Case Study: Re-Thinking Our Infrastructure Tooling
summary
Fairwinds replaces an old Python templating pipeline with a Terraform-driven system that generates per-customer repos and runs via Atlantis, moving from interactive edits to automated rollouts across hundreds of clusters in three clouds. Goals include git as single source of truth, Terraform generating Terraform, and split access from change so a CVE fix can roll out in under a day. Proofs of concept run one to two weeks; implementation starts in three to four months and is ~80% complete.
AI-written summary · about
chapters (11)
- 0:00Introduction and Speaker Background
- 0:42Overview of Fairwinds and Kubernetes Infrastructure
- 1:44Challenges with Current Infrastructure Tooling
- 2:49Reasons for Rewriting Infrastructure Tooling
- 5:04Process of Rethinking Infrastructure Tooling
- 6:02Existing Solution and Its Problems
- 8:40Defining Requirements for New Tooling
- 10:38Brainstorming and Evaluating Solutions
- 18:14Decision Making and Proofs of Concept
- 19:49Implementation and Iteration
- 22:46Final Thoughts and Q&A
related talks
22mCase Study: Re-Thinking Our Infrastructure Tooling
44mTaming the cloud sprawl
14mHow Global Discovery tribe approaches Infrastructure Changes at scale
25mRebuilding Infrastructure (as Code)
27mAdvanced Multipurpose Infrastructure: A Self-Managed Cloud Environment with HashiCorp Terraform
19mSynchronization of Infrastructure Provisioning & Application Deployments
22mTerraform Hybrid: Unifying Your Physical and Cloud Datacenters
44mEvolution of Infrastructure Management