def main(): """Main function for turbiniactl""" # TODO(aarontp): Allow for single run mode when # by specifying evidence which will also terminate the task manager after # evidence has been processed. parser = argparse.ArgumentParser() parser.add_argument( '-q', '--quiet', action='store_true', help='Show minimal output') parser.add_argument( '-v', '--verbose', action='store_true', help='Show verbose output', default=True) parser.add_argument( '-d', '--debug', action='store_true', help='Show debug output', default=False) parser.add_argument( '-a', '--all_fields', action='store_true', help='Show all task status fields in output', required=False) parser.add_argument( '-c', '--config_file', help='Load explicit config file. If specified it ' 'will ignore config files in other default locations ' '(/etc/turbinia.conf, ~/.turbiniarc, or in paths referenced in ' 'environment variable TURBINIA_CONFIG_PATH)', required=False) parser.add_argument( '-C', '--recipe_config', help='Recipe configuration data passed in as ' 'comma separated key=value pairs (e.g. ' '"-C key=value,otherkey=othervalue"). These will get passed to tasks ' 'as evidence config, and will also be written to the metadata.json file ' 'for Evidence types that write it', default=[], type=csv_list) parser.add_argument( '-f', '--force_evidence', action='store_true', help='Force evidence processing request in potentially unsafe conditions', required=False) parser.add_argument('-o', '--output_dir', help='Directory path for output') parser.add_argument('-L', '--log_file', help='Log file') parser.add_argument( '-r', '--request_id', help='Create new requests with this Request ID', required=False) parser.add_argument( '-R', '--run_local', action='store_true', help='Run completely locally without any server or other infrastructure. ' 'This can be used to run one-off Tasks to process data locally.') parser.add_argument( '-S', '--server', action='store_true', help='Run Turbinia Server indefinitely') parser.add_argument( '-V', '--version', action='version', version=__version__, help='Show the version') parser.add_argument( '-D', '--dump_json', action='store_true', help='Dump JSON output of Turbinia Request instead of sending it') parser.add_argument( '-F', '--filter_patterns_file', help='A file containing newline separated string patterns to filter ' 'text based evidence files with (in extended grep regex format). ' 'This filtered output will be in addition to the complete output') parser.add_argument( '-Y', '--yara_rules_file', help='A file containing Yara rules.') parser.add_argument( '-j', '--jobs_allowlist', default=[], type=csv_list, help='An allowlist for Jobs that will be allowed to run (in CSV format, ' 'no spaces). This will not force them to run if they are not configured ' 'to. This is applied both at server start time and when the client makes ' 'a processing request. When applied at server start time the change is ' 'persistent while the server is running. When applied by the client, it ' 'will only affect that processing request.') parser.add_argument( '-J', '--jobs_denylist', default=[], type=csv_list, help='A denylist for Jobs we will not allow to run. See ' '--jobs_allowlist help for details on format and when it is applied.') parser.add_argument( '-p', '--poll_interval', default=60, type=int, help='Number of seconds to wait between polling for task state info') parser.add_argument( '-t', '--task', help='The name of a single Task to run locally (must be used with ' '--run_local.') parser.add_argument( '-w', '--wait', action='store_true', help='Wait to exit until all tasks for the given request have completed') subparsers = parser.add_subparsers( dest='command', title='Commands', metavar='<command>') # Action for printing config parser_config = subparsers.add_parser('config', help='Print out config file') parser_config.add_argument( '-f', '--file_only', action='store_true', help='Print out file path only') #Sends Test Notification parser_testnotify = subparsers.add_parser( 'testnotify', help='Sends test notification') # TODO(aarontp): Find better way to specify these that allows for multiple # pieces of evidence to be submitted. Maybe automagically create different # commands based on introspection of evidence objects? # RawDisk parser_rawdisk = subparsers.add_parser( 'rawdisk', help='Process RawDisk as Evidence') parser_rawdisk.add_argument( '-l', '--source_path', help='Local path to the evidence', required=True) parser_rawdisk.add_argument( '-P', '--mount_partition', default=1, type=int, help='The partition number to use when mounting this disk. Defaults to ' 'the entire raw disk. Only affects mounting, and not what gets ' 'processed.') parser_rawdisk.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_rawdisk.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for APFS Disk Evidence type parser_apfs = subparsers.add_parser( 'apfs', help='Process APFSEncryptedDisk as Evidence') parser_apfs.add_argument( '-l', '--source_path', help='Local path to the encrypted APFS evidence', required=True) parser_apfs.add_argument( '-r', '--recovery_key', help='Recovery key for the APFS evidence. ' 'Either recovery key or password must be specified.', required=False) parser_apfs.add_argument( '-p', '--password', help='Password for the APFS evidence. ' 'If a recovery key is specified concurrently, password will be ignored.', required=False) parser_apfs.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_apfs.