Custom Datadog metrics on Google App Engine

gae.md

How to use custom Datadog metrics on Google App Engine

Installation

We will use the dogapi module to instrument your code. There are a lot of ways to ship libraries to Google App Engine using virtualenvs, symlinks, etc. If you already have that, just add to your stack the latest version of dogapi from PyPi.

If not, here is an easy way to use dogapi:

cd $MY_GAE_PROJECT
echo "dogapi" >> requirements.txt
pip install -r requirements.txt -t lib/

Then in your application initalization (appengine_config.py) make sure to the libraries to your Python path:

import sys
import os.path
sys.path.insert(0, os.path.join(os.path.dirname(__file__), 'lib'))

Also the dogapi module sends metrics and events through a secure TLS connection, so you need to enable the ssl module in your app.yaml

libraries:
  - name: ssl
    version: "latest"

Instrumentation

1. set your API key from datadog when initializing dogapi. The flush_in_thread is needed right now due to threading constraints on GAE

from dogapi import dog_stats_api as dog
dog.start(
    api_key=PUT_YOUR_API_KEY_HERE,
    flush_in_thread=False
)

2. start sending your own metrics

@app.route('/datadog')
def datadog():
    dog.increment('myapp.requests', tags=['controller:datadog'])
    return 'Datadog'

If you do flush_in_thread=False, how will the metrics be send. I did follow the above steps but the dashboard never recieve metrics. if i do flush_in_thread=True, than get metrics in the dashboard. Can u clarify