Learning path

Move from a five-minute contract check to backend-specific operations and production drills.

Stage 1 — semantics

Run two local contenders and verify that exactly one action runs while the other returns null. Repeat with runIfLeaderResult() so Skipped and an elected null result are visibly different. Then throw from the action and confirm the failure reaches the caller.

Stage 2 — model and backend

Choose single, group, or strategic election from the concurrency requirement. Choose the execution API from the application’s threading model. Only then select a backend already covered by your operational skills. The module pages contain constructors, configuration, and failure notes; the runnable examples show the pieces together.

Stage 3 — production exercises

Measure action duration, set lease and wait budgets, enable metrics, and write alerts for skipped spikes, failures, and lease-extension errors. Rehearse backend loss, expired ownership, duplicate delivery, graceful shutdown, and restart. A manual is complete only when the team knows what to observe and what to do next.

Release sources

• leader-core/src/test/kotlin/io/bluetape4k/leader/LeaderRunResultTest.kt
• examples/batch-scheduler/README.md
• examples/prometheus-dashboard/README.md

Continue learning

• Bluetape4k Leader manual
• Choose an election model
• Testing leader election
• Observability and operations