Skip to content

namelessjon/airline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

74 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

airline

tests | PyPI version

Lightweight wide event logging to bring more observability to lambda functions.

This is very strongly inspired by honeycomb and their beeline library.

How?

Use the decorators!

import airline

airline.init(dataset='your_app_name')


@airline.evented()
def some_function(a, b)
    # do things

or

import airline
from airline.awslambda import airline_wrapper


airline.init(dataset='function_or_app_name')


@airline_wrapper
def handler(event, context):
    # do things

Wide event logging?

The idea is to build up the full context of a function/script into one wide event that gets emitted at the end. This puts all the context in one log message, making it very easy to run analytics on, find like errors, and a lot of other things. A full blown observability platform like honeycomb would be more informative, and allows for the notion of spans and distributed tracing (i.e. across different (micro) services and the like).

But this is a start.

Take this:

import time
import random

import airline

airline.init(dataset='example')



@airline.evented()
def main(a, b):
    airline.add_context_field("a", a)
    airline.add_context_field("b", b)

    with airline.timer('processing_a'):
        subfunction1(a)

    with airline.timer('processing_b'):
        subfunction1(b)


def subfunction1(input):
    time.sleep(random.uniform(0, len(input)))


main("foo", "example_long_thing")

And emit this at the end

{
  "time": "2020-03-09T09:49:43.376126Z",
  "dataset": "example",
  "client": "airline/0.1.0",
  "data": {
    "a": "foo",
    "b": "example_long_thing",
    "duration_ms": 9438.041,
    "timers": {
      "processing_a_ms": 255.11,
      "processing_b_ms": 9182.859
    }
  }
}