165 - Flaky Tests

Changelog

• I’ve sorted the Kubernetes ingress controller for the staging deployment of howtocodewell.net
• I’ve got a few problems with connecting services outside of the Kubernetes cluster but I am working my way through those.
• I’ve been updating the pipelines to allow for certain manual jobs in GitLab.

News/articles

Python ASGI The future of Python Web Development

A new asynchronous standard for Python web applications.

WSGI = Web Server Gateway Interface

ASGI = Asynchronous Server Gateway Interface

https://www.infoworld.com/article/3658336/asgi-explained-the-future-of-python-web-development.html

Podcast 2.0

I have my eye on the podcast 2.0 movement and may start to add some backlog tickets to investigate what is involved from a technical standpoint to updating the RSS feeds.

https://medium.com/@everywheretrip/an-introduction-to-podcasting-2-0-3c4f61ea17f4

• The Podcast Index
• Podcasting Namespace
• PodPing
• Value-For-Value

Send Symfony logs to Slack

This could be a neat little future video tutorial. I could also do something similar with a Discord bot.

https://manyou.blog/send-symfony-logs-to-slack-with-symfony-notifier-as-a-monolog-handler

What’s a flaky test

A flaky test is a test that sometimes passes and sometimes fails. This usual happens randomly and is incredibly annoying.

What causes flaky tests

• Under powered resources
• Tests running in random order
• Tests not cleaning themselves up afterwards
• External programs or requests interfering with the tests

How to fix flaky tests

• Run tests more regularly
• Separate flaky tests and run them in isolation until fixed
• Only test what is needed based on what has changed
• Don’t ignore or skip tests if you can help it
• Check the environment
• Check inputs for non-deterministic values
• Check timeouts
• Check system resources
• Check for race conditions
• Check if there is any fixed order to the tests
• Keep your tests maintained.