errgroup VS run

or https://github.com/StevenACoffman/errgroup if you don't use panic-recovery middleware

A co-worker of mine co-worker Ben Kraft, wrote some handy wrapper code around sync/errgroup to avoid that boilerplate (and required discipline). With his permission, I lightly modified it to lift it out of our private work repository for the more general Go community.

What I’ve done in the past is use a run group (like https://github.com/oklog/run ) and one of the goroutines in the group is just listening for a SIGINT/SIGTERM/etc. When it sees that it throws an error which cancels the root context and let’s each component handle its own graceful shutdown

Just a bit of feedback. https://github.com/oklog/run is great and simple way of handling goroutines. It doesn't bake in signal handing, health checking, retries, because those are technically all separate concerns, that can be independently implemented. When I read your Readme, the lifecycle part immediately made me think of oklog/run.

I would suggest to use this library https://github.com/oklog/run It is really helpful to manage goroutines lifecycles.

https://github.com/oklog/run Sounds like you’re looking for something like this.

How does it compare to https://github.com/oklog/run?

There is an interesting discussion which has an alternative view that, with few exceptions, panics should crash your program. I'm ok with that in development and testing, but would rather sleep soundly at night.