How does this skill prevent deadlocks?

The skill enforces the use of Time-To-Live (TTL) on all locks to ensure that even if a process crashes, the lock will eventually expire and be released for other instances.

What is fencing in distributed locking?

Fencing involves using a unique owner ID (like a UUID) when acquiring a lock. This ensures that a process can only release the lock it currently holds, preventing it from accidentally releasing a lock acquired by another process.

What is the Redlock algorithm?

Redlock is a distributed lock management algorithm that coordinates across multiple independent Redis nodes to provide a fault-tolerant, highly available locking mechanism for critical operations.

When should I use Redis vs PostgreSQL for locking?

Use Redis for high-speed, low-latency locking where speed is prioritized. Use PostgreSQL advisory locks if you are already using Postgres and require ACID compliance or want to avoid managing extra infrastructure.

Distributed Locking Patterns

Name: Distributed Locking Patterns
Author: yonatangross

byyonatangross

•

API 개발

Implements exclusive resource access and prevents race conditions across distributed service instances using Redis and PostgreSQL.

This skill provides standardized patterns for coordinating exclusive access to shared resources across multiple service instances. It equips Claude with production-ready implementations for Redis-based locks (including Redlock for high availability) and PostgreSQL advisory locks. By utilizing these patterns, developers can reliably prevent duplicate job processing, implement leader election for singleton processes, and protect critical sections in distributed architectures while avoiding common pitfalls like deadlocks or owner-mismatched releases.

주요 기능

01High-availability locking strategies using the Redlock algorithm

02PostgreSQL advisory locks for session and transaction-level coordination

03Redis-based locking with Lua scripts for atomic acquire/release operations

04Pre-configured retry strategies with exponential backoff and jitter

05Safety mechanisms including mandatory TTL management and owner-fencing

0669 GitHub stars

사용 사례

01Preventing duplicate processing of background jobs or events across multiple workers

02Implementing leader election for singleton processes like scheduled cron tasks

03Protecting critical sections and shared resources in high-concurrency distributed environments

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit distributed-locks

For use in Claude.ai and ChatGPT

Download Skill