Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

https://github.com/ankane/production_rails

Best practices for running Rails in production
https://github.com/ankane/production_rails

Last synced: 3 months ago
JSON representation

Best practices for running Rails in production

Lists

README 

        
# Production Rails



:rocket: Best practices for running Rails in production



This guide covers different concepts you should be familiar with. Recommendations come from personal experience and work at [Instacart](https://www.instacart.com/opensource). A number of open source projects are ones I’ve created. For a comprehensive list of gems, check out [Awesome Ruby](https://awesome-ruby.com).



## Security



Everyone writing code must be responsible for security. See [best practices](https://github.com/ankane/secure_rails) and [how to secure sensitive data](https://ankane.org/sensitive-data-rails).



## Errors



Use an error reporting service like [Rollbar](https://rollbar.com/).



Use [Safely](https://github.com/ankane/safely) to rescue and report exceptions in non-critical code.



## Logging



Use a centralized logging service like [Mezmo](https://www.mezmo.com/).



Use [Lograge](https://github.com/roidrage/lograge) to reduce volume. Configure it to add `request_id`, `user_id`, and `params`.



```ruby

# config/environments/production.rb

config.lograge.enabled = true

config.lograge.custom_options = lambda do |event|

  options = event.payload.slice(:request_id, :user_id)

  options[:params] = event.payload[:params].except("controller", "action")

  options

end



# app/controllers/application_controller.rb

def append_info_to_payload(payload)

  super

  payload[:request_id] = request.uuid

  payload[:user_id] = current_user.id if current_user

end

```



## Audits



Use an auditing library like [Audited](https://github.com/collectiveidea/audited).



## Migrations



Use [Strong Migrations](https://github.com/ankane/strong_migrations) to catch unsafe migrations at dev time.



## Web Requests



Use a high performance web server like [Puma](https://github.com/puma/puma).



Use [Rack::Deflater](https://www.schneems.com/2017/11/08/80-smaller-rails-footprint-with-rack-deflate/) for compression.



Use a [CDN](https://en.wikipedia.org/wiki/Content_delivery_network) like [Amazon CloudFront](https://aws.amazon.com/cloudfront/) to serve assets.



Use [Slowpoke](https://github.com/ankane/slowpoke) for request timeouts.



## Background Jobs



Use a high performance background processing framework like [Sidekiq](https://github.com/mperham/sidekiq) with Active Job.



```ruby

config.active_job.queue_adapter = :sidekiq

```



Use [ActiveJob::TrafficControl](https://github.com/nickelser/activejob-traffic_control) to:



- quickly disable jobs

- throttle

- limit concurrency



```ruby

BadJob.disable!

```



## Email



For transactional emails, use an email delivery service like [SendGrid](https://sendgrid.com/).



For marketing emails, use a service like [MailChimp](https://mailchimp.com/).



For styling, use a CSS inliner like [Roadie](https://github.com/Mange/roadie-rails).



```ruby

class ApplicationMailer < ActionMailer::Base

  include Roadie::Rails::Automatic

end

```



Add UTM parameters to links.



## Caching and Performance



Use [Memcached](https://memcached.org/) and [Dalli](https://github.com/petergoldstein/dalli) for caching.



```ruby

config.cache_store = :dalli_store

```



Use a library like [Memoist](https://github.com/matthewrudy/memoist) for memoizing.



```ruby

memoize :time_consuming_method

```



Add [Oj](https://github.com/ohler55/oj) to speed up JSON parsing.



## Monitoring



### Tracing



Use a performance monitoring service with transaction traces like [New Relic](https://newrelic.com/) or [AppSignal](https://appsignal.com/).



### Uptime



Use an uptime monitoring service like [Pingdom](https://www.pingdom.com/) or [Uptime Robot](https://uptimerobot.com/).



### Database



The database is a common bottleneck for Rails apps and deserves some special monitoring attention. There are some dedicated tools for this:



- If you use Postgres, [PgHero](https://github.com/ankane/pghero) can help identify issues

- Use [Active Record Query Logs](https://api.rubyonrails.org/classes/ActiveRecord/QueryLogs.html) (Rails 7+) or [Marginalia](https://github.com/basecamp/marginalia) (Rails < 7) to track the origin of SQL queries



### Notable Events



Use [Notable](https://github.com/ankane/notable) to track notable requests and background jobs.



- errors

- slow requests, jobs, and timeouts

- 404s

- validation failures

- CSRF failures

- unpermitted parameters

- blocked and throttled requests



## Timeouts



[Add timeouts](https://github.com/ankane/the-ultimate-guide-to-ruby-timeouts).



One very important place is Active Record. Add to `config/database.yml` and adjust as needed.



#### PostgreSQL



```yml

production:

  connect_timeout: 2

  checkout_timeout: 5

  variables:

    statement_timeout: 5000 # ms

```



#### MySQL and MariaDB



```yml

production:

  connect_timeout: 1

  read_timeout: 1

  write_timeout: 1

  checkout_timeout: 5

  variables:

    max_execution_time: 5000 # ms, for MySQL 5.7.8 or higher

    max_statement_time: 5 # sec, for MariaDB 10.1.1 or higher

```



## Analytics



Use an analytics tool to measure important events. Consider a first-party library like [Ahoy](https://github.com/ankane/ahoy), or use a third-party service like [Amplitude](https://amplitude.com/) or [Mixpanel](https://mixpanel.com/).



## New Features



Use a feature flipper library like [Rollout](https://github.com/FetLife/rollout) to easily enable and disable new features without pushing code.



To A/B test features, use a library like [Field Test](https://github.com/ankane/field_test).



## Lastly...



Have suggestions? [Help make this guide better for everyone](https://github.com/ankane/rails-best-practices/issues/new).



Also check out [Development Rails](Development.md) and [Scaling Rails](Scaling.md).