Twisted error in server with mqtt client - twisted

I'm running a Twisted TCP server. Each protocol instance has an mqtt pub/sub client. I've reduced the actual production code to the simplest possible form below. I've stripped out a lot of irrelevant complexity to simplify the bug-finding process. The server works for multiple client connections and produces the data received from the mqtt client, but after any client connects/disconnects/reconnects a few times I get an exception that I haven't been able to understand or trap. I'm hoping that Jean-Paul or someone can point me at the error of my ways.
Once the exception occurs, no new clients can connect to the server. Each new connect attempt produces the exception.
Clients that are already connected continue to receive data ok.
The exception is
Unhandled Error
Traceback (most recent call last):
File "/usr/lib/python2.7/dist-packages/twisted/python/log.py", line 73, in callWithContext
return context.call({ILogContext: newCtx}, func, *args, **kw)
File "/usr/lib/python2.7/dist-packages/twisted/python/context.py", line 118, in callWithContext
return self.currentContext().callWithContext(ctx, func, *args, **kw)
File "/usr/lib/python2.7/dist-packages/twisted/python/context.py", line 81, in callWithContext
return func(*args,**kw)
File "/usr/lib/python2.7/dist-packages/twisted/internet/posixbase.py", line 614, in _doReadOrWrite
why = selectable.doRead()
--- <exception caught here> ---
File "/usr/lib/python2.7/dist-packages/twisted/internet/tcp.py", line 1069, in doRead
transport = self.transport(skt, protocol, addr, self, s, self.reactor)
File "/usr/lib/python2.7/dist-packages/twisted/internet/tcp.py", line 786, in __init__
self.startReading()
File "/usr/lib/python2.7/dist-packages/twisted/internet/abstract.py", line 429, in startReading
self.reactor.addReader(self)
File "/usr/lib/python2.7/dist-packages/twisted/internet/epollreactor.py", line 256, in addReader
_epoll.EPOLLIN, _epoll.EPOLLOUT)
File "/usr/lib/python2.7/dist-packages/twisted/internet/epollreactor.py", line 240, in _add
self._poller.modify(fd, flags)
exceptions.IOError: [Errno 2] No such file or directory
The basic server code is:
(this example will run and does generate the exception)
from twisted.internet import reactor
from twisted.internet.protocol import Factory
from twisted.protocols import basic
from paho.mqtt import client # The most recent version of the legacy Mosquitto client
from random import randint
class MsgReceiver(basic.LineReceiver):
def __init__(self, factory): # new (factory)
self.factory = factory # new (factory)
def connectionMade(self):
self.mqClient = self.startMQ()
if self.mqClient:
self.factory.clients.append(self)
else:
self.transport.loseConnection()
def connectionLost(self, reason):
pass
def lineReceived(self, line):
pass
def on_message(self, mosq, obj, msg):
try:
self.sendLine(msg.payload)
except Exception, err:
print(err.message)
def startMQ(self):
mqName = "-".join(["myAppName", str(randint(0, 99999))])
mqClient = client.Client(mqName)
if mqClient.connect("localhost", 1883, 60) != 0:
print('Could not connect to mq server')
return False
mqClient.on_message = self.on_message
(success, mid) = mqClient.subscribe("myTopic", 0)
if success != 0:
return False
mqClient.loop_start()
return mqClient
class MsgReceiverFactory(Factory):
allow_reuse_address = True
def __init__(self, clients):
self.clients = clients
def buildProtocol(self, addr):
return MsgReceiver(self)
if __name__ == "__main__":
try:
clients = []
reactor.listenTCP(43217, MsgReceiverFactory(clients))
reactor.run()
except Exception, err:
print(err.message)
if reactor.running:
reactor.stop()
A simple client that will induce the error when run twice (the first time it runs fine):
Interesting that if I enable the time.sleep(3) it runs fine and doesn't seem to induce the error
#!/usr/bin/python
from __future__ import print_function
import socket
import time
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.connect(("localhost", 43217))
data = s.recv(1024)
print(data)
#time.sleep(3)
s.close()