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for Bitlocker Disk Evidence type parser_bitlocker = subparsers.add_parser( 'bitlocker', help='Process Bitlocker Disk as Evidence') parser_bitlocker.add_argument( '-l', '--source_path', help='Local path to the encrypted Bitlocker evidence', required=True) parser_bitlocker.add_argument( '-r', '--recovery_key', help='Recovery key for the Bitlocker evidence. ' 'Either recovery key or password must be specified.', required=False) parser_bitlocker.add_argument( '-p', '--password', help='Password for the Bitlocker evidence. ' 'If a recovery key is specified concurrently, password will be ignored.', required=False) parser_bitlocker.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_bitlocker.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for Google Cloud Disk Evidence type parser_googleclouddisk = subparsers.add_parser( 'googleclouddisk', help='Process Google Cloud Persistent Disk as Evidence') parser_googleclouddisk.add_argument( '-C', '--copy_only', action='store_true', help='Only copy disk and do ' 'not process with Turbinia. This only takes effect when a source ' '--project is defined and can be run without any Turbinia server or ' 'workers configured.') parser_googleclouddisk.add_argument( '-d', '--disk_name', help='Google Cloud name for disk', required=True) parser_googleclouddisk.add_argument( '-p', '--project', help='Project that the disk to process is associated ' 'with. If this is different from the project that Turbinia is running ' 'in, it will be copied to the Turbinia project.') parser_googleclouddisk.add_argument( '-P', '--mount_partition', default=1, type=int, help='The partition number to use when mounting this disk. Defaults to ' 'the entire raw disk. Only affects mounting, and not what gets ' 'processed.') parser_googleclouddisk.add_argument( '-z', '--zone', help='Geographic zone the disk exists in') parser_googleclouddisk.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_googleclouddisk.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for Google Cloud Persistent Disk Embedded Raw Image parser_googleclouddiskembedded = subparsers.add_parser( 'googleclouddiskembedded', help='Process Google Cloud Persistent Disk with an embedded raw disk ' 'image as Evidence') parser_googleclouddiskembedded.add_argument( '-C', '--copy_only', action='store_true', help='Only copy disk and do ' 'not process with Turbinia. This only takes effect when a source ' '--project is defined and can be run without any Turbinia server or ' 'workers configured.') parser_googleclouddiskembedded.add_argument( '-e', '--embedded_path', help='Path within the Persistent Disk that points to the raw image file', required=True) parser_googleclouddiskembedded.add_argument( '-d', '--disk_name', help='Google Cloud name for disk', required=True) parser_googleclouddiskembedded.add_argument( '-p', '--project', help='Project that the disk to process is associated ' 'with. If this is different from the project that Turbinia is running ' 'in, it will be copied to the Turbinia project.') parser_googleclouddiskembedded.add_argument( '-P', '--mount_partition', default=1, type=int, help='The partition number to use when mounting this disk. Defaults to ' 'the entire raw disk. Only affects mounting, and not what gets ' 'processed.') parser_googleclouddiskembedded.add_argument( '--embedded_mount_partition', default=1, type=int, help='The partition number to use when mounting this embedded disk image.' ' Defaults to the first partition') parser_googleclouddiskembedded.add_argument( '-z', '--zone', help='Geographic zone the disk exists in') parser_googleclouddiskembedded.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_googleclouddiskembedded.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # RawMemory parser_rawmemory = subparsers.add_parser( 'rawmemory', help='Process RawMemory as Evidence') parser_rawmemory.add_argument( '-l', '--source_path', help='Local path to the evidence', required=True) parser_rawmemory.add_argument( '-P', '--profile', help='Profile to use with Volatility', required=True) parser_rawmemory.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) parser_rawmemory.add_argument( '-m', '--module_list', type=csv_list, help='Volatility module(s) to execute', required=True) # Parser options for Directory evidence type parser_directory = subparsers.add_parser( 'directory', help='Process a directory as Evidence') parser_directory.add_argument( '-l', '--source_path', help='Local path to the evidence', required=True) parser_directory.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_directory.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for CompressedDirectory evidence type parser_directory = subparsers.add_parser( 'compressedirectory', help='Process a compressed tar file as Evidence') parser_directory.add_argument( '-l', '--local_path', help='Local path to the evidence', required=True) parser_directory.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_directory.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for ChromiumProfile evidence type parser_hindsight = subparsers.add_parser( 'hindsight', help='Process ChromiumProfile as Evidence') parser_hindsight.add_argument( '-l', '--source_path', help='Local path to the evidence', required=True) parser_hindsight.add_argument( '-f', '--format', help='Output format (supported types are ' 'xlsx, sqlite, jsonl)', default='sqlite') parser_hindsight.add_argument( '-b', '--browser_type', help='The type of browser the input files belong' 'to (supported types are Chrome, Brave)', default='Chrome') parser_hindsight.