######Cloud Control Console - Version 1.0

Atlas provides a centralized console to manage instances on the AWS (amazon Web Services) cloud aided by the integration of various tools and their functionalities through simple and extensible API (Application Program Interface).

Listed below are the back end, front end and databse requirements for Atlas.

#####Operating System

1. Linux (Developed and tested on Ubuntu 12.04 and above)

#####Backend

1. Python * Version 2.7 or above
2. Django * Version 1.6 or above

#####Frontend

1. Jquery 1.10 and above
2. Jquery UI - 1.10.4
3. Bootstrap 3
4. Metro UI
5. Flat UI
6. Datatables - 1.10
7. date.js
8. sessvars 1.01
9. mmenu

#####Database

1. Postgres 9.3 (for Ubuntu)

#####Browser Support

1. Google Chrome
2. Firefox
3. Safari

Atlas currently supports four modules in addition to the base module which is the AWS module. Modules can be added further through simple and extensible API.

#####Base Module

• aws_module - base module
  • Required for atlas to function.
  • Provides information about all the instances identified by the region, vpc and subnet.

#####Other Modules

• chef_module - Allows users to take/release ownership of any subnet or application.
• cloudability_module - Instance, storage costs on a per instance, per subnet and EC2 basis.
• jenkins_module - Display build information for specific jobs, trigger jenkins actions.
• graphite_module - Display status from graphite server on a per subnet basis.

Atlas is based on the cloud infrastructure of an organization in a yaml (sample - atlas.yaml) file which contains declaration of organizations with environments(production, stage, quality assurance, etc.), profiles for instances(master, slaves, etc.,) and stacks(combination of instance profiles). Instances and stacks can be deployed to subnets in a vpc belonging to a Amazon EC2 regions.

##Steps

The steps in setting up atlas are outlined below:

#####1-Package downloads

1. Refer to installation requirements above.
2. Download all the javascript and css requirements and place the folders and files into the static folder inside the appv1 folder. Verify if all the files and all dependencies confirm to path specifications in base.html template in templates in appv1 folder.

#####2-Create atlas configuration file

1. Create a file in yaml format (atlas_config.yaml) with configuration parameters for each module.
2. Attributes under 'global_config_data' are Required
3. Environment group all is required. Others are optional.
4. Under modules 'aws_module' is required. Other modules are optional
5. Every module can have status, aggregates, columns, tabs and actions.
6. Under actions, instance_actions, instance_group_actions, stack_actions and vpc_actions can be defined.
7. To add or remove any module, include the required attributes under each module. All attributes other than the ones described in 4 & 5 can be included under "other".

######Sample atlas_config.yaml This is a sample atlas_configuration_file.

global_config_data:

  memcache_server_location:
  atlas_log_file: 'none' 

  atlas_cache_variables:
    jobname: 'jenkins_job_name'
    databag: 'chef_databag'
    databag_item: 'chef_databag_item'

  atlas_yaml_databag:
    atlas:
      items:
        main:
          keys:
          - organizations
          - regions
          - stack
          - profile
      
  environment_groups:
  - all
  development:
  - development
  - integration-testing
  - aggregation
  quality_assurance:
  - quality_assurance
  - functional_testing

modules:

  aws_module:

    status:
      count:
        icon_file: '/static/icons/count_icon.ico'
        position: right
      running:
        icon_file: '/static/icons/running_icon.ico'
        position: right
      stopped:
        icon_file: '/static/icons/stopped_icon.ico'
        position: right
      
    columns:
    - instance
    - instance_id
    - subnet
    - status

    tabs:
    - aws_information
    - aws_tags

    actions:

      instance_actions:
      - start_instance
      - stop_instance
      - terminate_instance
      - edit_tags

      instance_group_actions:
      - start_instances
      - stop_instances
      - terminate_instances
    
      stack_action:
      - sample_action_1
      - sample_action_2
    
      vpc_action:
      - sample_action_1
      - sample_action_2
    
    stack_attributes:
    - attribute1:
      stack:
      - all
      editable: false
    - attribute2:
      stack:
      - stack1
      - stack2
      editable: true

    aggregates:
    - count
    - running
    - stopped

    others:
      instances-regex:
  
        knowledgebase:
        - 'kb1\\d{2}'
        cluster:
        - 'ms\\d{3}'
        - 'rs\\d{3}'
        quality-assurance:
        - 'qa\\d{3}'
        
  cloudability_module:
  
    status:
      cost:
        icon_file: '/static/icons/cost_icon.ico'
        position: right
        
    aggregates:
    - cost
      
    columns:
    - instance_cost
    - storage_cost
      
    stack_attributes:
    - total_cost:
      stack:
      - all
      editable: false
      
    others:
      cloudability_base_url: 'https://app.cloudability.com/api/1'
      cloudability_cost_url: '/reporting/cost/run.json?'

#####3-Store infrastructure data on the chef server 2. Store infrastructure data on the Chef server as a databag. (More flexibility and multiple file formats will be included in the next version)

#####4-Postgres setup

1. Ensure postgres database is setup and functional
2. Run "syncdb" to generate tables corresponding to Django models.

#####5-Add new site

1. Log in as admin
2. Add new site.

#####6-Google oauth

1. Using the client key and secret set up the google oauth through the admin UI.
2. Create admin or other user user accounts if needed.

#####7-Environment variables setup

1. set all the required environment variables. * ATLAS_CONFIG_FILEPATH - filepath for atlas configuration data (.yaml) * ATLAS_YAML_FILEPATH - filepath for infrastructure data (.yaml) * APPLICATION_URL - 'http://127.0.0.1:8000/' * POSTGRES_PASSWORD - Postgres database password * POSTGRES_USERNAME - Postgres database username * AWS_ACCESS_KEY_ID - AWS access key ID. * AWS_SECRET_ACCESS_KEY - AWS secret access key * CLOUDABILITY_AUTH_TOKEN - Cloudability API token * JENKINS_URL - Jenkins build server url * JENKINS_USERNAME - Jenkins build server username * JENKINS_PASSWORD - Jenkins build server password

#####8-Set up cron jobs

1. Atlas has a two level cache. The long term cache is 24 hours. The short term depends on the requirements and can be specified in each module. Both the caches are updated whenever data on short term cache expires.

*execute the setatlascache.py management command once every 24 hours. python manage.py setatlascache --options all

2. If chef_module is included, *execute expireownership.py management command with expiry in days, hours or minutes. python manage.py expireownership.py --days 2

Logging is provided at two levels. 1. atlas_debug.log, the log file generated using Django log settings under settings.py 2. Setting atlas_log_file attribute under global_config_data in the atlas_config.yaml configuration file. * If the filepath is set, a log file is generated every day. * If the filepath is set to 'none' then error logs are generated through Django log settings.

Once all the above steps are completed the Atlas console will be up and operational. Users are login using their accounts or their gmail account. The UI is self explanatory and has tool tips on each icon for guidance.

The MIT License (http://opensource.org/licenses/mit-license.php)