Comment on page
🧠
Redis Configuration
soketi is optimized to work in multi-node or multi-process environments like Kubernetes or PM2. However, as you may expect, the main concern when scaling horizontally is how to make the nodes communicate with each other. To be able to scale horizontally and efficiently enable node-to-node or process-to-process communication, soketi leverages a Redis connection.
You can configure soketi to run using Redis as a primary adapter for all typically local-persistent data using the
ADAPTER_DRIVER environment variable:$ SOKETI_ADAPTER_DRIVER=redis soketi start
Adapters like Redis are used for horizontal scaling only. Configuring and running Redis when running soketi using the
local adapter will have no effect.These are the Redis variables to configure the adapter for Redis. For the rest of them, like host, port, and clustering support, read Redis Configuration.
Name | Default | Possible values | Description |
|---|---|---|---|
SOKETI_ADAPTER_DRIVER | local | redis, nats, local, cluster | The adapter driver to use to store and retrieve each app and channel's persistent data. |
SOKETI_ADAPTER_REDIS_PREFIX | '' | Any string | The Redis adapter's Pub / Sub channels prefix. |
SOKETI_ADAPTER_REDIS_CLUSTER_MODE | false | true, false | Whether the client should be initialized for Redis Cluster. You have to specify the DB_REDIS_CLUSTER_NODES value. |
Last modified 1yr ago