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # List Jobs subparsers.add_parser( 'listjobs', help='List all available Jobs. These Job names can be used by ' '--jobs_allowlist and --jobs_denylist') # PSQ Worker parser_psqworker = subparsers.add_parser('psqworker', help='Run PSQ worker') parser_psqworker.add_argument( '-S', '--single_threaded', action='store_true', help='Run PSQ Worker in a single thread', required=False) # Celery Worker subparsers.add_parser('celeryworker', help='Run Celery worker') # Parser options for Turbinia status command parser_status = subparsers.add_parser( 'status', help='Get Turbinia Task status') parser_status.add_argument( '-c', '--close_tasks', action='store_true', help='Close tasks based on Request ID or Task ID', required=False) parser_status.add_argument( '-C', '--csv', action='store_true', help='When used with --statistics, the output will be in CSV format', required=False) parser_status.add_argument( '-d', '--days_history', default=0, type=int, help='Number of days of history to show', required=False) parser_status.add_argument( '-f', '--force', help='Gatekeeper for --close_tasks', action='store_true', required=False) parser_status.add_argument( '-r', '--request_id', help='Show tasks with this Request ID', required=False) # 20 == Priority.High. We are setting this manually here because we don't want # to load the worker module yet in order to access this Enum. parser_status.add_argument( '-p', '--priority_filter', default=20, type=int, required=False, help='This sets what report sections are shown in full detail in ' 'report output. Any tasks that have set a report_priority value ' 'equal to or lower than this setting will be shown in full detail, and ' 'tasks with a higher value will only have a summary shown. To see all ' 'tasks report output in full detail, set --priority_filter=100') parser_status.add_argument( '-R', '--full_report', help='Generate full markdown report instead of just a summary', action='store_true', required=False) parser_status.add_argument( '-s', '--statistics', help='Generate statistics only', action='store_true', required=False) parser_status.add_argument( '-t', '--task_id', help='Show task for given Task ID', required=False) parser_status.add_argument( '-u', '--user', help='Show task for given user', required=False) # Server subparsers.add_parser('server', help='Run Turbinia Server') args = parser.parse_args() # Load the config before final logger setup so we can the find the path to the # log file. try: if args.config_file: config.LoadConfig(config_file=args.config_file) else: config.LoadConfig() except TurbiniaException as exception: print( 'Could not load config file ({0!s}).\n{1:s}'.format( exception, config.CONFIG_MSG)) sys.exit(1) if args.log_file: config.LOG_FILE = args.log_file if args.output_dir: config.OUTPUT_DIR = args.output_dir # Run logger setup again to get file-handler now that we have the logfile path # from the config. logger.setup() if args.quiet: log.setLevel(logging.ERROR) elif args.debug: log.setLevel(logging.DEBUG) else: log.setLevel(logging.INFO) # Enable GCP Stackdriver Logging if config.STACKDRIVER_LOGGING and args.command in ('server', 'psqworker'): google_cloud.setup_stackdriver_handler(config.TURBINIA_PROJECT) log.info('Turbinia version: {0:s}'.format(__version__)) # Do late import of other needed Turbinia modules. This is needed because the # config is loaded by these modules at load time, and we want to wait to load # the config until after we parse the args so that we can use those arguments # to point to config paths. from turbinia import notify from turbinia import client as TurbiniaClientProvider from turbinia.client import TurbiniaCeleryClient from turbinia.client import TurbiniaServer from turbinia.client import TurbiniaCeleryWorker from turbinia.client import TurbiniaPsqWorker from turbinia import evidence from turbinia.message import TurbiniaRequest # Print out config if requested if args.command == 'config': if args.file_only: log.info('Config file path is {0:s}\n'.format(config.configSource)) sys.exit(0) try: with open(config.configSource, "r") as f: print(f.read()) sys.exit(0) except IOError as exception: log.info( "Failed to read config file {0:s}: {1!s}".format( config.configSource, exception)) sys.exit(1) #sends test notification if args.command == 'testnotify': notify.sendmail( config.EMAIL_ADDRESS, 'Turbinia test notification', 'This is a test notification') sys.exit(0) args.jobs_allowlist = [j.lower() for j in args.jobs_allowlist] args.jobs_denylist = [j.lower() for j in args.jobs_denylist] if args.jobs_allowlist and args.jobs_denylist: log.error( 'A Job filter allowlist and denylist cannot be specified at the same ' 'time') sys.exit(1) # Read set set filter_patterns filter_patterns = None if (args.filter_patterns_file and not os.path.exists(args.filter_patterns_file)): log.error('Filter patterns file {0:s} does not exist.') sys.exit(1) elif args.filter_patterns_file: try: filter_patterns = open(args.filter_patterns_file).read().splitlines() except IOError as e: log.warning( 'Cannot open file {0:s} [{1!s}]'.format(args.filter_patterns_file, e)) # Read yara rules yara_rules = None if (args.yara_rules_file and not os.path.exists(args.yara_rules_file)): log.error('Filter