Related

How to solve the consumer error in Rabbitmq using python

when I run Consumer.py for headersExchange in rabbitmq using python,it is getting error like below
I have mentioned consumer and publish program below
Traceback (most recent call last):
File "headersConsumer.py", line 32, in <module>
main()
File "headersConsumer.py", line 14, in main
channel.exchange_declare(exchange = 'headers_logs',exchange_type='headers',durable=True)
File "C:\Python38\lib\site-packages\pika\adapters\blocking_connection.py", line 2387, in
exchange_declare
self._flush_output(declare_ok_result.is_ready)
File "C:\Python38\lib\site-packages\pika\adapters\blocking_connection.py", line 1339, in
_flush_output
raise self._closing_reason # pylint: disable=E0702
pika.exceptions.ChannelClosedByBroker: (406, "PRECONDITION_FAILED
- inequivalent arg 'type' for exchange 'headers_logs' in vhost '/': received 'headers' but
current is 'fanout'")
I have written consumer code like this
#!/usr/bin/env python
import pika, sys, os
def main():
connection = pika.BlockingConnection(pika.ConnectionParameters(host='localhost'))
channel = connection.channel()
channel.exchange_declare(exchange = 'headers_logs',exchange_type='headers',durable=True)
channel.queue_declare(queue = "HeaderQueue1", durable=True)
channel.queue_bind(exchange = 'headers_logs', queue="HeadersQueue1", routing_key='',
arguments={'x-match': 'any', 'key1': 'one', 'key2': 'two'})
def callback(ch, method, properties, body):
print(" [x] %r" % body.decode())
print(' [*] Waiting for logs. To exit press CTRL+C')
channel.basic_consume(
queue="HeadersQueue1", on_message_callback=callback, auto_ack=True)
channel.start_consuming()
if __name__ == '__main__':
try:
main()
except KeyboardInterrupt:
print('Interrupted')
try:
sys.exit(0)
except SystemExit:
os._exit(0)
I have written publish program likethis
import pika
import sys
connection = pika.BlockingConnection(pika.ConnectionParameters(host='localhost'))
channel = connection.channel()
channel.exchange_declare(exchange='headers_logs',exchange_type='headers')
message = ' '.join(sys.argv[1:]) or "Hello World!"
channel.basic_publish((exchange='headers_logs',routing_key="",body=message,properties=pika.BasicProperties(
delivery_mode = 2, # make message persistent
headers = {'key1':'one', 'key2': 'three'}
))
print(" [x] Sent %r" % message)
connection.close()
I am not understanding this error,Can anyone please suggest this error
PRECONDITION_FAILED means that you declared an exchange with set of parameters then you are trying to create the same queue name using different parameters.
in your case:
headers_logs' in vhost '/': received 'headers' but
current is 'fanout'")
so you are trying to change the exchange type from fanout to headers
Se here for more detail (this is for the queues but exchanges work in the same way).
Before a queue can be used it has to be declared. Declaring a queue
will cause it to be created if it does not already exist. The
declaration will have no effect if the queue does already exist and
its attributes are the same as those in the declaration. When the
existing queue attributes are not the same as those in the declaration
a channel-level exception with code 406 (PRECONDITION_FAILED) will be
raised.

