Continuous Delivery: Tales from Windowsland
summary
Claims that Continuous Delivery works on Windows but is harder due to tooling gaps and packaging quirks; recommends use TeamCity or Go as orchestration, avoid TFS for SCM and Team Build, and PowerShell + Chef/Puppet for infra as code. Examples: a client with 300 developers and 300 testers on 70 million LOC splits devops into four teams that exchange artifacts over FTP (no visibility), and a custom deployment tool gets deleted, taking two days to rebuild and a week for the project to recover. Tools mentioned: NuGet, Octopus, Chocolatey.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 2:21releasing frequently
- 10:07A little story about TFS...
- 12:39Compile and unit test
- 14:23Package
- 18:37Deployment Pain
- 20:16MOAR PAIN
- 24:43Virtual Machine Images
- 26:34virtualization
- 27:55Infrastructure as Code
- 31:22Infrastructure Pain
- 39:31Happy endings?
related talks
32mContinuous Delivery Sounds Great But It Won’t Work Here
30mModern Continuous Delivery
47mContinuous Delivery Warstories
24mWhy, What, and How of Continuous Delivery
01h 01mA team's transition to Continuous Delivery
30mOur Continuous Delivery Journey
51mBetter software, faster: principles of Continuous Delivery and DevOps
53mDevOps and Agility with Visual Studio, Azure and Scrum