def run(self):
gateway = self._initialize_gateway(self.gateway_address)
if not gateway:
log_error('Failed to initialize java gateway')
return
# noinspection PyProtectedMember
callback_server_port = gateway._callback_server.server_socket.getsockname()[1]
spark_context, sql_context = self._initialize_spark_contexts(gateway)
code_executor = CodeExecutor(spark_context, sql_context, gateway.entry_point)
try:
gateway.entry_point.registerCallbackServerPort(callback_server_port)
gateway.entry_point.registerCodeExecutor(code_executor)
except Py4JError as e:
log_error('Exception while registering codeExecutor, or callback server port: {}'.format(e))
gateway.close()
return
# Wait for the end of the world
try:
while True:
time.sleep(1)
except KeyboardInterrupt:
log_debug('Exiting on user\'s request')
gateway.close()
def _initialize_gateway(gateway_address):
(host, port) = gateway_address
callback_params = CallbackServerParameters(address=host, port=0)
gateway = JavaGateway(GatewayClient(address=host, port=port),
start_callback_server=True,
auto_convert=True,
callback_server_parameters=callback_params)
try:
java_import(gateway.jvm, "org.apache.spark.SparkEnv")
java_import(gateway.jvm, "org.apache.spark.SparkConf")
java_import(gateway.jvm, "org.apache.spark.api.java.*")
java_import(gateway.jvm, "org.apache.spark.api.python.*")
java_import(gateway.jvm, "org.apache.spark.mllib.api.python.*")
java_import(gateway.jvm, "org.apache.spark.sql.*")
java_import(gateway.jvm, "org.apache.spark.sql.hive.*")
java_import(gateway.jvm, "scala.Tuple2")
java_import(gateway.jvm, "scala.collection.immutable.List")
except Py4JError as e:
log_error('Error while initializing java gateway: {}'.format(e))
gateway.close()
return None
log_debug('Java Gateway initialized {}'.format(gateway))
return gateway
def _initialize_spark_contexts(gateway):
java_spark_context = gateway.entry_point.getSparkContext()
java_spark_conf = java_spark_context.getConf()
spark_context = SparkContext(conf=SparkConf(_jvm=gateway.jvm,
_jconf=java_spark_conf),
gateway=gateway,
jsc=java_spark_context)
java_spark_sql_session = gateway.entry_point.getSparkSQLSession()
spark_version = spark_context.version
spark_sql_session = None
if spark_version in [
"2.0.0", "2.0.1", "2.0.2", "2.1.0", "2.1.1", "2.2.0"
]:
from pyspark.sql import SparkSession
java_spark_session = java_spark_sql_session.getSparkSession()
spark_sql_session = SparkSession(spark_context, java_spark_session)
else:
log_error(
"Spark version {} is not supported".format(spark_version))
raise ValueError(
"Spark version {} is not supported".format(spark_version))
return spark_context, spark_sql_session
def _initialize_gateway(gateway_address):
(host, port) = gateway_address
callback_params = CallbackServerParameters(address=host, port=0)
gateway = JavaGateway(GatewayClient(address=host, port=port),
start_callback_server=True,
auto_convert=True,
callback_server_parameters=callback_params)
try:
java_import(gateway.jvm, "org.apache.spark.SparkEnv")
java_import(gateway.jvm, "org.apache.spark.SparkConf")
java_import(gateway.jvm, "org.apache.spark.api.java.*")
java_import(gateway.jvm, "org.apache.spark.api.python.*")
java_import(gateway.jvm, "org.apache.spark.mllib.api.python.*")
java_import(gateway.jvm, "org.apache.spark.sql.*")
java_import(gateway.jvm, "org.apache.spark.sql.hive.*")
java_import(gateway.jvm, "scala.Tuple2")
java_import(gateway.jvm, "scala.collection.immutable.List")
except Py4JError as e:
log_error('Error while initializing java gateway: {}'.format(e))
gateway.close()
return None
return gateway
def run(self):
gateway = self._initialize_gateway(self.gateway_address)
if not gateway:
log_error('Failed to initialize java gateway')
return
# noinspection PyProtectedMember
callback_server_port = gateway._callback_server.server_socket.getsockname(
)[1]
spark_context, spark_session = self._initialize_spark_contexts(gateway)
code_executor = CodeExecutor(spark_context, spark_session,
gateway.entry_point)
try:
gateway.entry_point.registerCallbackServerPort(
callback_server_port)
gateway.entry_point.registerCodeExecutor(code_executor)
except Py4JError as e:
log_error(
'Exception while registering codeExecutor, or callback server port: {}'
.format(e))
gateway.close()
return
# Wait for the end of the world
try:
while True:
time.sleep(1)
except KeyboardInterrupt:
log_debug('Exiting on user\'s request')
gateway.close()
def _supervised_execution(self, workflow_id, node_id,
custom_operation_code):
# noinspection PyBroadException
try:
self._run_custom_code(workflow_id, node_id, custom_operation_code)
self.entry_point.executionCompleted(workflow_id, node_id)
except Exception as e:
log_error('AN ERROR OCCURED ==>')
stacktrace = traceback.format_exc()
log_error(stacktrace)
self.entry_point.executionFailed(workflow_id, node_id, stacktrace)
log_error('ERROR END')
