riemann_wrapper: time functions with riemann

Introductory blog post: http://www.spootnik.org/entries/2013/05/21_using-riemann-to-monitor-python-apps.html

This library is provided to allow timing and exception reporting of python functions using http://riemann.io.

Installing

Pip:

pip install riemann_wrapper

Pypi: https://pypi.python.org/pypi/riemann_wrapper

Manual:

python setup.py install

Usage

Provide a metric name and a bernhard client for timing:

import bernhard
from riemann_wrapper import wrap_riemann, riemann_wrapper

riemann = bernhard.Client()

@wrap_riemann("cpu-intensive-task", client=riemann)
def do_something_cpu_intensive():
   # [...]

Call options:

The following keyword args may be passed to wrap_riemann:

client: instance of bernhard.Client to send events with
tags: tags to attach to riemann events
host: override hostname for the event
logger: a standard python logger to which transport errors may be logged

Alternately, a new wrapping function can be created by calling riemann_wrapper like-so:

import bernhard
from riemann_wrapper import wrap_riemann, riemann_wrapper

riemann = bernhard.Client()
my_wrapper = riemann_wrapper(client=bernhard.Client(), prefix="myapp.")

@my_wrapper("cpu-intensive-task")
def do_something_cpu_intensive():
   # [...]

The following keyword args may be passed to riemann_wrapper:

client: instance of bernhard.Client to send events with
global_tags: tags present in all sent events. Default: ['python'].
host: override hostname for all events. Default: None.
logger: a standard python logger to which transport errors may be logged. Default: None.
prefix: prepend given string to all event services. Default: python.
exception_state: state sent for exceptions. Default: 'warning'.
send_exceptions: boolean or callable that takes an exceptions as a parameter and returns a boolean to specify whether to send that particular exception to riemann. Default: True.