Python Applications with Habitat
summary
Shows a successful production Sentry cluster built and run with Habitat, replacing virtualenvs by packaging Python and all C libraries into Habitat packages so runtime is immutable. Uses Habitat plans, the supervisor, port bindings, Raven for telemetry, and continuous build to resolve version churn. The Sentry stack ends up as 109 packages with 338 dependency nodes, uses --no-compile installs and a package environment separator, and rebuilds transitive deps (eg OpenSSL) when vulnerabilities appear.
AI-written summary · about
chapters (9)
- 0:00Intro
- 0:40Build Package . Deploy Application configuration and management Continuous Integration and Delivery Platform Agnostic
- 0:55The Typical Python Project
- 2:56Working around the problems Build Python from source or 3rd party package Build binary modules from source
- 3:571. One codebase, one application 2. API first 3. Dependency management 4. Design, build, release, and run 5. Configuration, credentials, and code
- 5:197. Disposability 8. Backing services 9. Environment parity 10. Administrative processes
- 6:1911. Port binding 12. Stateless processes 13. Concurrency 14. Telemetry 15. Authentication and authorization
- 9:20Paradigms 1 Habitat plan, 1 Python library 1 Habitat plan, multiple Python libraries
- 12:46Package Environments
related talks
39mAchieving Cloud Agnostic Application Deployment with Habitat
28mHabitat Automation that travels with the app
39mSimplifying Container Operations with Habitat
45mThe Evolution Of Automation
37mHabitat Essentials - An Absolute Beginner's Guide
05mIgnite: Introduction to Habitat
40mCan an Enterprise Run Habitat? Or Course!
36mMeshing Legacy Systems with Habitat Deployed Components to...