patterns file {0:s} does not exist.') sys.exit(1) elif args.yara_rules_file: try: yara_rules = open(args.yara_rules_file).read() except IOError as e: log.warning( 'Cannot open file {0:s} [{1!s}]'.format(args.yara_rules_file, e)) sys.exit(1) # Create Client object client = None if args.command not in ('psqworker', 'server'): client = TurbiniaClientProvider.get_turbinia_client(args.run_local) # Make sure run_local flags aren't conflicting with other server/client flags server_flags_set = args.server or args.command == 'server' worker_flags_set = args.command in ('psqworker', 'celeryworker') if args.run_local and (server_flags_set or worker_flags_set): log.error('--run_local flag is not compatible with server/worker flags') sys.exit(1) if args.run_local and not args.task: log.error('--run_local flag requires --task flag') sys.exit(1) # Set zone/project to defaults if flags are not set, and also copy remote # disk if needed. if args.command in ('googleclouddisk', 'googleclouddiskrawembedded'): if not args.zone and config.TURBINIA_ZONE: args.zone = config.TURBINIA_ZONE elif not args.zone and not config.TURBINIA_ZONE: log.error('Turbinia zone must be set by --zone or in config') sys.exit(1) if not args.project and config.TURBINIA_PROJECT: args.project = config.TURBINIA_PROJECT elif not args.project and not config.TURBINIA_PROJECT: log.error('Turbinia project must be set by --project or in config') sys.exit(1) if ((args.project and args.project != config.TURBINIA_PROJECT) or (args.zone and args.zone != config.TURBINIA_ZONE)): new_disk = gcp_forensics.CreateDiskCopy( args.project, config.TURBINIA_PROJECT, None, config.TURBINIA_ZONE, disk_name=args.disk_name) args.disk_name = new_disk.name if args.copy_only: log.info('--copy_only specified, so not processing with Turbinia') sys.exit(0) # Start Evidence configuration evidence_ = None if args.command == 'rawdisk': args.name = args.name if args.name else args.source_path source_path = os.path.abspath(args.source_path) evidence_ = evidence.RawDisk( name=args.name, source_path=source_path, mount_partition=args.mount_partition, source=args.source) elif args.command == 'apfs': if not args.password and not args.recovery_key: log.error('Neither recovery key nor password is specified.') sys.exit(1) args.name = args.name if args.name else args.source_path source_path = os.path.abspath(args.source_path) evidence_ = evidence.APFSEncryptedDisk( name=args.name, source_path=source_path, recovery_key=args.recovery_key, password=args.password, source=args.source) elif args.command == 'bitlocker': if not args.password and not args.recovery_key: log.error('Neither recovery key nor password is specified.') sys.exit(1) args.name = args.name if args.name else args.source_path source_path = os.path.abspath(args.source_path) evidence_ = evidence.BitlockerDisk( name=args.name, source_path=source_path, recovery_key=args.recovery_key, password=args.password, source=args.source) elif args.command == 'directory': args.name = args.name if args.name else args.source_path source_path = os.path.abspath(args.source_path) if not config.SHARED_FILESYSTEM: log.info( 'A Cloud Only Architecture has been detected. ' 'Compressing the directory for GCS upload.') source_path = archive.CompressDirectory(source_path, output_path='/tmp') args.name = args.name if args.name else source_path evidence_ = evidence.CompressedDirectory( name=args.name, source_path=source_path, source=args.source) else: evidence_ = evidence.Directory( name=args.name, source_path=source_path, source=args.source) elif args.command == 'compressedirectory': archive.ValidateTarFile(args.source_path) args.name = args.name if args.name else args.source_path source_path = os.path.abspath(args.source_path) evidence_ = evidence.CompressedDirectory( name=args.name, source_path=source_path, source=args.source) elif args.command == 'googleclouddisk': args.name = args.name if args.name else args.disk_name evidence_ = evidence.GoogleCloudDisk( name=args.name, disk_name=args.disk_name, project=args.project, mount_partition=args.mount_partition, zone=args.zone, source=args.source) elif args.command == 'googleclouddiskembedded': args.name = args.name if args.name else args.disk_name parent_evidence_ = evidence.GoogleCloudDisk( name=args.name, disk_name=args.disk_name, project=args.project, mount_partition=args.mount_partition, zone=args.zone, source=args.source) evidence_ = evidence.GoogleCloudDiskRawEmbedded( name=args.name, disk_name=args.disk_name, project=args.project, mount_partition=args.mount_partition, zone=args.zone, embedded_path=args.embedded_path, embedded_partition=args.embedded_mount_partition) evidence_.parent_evidence = parent_evidence_ elif args.command == 'hindsight': if args.format not in ['xlsx', 'sqlite', 'jsonl']: log.error('Invalid output format.') sys.exit(1) if args.browser_type not in ['Chrome', 'Brave']: log.error('Browser type not supported.') sys.exit(1) args.name = args.name if args.name else args.source_path source_path = os.path.abspath(args.source_path) evidence_ = evidence.ChromiumProfile( name=args.name, source_path=source_path, output_format=args.format, browser_type=args.browser_type) elif args.command == 'rawmemory': args.name = args.name if args.name else args.source_path source_path = os.path.abspath(args.source_path) evidence_ = evidence.RawMemory( name=args.name, source_path=source_path, profile=args.profile, module_list=args.module_list) elif