KEDA | Redis Streams

Scale applications based on Redis Streams.

Availability: v1.5+ Maintainer: Community

Trigger Specification

Redis 5.0 introduced Redis Streams which is an append-only log data structure.

One of its features includes Consumer Groups, that allows a group of clients to co-operate consuming a different portion of the same stream of messages.

This specification describes the redis-streams trigger that scales based on the Pending Entries List (see XPENDING) for a specific Consumer Group of a Redis Stream.

triggers:
- type: redis-streams
  metadata:
    address: localhost:6379 # Required if host and port are not provided. Format - host:port
    host: localhost # Required if address is not provided
    port: "6379" # Required if address is not provided and host has been provided.
    passwordFromEnv: REDIS_PASSWORD # optional (can also use authenticationRef)
    stream: my-stream # Required - name of the Redis Stream
    consumerGroup: my-consumer-group # Required - name of consumer group associated with Redis Stream
    pendingEntriesCount: "10" # Required - number of entries in the Pending Entries List for the specified consumer group in the Redis Stream
    enableTLS: "false" # optional
    databaseIndex: "0" # optional
    # Alternatively, you can use existing environment variables to read configuration from:
    # See details in "Parameter list" section
    addressFromEnv: REDIS_ADDRESS # Optional. You can use this instead of `address` parameter
    hostFromEnv: REDIS_HOST # Optional. You can use this instead of `host` parameter
    portFromEnv: REDIS_PORT # Optional. You can use this instead of `port` parameter

Parameter list:

address: The host and port of the Redis server in the format host:port, for example my-redis:6379

As an alternative to the address field, the user can specify host and port parameters.

host: The host of the Redis server

It is not required if address has been provided

port: The port of the Redis server

It is only to be used along with the host/hostFromEnv attribute and not required if address has been provided

passwordFromEnv (optional): Name of the environment variable your deployment uses to get the Redis password.
stream: Name of the Redis Stream
consumerGroup: Name of the Consumer group associated with Redis Stream
pendingEntriesCount: Threshold for the number of Pending Entries List. This is the average target value to scale the workload. Defaults to 5
databaseIndex: The Redis database index. Defaults to 0 if not specified
enableTLS: Set this to true if TLS connection to Redis is required. Defaults to false

Some parameters could be provided using environmental variables, instead of setting them directly in metadata. Here is a list of parameters you can use to retrieve values from environment variables:

addressFromEnv: The host and port of the Redis server, similar to address, but reads it from an environment variable on the scale target. Name of the environment variable your deployment uses to get the Redis server URL. The resolved host should follow a format like my-redis:6379.
hostFromEnv: The host of the Redis server, similar to host, but reads it from an environment variable on the scale target.
portFromEnv: The port of the Redis server, similar to port, but reads it from an environment variable on the scale target.

Authentication Parameters

The scaler supports two modes of authentication:

Using password authentication

Use the password field in the metadata to specify the name of an environment variable that your deployment uses to get the Redis password.

This is usually resolved from a Secret V1 or a ConfigMap V1 collections. env and envFrom are both supported.

Here is an example:

apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: redis-streams-scaledobject
  namespace: default
spec:
  scaleTargetRef:
    name: redis-streams-consumer
  pollingInterval: 15
  cooldownPeriod: 200
  maxReplicaCount: 25
  minReplicaCount: 1
  triggers:
    - type: redis-streams
      metadata:
        addressFromEnv: REDIS_HOST
        passwordFromEnv: REDIS_PASSWORD # name of the environment variable in the Deployment
        stream: my-stream
        consumerGroup: consumer-group-1
        pendingEntriesCount: "10"

Using `TriggerAuthentication`

You can use TriggerAuthentication CRD to configure the authentication. For example:

apiVersion: v1
kind: Secret
metadata:
  name: redis-streams-password
type: Opaque
data:
  redis_password: <encoded redis password>
---
apiVersion: keda.sh/v1alpha1
kind: TriggerAuthentication
metadata:
  name: keda-redis-stream-triggerauth
spec:
  secretTargetRef:
    - parameter: password
      name: redis-streams-password # name of the Secret
      key: redis_password # name of the key in the Secret
---
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: redis-streams-scaledobject
  namespace: default
spec:
  scaleTargetRef:
    name: redis-streams-consumer
  pollingInterval: 15
  cooldownPeriod: 200
  maxReplicaCount: 25
  minReplicaCount: 1
  triggers:
    - type: redis-streams
      metadata:
        address: localhost:6379
        stream: my-stream
        consumerGroup: consumer-group-1
        pendingEntriesCount: "10"
      authenticationRef:
        name: keda-redis-stream-triggerauth # name of the TriggerAuthentication resource

Trigger Specification

Authentication Parameters

Using password authentication

Using TriggerAuthentication

Currently available scalers for KEDA

Using `TriggerAuthentication`