10 Tips for failing at microservices
summary
The talk claims that common microservice practices break in practice and shows concrete failures: mixing dozens of runtimes and build tools leads to poly WTF, a single shared Oracle database causes 20 dependent services to crash because one schema change propagates, a self-built event-sourcing framework ships breaking ID changes and forces coordinated releases, and an auth service outage turns many services into a distributed monolith. Tools named include Kubernetes Terraform Chef Puppet Jenkins Ansible Docker React Spring Boot.
AI-written summary · about
chapters (23)
- 0:00Start
- 2:54Sabotage countdown
- 6:59The data monolith
- 11:512 months later...
- 15:18Day rates!
- 15:36Building frameworks is fun!
- 16:06The meat cloud
- 17:27Cattle not pets
- 18:31Ops communication?
- 20:14The distributed monolith
- 26:40Business people cannot understand complex any architecture
- 27:38Requirements Ping Pong
- 27:50Change management
- 28:08Bonus: works with ops, too
- 29:11The staffing monolith
- 30:29Dysfunctional setup in 3 easy steps
- 31:15Protect the teams
- 31:32Establish bureaucracy
- 32:41Even sounds "agile"!
- 34:41Wrapping up....
- 35:41The key to failure is the hidden monolith
- 36:29Brownfield - The perfect excuse
- 37:39Embrace the monolith both in mind and way of working
related talks
40m10 tips for failing at microservices
30mHow to Fail at Building Modern Microservices
57mEmbracing Microservices
43mHow to Avoid Doing Microservices Completely Wrong
44mMicroservices, Where Did It All Go Wrong
43mPanel: Microservices, Where Did It All Go Wrong
55mMicroservices, Where Did It All Go Wrong
01h 00mMicroservices, Where Did It All Go Wrong?