args.command == 'psqworker': # Set up root logger level which is normally set by the psqworker command # which we are bypassing. logger.setup() worker = TurbiniaPsqWorker( jobs_denylist=args.jobs_denylist, jobs_allowlist=args.jobs_allowlist) worker.start() elif args.command == 'celeryworker': logger.setup() worker = TurbiniaCeleryWorker( jobs_denylist=args.jobs_denylist, jobs_allowlist=args.jobs_allowlist) worker.start() elif args.command == 'server': server = TurbiniaServer( jobs_denylist=args.jobs_denylist, jobs_allowlist=args.jobs_allowlist) server.start() elif args.command == 'status': region = config.TURBINIA_REGION if args.close_tasks: if args.user or args.request_id or args.task_id: print( client.close_tasks( instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=args.request_id, task_id=args.task_id, user=args.user, requester=getpass.getuser())) sys.exit(0) else: log.info( '--close_tasks (-c) requires --user, --request_id, or/and --task_id' ) sys.exit(1) if args.statistics: print( client.format_task_statistics( instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, days=args.days_history, task_id=args.task_id, request_id=args.request_id, user=args.user, csv=args.csv)) sys.exit(0) if args.wait and args.request_id: client.wait_for_request( instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=args.request_id, user=args.user, poll_interval=args.poll_interval) elif args.wait and not args.request_id: log.info( '--wait requires --request_id, which is not specified. ' 'turbiniactl will exit without waiting.') print( client.format_task_status( instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, days=args.days_history, task_id=args.task_id, request_id=args.request_id, user=args.user, all_fields=args.all_fields, full_report=args.full_report, priority_filter=args.priority_filter)) elif args.command == 'listjobs': log.info('Available Jobs:') client.list_jobs() else: log.warning('Command {0!s} not implemented.'.format(args.command)) if evidence_ and not args.force_evidence: if config.SHARED_FILESYSTEM and evidence_.cloud_only: log.error( 'The evidence type {0:s} is Cloud only, and this instance of ' 'Turbinia is not a cloud instance.'.format(evidence_.type)) sys.exit(1) elif not config.SHARED_FILESYSTEM and not evidence_.cloud_only: log.error( 'The evidence type {0:s} cannot run on Cloud instances of ' 'Turbinia. Consider wrapping it in a ' 'GoogleCloudDiskRawEmbedded or other Cloud compatible ' 'object'.format(evidence_.type)) sys.exit(1) # If we have evidence to process and we also want to run as a server, then # we'll just process the evidence directly rather than send it through the # PubSub frontend interface. If we're not running as a server then we will # create a new TurbiniaRequest and send it over PubSub. request = None if evidence_ and args.server: server = TurbiniaServer() server.add_evidence(evidence_) server.start() elif evidence_: request = TurbiniaRequest( request_id=args.request_id, requester=getpass.getuser()) request.evidence.append(evidence_) if filter_patterns: request.recipe['filter_patterns'] = filter_patterns if args.jobs_denylist: request.recipe['jobs_denylist'] = args.jobs_denylist if args.jobs_allowlist: request.recipe['jobs_allowlist'] = args.jobs_allowlist if yara_rules: request.recipe['yara_rules'] = yara_rules if args.recipe_config: for pair in args.recipe_config: try: key, value = pair.split('=') except ValueError as exception: log.error( 'Could not parse key=value pair [{0:s}] from recipe config ' '{1!s}: {2!s}'.format(pair, args.recipe_config, exception)) sys.exit(1) request.recipe[key] = value if args.dump_json: print(request.to_json().encode('utf-8')) sys.exit(0) else: log.info( 'Creating request {0:s} with evidence {1:s}'.format( request.request_id, evidence_.name)) log.info( 'Run command "turbiniactl status -r {0:s}" to see the status of' ' this request and associated tasks'.format(request.request_id)) if not args.run_local: client.send_request(request) else: log.debug('--run_local specified so not sending request to server') if args.wait: log.info( 'Waiting for request {0:s} to complete'.format(request.request_id)) region = config.TURBINIA_REGION client.wait_for_request( instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=request.request_id, poll_interval=args.poll_interval) print( client.format_task_status( instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=request.request_id, all_fields=args.all_fields)) if args.run_local and not evidence_: log.error('Evidence must be specified if using --run_local') sys.exit(1) if args.run_local and evidence_.cloud_only: log.error('--run_local cannot be used with Cloud only Evidence types') sys.exit(1) if args.run_local and evidence_: result = client.run_local_task(args.task, request) log.info('Task execution result: {0:s}'.format(result)) log.info('Done.') sys.exit(0)
