ebr-board

RESTful interface for Elastic Build Results.

Usage

To view the API documentation, start the server and go to to <url>/api/docs.

Configuration

ebr-board uses Vault-Anyconfig to read in its configuration, allowing it to access a Hashicorp Vault instance for loading secrets. For more details refer to its documentation.

In order to deploy a simple ebr-board instance, you will need two files: config.yaml and vault.yaml. Leave vault.yaml empty, as we will not use secret loading from Vault-Anyconfig in this instance.

config.yaml should be formatted as follows:

elastic:
  host: <elastic_url>
  port: 9200
  timeout: 20
  use_ssl: true
  verify_certs: true
  ca_certs: /etc/ebr-board/elastic.crt
  index: testspipeline*
  user: <elastic_user>
  pwd: <elastic_password>

Dev Mode

To start in dev mode, run python ebr_board/ebr_board.py

Production Mode

Can be invoked with ebr_board:create_app(config_filename='/etc/ebr-board/config.yaml', vault_config_filename='/etc/ebr-board/vault.yaml', vault_creds_filename='/etc/ebr-board/vault.yaml', load_certs=True, reverse_proxy=True), for example from Gunicorn. You should configure it behind a reverse proxy - for more details see any guide on configuring Flask servers for deployment. A Dockerfile pre-configuring Gunicorn is available in the root of the repository.

AWS Lambda Support

The application can be run in AWS Lambda by using the handler function in the aws_lambda module. It expects that the configuration (the main configuration, vault configuration and vault creds) will be stored entirely as strings in the parameter store. The way it processes these parameters can be configured with environmental variables:

• config_name: defaults to ebr_board_config
• vault_config_name: defaults to ebr_board_vault_config
• vault_creds_name: defaults to ebr_board_vault_creds
• config_format: defaults to yaml

Features

• Provides abstraction to fetch:
  • a list of builds from a given job
  • tests from a given job
  • aggregations of tests failures
• AWS Lambda support

Todo:

• Improve test coverage
• Fill in coverage of resources
• Expand aggregation/search functionality

Credits

This package was created with Cookiecutter and the tomtom-international/cookiecutter-python project template.

Changelog

0.1.0 (2019-08-22)

• Add aws lambda handler
• Clean up code to contain state better (for lambda compatability)

0.0.1-dev (2019-07-05)

• First release on PyPI.