application by reducing the number of database queries needed to retrieve the schema information.

The CYCLE_SCHEMA_WARMUP variable controls whether the ORM should warm up the schema cache when the application starts. When set to true, the ORM will pre-populate the schema cache with the schema information, which can further improve the performance of the application by reducing the time it takes to retrieve the schema information.

dotenv
.env
CYCLE_SCHEMA_CACHE=true
CYCLE_SCHEMA_WARMUP=true

Composer

The following composer commands can be used to install vendor packages for a Spiral application:

composer install --optimize-autoloader --no-dev --no-scripts

This command installs the application's dependencies, generates the autoloader files for the application and optimizes them for performance.

  • --no-dev option is used to prevent any development dependencies from being installed ad
  • --no-scripts option is used to prevent any post-install scripts from being executed.

Nginx

Proxy to RoadRunner

If you want to use Spiral with RoadRunner and have Nginx server in front of it, you will need to configure Nginx to act as a reverse proxy.

Here is an example of how to set up Nginx as a reverse proxy.

nginx
/etc/nginx/sites-enabled/roadrunner.conf
server {
    listen 80;

    server_name _;

    location / {
        proxy_pass http://127.0.0.1:8080;
    }
}

This configuration instructs Nginx to listen on port 80 and forward all incoming requests to 127.0.0.1:8080, where RoadRunner is running.

You can place this configuration file in the /etc/nginx/sites-available/ directory and then create a symbolic link to it in the /etc/nginx/sites-enabled/ directory.

Here is an example of a configuration that uses a RoadRunner HTTP server:

yaml
.rr.yaml
http:
  address: 127.0.0.1:8080

Warning:
Avoid using address: 0.0.0.0:8080 in the RoadRunner configuration as it would block direct access to the RoadRunner HTTP server, allowing access only through the Nginx reverse proxy.

PHP-FPM

To use Spiral with PHP-FPM, you'll need to follow these steps:

  1. Install the spiral/sapi-bridge package, which provides a dispatcher that PHP-FPM will use to handle requests.

Use the following command to install the package:

composer require spiral/sapi-bridge

Once the package is installed, you'll need to register the Spiral\Sapi\Bootloader\SapiBootloader bootloader in your application's list of bootloaders.

php
app/src/Application/Kernel.php
public function defineBootloaders(): array
{
    return [
        // ...
        \Spiral\Sapi\Bootloader\SapiBootloader::class,
        // ...
    ];
}

Read more about bootloaders in the Framework — Bootloaders section.

  1. Configure Nginx to use PHP-FPM.

Here's an example of how you can set up Nginx to work with PHP-FPM:

nginx
/etc/nginx/sites-enabled/spiral.conf
server {
    listen 80;
    listen [::]:80;
    server_name example.com;
    root /srv/example.com/public;
 
    add_header X-Frame-Options "SAMEORIGIN";
    add_header X-Content-Type-Options "nosniff";
 
    index app.php;
 
    charset utf-8;
 
    location / {
        try_files $uri $uri/ /app.php?$query_string;
    }
 
    location = /favicon.ico { access_log off; log_not_found off; }
    location = /robots.txt  { access_log off; log_not_found off; }
 
    error_page 404 /app.php;
 
    location ~ \.php$ {
        fastcgi_pass unix:/var/run/php/php8.1-fpm.sock;
        fastcgi_param SCRIPT_FILENAME $realpath_root$fastcgi_script_name;
        include fastcgi_params;
    }
 
    location ~ /\.(?!well-known).* {
        deny all;
    }
}

Deployer

Using automation tools like Deployer can automate the deployment process and handle tasks such as database migrations, cache clearing and more. They also allow for faster deployment and rollback processes, and can be integrated with other tools like load balancers and monitoring systems. This can make the deployment process more efficient and less prone to errors, and makes it easier to keep track of which version of the code is currently on the production server, making it hard to rollback in case of errors.

Deployer is a popular deployment tool that can be used to automate the deployment process for a Spiral application. One of the advantages of using Deployer is that it can provide "zero downtime" deployment.

Zero downtime deployment is a technique that allows you to update your application without interrupting the service to your users. This is achieved by deploying the new version of the application alongside the old version, and then swapping them out once the new version is ready. This allows for a seamless transition, as users are not affected by any downtime during the deployment process.

Deployer provides the capability of Zero Downtime deployment by creating a new release of the application in a different directory and switching the symlink to the new release.

Installing

To install Deployer, run next command in your project dir:

composer require --dev deployer/deployer

Configuration

To initialize deployer in your project run:

./vendor/bin/dep init

Deployer will ask you a few questions and after finishing you will have a deploy.php or deploy.yaml file. This is your deployment recipe. It contains hosts, tasks and requires other recipes.

