Server

Cache

Caching backends for app lookups, rate limiting, and general key-value storage.

The cache system provides a key-value store with TTL support. It is used internally for rate limiting state and can be used by other subsystems that need fast lookups.

This is the global cache driver configured under cache. The App Manager has its own separate caching layer configured under app_manager.cache.

Drivers

Driver	Best For	Persistence	Shared Across Nodes
`memory`	Single-node, development	None	No
`redis`	Multi-node production	Yes	Yes
`redis-cluster`	Large-scale Redis Cluster	Yes	Yes
`none`	Disable caching	N/A	N/A

Set the driver:

CACHE_DRIVER=memory

{ "cache": { "driver": "memory" } }

Memory Cache

In-process cache using Moka, a high-performance concurrent cache.

{
  "cache": {
    "driver": "memory",
    "memory": {
      "ttl": 300,
      "cleanup_interval": 60,
      "max_capacity": 10000
    }
  }
}

Setting	Default	Description
`ttl`	`300`	Global TTL in seconds (0 = no expiry)
`cleanup_interval`	`60`	Cleanup interval in seconds
`max_capacity`	`10000`	Maximum number of entries

The memory cache is fast but node-local - each Sockudo instance has its own independent cache. This is fine for single-node deployments but means multi-node setups won't share cache state.

Redis Cache

Shared cache using Redis. All Sockudo nodes read from and write to the same Redis instance, ensuring consistency across the cluster.

{
  "cache": {
    "driver": "redis",
    "redis": {
      "prefix": "sockudo_cache:",
      "url_override": null,
      "cluster_mode": false
    }
  }
}

Setting	Default	Description
`prefix`	`sockudo_cache:`	Redis key prefix
`url_override`	`null`	Override Redis URL (otherwise uses `database.redis`)
`cluster_mode`	`false`	Use cluster-aware connections

Redis cache supports per-key TTL via SET EX and additional bulk operations (MGET, MSET).

Redis Cluster Cache

Same as Redis cache but uses Redis Cluster connections. Use this when your Redis infrastructure runs in cluster mode.

{
  "cache": {
    "driver": "redis-cluster",
    "redis": {
      "prefix": "sockudo_cache:",
      "cluster_mode": true
    }
  }
}

Global Cache Tuning

These environment variables adjust cache behavior across multiple subsystems at once:

Env Var	Default	Affects
`CACHE_TTL_SECONDS`	`300`	App manager cache, channel limits, DB caches, memory cache
`CACHE_CLEANUP_INTERVAL`	`60`	MySQL/PostgreSQL caches, memory cache
`CACHE_MAX_CAPACITY`	`10000`	MySQL/PostgreSQL caches, memory cache

Choosing a Driver

Scenario	Recommended Driver
Development / single-node	`memory`
Multi-node production	`redis`
Redis Cluster infrastructure	`redis-cluster`
Cache not needed	`none`

When using redis for the adapter, use redis for the cache driver too. The REDIS_URL env var configures all Redis connections at once.

Queue

Asynchronous job processing for webhooks and background tasks.

Webhooks

Pusher-compatible webhook delivery with batching, filtering, and Lambda support.