c7dbe3585b
Update ingestor and data manager for improved shutdown handling and logging - Incremented project version in quality gate configuration. - Commented out ingestor service in docker-compose for clarity. - Enhanced main loop in app.py to handle exit signals and exceptions. - Added shutdown methods in Ingestor and DataManager classes for graceful resource cleanup. - Updated unit tests to validate shutdown behavior and exception handling. - Introduced coverage configuration to omit specific files.
115 lines
4.4 KiB
Python
115 lines
4.4 KiB
Python
import json
|
|
from logging import Logger
|
|
from time import sleep
|
|
from kafka import KafkaProducer
|
|
from kafka.errors import NoBrokersAvailable
|
|
from sientia_do.notifications.handlers import NotificationHandler
|
|
from sientia_do.notifications.models import NotificationLevel
|
|
import traceback
|
|
|
|
|
|
class DataManager():
|
|
def __init__(self, kafka_servers: str, logger: Logger,
|
|
notification_handler: NotificationHandler) -> None:
|
|
"""
|
|
Initializes the DataManager instance with a Kafka producer.
|
|
This constructor attempts to establish a connection to the specified Kafka servers
|
|
and initializes a Kafka producer for sending messages. It retries the connection
|
|
up to 3 times if the Kafka servers are unavailable.
|
|
Args:
|
|
kafka_servers (str): A comma-separated string of Kafka server addresses.
|
|
logger (Logger): A logger instance for logging messages.
|
|
Raises:
|
|
NoBrokersAvailable: If the connection to Kafka servers fails after 3 attempts.
|
|
"""
|
|
|
|
self.kafka_producer = None
|
|
for i in range(0, 3):
|
|
logger.info(
|
|
f"Trying ({i}) to initializing DataManager with Kafka servers: {kafka_servers}")
|
|
try:
|
|
self.kafka_producer = KafkaProducer(
|
|
bootstrap_servers=kafka_servers,
|
|
value_serializer=lambda v: json.dumps(v).encode(
|
|
'utf-8'), # Serialize JSON messages
|
|
key_serializer=lambda k: str(
|
|
k).encode('utf-8') if k else None,
|
|
)
|
|
break
|
|
except NoBrokersAvailable:
|
|
logger.error(
|
|
f"Kafka servers {kafka_servers} are not available. Retrying...")
|
|
sleep(5)
|
|
else:
|
|
logger.error(
|
|
f"Failed to connect to Kafka servers {kafka_servers} after 3 attempts.")
|
|
raise NoBrokersAvailable(
|
|
f"Failed to connect to Kafka servers {kafka_servers} after 3 attempts.")
|
|
|
|
logger.info(
|
|
f"DataManager initialized with Kafka servers: {kafka_servers}")
|
|
self.logger = logger
|
|
self.notification_handler = notification_handler
|
|
|
|
def shutdown(self):
|
|
"""Closes the Kafka producer connection."""
|
|
if self.kafka_producer:
|
|
try:
|
|
self.kafka_producer.flush(timeout=10)
|
|
self.kafka_producer.close()
|
|
except Exception as e:
|
|
self.logger.error(
|
|
f"Error closing Kafka producer: {e}")
|
|
else:
|
|
self.logger.warning(
|
|
"Kafka producer is already closed or not initialized.")
|
|
|
|
def __del__(self):
|
|
self.shutdown()
|
|
|
|
def delivery_report(self, msg: str):
|
|
"""Callback for delivery reports from Kafka."""
|
|
self.logger.debug(
|
|
f"Record successfully produced to {msg.topic} [{msg.partition}] at offset {msg.offset}")
|
|
|
|
def delivery_error(self, err: str):
|
|
"""Callback for delivery reports from Kafka."""
|
|
self.logger.error(f"Delivery failed for record : {err}")
|
|
|
|
def publish(self, topic: str, data: dict) -> None:
|
|
"""
|
|
Publishes a message to a specified Kafka topic.
|
|
|
|
Args:
|
|
topic (str): The name of the Kafka topic to which the message will be published.
|
|
data (dict): The message data to be sent to the Kafka topic.
|
|
|
|
Returns:
|
|
None
|
|
|
|
Raises:
|
|
Exception: If there is an error during message delivery, it will be handled by the `delivery_error` callback.
|
|
"""
|
|
|
|
try:
|
|
|
|
self.logger.debug(
|
|
f"Publishing message to topic {topic}: {data}")
|
|
self.kafka_producer.send(
|
|
topic=topic, value=data).add_callback(
|
|
self.delivery_report).add_errback(
|
|
self.delivery_error)
|
|
|
|
self.kafka_producer.flush(timeout=10)
|
|
|
|
except Exception as e:
|
|
trace = traceback.format_exc()
|
|
self.notification_handler.build_and_send_notification(
|
|
notification_id=f"KAFKA_PRODUCER_ERROR_{topic}",
|
|
message=f"Error publishing message to topic {topic}: {e}",
|
|
block="kafka_producer",
|
|
level=NotificationLevel.ERROR,
|
|
attachment_content=trace
|
|
)
|
|
self.logger.error(trace)
|