zlacker

This is still a pretty common misconception (and one that I held until recently when looking at Redis configuration options again), but Redis does have durable operating modes to reload the datastore after going offline (which do significantly trade off speed for durability). If your task queue requires a high enough volume of workload that running Redis in a durable fashion is insufficient, then it's probably time to think about a different queuing system than WakaQ.

>>belthe+(OP)
I'm talking about if the VM/Instance/Disk goes away. It looks like Redis is a single point of failure unless you run Redis in a redundant way, but skimming the code it looks like that is at best left as an exercise to the reader. ie. Redundancy in the data store doesn't appear to be handled by this code.

>>jedber+h
Correct, this setup wouldn't survive chaos monkey.

>>welder+ua
Then I wouldn't call it distributed, since it's centralized on redis.

Otherwised good product.