Grab a Life Raft and Migrate to Integrated Storage
summary
Claim: migrate Vault from Consul to Integrated Storage (RAFT) to cut operational complexity and nodes. Integrated Storage (introduced in Vault 1.4) replicates data locally via RAFT, requires a typical five-node cluster with quorum of three, uses only ports 8200 and 8201, and moves from memory-bound Consul to disk-bound storage needing high IOPS. Features added across 1.6–1.8 include auto snapshots, cloud auto join and autopilot; missing enterprise features are automated upgrades and redundancy zones. Migration paths: single-cluster snapshot + migrate command, or phased replicated replacement and promote/retire.
AI-written summary · about
chapters (26)
- 0:00Introduction
- 0:26About me
- 1:12Agenda
- 2:14Vault Storage Backends
- 2:41Enterprise Storage Backends
- 4:02Console Architecture
- 5:21Benefits
- 6:59Integrated Storage Architecture
- 7:29Benefits of Integrated Storage
- 8:54Introduction to Integrated Storage
- 11:05Integrated Storage Features
- 13:01Cloud Auto Join
- 13:48Vault 17 Released
- 14:24Vault 18 Released
- 14:58Integrated Storage Features Overview
- 15:58Automated Upgrades
- 17:30Redundancy Zones
- 18:40Comparison Summary
- 19:26Architecture
- 21:34Network Requirements
- 23:25Single Cluster Migration
- 24:42Tips for Single Cluster Migration
- 25:41How do we migrate our replicated environment
- 26:10Migration workflow
- 28:23Migration considerations
- 28:58Recap
related talks
25mLessons learned from deploying Vault at scale
30mHow to Migrate from Vault Open Source to Vault Enterprise
27mScaling Vault Enterprise with a cell architecture
29mDesigning High Availability for HashiCorp Vault in AWS
27mVault At Scale
29mAdopting HashiCorp Vault: Deployment, Adoption, and Beyond
45mSecure Together: Consul + Vault
29mAutomating Vault deployments: Patterns, pitfalls, and practical lessons