Connection drops when handling tasks with a long duration

I’m starting to get the same error as #418 even when I adjust the heartbeat connection parameter. My tasks take usually around 5 mins, but even then, when I set the heartbeat to both 0 and some very large number (ie: 60000)

My code Is essentially identical to the issue I mentioned above and the error I receive is this: Traceback (most recent call last): File "instagram_worker.py", line 43, in <module> channel.start_consuming() File "/usr/local/lib/python2.7/dist-packages/pika/adapters/blocking_connection.py", line 1780, in start_consuming self.connection.process_data_events(time_limit=None) File "/usr/local/lib/python2.7/dist-packages/pika/adapters/blocking_connection.py", line 716, in process_data_events self._dispatch_channel_events() File "/usr/local/lib/python2.7/dist-packages/pika/adapters/blocking_connection.py", line 518, in _dispatch_channel_events impl_channel._get_cookie()._dispatch_events() File "/usr/local/lib/python2.7/dist-packages/pika/adapters/blocking_connection.py", line 1403, in _dispatch_events evt.body) File "instagram_worker.py", line 37, in dataHandler channel.basic_ack(delivery_tag=method.delivery_tag) File "/usr/local/lib/python2.7/dist-packages/pika/adapters/blocking_connection.py", line 1988, in basic_ack self._flush_output() File "/usr/local/lib/python2.7/dist-packages/pika/adapters/blocking_connection.py", line 1250, in _flush_output *waiters) File "/usr/local/lib/python2.7/dist-packages/pika/adapters/blocking_connection.py", line 474, in _flush_output result.reason_text) pika.exceptions.ConnectionClosed: (-1, "error(104, 'Connection reset by peer')")

[gmr]

Try something like this:

import re
import json
import threading

from google.cloud import bigquery
import pandas as pd
import pika
from unidecode import unidecode


def process_export(url, tablename):
    df = pd.read_csv(csvURL, encoding="utf-8")
    print("read in the csv")
    columns = list(df)
    ascii_only_name = [unidecode(name) for name in columns]
    cleaned_column_names = [re.sub("[^a-zA-Z0-9_ ]", "", name) for name in ascii_only_name]
    underscored_names = [name.replace(" ", "_") for name in cleaned_column_names]
    valid_gbq_tablename = "test." + tablename
    df.columns = underscored_names

    # try:
    df.to_gbq(valid_gbq_tablename, "some_project", if_exists="append", verbose=True, chunksize=10000)
    # print("Finished Exporting")
    # except Exception as error:
    #     print("unable to export due to: ")
    #     print(error)
    #     print()


def data_handler(channel, method, properties, body):
    body = json.loads(body)

    thread = threading.Thread(target=process_export, args=(body["csvURL"], body["tablename"]))
    thread.start()
    while thread.is_alive():  # Loop while the thread is processing
        channel._connection.sleep(1.0)
    print('Back from thread')
    channel.basic_ack(delivery_tag=method.delivery_tag)


def main():
    params = pika.ConnectionParameters(host='localhost', heartbeat=60)
    connection = pika.BlockingConnection(params)
    channel = connection.channel()
    channel.queue_declare(queue="some_queue", durable=True)
    channel.basic_qos(prefetch_count=1)
    channel.basic_consume(data_handler, queue="some_queue")
    try:
        channel.start_consuming()
    except KeyboardInterrupt:
        channel.stop_consuming()
    channel.close()


if __name__ == '__main__':
    main()

[lukebakken]

Thanks @gmr