If you have production traffic, you are testing in production right now. Don’t worry though, you are in good company. It is economically irresponsible to do anything else. Could you imagine imitating all of your production load in a non-production environment to test every change including the full data history, global Content Delivery Network and heavy traffic load across devices, countries and internet speeds?
We all know that this is not feasible. That’s why we have techniques such as unit testing and exploratory testing to target the highest risk issues before we stand at the door waving goodbye to our features on their way to their first day of school. But when kids start school, it’s via introduction days that include parents, and shorter solo days with ongoing feedback loops like daily journals and parent/teacher meetings. We should be providing our software with a similarly soft entrance to the world, through more measured practices like dark deploys and feedback loops with observability.
If we all agree that replicating production in test environments before release is infeasible, then we might as well start talking about how to gain insights into the “tests” our users are running on production every second of every day. So how do we go about listening to their feedback? Sure, customer support teams are an amazing avenue, but not all users will reach out. By the time enough do raise alarms, who knows what the impact of an issue would be. We need to empower our delivery team members to contain and track the impact of changes as they make them.
Instead, what if we could say that at first, only our company employees could see the change? Or only users on a certain device type? Or only users who are part of the beta program? Now the “blast radius” of the change impact is better defined, and therefore control over the scale of impact. But what about feedback on the reaction to this change? Just as testability provides hooks to support discoveries through exploratory testing, observability enables exploratory testing on real time user experiences.
Join Abby for more examples and explanations about how leveraging the dynamic duo of observability and testing in production is the next step for empowering software teams to deliver high quality, high impact software with speed and confidence.
Timings
6:00 - 6:10 Introduction & News
6:00 - 6:45 Abby's Talk
6:45 - 7:00 Q&A
7:00 - End
Abby Bangser - Senior Engineer, Duffel, UK
Abby Bangser is a software tester with a keen interest in working on products where fellow engineers are the users. Abby brings the techniques of analysing and testing customer facing products to tools like delivery pipelines and logging so as to generate clearer feedback and greater value. Currently Abby is a Test Engineer on the Platform Engineering team at MOO which supports the shared infrastructure and tooling needs of the organisation.
Outside of work Abby is active in the community by co-leading TechVoices which mentors new and diverse speakers, co-hosting the London Essentials which brings together mentors and new joiners to the software testing industry, and hosts the London chapter of #CoffeeOps meetup which provides a more interactive space for DevOps professionals to discuss relevant topics.
