Project: ssm-parameter-store

Simple Python wrapper for getting values from AWS Systems Manager Parameter Store

Project Details

Latest version
19.11.0
Home Page
http://github.com/christippett/ssm-parameter-store
PyPI Page
https://pypi.org/project/ssm-parameter-store/

Project Popularity

PageRank
0.0015543580435259964
Number of downloads
80579

SSM Parameter Store

PyPI version Build status Coverage Python versions Github license

Description

This is a simple Python wrapper for getting values from AWS Systems Manager Parameter Store.

The module supports getting a single parameter, multiple parameters or all parameters matching a particular path.

All parameters are returned as a Python dict.

Installation

Install with pip:

pip install ssm-parameter-store

Usage

Import the module and create a new instance of EC2ParameterStore.

from ssm_parameter_store import EC2ParameterStore

store = EC2ParameterStore()

AWS Credentials

ssm-parameter-store uses boto3 under the hood and therefore inherits the same mechanism for looking up AWS credentials. See configuring credentials in the Boto 3 documentation for more information.

EC2ParameterStore accepts all boto3 client parameters as keyword arguments.

For example:

from ssm_parameter_store import EC2ParameterStore

store = EC2ParameterStore(
    aws_access_key_id=ACCESS_KEY,
    aws_secret_access_key=SECRET_KEY,
    aws_session_token=SESSION_TOKEN,  # optional
    region_name='us-west-2'
)

Examples

Given the following parameters:

# set default AWS region
AWS_DEFAULT_REGION=us-west-2

# add parameters
aws ssm put-parameter --name "param1" --value "value1" --type SecureString
aws ssm put-parameter --name "param2" --value "value2" --type SecureString

# add parameters organised by hierarchy
aws ssm put-parameter --name "/dev/app/secret" --value "dev_secret" --type SecureString
aws ssm put-parameter --name "/dev/db/postgres_username" --value "dev_username" --type SecureString
aws ssm put-parameter --name "/dev/db/postgres_password" --value "dev_password" --type SecureString
aws ssm put-parameter --name "/prod/app/secret" --value "prod_secret" --type SecureString
aws ssm put-parameter --name "/prod/db/postgres_username" --value "prod_username" --type SecureString
aws ssm put-parameter --name "/prod/db/postgres_password" --value "prod_password" --type SecureString

Get a single parameter

parameter = store.get_parameter('param1', decrypt=True)

assert parameter == {
   'param1': 'value1'
}

Get multiple parameters

parameters = store.get_parameters(['param1', 'param2'])

assert parameters == {
   'param1': 'value1',
   'param2': 'value2',
}

Get parameters by path

parameters = store.get_parameters_by_path('/dev/', recursive=True)

assert parameters == {
    'secret': 'dev_secret',
    'postgres_username': 'dev_username',
    'postgres_password': 'dev_password',
}

By default get_parameters_by_path strips the path from each parameter name. To return a parameter's full name, set strip_path to False.

parameters = store.get_parameters_by_path('/dev/', strip_path=False, recursive=True)

assert parameters == {
    '/dev/app/secret': 'dev_secret',
    '/dev/db/postgres_username': 'dev_username',
    '/dev/db/postgres_password': 'dev_password'
}

Get parameters with original hierarchy

You can also get parameters by path, but in a nested structure that models the path hierarchy.

parameters = store.get_parameters_with_hierarchy('/dev/')

assert parameters == {
    'app': {
        'secret': 'dev_secret',
    },
    'db': {
        'postgres_username': 'dev_username',
        'postgres_password': 'dev_password',
    },
}

By default get_parameters_with_hierarchy strips the leading path component. To return the selected parameters with the full hierarchy, set strip_path to False.

parameters = store.get_parameters_with_hierarchy('/dev/', strip_path=False)

assert parameters == {
    'dev': {
        'app': {
            'secret': 'dev_secret',
        },
        'db': {
            'postgres_username': 'dev_username',
            'postgres_password': 'dev_password',
        },
    },
}

Populating Environment Variables

The module includes a static method on EC2ParameterStore to help populate environment variables. This can be helpful when integrating with a library like django-environ.

Example

Given the following parameters:

aws ssm put-parameter --name "/prod/django/SECRET_KEY" --value "-$y_^@69bm69+z!fawbdf=h_10+zjzfwr8_c=$$&j@-%p$%ct^" --type SecureString
aws ssm put-parameter --name "/prod/django/DATABASE_URL" --value "psql://user:pass@db-prod.xyz123.us-west-2.rds.amazonaws.com:5432/db" --type SecureString
aws ssm put-parameter --name "/prod/django/REDIS_URL" --value "redis://redis-prod.edc1ba.0001.usw2.cache.amazonaws.com:6379" --type SecureString
import environ
from ssm_parameter_store import EC2ParameterStore

env = environ.Env(
    DEBUG=(bool, False)
)

# Get parameters and populate os.environ (region not required if AWS_DEFAULT_REGION environment variable set)
parameter_store = EC2ParameterStore(region_name='us-west-2')
django_parameters = parameter_store.get_parameters_by_path('/prod/django/', strip_path=True)
EC2ParameterStore.set_env(django_parameters)

# False if not in os.environ
DEBUG = env('DEBUG')

# Raises django's ImproperlyConfigured exception if SECRET_KEY not in os.environ
SECRET_KEY = env('SECRET_KEY')

DATABASES = {
    # read os.environ['DATABASE_URL'] and raises ImproperlyConfigured exception if not found
    'default': env.db(),
}

CACHES = {
    'default': env.cache('REDIS_URL'),
}

Related Projects

  • param-store – Python module to store secrets in secret stores
  • ssm-cache – AWS System Manager Parameter Store caching client for Python