asyncio example

This repo contains an example program demonstrating a use of asyncio, and accompanies my blog post on the topic. The program implements an example job runner agent which runs another command while concurrently sending a regular heartbeat to a tracking server. Status information about the job is also sent to the tracking server on start and completion.

Usage

This example requires Python 3.7 or higher. If you don't have that available on your system, I recommend looking into pyenv or conda.

Install the dependencies with:

pip install -r requirements.txt

The program sends requests to a compatible tracking server. I've included a simple Flask tracking server that prints out information on the requests it receives. Run it with:

python dummy_tracking_server.py

With the dummy server running, run the job agent in another shell. For example, to have it run the command sleep 3, sending tracking information to our dummy tracking server:

python agent.py "sleep 3" http://localhost:5000

In the output of the tracking server, you should see it receive first the 'started' status, then a heartbeat once per second, plus a 'completed' status when sleep 3 finishes:

Received status started
127.0.0.1 - - [15/Feb/2019 13:55:27] "PUT /status HTTP/1.1" 200 -
Received heartbeat
127.0.0.1 - - [15/Feb/2019 13:55:27] "PUT /heartbeat HTTP/1.1" 200 -
Received heartbeat
127.0.0.1 - - [15/Feb/2019 13:55:28] "PUT /heartbeat HTTP/1.1" 200 -
Received heartbeat
127.0.0.1 - - [15/Feb/2019 13:55:29] "PUT /heartbeat HTTP/1.1" 200 -
Received status completed
127.0.0.1 - - [15/Feb/2019 13:55:30] "PUT /status HTTP/1.1" 200 -

The program will also log a 'failed' status with the tracking server when the executed command finishes with a non-zero exit code:

python agent.py "python -c 'raise Exception()'" http://localhost:5000

Received status failed
127.0.0.1 - - [15/Feb/2019 13:57:33] "PUT /status HTTP/1.1" 200 -