User Story

As an engineer I want our worker to survive cluster upgrades so we can continue processing.

Backend Requirements

• We recently switched to using AWS Elasticache Redis in staging/prod and have redis deployed in our lower envs.
• We can swap the backend for celery to use redis with a config

Additional Information and Assumptions

• We were considering HA Rabbit, but using redis will reduce our dependencies + use a highly available service we already must use.
• https://www.rabbitmq.com/ha.html
• https://stackoverflow.com/questions/31662727/define-rabbitmq-policies-in-configuration-file
• https://bhoey.com/blog/high-availability-rabbitmq-with-mirrored-queues/
• We see an error like this in some of the rabbit pods "rabbitmq home node (node name) is down or inaccessible"

Acceptance Criteria

• Switch celery backend to redis
• Remove rabbit deployments from e2e-deploy
• Remove rabbit from app-interface
• Remove rabbit from docker-compose