Socket error while connecting to hive through: hiveserver2-client.py

Has anyone experienced this error before when trying to connect to hive.
Sample code used (https://github.com/telefonicaid/fiware-cygnus/blob/master/cygnus-ngsi/resources/hiveclients/python/hiveserver2-client.py):
import sys
import pyhs2
from pyhs2.error import Pyhs2Exception
# get the input parameters
if len(sys.argv) != 6:
print 'Usage: python hiveserver2-client.py <hive_host> <hive_port> <db_name> <hadoop_user> <hadoop_password>'
sys.exit()
hiveHost = sys.argv[1]
hivePort = sys.argv[2]
dbName = sys.argv[3]
hadoopUser = sys.argv[4]
hadoopPassword = sys.argv[5]
# do the connection
with pyhs2.connect(host=hiveHost,
port=hivePort,
authMechanism="PLAIN",
user=hadoopUser,
password=hadoopPassword,
database=dbName) as conn:
# get a client
with conn.cursor() as client:
# create a loop attending HiveQL queries
while (1):
query = raw_input('remotehive> ')
try:
if not query:
continue
if query == 'exit':
sys.exit()
# execute the query
client.execute(query)
# get the content
for row in client.fetch():
print row
except Pyhs2Exception, ex:
print ex.errorMessage
Error displayed:
[centos#test]$ sudo python hiveserver2-client.py computing.cosmos.lab.fiware.org 10000 default USERNAME TOKEN
Traceback (most recent call last):
File "hiveserver2-client.py", line 42, in <module>
database=dbName) as conn:
File "/usr/lib/python2.7/site-packages/pyhs2/__init__.py", line 7, in connect
return Connection(*args, **kwargs)
File "/usr/lib/python2.7/site-packages/pyhs2/connections.py", line 46, in __init__
transport.open()
File "/usr/lib/python2.7/site-packages/pyhs2/cloudera/thrift_sasl.py", line 74, in open
status, payload = self._recv_sasl_message()
File "/usr/lib/python2.7/site-packages/pyhs2/cloudera/thrift_sasl.py", line 92, in _recv_sasl_message
header = self._trans.readAll(5)
File "/usr/lib/python2.7/site-packages/thrift/transport/TTransport.py", line 60, in readAll
chunk = self.read(sz - have)
File "/usr/lib/python2.7/site-packages/thrift/transport/TSocket.py", line 132, in read
message='TSocket read 0 bytes')
thrift.transport.TTransport.TTransportException: TSocket read 0 bytes
can you post your piece of code ? This looks like some auth mechanism or credentials sent are not Valid.
authMechanism= can be "PLAIN" or "KERBEROS" as per your setup .

Wrote an errback for my Scrapy spider, but tracebacks also keep happening, why?

I am using Scrapy 1.1 and I call Scrapy from within a script. My spider launching method looks like this:
def run_spider(self):
runner = CrawlerProcess(get_project_settings())
spider = SiteSpider()
configure_logging()
d = runner.crawl(spider, websites_file=self.raw_data_file)
d.addBoth(lambda _: reactor.stop())
reactor.run()
Here is an extract of my spider with an errback written as in the documentation, but it only prints when catches a failure.
class SiteSpider(scrapy.Spider):
name = 'SiteCrawler'
custom_settings = {
'FEED_FORMAT': 'json',
'FEED_URI': 'result.json',
}
def __init__(self, websites_file=None, *args, **kwargs):
super().__init__(*args, **kwargs)
self.websites_file = websites_file
print('***********')
print(self.websites_file)
def start_requests(self):
.....
if is_valid_url(website_url):
yield scrapy.Request(url=website_url, callback=self.parse, errback=self.handle_errors, meta={'url': account_id})
def parse(self, response):
.....
yield item
def handle_errors(self, failure):
if failure.check(HttpError):
# these exceptions come from HttpError spider middleware
# you can get the non-200 response
response = failure.value.response
print('HttpError on ' + response.url)
elif failure.check(DNSLookupError):
# this is the original request
request = failure.request
print('DNSLookupError on ' + request.url)
elif failure.check(TimeoutError, TCPTimedOutError):
request = failure.request
print('TimeoutError on ' + request.url)
My problem is that I get errors I expect, like:
TimeoutError on http://www.example.com
But also get tracebacks for the same websites:
2016-08-05 13:40:55 [scrapy] ERROR: Error downloading <GET http://www.example.com/robots.txt>: TCP connection timed out: 60: Operation timed out.
Traceback (most recent call last):
File ".../anaconda/lib/python3.5/site-packages/twisted/internet/defer.py", line 1126, in _inlineCallbacks
result = result.throwExceptionIntoGenerator(g)
File ".../anaconda/lib/python3.5/site-packages/twisted/python/failure.py", line 389, in throwExceptionIntoGenerator
return g.throw(self.type, self.value, self.tb)
File ".../anaconda/lib/python3.5/site-packages/scrapy/core/downloader/middleware.py", line 43, in process_request
defer.returnValue((yield download_func(request=request,spider=spider)))
twisted.internet.error.TCPTimedOutError: TCP connection timed out: 60: Operation timed out.
The written exception handling messages and the tracebacks can often be traced to the same websites. After searching a lot on stackoverflow, in the docs and the likes I still dont know why I see the tracebacks.
This also occurs with DNSLookupErrors for example.
Excuse me, my Scrapy knowledge is juvenile. Is this normal behavior?
Also, I added this to settings.py, which is under my crawler. Other entires (for example the item_pipelines) most exactly work.
LOG_LEVEL = 'WARNING'
But I still see debug messages, not only warnings and everything above that. (if configure_logging() is added to the spider launch) I am running this from terminal on mac os x.
I would be very happy to get any help with this.
Try this in a script:
if __name__ == '__main__':
runner = CrawlerProcess(get_project_settings())
spider = SiteSpider()
configure_logging()
d = runner.crawl(spider, websites_file=self.raw_data_file)
d.addBoth(lambda _: reactor.stop())
reactor.run()