def main(): """Main function for turbiniactl""" # TODO(aarontp): Allow for single run mode when # by specifying evidence which will also terminate the task manager after # evidence has been processed. parser = argparse.ArgumentParser() parser.add_argument('-q', '--quiet', action='store_true', help='Show minimal output') parser.add_argument('-v', '--verbose', action='store_true', help='Show verbose output') # TODO(aarontp): Turn off debug by default parser.add_argument('-d', '--debug', action='store_true', help='Show debug output', default=True) parser.add_argument('-a', '--all_fields', action='store_true', help='Show all task status fields in output', required=False) parser.add_argument( '-f', '--force_evidence', action='store_true', help= 'Force evidence processing request in potentially unsafe conditions', required=False) parser.add_argument('-o', '--output_dir', help='Directory path for output') parser.add_argument('-L', '--log_file', help='Log file') parser.add_argument('-r', '--request_id', help='Create new requests with this Request ID', required=False) parser.add_argument( '-R', '--run_local', action='store_true', help= 'Run completely locally without any server or other infrastructure. ' 'This can be used to run one-off Tasks to process data locally.') parser.add_argument('-S', '--server', action='store_true', help='Run Turbinia Server indefinitely') parser.add_argument( '-C', '--use_celery', action='store_true', help='Pass this flag when using Celery/Kombu for task queuing and ' 'messaging (instead of Google PSQ/pubsub)') parser.add_argument('-V', '--version', action='version', version=__version__, help='Show the version') parser.add_argument( '-D', '--dump_json', action='store_true', help='Dump JSON output of Turbinia Request instead of sending it') parser.add_argument( '-F', '--filter_patterns_file', help='A file containing newline separated string patterns to filter ' 'text based evidence files with (in extended grep regex format). ' 'This filtered output will be in addition to the complete output') parser.add_argument( '-j', '--jobs_whitelist', default=[], type=csv_list, help='A whitelist for Jobs that will be allowed to run (in CSV format, ' 'no spaces). This will not force them to run if they are not configured ' 'to. This is applied both at server start time and when the client makes ' 'a processing request. When applied at server start time the change is ' 'persistent while the server is running. When applied by the client, it ' 'will only affect that processing request.') parser.add_argument( '-J', '--jobs_blacklist', default=[], type=csv_list, help='A blacklist for Jobs we will not allow to run. See ' '--jobs_whitelist help for details on format and when it is applied.') parser.add_argument( '-p', '--poll_interval', default=60, type=int, help='Number of seconds to wait between polling for task state info') parser.add_argument( '-t', '--task', help='The name of a single Task to run locally (must be used with ' '--run_local.') parser.add_argument( '-w', '--wait', action='store_true', help='Wait to exit until all tasks for the given request have completed' ) subparsers = parser.add_subparsers(dest='command', title='Commands', metavar='<command>') # TODO(aarontp): Find better way to specify these that allows for multiple # pieces of evidence to be submitted. Maybe automagically create different # commands based on introspection of evidence objects? # RawDisk parser_rawdisk = subparsers.add_parser('rawdisk', help='Process RawDisk as Evidence') parser_rawdisk.add_argument('-l', '--local_path', help='Local path to the evidence', required=True) parser_rawdisk.add_argument( '-P', '--mount_partition', default=1, type=int, help='The partition number to use when mounting this disk. Defaults to ' 'the entire raw disk. Only affects mounting, and not what gets ' 'processed.') parser_rawdisk.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_rawdisk.add_argument('-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for Bitlocker Disk Evidence type parser_bitlocker = subparsers.add_parser( 'bitlocker', help='Process Bitlocker Disk as Evidence') parser_bitlocker.add_argument( '-l', '--local_path', help='Local path to the encrypted Bitlocker evidence', required=True) parser_bitlocker.add_argument( '-r', '--recovery_key', help='Recovery key for the Bitlocker evidence. ' 'Either recovery key or password must be specified.', required=False) parser_bitlocker.add_argument( '-p', '--password', help='Password for the Bitlocker evidence. ' 'If a recovery key is specified concurrently, password will be ignored.', required=False) parser_bitlocker.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_bitlocker.add_argument('-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for Google Cloud Disk Evidence type parser_googleclouddisk = subparsers.add_parser( 'googleclouddisk', help='Process Google Cloud Persistent Disk as Evidence') parser_googleclouddisk.add_argument('-d', '--disk_name', help='Google Cloud name for disk', required=True) parser_googleclouddisk.add_argument( '-p', '--project', help='Project that the disk is associated with', required=True) parser_googleclouddisk.add_argument( '-P', '--mount_partition', default=0, type=int, help='The partition number to use when mounting this disk. Defaults to ' 'the entire raw disk. Only affects mounting, and not what gets ' 'processed.') parser_googleclouddisk.add_argument( '-z', '--zone', help='Geographic zone the disk exists in', required=True) parser_googleclouddisk.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_googleclouddisk.