Developing for Multi-Component Environments While Keeping Your Sanity
summary
A production SaaS for binaries (Bintray) is built from many live pieces—CouchDB/Cloudant, Grails, MongoDB, EHCache, Elasticsearch, Redis and nginx—and the infrastructure is declared as code so developers run the same stack locally. Key choices and outcomes: 300 Chef nodes manage environments, Vagrant + Chef + Jenkins move commits to production with full traceability, a user feedback report becomes a site change in about 2–4 hours, and a local Artifactory for binaries cuts VM setup from minutes to seconds. Docker replaces heavy Vagrant boxes for speed.
AI-written summary · about
chapters (18)
- 0:00Start
- 1:27What Frog?
- 2:10Agenda
- 4:46We love our tools (and frameworks)
- 5:49mongoDB
- 13:37Chef what?
- 14:17Chef works with cookbooks
- 14:56Install MySQL in Ruby
- 16:57Vagrant what?
- 17:33Setup Vagrant with Chef
- 19:38Development
- 20:46DevOps in Dev Env
- 22:171. VAGRANT BOOTS
- 23:43Going to Production...
- 24:12The Solution
- 26:37Production - cloud all the way
- 28:21THE GOOD
- 31:08What's next?
related talks
33mDeveloping for Multi-Component Environments While Keeping Your Sanity
51mApplication Stacks for the Modern World
27mTaming the Kraken - How Operations enables developer productivity
53mChange your tools, change your outcome
49mAutomation in Deployment on Hybrid Hosting and Private Cloud Environments
37mLessons learned from dockerizing complex Web Apps
54mBuilding and Managing a Ruby Infrastructure
39mRunning Ruby Apps