Skip to content

Commit

Permalink
Merge pull request consuldemocracy#78 from PierreMesure/patch-4
Browse files Browse the repository at this point in the history
Updating the heroku guide for CONSUL v1.0.0
  • Loading branch information
javierm committed Nov 30, 2021
2 parents a656cac + 838ea32 commit b3378cb
Show file tree
Hide file tree
Showing 3 changed files with 186 additions and 196 deletions.
190 changes: 93 additions & 97 deletions en/installation/deploying-on-heroku.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,163 +7,159 @@ This tutorial assumes that you have already managed to clone CONSUL on your mach
1. First, create a [Heroku](https://www.heroku.com) account if it isn't already done.
2. Install the [Heroku CLI](https://devcenter.heroku.com/articles/heroku-cli) and sign in using

```
heroku login
```
```bash
heroku login
```

3. Go to your CONSUL repository and instantiate the process

```
cd consul
heroku create your-app-name
```
```bash
cd consul
heroku create your-app-name
```

You can add the flag `--region eu` if you want to use their European servers instead of the US ones.
You can add the flag `--region eu` if you want to use their European servers instead of the US ones.

If _your-app-name_ is not already taken, Heroku should now create your app.
If _your-app-name_ is not already taken, Heroku should now create your app.

4. Create a database using

```
heroku addons:create heroku-postgresql
```
```bash
heroku addons:create heroku-postgresql
```

You should now have access to an empty Postgres database whose address was automatically saved as an environment variable named _DATABASE\_URL_. CONSUL will automatically connect to it when deployed.
You should now have access to an empty Postgres database whose address was automatically saved as an environment variable named _DATABASE\_URL_. CONSUL will automatically connect to it when deployed.

5. Add a file name _heroku.yml_ at the root of your project and paste the following in it
5. **(Not needed)** Add a file name _heroku.yml_ at the root of your project and paste the following in it

```
build:
languages:
- ruby
packages:
- imagemagick
run:
web: bundle exec rails server -e ${RAILS_ENV:-production}
```
```yml
build:
languages:
- ruby
packages:
- imagemagick
run:
web: bundle exec rails server -e ${RAILS_ENV:-production}
```

6. Now, generate a secret key and save it to an ENV variable named SECRET\_KEY\_BASE using

```
heroku config:set SECRET_KEY_BASE=`ruby -rsecurerandom -e "puts SecureRandom.hex(64)"`
```
```bash
heroku config:set SECRET_KEY_BASE=$(rails secret)
```

You need to let the app know where the secret key is stored by adding a link to the ENV variable in _config/secrets.yml_
Also add your server address:

```
production:
secret_key_base: <%= ENV["SECRET_KEY_BASE"] %>
```
```bash
heroku config:set SERVER_NAME=myserver.address.com
```

You need to let the app know where the configuration variables are stored by adding a link to the ENV variables in _config/secrets.yml_

and commit this file in the repo by commenting out the corresponding line in the _.gitignore_.
```yml
production:
secret_key_base: <%= ENV["SECRET_KEY_BASE"] %>
server_name: <%= ENV["SERVER_NAME"] %>
```

```
#/config/secrets.yml
```
and commit this file in the repo by commenting out the corresponding line in the _.gitignore_.

**Remember not to commit the file if you have any sensitive information in it!**
```gitignore
#/config/secrets.yml
```

**Remember not to commit the file if you have any sensitive information in it!**

7. You can now push your app using

```
git push heroku your-branch:master
```
```bash
git push heroku your-branch:master
```

8. It won't work straight away because the database doesn't contain the tables needed. To create them, run

```
heroku run rake db:migrate
heroku run rake db:seed
```
```bash
heroku run rake db:migrate
heroku run rake db:seed
```

If you want to add the test data in the database, move `gem 'faker', '~> 1.8.7'` outside of `group :development` and run
If you want to add the test data in the database, move `gem 'faker', '~> 1.8.7'` outside of `group :development` and run

```
heroku config:set DATABASE_CLEANER_ALLOW_REMOTE_DATABASE_URL=true
heroku config:set DATABASE_CLEANER_ALLOW_PRODUCTION=true
heroku run rake db:dev_seed
```
```bash
heroku config:set DATABASE_CLEANER_ALLOW_REMOTE_DATABASE_URL=true
heroku config:set DATABASE_CLEANER_ALLOW_PRODUCTION=true
heroku run rake db:dev_seed
```

9. Your app should now be ready to use. You can open it with

```
heroku open
```

You also can run the console on heroku using

```
heroku console --app your-app-name
```

10. Heroku doesn't allow to save images or documents in its servers, so it's necessary make this changes
```bash
heroku open
```

On `app/models/image.rb:47` and `app/models/document.rb:39`
You also can run the console on heroku using

Change `URI.parse(cached_attachment)` to `URI.parse("http:" + cached_attachment)`
```bash
heroku console --app your-app-name
```

Create a new file on `config/initializers/paperclip.rb` with the following content
10. Heroku doesn't allow to save images or documents in its servers, so it's necessary to setup a permanent storage space.

```
Paperclip::UriAdapter.register
```
See [our S3 guide](./using-aws-s3-as-storage.md) for more details about configuring Paperclip with S3.

