django-vault-helpers 0.3.3

Authenticating to Vault

Configure connection settings using environment variables to authenticate to Vault.

In addition to the settings above, you must provide environment variables for one of the authentication methods below.

Database Connection Secrets

To use Vault to load database connection configuration and credentials, configure the Vault database secret backend as described in the Database secret backend documentation. For example:

$ vault mount database
Successfully mounted 'database' at 'database'!
$ CONNECTION_NAME='myapplication'
$ CONNECTION_URL='postgresql://vaultuser:FOO@mydb:5432/myapplication'
$ PARENT_ROLE_NAME='myapplication'
$ vault write "database/config/$CONNECTION_NAME" \
        plugin_name="postgresql-database-plugin" \
        allowed_roles="$CONNECTION_NAME" \
        connection_url="$CONNECTION_URL"
$ vault write "database/roles/$CONNECTION_NAME" \
        db_name="$CONNECTION_NAME" \
        creation_statements="CREATE ROLE \"{{name}}\" WITH LOGIN ENCRYPTED PASSWORD '{{password}}' VALID UNTIL '{{expiration}}' IN ROLE \"${PARENT_ROLE_NAME}\" INHERIT NOCREATEROLE NOCREATEDB NOSUPERUSER NOREPLICATION;" \
        default_ttl="1h" \
        max_ttl="24h"

Next, add settings via the following environment variables.

Finally, edit your projects settings.py file to load database configuration using Vault.

import vaulthelpers

# Load database credentials from Vault
DATABASES = {
    'default': vaulthelpers.database.get_config(),
}

To add additional keys to the database configuration, pass in a dictionary to the get_config call. For example:

import vaulthelpers

# Load database credentials from Vault
DATABASES = {
    'default': vaulthelpers.database.get_config({
        'ATOMIC_REQUESTS': True,
        'CONN_MAX_AGE': 3600,
    }),
}

AWS Credentials

To use Vault to load IAM or STS credentials for AWS, configure the Vault AWS secret backend as described in the AWS secret backend documentation.

$ vault mount aws
Successfully mounted 'aws' at 'aws'!
$ vault write aws/config/root \
        access_key=AKIAJWVN5Z4FOFT7NLNA \
        secret_key=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 \
        region=us-east-1
$ vault write aws/roles/myapplication \
        arn=arn:aws:iam::ACCOUNT-ID-WITHOUT-HYPHENS:role/MyApplicationRoleName

Next, add settings via the following environment variables.

Finally, configure you Django project to load AWS credentials using Vault. To do this, edit the settings.py file to include the following line.

import vaulthelpers

# Load AWS credentials from Vault
vaulthelpers.aws.init_boto3_credentials()

This will override the credential resolve code in boto3 and botocore so that it will fetch credentials from Vault instead of the usual means, like environment variables or the EC2 metadata service.

Direct HVAC Client Access

To directly access the authentication hvac client connector, fetch it from the vaulthelpers.common module.

import vaulthelpers

vault_auth = vaulthelpers.common.get_vault_auth()
verify = vaulthelpers.common.VAULT_CACERT or vaulthelpers.common.VAULT_SSL_VERIFY
vcl = vault_auth.authenticated_client(vaulthelpers.common.VAULT_URL, verify=verify)
result = vcl.read('secret/apps/myaplication')
print(result)

0.3.3

• Fix bug in with passing url parameter to HVAC client in common.EnvironmentConfig

• Improve testing.

• Support Django 2.0

0.3.2

• Prevent recycling TCP connections after forking a process.

0.3.1

• Fixed TCP connection issue by caching VaultAuthenticator instance in thread local storage.

0.3.0

• Add file system caching of vault tokens.

0.2.0

• Add S3 storage backend based on storages.backends.s3boto3.S3Boto3Storage.

0.1.0

• Initial release.