db-facts

db-facts

db-facts translates from user-familiar database coordinates (e.g., "redshift", "corppostgres dbadmin", "productoracle impl juser") into detailed instructions on how to access the database in question, providing configuration and templating mechnisms to wrap any credential management tools involved in providing those details and credentials.

Example:

$ db-facts sh redshift
export CONNECTION_TYPE
CONNECTION_TYPE=direct
export LASTPASS_SHARE_NAME_SUFFIX
LASTPASS_SHARE_NAME_SUFFIX='blue labs redshift'
export DB_PASSWORD
DB_PASSWORD='hunter2'
DB_HOST=whatevs.whatevs.us-east-1.redshift.amazonaws.com
export DB_HOST
export DB_DATABASE
DB_DATABASE=analytics
export DB_USERNAME
DB_USERNAME=vbroz
export DB_PORT
DB_PORT=5439
export DB_TYPE
DB_TYPE=redshift
export DB_PROTOCOL
DB_PROTOCOL=postgres

For details on the potential facts returned, see the API reference.

db-facts relies on a config file ("dbcli.yml") which teaches it how to parse the user-friendly coordinates. Much of the heavy lifting in the parsing part is done by jinja_context.py, which sets some variables and functions that can be used in jinja templates within the config file.

If you need to set the instructions immediately to your environment variables, you can do this with the command:

eval $(db-facts sh redshift)

For AWS IAM credentials in Lastpass, ensure that this block exists under exports_from: in your db-facts config:

  lpass_aws_iam:
    pull_lastpass_aws_iam: "{{ lastpass_entry }}"

and set up a db credential with export_type lpass_aws_iam, for example:

  aws_user_example:
    exports_from: lpass_aws_iam
    lastpass_entry: 'AWS IAM: example_user'

You can then credential your shell with this one-liner: eval $(db-facts sh aws_user_example)

You can also access db-facts via a Python API; for details, see the API reference.

Configuration

You can configure db-facts to connect to your databases. See CONFIGURATION.md for details.

Extensions

You can extend db-facts to pull configuration from other systems. See EXTENSIONS.md for details.

Library

To use as a library:

$ python
Python 3.5.2 (default, Sep 12 2016, 09:31:17)
[GCC 4.2.1 Compatible Apple LLVM 7.3.0 (clang-703.0.31)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import bluelabs_db_facts
>>> db_facts.db(['redshift'])
{'protocol': 'postgres', 'lastpass_share_name_suffix': 'blue labs redshift', 'host': 'bl-int-analytics1.cxtyzogmmhiv.us-east-1.redshift.amazonaws.com', 'connection_type': 'direct', 'user': 'vbroz', 'database': 'analytics', 'password': 'hunter1', 'port': 5439, 'type': 'redshift'}
>>>

Development

See DEVELOPMENT.md