Here is an example of a deploy.php file:

php
deploy.php
namespace Deployer;

require 'recipe/spiral.php';

set('repository', 'https://github.com/xxx/my-app');

add('shared_files', []);
add('shared_dirs', []);
add('writable_dirs', []);

host('example.org')
    ->set('remote_user', 'deployer')
    ->set('deploy_path', '/var/www/my-app');

after('deploy:failed', 'deploy:unlock');

desc('Deploys your project');
task('deploy', [
    'deploy:prepare',
    'deploy:environment',
    'deploy:vendors',
    'spiral:encrypt-key',
    'spiral:configure',
    'deploy:download-rr',
    'deploy:publish',
    'deploy:restart-rr'
]);

To connect to the remote host we need to specify an identity key or private key. We can add our identity key directly into the host definition, but it's better to put it in the ~/.ssh/config file:

~/.ssh/config
Host example.org
  IdentityFile ~/.ssh/id_rsa

Now let's provision our server. As our host doesn't have user deployer, we are going to override remote_user for provision via -o remote_user=root.

dep provision -o remote_user=root

Deployer will ask you a few questions during provisioning: php version, database type, etc. Next Deployer will configure our server and create the deployer user. Provision takes around 5 minutes and will install everything we need to run a website. A new website will be configured at deploy_path.

Deploying

To deploy your application, run the following command:

./vendor/bin/dep deploy

Note
If deploy failed, Deployer will print an error message and which command was unsuccessful. Most likely we need to configure the correct database credentials in .env file or similar.

CI/CD

Deployer can be used in CI/CD pipelines.

See more
Read more about it in Deployer CI/CD section.


Docker

Docker allows for easy deployment, scaling, and management of application containers. It also makes it easy to replicate the production environment and test applications locally, which makes it easier to find and fix issues.

This method involves building a Docker image of the application, and then running the image on the production server.

Dockerfile

Here is an example of a Dockerfile that can be used to build a Docker image for a Spiral application:

dockerfile
# This example will work with application root directory as docker context
FROM php:8.2-cli-alpine3.17 as backend

RUN  --mount=type=bind,from=mlocati/php-extension-installer:1.5,source=/usr/bin/install-php-extensions,target=/usr/local/bin/install-php-extensions \
      install-php-extensions opcache zip xsl dom exif intl pcntl bcmath sockets && \
     apk del --no-cache  ${PHPIZE_DEPS} ${BUILD_DEPENDS}

WORKDIR /app

ENV COMPOSER_ALLOW_SUPERUSER=1
COPY --from=composer:2.3 /usr/bin/composer /usr/bin/composer
COPY ./composer.* .
RUN composer config --no-plugins allow-plugins.spiral/composer-publish-plugin false && \
    composer install --optimize-autoloader --no-dev

COPY --from=spiralscout/roadrunner:latest /usr/bin/rr /app

EXPOSE 8080/tcp

COPY ./ .

CMD ./rr serve -c .rr.yaml

You can build the image by running the following command:

docker build . -t my-application:latest

After the image is built, you need to push it to a Docker registry or hub.

docker push my-application:latest

You can also tag the image with a version number, such as my-application:1.0.

docker build . -t my-application:latest -t my-application:1.0
docker push my-application:latest
docker push my-application:1.0

Docker Compose

One of the advantages of using Docker is that you can manage environment variables for your application through the docker-compose file and external .env files, rather than having to create .env files inside the container.

Here is an example of a docker-compose.yml file that can be used to start the application:

yaml
docker-compose.yaml
version: '3'
services:
  app:
    image: my-application:1.0
    ports:
      - "8080:8080"
    environment:
      - DEBUG=false
      - APP_ENV=production
      - ...
...

Starting and Stopping the Application

You can start the application by running the following command:

docker-compose up -d

and stop it by running

docker-compose down

Source Control

A popular way to deploy an application is by using source control. This involves connecting your production server to a remote repository, such as Git or SVN, and then deploying the application by pulling in the latest changes from the repository.

Example

  1. Connect to the production server via SSH
  2. Navigate to the root directory of your application on the server
  3. Run the command git pull origin master to pull in the latest changes from the remote repository
  4. Once the changes are pulled in, run composer install --optimize-autoloader --no-dev to install the dependencies
  5. Run database migrations and other necessary console commands
  6. Restart the RoadRunner server workers ./rr reset

Warning:
Using this strategy without automation can make the deployment process more time-consuming and prone to errors. Every time a deployment is needed, developers would need to manually pull the code from the repository and run any necessary console commands in a specific order. This increases the risk of human error, such as running the commands in the wrong order or forgetting to run a command, which can lead to issues with the application.


File Transfer

One of the simplest ways to deploy a Spiral application is through basic file transfer. This can be done using FTP, or SCP. The process involves uploading the application files from your local machine to the production server.

Example

  1. Connect to the production server using FTP client (FileZilla, WinSCP)
  2. Navigate to the root directory of your application on the server
  3. Select all of the application files on your local machine and upload them to the server
  4. Once the upload is complete, run composer install --optimize-autoloader --no-dev to install the dependencies
  5. Run database migrations and other necessary console commands
  6. Restart the RoadRunner server workers ./rr reset

Warning:
This strategy is considered less efficient and less secure than other methods. It can take a long time to upload all the files manually, and there is a risk of data loss or corruption if the transfer is interrupted. Additionally, it can be difficult to keep track of which version of the code is currently on the production server, making it hard to rollback in case of errors.