SELECT is not allowed in cluster mode

[AHulshoff]

I tried to install Superset on kubernetes with the helm chart. As I already had a redis and postgres environment, I decided to reuse them. So I configured values.yml to use the correct parameters to setup the environments. But after deploying the helm chart, I found out that Superset is not working properly. After some investigation I found out that Superset worker is trying to start, but fails a little after with the following error message:

[2024-03-26 17:11:13,020: ERROR/MainProcess] consumer: Cannot connect to redis://:**@redis-redis-cluster-headless.redis.svc.cluster.local:6379/2: SELECT is not allowed in cluster mode.
Trying again in 6.00 seconds... (3/100)

On Slack, I asked a similar question where someone thought it might be caused by the redis client not be configured to use a redis cluster. I have done some research and he might be right. If I look at the code of Superset, I cannot find a reference to redis cluster. Unfortunately, I am not an expert on using redis in python... Or redis in general for that matter. In other investigations I have done, I couldn't find any references to a similar problem. So I am not sure it is a bug and therefore I assume I made a mistake somewhere.
Both redis and postgres clusters are also running on kubernetes and are installed with the Bitnami helm charts.

My values.yml:

service:
  type: LoadBalancer
  port: 8088
  nodePort:
    http: 30111
supersetNode:
  connections:
    redis_host: redis-redis-cluster-headless.redis.svc.cluster.local
    redis_port: 6379
    redis_password: [redacted]
    redis_cache_db: 1
    redis_celery_db: 2
    db_host: [redacted]
    db_port: 5432
    db_user: [redacted]
    db_pass: [redacted]
    db_name: superset
postgresql:
  enabled: false
redis:
  enabled: false
configOverrides:
  secret: |
    SECRET_KEY = [redacted]
    CACHE_KEY_PREFIX = 'superset_'

Can someone help me to solve this?

[poorTANY]

tricky skip:

set env or in .env-local
REDIS_RESULTS_DB=0

maybe in your values.yml:
redis_cache_db: 0
redis_celery_db: 0

[szymanmi]

@AHulshoff Did you manage to solve this problem?
@poorTANY I set everything to '0'

redis_cache_db: 0
redis_results_db: 0
redis_celery_db: 0

But it didnt solve my issue
redis.exceptions.ResponseError: MOVED 14663 10.244.9.250:6379

[AHulshoff]

Well, I got it to work. But I made 2 changes and I am not sure which one fixed it. The first one is that updated the version of the helm chart and the second one was throwing out redis and replace it with keydb.