def __init__(self, config=None, runtime=None, log_level=None, job_max_runtime=JOB_MAX_RUNTIME):
"""
Initialize and return an executor class.
:param config: Settings passed in here will override those in `pywren_config`. Default None.
:param runtime: Runtime name to use. Default None.
:param job_max_runtime: Max time per lambda. Default 300
:return `executor` object.
Usage
>>> import pywren
>>> pw = pywren.ibm_cf_executor()
"""
self._state = ExecutorState.new
if config is None:
self.config = wrenconfig.default()
else:
self.config = wrenconfig.default(config)
if runtime:
self.config['ibm_cf']['action_name'] = runtime
if log_level:
wrenlogging.default_config(log_level)
self._openwhisk = False
if any([k.startswith('__OW_') for k in os.environ.keys()]):
# OpenWhisk execution
self._openwhisk = True
wrenlogging.ow_config(logging.INFO)
self.runtime = self.config['ibm_cf']['action_name']
ibm_cf_config = self.config['ibm_cf']
invoker = invokers.IBMCloudFunctionsInvoker(ibm_cf_config)
self.storage_config = wrenconfig.extract_storage_config(self.config)
self.storage_handler = storage.Storage(self.storage_config)
self.executor = Executor(invoker, self.config, self.storage_handler, job_max_runtime)
self.executor_id = self.executor.executor_id
self.futures = None
self.reduce_future = None
log_msg='IBM Cloud Functions executor created with ID {}'.format(self.executor_id)
logger.info(log_msg)
if(logger.getEffectiveLevel() == logging.WARNING):
print(log_msg)
def __init__(self, jr_config, result_queue):
super().__init__()
start_time = time.time()
self.result_queue = result_queue
self.config = jr_config
self.stats = stats(self.config['stats_filename'])
self.stats.write('jobrunner_start', start_time)
pw_config = json.loads(os.environ.get('PYWREN_CONFIG', ''))
self.storage_config = extract_storage_config(pw_config)
if 'SHOW_MEMORY_USAGE' in os.environ:
self.show_memory = eval(os.environ['SHOW_MEMORY_USAGE'])
else:
self.show_memory = False
log_level = self.config['log_level']
wrenlogging.ow_config(log_level)
self.func_key = self.config['func_key']
self.data_key = self.config['data_key']
self.data_byte_range = self.config['data_byte_range']
self.output_key = self.config['output_key']
from multiprocessing import Process
from pywren_ibm_cloud import wrenlogging
from pywren_ibm_cloud.storage import storage
from pywren_ibm_cloud.utils import sizeof_fmt, b64str_to_bytes
from pywren_ibm_cloud.wrenconfig import extract_storage_config
from pywren_ibm_cloud.future import ResponseFuture
from pywren_ibm_cloud.libs.tblib import pickling_support
from pywren_ibm_cloud.utils import get_current_memory_usage
from pywren_ibm_cloud.storage.backends.cos import COSBackend
from pywren_ibm_cloud.storage.backends.swift import SwiftBackend
pickling_support.install()
level = logging.INFO
logger = logging.getLogger('jobrunner')
logger.setLevel(level)
wrenlogging.ow_config(level)
class stats:
def __init__(self, stats_filename):
self.stats_filename = stats_filename
self.stats_fid = open(stats_filename, 'w')
def write(self, key, value):
self.stats_fid.write("{} {}\n".format(key, value))
self.stats_fid.flush()
def __del__(self):
self.stats_fid.close()
from __future__ import print_function
from pywren_ibm_cloud.wrenhandler import ibm_cloud_function_handler
import logging
from pywren_ibm_cloud import wrenlogging
logger = logging.getLogger('pywren')
wrenlogging.ow_config(logging.INFO)
def main(args):
logger.info("Starting IMB Cloud Function")
ibm_cloud_function_handler(args)
return {"greeting": "Finished"}
#
# (C) Copyright IBM Corp. 2018
#
# 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
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
import logging
from pywren_ibm_cloud import wrenlogging
from pywren_ibm_cloud.action.wrenhandler import ibm_cloud_function_handler
logger = logging.getLogger('__main__')
wrenlogging.ow_config(logging.DEBUG)
def main(args):
logger.info("Starting IBM Cloud Function execution")
ibm_cloud_function_handler(args)
return {"greeting": "Finished"}
def function_handler(event):
start_time = time.time()
logger.debug("Action handler started")
response_status = {'exception': False}
response_status['host_submit_time'] = event['host_submit_time']
response_status['start_time'] = start_time
context_dict = {
'ibm_cf_request_id': os.environ.get("__OW_ACTIVATION_ID"),
'ibm_cf_python_version': os.environ.get("PYTHON_VERSION"),
}
config = event['config']
storage_config = wrenconfig.extract_storage_config(config)
log_level = event['log_level']
wrenlogging.ow_config(log_level)
call_id = event['call_id']