Getting URLError Exception when caching webdriver instances

I am attempting to cache webdriver instances across test case classes. I do not need a "clean" webdriver since I am simply using PhantomJS to query the DOM (I do need JavaScript enabled, which is why I am not simply fetching the source and parsing that).
The cache is a dictionary with the URL as a key and the driver instance as value. The cache is in the base test case, and I call get() which is a method on the base test case. This method instantiates webdriver, and goes to the url if the driver is not in the cache already.
It appears there's some kind of socket issue when trying to access driver properties on the cached instance in the second test case (derivedb.py). I'd appreciate if someone could tell how to get this work.
I am getting the following output:
$ python launcher.py
test_a (deriveda.DerivedTestCaseA) ... Instantiate new driver
Title is: Google
ok
test_b (deriveda.DerivedTestCaseA) ... Retrieve driver from cache
Title is: Google
ok
test_a (derivedb.DerivedTestCaseB) ... Retrieve driver from cache
ERROR
======================================================================
ERROR: test_a (derivedb.DerivedTestCaseB)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/Users/cohenaa/PycharmProjects/sanity/derivedb.py", line 7, in test_a
print "Title is: %s" % self.driver.title
File "/Users/cohenaa/sanity-env/lib/python2.7/site-packages/selenium/webdriver/remote/webdriver.py", line 185, in title
resp = self.execute(Command.GET_TITLE)
File "/Users/cohenaa/sanity-env/lib/python2.7/site-packages/selenium/webdriver/remote/webdriver.py", line 162, in execute
response = self.command_executor.execute(driver_command, params)
File "/Users/cohenaa/sanity-env/lib/python2.7/site-packages/selenium/webdriver/remote/remote_connection.py", line 349, in execute
return self._request(url, method=command_info[0], data=data)
File "/Users/cohenaa/sanity-env/lib/python2.7/site-packages/selenium/webdriver/remote/remote_connection.py", line 410, in _request
resp = opener.open(request)
File "/sw/lib/python2.7/urllib2.py", line 404, in open
response = self._open(req, data)
File "/sw/lib/python2.7/urllib2.py", line 422, in _open
'_open', req)
File "/sw/lib/python2.7/urllib2.py", line 382, in _call_chain
result = func(*args)
File "/sw/lib/python2.7/urllib2.py", line 1214, in http_open
return self.do_open(httplib.HTTPConnection, req)
File "/sw/lib/python2.7/urllib2.py", line 1184, in do_open
raise URLError(err)
URLError: <urlopen error [Errno 61] Connection refused>
----------------------------------------------------------------------
Ran 3 tests in 1.271s
FAILED (errors=1)
launcher.py
import unittest
from deriveda import DerivedTestCaseA
from derivedb import DerivedTestCaseB
suite = unittest.TestSuite()
testclasses = [DerivedTestCaseA, DerivedTestCaseB]
testloader = unittest.TestLoader()
classes_to_names = {}
for tc in testclasses:
classes_to_names[tc] = testloader.getTestCaseNames(tc)
for tc in classes_to_names:
for testname in classes_to_names[tc]:
suite.addTest(tc(testname))
unittest.TextTestRunner(verbosity=10).run(suite)
deriveda.py
from basetestcase import BaseTestCase
from unittest import main
class DerivedTestCaseA(BaseTestCase):
def test_a(self):
self.get("http://www.google.com")
print "Title is: %s" % self.driver.title
def test_b(self):
self.get("http://www.google.com")
print "Title is: %s" % self.driver.title
derivedb.py
from basetestcase import BaseTestCase
class DerivedTestCaseB(BaseTestCase):
def test_a(self):
self.get("http://www.google.com")
print "Title is: %s" % self.driver.title
basetestcase.py:
import unittest
from selenium import webdriver
class BaseTestCase(unittest.TestCase):
cache = {}
def get(self, url):
if url not in self.cache:
print "Instantiate new driver"
self.driver = webdriver.PhantomJS()
self.driver.get(url)
self.cache[url] = self.driver
else:
print "Retrieve driver from cache"
self.driver = self.cache[url]
#classmethod
def tearDownClass(cls):
for url in BaseTestCase.cache:
BaseTestCase.cache[url].quit()
Ahhh. I see now. The driver is quitting after each test case. If I quit after the suite is run instead, no error.

