In this talk, I will discuss the next steps after you have written a web application. Python frameworks and libraries makes it really easy to write a highly performant web application out of the box, but the problems that come next needs a number of building blocks and tools to come together.
We will talk about deploying your web application and discuss scaling up and down policies so that we can cater to peak traffic demands. We will discuss how we can peak into our application’s performance by exporting “health” status and metrics, and learn about implementing distributed tracing into our web application.
You have just implemented your next big feature, your pull request has been merged, testing looks good and you want to roll it out to your users. But, have you thought about monitoring, observability, and logging? These three things are usually an afterthought. They shouldn't be. On this page, I will point to resources and discuss about the things that we should always be thinking about before rolling out our code change into any kind of production environment. We will learn about good logging practices, effective monitoring strategies and writing code which allows us to find out what's going on when something goes wrong.
As far as possible, the talk will discuss generic strategies, only using specific tools/software for hands-on examples.
Outline of topics
- Correlated logging/tracing
- Metrics
- Web application monitoring - WSGI, asyncio applications
- Worker monitoring
- Deployment markers
- Scaling