Using bottleneck options to recover from instance's crash

[yarinvak]

Hello, I would like to further understand what do you mean by "inactivity" when you talk about the timeout option.

And would appreciate if you can help me understand how can I use bottleneck options in the following scenario:
I have a service which runs on multiple instances, and a bottleneck limiter that is configured to store the keys on redis.
I have jobs, which I don't want to set expiration on.

In case one of the instances started handling the job, but then it was restarted and the job wasn't completed successfully -
how can I make sure that when the instance will be restarted, the limiter will not be stuck on a job that can't run anymore? (because only the key exists on redis, but the job itself doesn't exist).
I thought about using the clearDataStore:true option, but I'm afraid that if only one instance is restarted, it will delete the bottleneck state that is used by the other instances.

Maybe I'm missing something in the way I use bottleneck..