Celery: abort task on connection error

I have to implement a Task subclass that gracefully fails if the broker is not running - currently I'm using RabbitMQ.
I could probably just use a try statement to catch the exception:
try:
Mytask.delay(arg1, arg2)
except socket.error:
# Send an notice to an admin
pass
but I'd like to create a subclass of Task that can handle that.
I've tried something like that:
class MyTask(Task):
ignore_result = True
def __call__(self, *args, **kwargs):
try:
return self.run(*args, **kwargs)
except socket.error:
# Send an notice to an admin
return None
but the workflow is clearly wrong. I think I need to inject maybe a backend subclass or a failure policy somehow.
Do you have any suggestion?
A possible solution I came up with:
import socket
from celery.decorators import task
from celery.task import Task
from celery.backends.base import BaseBackend
UNDELIVERED = 'UNDELIVERED'
class DummyBackend(BaseBackend):
"""
Dummy queue backend for undelivered messages (due to the broker being down).
"""
def store_result(self, *args, **kwargs):
pass
def get_status(self, *args, **kwargs):
return UNDELIVERED
def _dummy(self, *args, **kwargs):
return None
wait_for = get_result = get_traceback = _dummy
class SafeTask(Task):
"""
A task not raising socket errors if the broker is down.
"""
abstract = True
on_broker_error = None
errbackend = DummyBackend
#classmethod
def apply_async(cls, *args, **kwargs):
try:
return super(SafeTask, cls).apply_async(*args, **kwargs)
except socket.error, err:
if cls.on_broker_error is not None:
cls.on_broker_error(err, cls, *args, **kwargs)
return cls.app.AsyncResult(None, backend=cls.errbackend(),
task_name=cls.name)
def safetask(*args, **kwargs):
"""
Task factory returning safe tasks handling socket errors.
When a socket error occurs, the given callable *on_broker_error*
is called passing the exception object, the class of the task
and the original args and kwargs.
"""
if 'base' not in kwargs:
on_broker_error = kwargs.pop('on_broker_error', SafeTask.on_broker_error)
errbackend = kwargs.pop('errbackend', SafeTask.errbackend)
kwargs['base'] = type('SafeTask', (SafeTask,), {
'on_broker_error': staticmethod(on_broker_error),
'errbackend': errbackend,
'abstract': True,
})
return task(*args, **kwargs)
You can both subclass SafeTask or use the decorator #safetask.
If you can think of an improvement, don't hesitate to contribute.