search

cloudfoundry / cloud_controller_ng

Cloud Foundry Cloud Controller
Apache License 2.0
194 stars 360 forks source link
cloudfoundry ruby

readme

Code Climate Code Climate slack.cloudfoundry.org

Welcome to the Cloud Controller

Helpful Resources

Components

Cloud Controller

The Cloud Controller provides REST API endpoints to create and manage apps, services, user roles, and more!

Database

The Cloud Controller supports Postgres and Mysql.

Blobstore

The Cloud Controller manages a blobstore for:

All Platforms:

When deployed via capi-release only:

Cloud Controller currently supports webdav and the following fog connectors:

Runtime

The Cloud Controller on VMs uses Diego to stage and run apps and tasks. See Diego Design Notes for more details.

Contributing

Please read the contributors' guide and the Cloud Foundry Code of Conduct

Predefined Development Environment

To commence your work in a fully equipped development environment, you have two main options:

  1. GitHub Codespaces: GitHub Codespaces provisions a virtual machine with essential core services, such as S3 Blobstore, Database, and NGINX. It also establishes a connection that your IDE can use (VSCode is recommended). To initiate a codespace, click on the green button within the GitHub UI(upper right corner) and select the 'Codespaces' tab.

  2. Local Environment: This option allows you to establish an environment on your local machine with the same core services as GitHub Codespaces, using Docker.

A script in the project's root directory provides convenient shortcuts to set up an environment locally:

Usage: ./devenv.sh COMMAND

Commands:
  create     - Setting up the development environment(containers)
  start      - Starting the development environment(containers), an existing fully set up set of containers must exist.
  stop       - Stopping but not removing the development environment(containers)
  destroy    - Stopping and removing the development environment(containers)
  runconfigs - Copies matching run configurations for intellij and vscode into the respective folders
  help       - Print this help text

To run this script, ensure the following are installed on your local system:

Upon executing ./devenv.sh create, the necessary containers will be set up and the databases will be initialized and migrated.

As an optional step, execute ./devenv.sh runconfigs to copy predefined settings and run configurations for this project into .vscode and .idea directories for VSCode and IntelliJ/RubyMine/JetBrains IDEs. These configurations are opinionated and, hence, not provided by default, but they do offer common configurations to debug rspecs, cloud_controller, local_worker, and generic_worker.

Credentials

This Setup automatically creates a user in UAA for login in the cloud_controller, and sets Passwords for Postgres and Mysql. In case you need them to configure them somewhere else (e.g. database visualizers):

Starting the Cloud Controller locally

When the Docker containers have been set up as described above, you can start the cloud controller locally. Start the main process with:

DB_CONNECTION_STRING=mysql2://root:supersecret@127.0.0.1:3306/ccdb ./bin/cloud_controller -c ./tmp/cloud_controller.yml

Then start a local worker:

DB_CONNECTION_STRING=mysql2://root:supersecret@127.0.0.1:3306/ccdb CLOUD_CONTROLLER_NG_CONFIG=./tmp/cloud_controller.yml bundle exec rake jobs:local

Start a delayed_job worker:

DB_CONNECTION_STRING=mysql2://root:supersecret@127.0.0.1:3306/ccdb CLOUD_CONTROLLER_NG_CONFIG=./tmp/cloud_controller.yml bundle exec rake jobs:generic

And finally start the scheduler:

DB_CONNECTION_STRING=mysql2://root:supersecret@127.0.0.1:3306/ccdb CLOUD_CONTROLLER_NG_CONFIG=./tmp/cloud_controller.yml bundle exec rake clock:start

Known limitations:

Unit Tests

TLDR: Always run bundle exec rake before committing

To maintain a consistent and effective approach to testing, please refer to the spec README and keep it up to date, documenting the purpose of the various types of tests.

By default rspec will randomly pick between postgres and mysql.

If postgres is not running on your OSX machine, you can start up a server by doing the following:

brew services start postgresql
createuser -s postgres
DB=postgres rake db:create

It will try to connect to those databases with the following connection string:

To specify a custom username, password, host, or port for either database type, you can override the default connection string prefix (the part before the cc_test database name) by setting the MYSQL_CONNECTION_PREFIX and/or POSTGRES_CONNECTION_PREFIX variables. Alternatively, to override the full connection string, including the database name, you can set the DB_CONNECTION_STRING environment variable. This will restrict you to only running tests in serial, however.

For example, to run unit tests in parallel with a custom mysql username and password, you could execute:

MYSQL_CONNECTION_PREFIX=mysql2://custom_user:custom_password@localhost:3306 bundle exec rake

The following are examples of completely fully overriding the database connection string:

DB_CONNECTION_STRING="postgres://postgres@localhost:5432/cc_test" DB=postgres rake spec:serial
DB_CONNECTION_STRING="mysql2://root:password@localhost:3306/cc_test" DB=mysql rake spec:serial

If you are running the integration specs (which are included in the full rake), and you are specifying DB_CONNECTION_STRING, you will also need to have a second test database with _integration_cc as the name suffix.

For example, if you are using:

DB_CONNECTION_STRING="postgres://postgres@localhost:5432/cc_test"

You will also need a database called:

`cc_test_integration_cc`

The command

rake db:create

will create the above database when the DB environment variable is set to postgres or mysql. You should run this before running rake in order to ensure that the cc_test database exists.

Running tests on a single file

The development team typically will run the specs to a single file as (e.g.)

bundle exec rspec spec/unit/controllers/runtime/users_controller_spec.rb

Running all the unit tests

bundle exec rake spec

Note that this will run all tests in parallel by default. If you are setting a custom DB_CONNECTION_STRING, you will need to run the tests in serial instead:

bundle exec rake spec:serial

To be able to run the unit tests in parallel and still use custom connection strings, use the MYSQL_CONNECTION_PREFIX and POSTGRES_CONNECTION_PREFIX environment variables described above.

Running static analysis

bundle exec rubocop

Running both unit tests and rubocop

By default, bundle exec rake will run the unit tests first, and then rubocop if they pass. To run rubocop first, run:

RUBOCOP_FIRST=1 bundle exec rake

Logs

Cloud Controller uses Steno to manage its logs. Each log entry includes a "source" field to designate which module in the code the entry originates from. Some of the possible sources are 'cc.app', 'cc.app_stager', and 'cc.healthmanager.client'.

Here are some use cases for the different log levels:

Configuration

The Cloud Controller uses a YAML configuration file. For an example, see config/cloud_controller.yml.