class Collector(object):
def __init__(self, hdfs_app_path, kafka_topic, conf_type):
self._initialize_members(hdfs_app_path, kafka_topic, conf_type)
def _initialize_members(self, hdfs_app_path, kafka_topic, conf_type):
# getting parameters.
self._logger = logging.getLogger('SPOT.INGEST.PROXY')
self._hdfs_app_path = hdfs_app_path
self._kafka_topic = kafka_topic
# get script path
self._script_path = os.path.dirname(os.path.abspath(__file__))
# read proxy configuration.
conf_file = "{0}/ingest_conf.json".format(
os.path.dirname(os.path.dirname(self._script_path)))
conf = json.loads(open(conf_file).read())
self._message_size = conf["kafka"]["message_size"]
self._conf = conf["pipelines"][conf_type]
# get collector path.
self._collector_path = self._conf['collector_path']
#get supported files
self._supported_files = self._conf['supported_files']
# create collector watcher
self._watcher = FileWatcher(self._collector_path,
self._supported_files)
# Multiprocessing.
self._processes = conf["collector_processes"]
self._ingestion_interval = conf["ingestion_interval"]
self._pool = Pool(processes=self._processes)
def start(self):
self._logger.info("Starting PROXY collector")
self._watcher.start()
try:
while True:
#self._ingest_files()
self._ingest_files_pool()
time.sleep(self._ingestion_interval)
except KeyboardInterrupt:
self._logger.info("Stopping Proxy collector...")
Util.remove_kafka_topic(self._kafka_topic.Zookeeper,
self._kafka_topic.Topic, self._logger)
self._watcher.stop()
self._pool.terminate()
self._pool.close()
self._pool.join()
def _ingest_files_pool(self):
if self._watcher.HasFiles:
for x in range(0, self._processes):
file = self._watcher.GetNextFile()
resutl = self._pool.apply_async(
ingest_file,
args=(file, self._message_size, self._kafka_topic.Topic,
self._kafka_topic.BootstrapServers))
#resutl.get() # to debug add try and catch.
if not self._watcher.HasFiles: break
return True
class Collector(object):
def __init__(self, hdfs_app_path, kafka_topic, conf_type):
self._initialize_members(hdfs_app_path, kafka_topic, conf_type)
def _initialize_members(self, hdfs_app_path, kafka_topic, conf_type):
# getting parameters.
self._logger = logging.getLogger('SPOT.INGEST.DNS')
self._hdfs_app_path = hdfs_app_path
self._kafka_topic = kafka_topic
# get script path
self._script_path = os.path.dirname(os.path.abspath(__file__))
# read dns configuration.
conf_file = "{0}/ingest_conf.json".format(os.path.dirname(os.path.dirname(self._script_path)))
conf = json.loads(open(conf_file).read())
self._conf = conf["pipelines"][conf_type]
# set configuration.
self._collector_path = self._conf['collector_path']
self._dsource = 'dns'
self._hdfs_root_path = "{0}/{1}".format(hdfs_app_path, self._dsource)
# set configuration.
self._pkt_num = self._conf['pkt_num']
self._pcap_split_staging = self._conf['pcap_split_staging']
self._supported_files = self._conf['supported_files']
# create collector watcher
self._watcher = FileWatcher(self._collector_path, self._supported_files)
# Multiprocessing.
self._processes = conf["collector_processes"]
self._ingestion_interval = conf["ingestion_interval"]
self._pool = Pool(processes=self._processes)
def start(self):
self._logger.info("Starting DNS ingest")
self._watcher.start()
try:
while True:
self._ingest_files_pool()
time.sleep(self._ingestion_interval)
except KeyboardInterrupt:
self._logger.info("Stopping DNS collector...")
Util.remove_kafka_topic(self._kafka_topic.Zookeeper, self._kafka_topic.Topic, self._logger)
self._watcher.stop()
self._pool.terminate()
self._pool.close()
self._pool.join()
SystemExit("Ingest finished...")
def _ingest_files_pool(self):
if self._watcher.HasFiles:
for x in range(0, self._processes):
file = self._watcher.GetNextFile()
resutl = self._pool.apply_async(ingest_file, args=(file, self._pkt_num, self._pcap_split_staging, self._kafka_topic.Partition, self._hdfs_root_path, self._kafka_topic.Topic, self._kafka_topic.BootstrapServers, ))
#resutl.get() # to debug add try and catch.
