From database-reliability-engineering
Design a production database's high-availability topology from its RPO/RTO targets — replica type (sync/async/quorum), failure-domain spread (multi-AZ/region), and the failover mechanism with a split-brain guard. Reach for this when a DB needs to survive an AZ/region outage, when failover has never been designed, or when 'we have a replica' is mistaken for HA. Pairs with backup-and-restore-verification.
How this skill is triggered — by the user, by Claude, or both
Slash command
/database-reliability-engineering:ha-topology-and-failoverThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Design availability from the **targets down**, and design the *failover*, not just
Design availability from the targets down, and design the failover, not just the replica.
A replica is not HA; a tested failover is. A standby you can't promote safely under load is decoration. The deliverable is the promotion path + split-brain guard, not the replica count.
Write down RTO (how long can we be down?), RPO (how much data can we lose?), and the failure domains you must survive (node → AZ → region). No targets, no design.
Traverse ../../knowledge/dbre-decision-trees.md §1
to a leaf. Record the path and the runner-up. Key forks: async vs sync (RPO vs write
latency), single-AZ vs multi-AZ vs multi-region (which outage you survive),
automatic vs manual failover (RTO vs complexity).
Specify explicitly:
Confirm primary, replicas, and backups live in different domains. A replica in the primary's AZ doesn't survive an AZ outage.
A topology is a hypothesis until failover is rehearsed. Hand the drill to
database-operations-engineer (the ha-topology-and-failover design pairs with an
operational failover drill).
backup-and-restore-verification.terraform-iac / cloud plugins.database-engineering.An HA topology with the tree path + runner-up, RPO/RTO it meets, the failover mechanism (promotion + split-brain guard + trigger + app reconnection), the failure-domain placement, and a game-day plan to prove it.
npx claudepluginhub mcorbett51090/ravenclaude --plugin database-reliability-engineeringGuides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.
Dispatches multiple subagents concurrently for independent tasks without shared state. Use when facing 2+ unrelated failures or subsystems that can be investigated in parallel.