Getting from Grizzly to Havana: A DevOps Upgrade Pattern
Talkat OpenStack Summit Portland 2013 · on Monday, 2013-04-15· at 09:50· A105 (Portland Convention Center), 777 NE Martin Luther King Jr Blvd Portland, OR 97232·
summary
Crowbar-backed deployments aim to move clouds from Grizzly to Havana by enforcing n minus one compatibility and Milestone-to-Milestone upgrades, using pull-from-source (git clones) and continuous deployment instead of waiting for packs. The team runs automated Tempest suites and deploys every cookbook PR, has Grizzly in QA, and warns against mixing release cookbooks—use dedicated orchestration cookbooks. Releases come every three months; current user split is ~25% Essex, 50% Folsom, 25% Grizzly, and full multi-node upgrade automation is still a work in progress.
AI-written summary · about
chapters (15)
- 0:00Intro
- 2:50Getting Ready
- 4:27Introduction
- 8:24Table of Contents
- 14:26Whats New
- 16:42Upgrade Example
- 18:11Operators Responsibility
- 20:58Going from Grizzly to Havana
- 23:23Why Continuous Deployment
- 25:18The Migration Cube
- 32:18The Benefit
- 33:33Work in Cycles
- 39:11Community Demand
- 41:47Question Time
- 44:41Python Package Management
related talks
01h 12mPitfalls of Continuous Deployment
50mOps issues
56mFast & Simple Deployments for Code & Infra
34mEnabling a Continuous Deployment Mindset
46mRethinking Continuous Delivery
49mAutomation in Deployment on Hybrid Hosting and Private Cloud Environments
31mHow we evolved our cloud infrastructure
33mUnleashing the Gorillacorn - DevOps in Very Large Enterprises