ansible-anonymizer 1.5.0

Ansible Anonymizer

Library to clean up Ansible tasks from any Personally Identifiable Information (PII)

• Free software: Apache Software License 2.0

Anonymized fields

• Credit Card number

• email address

• IP address

• MAC address

• US SSN

• US phone number

• YAML comment

• password value, when the field name is identified as being sensitive

• user name from home directory path

Usage

The library can be used to remove the PII from a multi level structure:

from ansible_anonymizer.anonymizer import anonymize_struct

example = [{"name": "foo bar", "email": "[e-mail]"}]

anonymize_struct(example)
# [{'name': 'foo bar', 'email': '[e-mail]'}]

But you can also anonymize a block of text:

from ansible_anonymizer.anonymizer import anonymize_text_block

some_text = """
- name: a task
  a_module:
    secret: foobar
"""

anonymize_text_block(some_text)
# '\n- name: a task\n  a_module:\n    secret: "{{ secret }}"\n'

You can also use the ansible-anonymizer command:

ansible-anonymizer my-secret-file

Customize the anonymized strings

By default, the variables are anonymized with a string based on the name of the field. You can customize it with the value_template parameter:

from ansible_anonymizer.anonymizer import anonymize_struct
from string import Template

original = {"password": "$RvEDSRW#R"}
value_template = Template("_${variable_name}_")
anonymize_struct(original, value_template=value_template)
#  {'password': '_password_'}

Limitations

• anonymize_text_block() relies on its own text parser which only support a subset of YAML features. Because of this, it may not be able to identify some PII. When possible, use anonymize_struct which accepts a Python structure instead.

• The Anonymizer is not a silver bullet and it’s still possible to see PII going through the filters.