skip to navigation
skip to content

python-rush 1.0.0

Create workers, wait for them to be ready, then run them all at once

This python module is for making simple tools that rush a resource, such as an API endpoint or web UI. This runs with both Python 2.6+ or 3.0+.

This was originally created to test the throttling of authentication attempts. It is quite basic, but was enough to get good results from a Django based web app. I imagine that this is good enough to indicate if rate limiting is reasonably implemented on faster systems (more likely to come down to network?).


This example attempts to connect to an API endpoint which will throttle after 10 requests:

    # Python 2
    from xmlrpclib import ServerProxy, Fault
except ImportError:
    # Python 3
    from xmlrpc.client import ServerProxy, Fault

from rush import Rusher

def api_auth_tester(index, thread_count):
    Rush the API with invalid authentication attempts.
    # prepare the worker
    proxy = ServerProxy('')
    yield  # yield to indicate that the worker is ready to rush
    # rush
    except Fault as error:
        # yield a string indicating the result
        if error.faultCode == 4:  # bad username/pass
            yield 'attempted'
        elif error.faultCode == 12:  # throttled
            yield 'throttled'

print("API rate limiting test:")
# the API will throttle after 10 requests, so make 9 requests first, then rush two calls
rusher = Rusher(api_auth_tester, 9)
# preform the first 9 requests so the next request should set a throttling indicator
duration, results = rusher.rush_and_report()
# change the number of threads we want to make
rusher.thread_count = 2
# only one call should not be throttled

This will produce output like the following:

API rate limiting test:
9 threads completed in .747115, results:
    attempted: 9
2 threads completed in .127208, results:
    attempted: 1
    throttled: 1
File Type Py Version Uploaded on Size
python_rush-1.0.0-py2.py3-none-any.whl (md5) Python Wheel py2.py3 2017-04-15 6KB