How did you implement caching? Did you use Redis locks?

For caching in Go projects, Redis is often used due to its high performance and support for various data structures. In my projects, caching was implemented through wrappers around Redis, where data was serialized into JSON or binary format for quick access.

To prevent race conditions during cache updates, a locking mechanism based on Redis was used — for example, using the SET command with the NX option and a TTL. This allowed one process to acquire the lock and update the cache, while others waited or used old data.

An example of setting a lock in Redis in Go using the go-redis library:

ctx := context.Background()
lockKey := "cache_lock"
success, err := rdb.SetNX(ctx, lockKey, "locked", 10*time.Second).Result()
if err != nil {
    // handle error
}
if success {
    // lock acquired, cache can be updated
} else {
    // lock not acquired, wait or use old data
}

Thus, Redis locks help prevent simultaneous cache updates by multiple processes and ensure data consistency.