See [our S3 guide](../getting_started/using-aws-s3-as-storage.md) for more details about configuring Paperclip with S3.
## Optional but recommended:

### Optional but recommended:
### Install rails\_12factor and specify the Ruby version

**Install rails\_12factor and specify the Ruby version**
**The rails\_12factor is only useful if you use a version of CONSUL older than 1.0.0. The latter uses Rails 5 which includes the changes.**

As recommended by Heroku, you can add the gem rails\_12factor and specify the version of Ruby you want to use. You can do so by adding

```
```ruby
gem 'rails_12factor'

ruby 'x.y.z'
```

in the file _Gemfile\_custom_, where `x.y.z` is the version defined in the `.ruby-version` file in the CONSUL repository. Don't forget to run

```
```bash
bundle install
```

to generate _Gemfile.lock_ before commiting and pushing to the server.

### Optional but recommended:
### Use Puma as a web server

**Use Puma as a web server**
Heroku recommends to use Puma to improve the responsiveness of your app on [a number of levels](http://blog.scoutapp.com/articles/2017/02/10/which-ruby-app-server-is-right-for-you).

Heroku recommends to use Puma instead of the default web server to improve the responsiveness of your app on [a number of levels](http://blog.scoutapp.com/articles/2017/02/10/which-ruby-app-server-is-right-for-you). First, add the gem in your _Gemfile\_custom_ file:
If you want to allow more concurrency, uncomment the line:

```
gem 'puma'
```
```ruby
workers ENV.fetch("WEB_CONCURRENCY") { 2 }
```

Then you need to create a new file named _puma.rb_ \(your _config_ folder is a good place to store it\). Here is a standard content for this file:
You can find an explanation for each of these settings in the [Heroku tutorial](https://devcenter.heroku.com/articles/deploying-rails-applications-with-the-puma-web-server).

```
workers Integer(ENV['WEB_CONCURRENCY'] || 1)
threads_count = Integer(ENV['RAILS_MAX_THREADS'] || 5)
threads threads_count, threads_count
The last part is to change the _web_ task to use Puma by changing it to this in your _heroku.yml_ file:

preload_app!
```yml
web: bundle exec puma -C config/puma.rb
```

rackup DefaultRackup
port ENV['PORT'] || 3000
environment ENV['RACK_ENV'] || 'production'
### Add configuration variables to tune your app from the dashboard

on_worker_boot do
# Worker specific setup for Rails 4.1+
# See: https://devcenter.heroku.com/articles/deploying-rails-applications-with-the-puma-web-server#on-worker-boot
ActiveRecord::Base.establish_connection
end
```
The free and hobby versions of Heroku are barely enough to run an app like CONSUL. To optimise the response time and make sure the app doesn't run out of memory, you can [change the number of workers and threads](https://devcenter.heroku.com/articles/deploying-rails-applications-with-the-puma-web-server#workers) that Puma uses.

You can find an explanation for each of these settings in the [Heroku tutorial](https://devcenter.heroku.com/articles/deploying-rails-applications-with-the-puma-web-server).
My recommended settings are one worker and three threads. You can set it by running these two commands:

The last part is to change the _web_ task to use Puma by changing it to this in your _heroku.yml_ file:
```bash
heroku config:set WEB_CONCURRENCY=1
heroku config:set RAILS_MAX_THREADS=3
```

```
web: bundle exec puma -C config/puma.rb
```
I also recommend to set the following:
```bash
heroku config:set RAILS_SERVE_STATIC_FILES=enabled
heroku config:set RAILS_ENV=production
```
2 changes: 0 additions & 2 deletions en/installation/using-aws-s3-as-storage.md
Original file line number Diff line number Diff line change
Expand Up @@ -84,5 +84,3 @@ config.paperclip_defaults = {
```

You will need to restart to apply the changes.

**WARNING:** There is currently a bug which causes an error 500 when creating a budget investment with an image stored on S3. You can find the fix on that pull request. If it isn't merged yet or your fork doesn't include it, you will have to apply the changes to your fork manually.
Loading

0 comments on commit b3378cb

Please sign in to comment.