Alauda Cache Service for Redis OSS

How to Initialize Cluster Shard Slots

By default, a created Redis cluster instance will evenly distribute all 16384 hash slots to all shards. However, in specific scenarios such as disaster recovery, a more flexible slot layout is often required. To this end, we support providing a custom slot distribution plan when creating an instance to meet such advanced requirements.

CLI
cat << EOF | kubectl -n default create -f -
apiVersion: middleware.alauda.io/v1
kind: Redis
metadata:
  name: c6
spec:
  activeRedis:
    proxy:
      replicas: 1
      resources:
        limits:
          cpu: "1"
          memory: 1Gi
        requests:
          cpu: "1"
          memory: 1Gi
      service:
        type: LoadBalancer
    serviceID: 0
  affinityPolicy: AntiAffinityInSharding
  arch: cluster
  customConfig:
    save: 60 10000 300 100 600 1
  passwordSecret: redis-c6-password
  persistent:
    storageClassName: sc-topolvm
  persistentSize: 1Gi
  replicas:
    cluster:
      shard: 3
      slave: 1
      shards:
      - slots: 0-5460
      - slots: 5461-10922
      - slots: 10923-16383
  resources:
    limits:
      cpu: 1
      memory: 1Gi
    requests:
      cpu: 1
      memory: 1Gi
  version: "6.0"
EOF

The primary field configured here is spec.replicas.cluster.shards. The number of items must match the value in spec.replicas.cluster.shard. The spec.replicas.cluster.shards[].slots field supports: 0-5460, 5461-10922, 10923-16383, single slots like 5461, and mixed lists like 0-1000, 1001, 1002-5461.

For the specific meaning of other fields, please refer to the API documentation.