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for Google Cloud Persistent Disk Embedded Raw Image parser_googleclouddiskembedded = subparsers.add_parser( 'googleclouddiskembedded', help='Process Google Cloud Persistent Disk with an embedded raw disk ' 'image as Evidence') parser_googleclouddiskembedded.add_argument( '-e', '--embedded_path', help= 'Path within the Persistent Disk that points to the raw image file', required=True) parser_googleclouddiskembedded.add_argument( '-d', '--disk_name', help='Google Cloud name for disk', required=True) parser_googleclouddiskembedded.add_argument( '-p', '--project', help='Project that the disk is associated with', required=True) parser_googleclouddiskembedded.add_argument( '-P', '--mount_partition', default=0, type=int, help='The partition number to use when mounting this disk. Defaults to ' 'the entire raw disk. Only affects mounting, and not what gets ' 'processed.') parser_googleclouddiskembedded.add_argument( '-z', '--zone', help='Geographic zone the disk exists in', required=True) parser_googleclouddiskembedded.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_googleclouddiskembedded.add_argument( '-n', '--name', help='Descriptive name of the evidence', required=False) # Parser options for Directory evidence type parser_directory = subparsers.add_parser( 'directory', help='Process a directory as Evidence') parser_directory.add_argument('-l', '--local_path', help='Local path to the evidence', required=True) parser_directory.add_argument( '-s', '--source', help='Description of the source of the evidence', required=False) parser_directory.add_argument('-n', '--name', help='Descriptive name of the evidence', required=False) # List Jobs subparsers.add_parser( 'listjobs', help='List all available Jobs. These Job names can be used by ' '--jobs_whitelist and --jobs_blacklist') # PSQ Worker parser_psqworker = subparsers.add_parser('psqworker', help='Run PSQ worker') parser_psqworker.add_argument('-S', '--single_threaded', action='store_true', help='Run PSQ Worker in a single thread', required=False) # Celery Worker subparsers.add_parser('celeryworker', help='Run Celery worker') # Parser options for Turbinia status command parser_status = subparsers.add_parser('status', help='Get Turbinia Task status') parser_status.add_argument( '-c', '--close_tasks', action='store_true', help='Close tasks based on Request ID or Task ID', required=False) parser_status.add_argument('-d', '--days_history', default=0, type=int, help='Number of days of history to show', required=False) parser_status.add_argument('-f', '--force', help='Gatekeeper for --close_tasks', action='store_true', required=False) parser_status.add_argument('-r', '--request_id', help='Show tasks with this Request ID', required=False) parser_status.add_argument( '-p', '--priority_filter', default=20, type=int, required=False, help='This sets what report sections are shown in full detail in ' 'report output. Any tasks that have set a report_priority value ' 'equal to or lower than this setting will be shown in full detail, and ' 'tasks with a higher value will only have a summary shown. To see all ' 'tasks report output in full detail, set --priority_filter=100') parser_status.add_argument( '-R', '--full_report', help='Generate full markdown report instead of just a summary', action='store_true', required=False) parser_status.add_argument('-t', '--task_id', help='Show task for given Task ID', required=False) parser_status.add_argument('-u', '--user', help='Show task for given user', required=False) # Server subparsers.add_parser('server', help='Run Turbinia Server') args = parser.parse_args() if args.quiet: log.setLevel(logging.ERROR) elif args.verbose: log.setLevel(logging.INFO) elif args.debug: log.setLevel(logging.DEBUG) else: log.setLevel(logging.WARNING) log.info('Turbinia version: {0:s}'.format(__version__)) if args.jobs_whitelist and args.jobs_blacklist: log.error( 'A Job filter whitelist and blacklist cannot be specified at the same ' 'time') sys.exit(1) filter_patterns = None if (args.filter_patterns_file and not os.path.exists(args.filter_patterns_file)): log.error('Filter patterns file {0:s} does not exist.') sys.exit(1) elif args.filter_patterns_file: try: filter_patterns = open( args.filter_patterns_file).read().splitlines() except IOError as e: log.warning('Cannot open file {0:s} [{1!s}]'.format( args.filter_patterns_file, e)) # Client config.LoadConfig() if args.command not in ('psqworker', 'server'): if args.use_celery: client = TurbiniaCeleryClient() elif args.run_local: client = TurbiniaClient(run_local=True) else: client = TurbiniaClient() else: client = None server_flags_set = args.server or args.use_celery or args.command == 'server' worker_flags_set = (args.use_celery or args.command in ('psqworker', 'celeryworker')) if args.run_local and (server_flags_set or worker_flags_set): log.error( '--run_local flag is not compatible with server/worker flags') sys.exit(1) if args.run_local and not args.task: log.error('--run_local flag requires --task flag') sys.exit(1) if args.output_dir: config.OUTPUT_DIR = args.output_dir if args.log_file: config.LOG_FILE = args.log_file evidence_ = None is_cloud_disk = False if args.command == 'rawdisk': args.name = args.name if args.name else args.local_path local_path = os.path.abspath(args.local_path) evidence_ = evidence.RawDisk(name=args.name, local_path=local_path, mount_partition=args.mount_partition, source=args.source) elif args.command == 'bitlocker': if not args.password and not args.recovery_key: log.error('Neither recovery key nor password is specified.') sys.exit(1) args.name = args.name if args.name else args.local_path local_path = os.path.abspath(args.local_path) evidence_ = evidence.BitlockerDisk(name=args.name, local_path=local_path, recovery_key=args.recovery_key, password=args.password, source=args.source) elif args.command == 