Issue in Redis Lockstore

Rasa Open Source
1

I am facing issues in handling requests to Rasa server, I facing mainly issue in Redis Lockstore, it throws

"Failed to acquire lock for conversation ID '{conversation_id}'. Retrying..."

Rasa Configuration:

  • Rasa version: 1.10.7
  • Running in Docker Compose
  • CPU: 4 core
  • Python version: 3.6.9
  • Redis latest docker image

I have connected to Redis Lockstore using Redis docker image:

also I have set up the SANIC_WORKER TO 3 in the docker file of Rasa server:

ENV SANIC_WORKERS=3

Lockstore configuration:

lock_store:
  type: redis
  url: redis
  port: 6379
  db: 3

Thanks, Jitesh

2

@Tobias_Wochinger can you help me?

3

@JiteshGaikwad do you get this issue when you only have one sanic worker?

4

@erohmensing even if I set 1 sanic workers I face the same issue.

5

@erohmensing I am getting the issues frequently, it keeps not throwing the below message on the console:

“Failed to acquire lock for conversation ID ‘{conversation_id}’. Retrying…”

and keeps throwing the issue retrying forvever and for every other user and the bot stops responding.

and I have 4 web services call from my custom actions, out of which 1 of the webservice takes around 3-4 seconds for response.

I tried all the services by making async and without async but the issue still occurs.

6

@JiteshGaikwad How do you run the redis docker image?

7

I am using docker-compose, and below is the content for redis image:

    redis:
        image: redis:latest
        container_name: redis
        ports:
            - 6379:6379
        volumes:
            - ./redis.conf:/redis.conf
        environment:
            - ALLOW_EMPTY_PASSWORD=yes
        command: ["redis-server","--maxmemory","1gb"]

also my redis.conf is an empty file.

and I faced the same issue even if I don’t use redis lockstore

8

Any updates?