callgroup_id = event['callgroup_id']
executor_id = event['executor_id']
logger.info("Execution ID: {}/{}/{}".format(executor_id, callgroup_id, call_id))
job_max_runtime = event.get("job_max_runtime", 590) # default for CF
status_key = event['status_key']
func_key = event['func_key']
data_key = event['data_key']
data_byte_range = event['data_byte_range']
output_key = event['output_key']
extra_env = event.get('extra_env', {})
response_status['call_id'] = call_id
response_status['callgroup_id'] = callgroup_id
response_status['executor_id'] = executor_id
# response_status['func_key'] = func_key
# response_status['data_key'] = data_key
# response_status['output_key'] = output_key
# response_status['status_key'] = status_key
try:
if version.__version__ != event['pywren_version']:
raise Exception("WRONGVERSION", "PyWren version mismatch",
version.__version__, event['pywren_version'])
# response_status['free_disk_bytes'] = free_disk_space("/tmp")
custom_env = {'PYWREN_CONFIG': json.dumps(config),
'PYWREN_EXECUTOR_ID': executor_id,
'PYTHONPATH': "{}:{}".format(os.getcwd(), PYWREN_LIBS_PATH),
'PYTHONUNBUFFERED': 'True'}
os.environ.update(custom_env)
os.environ.update(extra_env)
# pass a full json blob
jobrunner_config = {'func_key': func_key,
'data_key': data_key,
'log_level': log_level,
'data_byte_range': data_byte_range,
'python_module_path': PYTHON_MODULE_PATH,
'output_key': output_key,
'stats_filename': JOBRUNNER_STATS_FILENAME}
if os.path.exists(JOBRUNNER_STATS_FILENAME):
os.remove(JOBRUNNER_STATS_FILENAME)
setup_time = time.time()
response_status['setup_time'] = round(setup_time - start_time, 8)
result_queue = multiprocessing.Queue()
jr = jobrunner(jobrunner_config, result_queue)
jr.daemon = True
logger.info("Starting jobrunner process")
jr.start()
jr.join(job_max_runtime)
response_status['exec_time'] = round(time.time() - setup_time, 8)
if jr.is_alive():
# If process is still alive after jr.join(job_max_runtime), kill it
logger.error("Process exceeded maximum runtime of {} seconds".format(job_max_runtime))
# Send the signal to all the process groups
jr.terminate()
raise Exception("OUTATIME", "Process executed for too long and was killed")
try:
# Only 1 message is returned by jobrunner
result_queue.get(block=False)
except Exception:
# If no message, this means that the process was killed due memory usage
raise Exception("OUTOFMEMORY", "Process exceeded maximum memory and was killed")
# print(subprocess.check_output("find {}".format(PYTHON_MODULE_PATH), shell=True))
# print(subprocess.check_output("find {}".format(os.getcwd()), shell=True))
if os.path.exists(JOBRUNNER_STATS_FILENAME):
with open(JOBRUNNER_STATS_FILENAME, 'r') as fid:
for l in fid.readlines():
key, value = l.strip().split(" ", 1)
try:
response_status[key] = float(value)
except Exception:
response_status[key] = value
if key == 'exception' or key == 'exc_pickle_fail' \
or key == 'result':
response_status[key] = eval(value)
# response_status['server_info'] = get_server_info()
response_status.update(context_dict)
response_status['end_time'] = time.time()
except Exception as e:
# internal runtime exceptions
logger.error("There was an exception: {}".format(str(e)))
response_status['end_time'] = time.time()
response_status['exception'] = True
pickled_exc = pickle.dumps(sys.exc_info())
pickle.loads(pickled_exc) # this is just to make sure they can be unpickled
response_status['exc_info'] = str(pickled_exc)
finally:
store_status = strtobool(os.environ.get('STORE_STATUS', 'True'))
rabbit_amqp_url = config['rabbitmq'].get('amqp_url')
dmpd_response_status = json.dumps(response_status)
drs = sizeof_fmt(len(dmpd_response_status))
if rabbit_amqp_url and store_status:
status_sent = False
output_query_count = 0
while not status_sent and output_query_count < 5:
output_query_count = output_query_count + 1
try:
params = pika.URLParameters(rabbit_amqp_url)
connection = pika.BlockingConnection(params)
channel = connection.channel()
channel.queue_declare(queue=executor_id, auto_delete=True)
channel.basic_publish(exchange='', routing_key=executor_id,
body=dmpd_response_status)
connection.close()
logger.info("Execution stats sent to rabbitmq - Size: {}".format(drs))
status_sent = True
except Exception as e:
logger.error("Unable to send status to rabbitmq")
logger.error(str(e))
logger.info('Retrying to send stats to rabbitmq...')
time.sleep(0.2)
if store_status:
internal_storage = storage.InternalStorage(storage_config)
logger.info("Storing execution stats - status.json - Size: {}".format(drs))
internal_storage.put_data(status_key, dmpd_response_status)
