We are in the process of migrating from Heroku to AWS, and I am noticing the Sidekiq stats mysteriously resetting for no apparent reason.
This is happening in several different applications that are connected to the same Redis instance, each with its own namespace set in initializers/sidekiq.rb.
The stats reset across all of the Sidekiq counters at the same time. It seems like perhaps we are momentarily dropping the Redis connection, but that is just wild conjecture and at any rate I'm not sure how to mitigate it.
Is this a common problem? Is there a setting I can tweak?
Someone is running the FLUSHDB or FLUSHALL command and clearing out data in Redis. Perhaps one of the apps when it starts.