Source code for

# Copyright 2017 The Forseti Security Authors. All rights reserved.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# See the License for the specific language governing permissions and
# limitations under the License.

"""External project access scanner."""
# pylint: disable=line-too-long
import time

from google.auth.exceptions import RefreshError
from import logger
from import api_helpers # noqa=E501
from import resource_util # noqa=E501
from import CloudResourceManagerClient # noqa=E501
from import DataAccess
from import Storage
from import external_project_access_rules_engine as epa_rules_engine # noqa=E501
from import base_scanner

# pylint: enable=line-too-long

LOGGER = logger.get_logger(__name__)

SCOPES = ['']

[docs]def _get_inventory_storage(session, inventory_index_id): """Creates an open inventory. Args: session (object): db session. inventory_index_id (int): The inventory index Returns: Storage: storage object """ inventory_storage = Storage(session, inventory_index_id, True) return inventory_storage
[docs]def get_user_emails(service_config, member_types=None): """Retrieves the list of user email addresses from inventory. Args: service_config (dict): The service configuration member_types (list): Member types to query in storage. This defaults to 'gsuite_user'. Returns: list: List of list of user e-mail addresses. """ if not member_types: member_types = ['gsuite_user'] emails = [] with service_config.scoped_session() as session: inventory_index_id = ( DataAccess.get_latest_inventory_index_id(session)) inventory_storage = _get_inventory_storage(session, inventory_index_id) for inventory_row in inventory_storage.iter(type_list=member_types): emails.append(inventory_row.get_resource_data()['primaryEmail']) return emails
[docs]def extract_project_ids(crm_client): """Extract a list of project ID's Args: crm_client (CloudResourceManagerClient): An authenticated CRM client Returns: list: Project ID's as strings """ project_response = crm_client.get_projects() projects = api_helpers.flatten_list_results(project_response, 'projects') return [project['projectId'] for project in projects]
[docs]def memoize_ancestry(ancestry_function): # pylint: disable=C0301 """A decorator function to intelligently retrieve project ancestries, only if necessary. Args: ancestry_function (function): The ancestry retrieval function. Returns: function: The helper """ discovered_ancestries = {} # pylint: disable=W9011,W9012,C0111 def helper(crm_client, project_id): if project_id not in discovered_ancestries: discovered_ancestries[project_id] = ( ancestry_function(crm_client, project_id)) return discovered_ancestries[project_id] return helper
[docs]@memoize_ancestry def get_project_ancestry(crm_client, project_id): """Get project ancestry as a list of type Resource. Args: crm_client (CloudResourceManagerClient): crm client project_id (str): A project ID Returns: list: Resource objects defining the ancestry chain from the Project to the Organization """ ancestries = crm_client.get_project_ancestry(project_id) ancestry_resources = ( resource_util.cast_to_gcp_resources(ancestries)) return ancestry_resources
[docs]def get_project_ancestries(crm_client, project_id_list): """Get the ancestries from a list of project ID's Args: crm_client (CloudResourceManagerClient): crm client project_id_list (list): A list of project ID's as strings Returns: list: A list of lists ofResource objects defining the ancestrychain from the Project to the Organization """ ancestry_list = [] for project_id in project_id_list: ancestry_list.append(get_project_ancestry(crm_client, project_id)) return ancestry_list
[docs]class ExternalProjectAccessScanner(base_scanner.BaseScanner): """Scanner for external project access.""" def __init__(self, global_configs, scanner_configs, service_config, model_name, snapshot_timestamp, rules): """Initialization. Args: global_configs (dict): Global configurations. scanner_configs (dict): Scanner configurations. service_config (ServiceConfig): Forseti 2.0 service configs model_name (str): name of the data model snapshot_timestamp (str): Timestamp, formatted as YYYYMMDDTHHMMSSZ. rules (str): Fully-qualified path and filename of the rules file. """ super(ExternalProjectAccessScanner, self).__init__( global_configs, scanner_configs, service_config, model_name, snapshot_timestamp, rules) self.inventory_configs = self.service_config.get_inventory_config() self.rules_engine = ( epa_rules_engine.ExternalProjectAccessRulesEngine( rules_file_path=self.rules, snapshot_timestamp=self.snapshot_timestamp)) self.rules_engine.build_rule_book(self.inventory_configs) self._ancestries = dict()
[docs] def _output_results(self, all_violations): """Output results. Args: all_violations (list): A list of violations. """ all_violations = self._flatten_violations(all_violations) self._output_results_to_db(all_violations)
[docs] def _find_violations(self, ancestries_by_user): """Find violations in the policies. Args: ancestries_by_user (dict): The project ancestries collected from the scanner Returns: list: A list of ExternalProjectAccess violations """ all_violations = []'Finding project access violations...') for user_mail, project_ancestries in ancestries_by_user.iteritems(): for project_ancestry in project_ancestries: violations = ( self.rules_engine.find_violations( user_mail, project_ancestry)) all_violations.extend(violations) return all_violations
[docs] @staticmethod def _flatten_violations(violations): """Flatten RuleViolations into a dict for each RuleViolation member. Args: violations (list): The RuleViolations to flatten. Yields: dict: Iterator of RuleViolations as a dict per member. """ for violation in violations: rule_ancestors_names = [] for ancestor in violation.rule_ancestors: rule_ancestors_names.append( violation_data = { 'full_name': violation.full_name, 'member': violation.member, 'rule_ancestors': rule_ancestors_names } yield { 'resource_id': violation.resource_id, 'resource_type': violation.resource_type, 'full_name': violation.full_name, 'rule_index': violation.rule_index, 'rule_name': violation.rule_name, 'violation_type': violation.violation_type, 'violation_data': violation_data, 'resource_data': violation.resource_data }
[docs] def _get_crm_client(self, user_email): """Get a user scoped CloudResourceManagerClient. Args: user_email (str): The e-mail address of the user. Returns: CloudResourceManagerClient: crm client """ user_scoped_credential = ( api_helpers.get_delegated_credential(user_email, SCOPES)) client = CloudResourceManagerClient( global_configs=self.inventory_configs, credentials=user_scoped_credential) return client
[docs] def _retrieve(self): # pylint: disable=line-too-long """Retrieve the project ancestries for all users. Returns: dict: User project relationship. {"": [[Project("1234"), Organization("1234567")], [Project("12345"), Folder("ABCDEFG"), Organization("1234567")]], "": [[Project("1234"), Organization("1234567")], [Project("12345"), Folder("ABCDEFG"), Organization("1234567")]]} """ # pylint: enable=line-too-long # This dictionary is the result of the scan. The key # is the user ID. The value is a list of lists of ancestries. user_to_project_ancestries_map = {} user_count = 0 start_time = time.time() user_emails = get_user_emails(self.service_config) for user_email in user_emails: user_count += 1 try: user_crm_client = self._get_crm_client(user_email) project_ids = extract_project_ids(user_crm_client) ancestries = get_project_ancestries(user_crm_client, project_ids) user_to_project_ancestries_map[user_email] = ancestries except RefreshError: LOGGER.debug('Unable to access project ancestry %s.', user_email) # TODO: Remove when instrumentation is implemented. elapsed_time = time.time() - start_time LOGGER.debug('It took %f seconds to query projects for %d users', elapsed_time, user_count) return user_to_project_ancestries_map
[docs] def run(self): """Entry point to run the scanner.""" user_to_project_ancestries_map = self._retrieve() all_violations = self._find_violations(user_to_project_ancestries_map) self._output_results(all_violations)