if not self._watcher.HasFiles: break
return True
class Collector(object):
def __init__(self, hdfs_app_path, kafkaproducer, conf_type):
self._initialize_members(hdfs_app_path, kafkaproducer, conf_type)
def _initialize_members(self, hdfs_app_path, kafkaproducer, conf_type):
# getting parameters.
self._logger = logging.getLogger('SPOT.INGEST.DNS')
self._hdfs_app_path = hdfs_app_path
self._producer = kafkaproducer
# get script path
self._script_path = os.path.dirname(os.path.abspath(__file__))
# read dns configuration.
conf_file = "{0}/ingest_conf.json".format(
os.path.dirname(os.path.dirname(self._script_path)))
conf = json.loads(open(conf_file).read())
self._conf = conf["pipelines"][conf_type]
# set configuration.
self._collector_path = self._conf['collector_path']
self._dsource = 'dns'
self._hdfs_root_path = "{0}/{1}".format(hdfs_app_path, self._dsource)
# set configuration.
self._pkt_num = self._conf['pkt_num']
self._pcap_split_staging = self._conf['pcap_split_staging']
self._supported_files = self._conf['supported_files']
# create collector watcher
self._watcher = FileWatcher(self._collector_path,
self._supported_files)
# Multiprocessing.
self._processes = conf["collector_processes"]
self._ingestion_interval = conf["ingestion_interval"]
self._pool = Pool(processes=self._processes)
# TODO: review re-use of hdfs.client
self._hdfs_client = hdfs.get_client()
def start(self):
self._logger.info("Starting DNS ingest")
self._watcher.start()
try:
while True:
self._ingest_files_pool()
time.sleep(self._ingestion_interval)
except KeyboardInterrupt:
self._logger.info("Stopping DNS collector...")
Util.remove_kafka_topic(self._producer.Zookeeper,
self._producer.Topic, self._logger)
self._watcher.stop()
self._pool.terminate()
self._pool.close()
self._pool.join()
SystemExit("Ingest finished...")
def _ingest_files_pool(self):
if self._watcher.HasFiles:
for x in range(0, self._processes):
self._logger.info('processes: {0}'.format(self._processes))
new_file = self._watcher.GetNextFile()
if self._processes <= 1:
_ingest_file(self._hdfs_client, new_file, self._pkt_num,
self._pcap_split_staging,
self._hdfs_root_path, self._producer,
self._producer.Topic)
else:
resutl = self._pool.apply_async(
_ingest_file,
args=(self._hdfs_client, new_file, self._pkt_num,
self._pcap_split_staging, self._hdfs_root_path,
self._producer, self._producer.Topic))
# resutl.get() # to debug add try and catch.
if not self._watcher.HasFiles:
break
return True
class Collector(object):
def __init__(self,hdfs_app_path,kafka_topic,conf_type):
self._initialize_members(hdfs_app_path,kafka_topic,conf_type)
def _initialize_members(self,hdfs_app_path,kafka_topic,conf_type):
# getting parameters.
self._logger = logging.getLogger('SPOT.INGEST.PROXY')
self._hdfs_app_path = hdfs_app_path
self._kafka_topic= kafka_topic
# get script path
self._script_path = os.path.dirname(os.path.abspath(__file__))
# read proxy configuration.
conf_file = "{0}/ingest_conf.json".format(os.path.dirname(os.path.dirname(self._script_path)))
conf = json.loads(open(conf_file).read())
self._message_size = conf["kafka"]["message_size"]
self._conf = conf["pipelines"][conf_type]
# get collector path.
self._collector_path = self._conf['collector_path']
#get supported files
self._supported_files = self._conf['supported_files']
# create collector watcher
self._watcher = FileWatcher(self._collector_path,self._supported_files)
# Multiprocessing.
self._processes = conf["collector_processes"]
self._ingestion_interval = conf["ingestion_interval"]
self._pool = Pool(processes=self._processes)
def start(self):
self._logger.info("Starting PROXY collector")
self._watcher.start()
try:
while True:
#self._ingest_files()
self._ingest_files_pool()
time.sleep(self._ingestion_interval)
except KeyboardInterrupt:
self._logger.info("Stopping Proxy collector...")
Util.remove_kafka_topic(self._kafka_topic.Zookeeper,self._kafka_topic.Topic,self._logger)
self._watcher.stop()
self._pool.terminate()
self._pool.close()
self._pool.join()
def _ingest_files_pool(self):
if self._watcher.HasFiles:
for x in range(0,self._processes):
file = self._watcher.GetNextFile()
resutl = self._pool.apply_async(ingest_file,args=(file,self._message_size,self._kafka_topic.Topic,self._kafka_topic.BootstrapServers))
#resutl.get() # to debug add try and catch.
if not self._watcher.HasFiles: break
return True