'directory': args.name = args.name if args.name else args.local_path local_path = os.path.abspath(args.local_path) evidence_ = evidence.Directory(name=args.name, local_path=local_path, source=args.source) elif args.command == 'googleclouddisk': is_cloud_disk = True args.name = args.name if args.name else args.disk_name evidence_ = evidence.GoogleCloudDisk( name=args.name, disk_name=args.disk_name, project=args.project, mount_partition=args.mount_partition, zone=args.zone, source=args.source) elif args.command == 'googleclouddiskembedded': is_cloud_disk = True args.name = args.name if args.name else args.disk_name evidence_ = evidence.GoogleCloudDiskRawEmbedded( name=args.name, disk_name=args.disk_name, embedded_path=args.embedded_path, mount_partition=args.mount_partition, project=args.project, zone=args.zone, source=args.source) elif args.command == 'psqworker': # Set up root logger level which is normally set by the psqworker command # which we are bypassing. logger.setup() worker = TurbiniaPsqWorker() worker.start() elif args.command == 'celeryworker': logger.setup() worker = TurbiniaCeleryWorker() worker.start() elif args.command == 'server': server = TurbiniaServer(jobs_blacklist=args.jobs_blacklist, jobs_whitelist=args.jobs_whitelist) server.start() elif args.command == 'status': region = config.TURBINIA_REGION if args.close_tasks: if args.user or args.request_id or args.task_id: print( client.close_tasks(instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=args.request_id, task_id=args.task_id, user=args.user, requester=getpass.getuser())) sys.exit(0) else: log.info( '--close_tasks (-c) requires --user, --request_id, or/and --task_id' ) sys.exit(1) if args.wait and args.request_id: client.wait_for_request(instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=args.request_id, user=args.user, poll_interval=args.poll_interval) elif args.wait and not args.request_id: log.info('--wait requires --request_id, which is not specified. ' 'turbiniactl will exit without waiting.') print( client.format_task_status(instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, days=args.days_history, task_id=args.task_id, request_id=args.request_id, user=args.user, all_fields=args.all_fields, full_report=args.full_report, priority_filter=args.priority_filter)) elif args.command == 'listjobs': log.info('Available Jobs:') client.list_jobs() else: log.warning('Command {0:s} not implemented.'.format(args.command)) if evidence_ and not args.force_evidence: if config.SHARED_FILESYSTEM and evidence_.cloud_only: log.error( 'The evidence type {0:s} is Cloud only, and this instance of ' 'Turbinia is not a cloud instance.'.format(evidence_.type)) sys.exit(1) elif not config.SHARED_FILESYSTEM and not evidence_.cloud_only: log.error( 'The evidence type {0:s} cannot run on Cloud instances of ' 'Turbinia. Consider wrapping it in a ' 'GoogleCloudDiskRawEmbedded or other Cloud compatible ' 'object'.format(evidence_.type)) sys.exit(1) if is_cloud_disk and evidence_.project != config.TURBINIA_PROJECT: msg = ( 'Turbinia project {0:s} is different from evidence project {1:s}. ' 'This processing request will fail unless the Turbinia service ' 'account has permissions to this project.'.format( config.TURBINIA_PROJECT, evidence_.project)) if args.force_evidence: log.warning(msg) else: msg += ' Use --force_evidence if you are sure you want to do this.' log.error(msg) sys.exit(1) # If we have evidence to process and we also want to run as a server, then # we'll just process the evidence directly rather than send it through the # PubSub frontend interface. If we're not running as a server then we will # create a new TurbiniaRequest and send it over PubSub. request = None if evidence_ and args.server: server = TurbiniaServer() server.add_evidence(evidence_) server.start() elif evidence_: request = TurbiniaRequest(request_id=args.request_id) request.evidence.append(evidence_) if filter_patterns: request.recipe['filter_patterns'] = filter_patterns if args.jobs_blacklist: request.recipe['jobs_blacklist'] = args.jobs_blacklist if args.jobs_whitelist: request.recipe['jobs_whitelist'] = args.jobs_whitelist if args.dump_json: print(request.to_json().encode('utf-8')) else: log.info('Creating request {0:s} with evidence {1:s}'.format( request.request_id, evidence_.name)) log.info( 'Run command "turbiniactl {0:s} status -r {1:s}" to see the status of' ' this request and associated tasks'.format( '-C' if args.use_celery else '', request.request_id)) if not args.run_local: client.send_request(request) else: log.debug( '--run_local specified so not sending request to server') if args.wait: log.info('Waiting for request {0:s} to complete'.format( request.request_id)) region = config.TURBINIA_REGION client.wait_for_request(instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=request.request_id, poll_interval=args.poll_interval) print( client.format_task_status(instance=config.INSTANCE_ID, project=config.TURBINIA_PROJECT, region=region, request_id=request.request_id, all_fields=args.all_fields)) if args.run_local and not evidence_: log.error('Evidence must be specified if using --run_local') sys.exit(1) if args.run_local and evidence_.cloud_only: log.error('--run_local cannot be used with Cloud only Evidence types') sys.exit(1) if args.run_local and evidence_: result = client.run_local_task(args.task, request) log.info('Task execution result: {0:s}'.format(result)) log.info('Done.') sys.exit(0)