Thomas Eizinger
Thomas Eizinger
> I find another problem: dependency version. There has been multiple instances of `tracing` (and `tracing_subscribers`) in the manifests of all crates that depend on it, which will make depencency...
> > Re-exports don't work that well because we'd have to make `logging` a non-dev dependency then which means we'd have to publish it to crates.io. > > Not sure...
This is very interesting. I don't have any resources to debug this but it is definitely something that shouldn't happen! :)
Does `PeerA` subscribe to the topic on startup? Are these the logs of `PeerB`?
To further isolate the issue, could somebody try reproducing this with just `quinn` itself? We don't have a lot of code on top of `quinn` to provide the QUIC transport...
> @thomaseizinger I added `!` to the title, given that this is a breaking change. Let me know in case I misunderstand the conventions of conventional commits. I guess this...
> > Thank you! > > Can we add a test please to ensure we don't break this in the future? > > @thomaseizinger Sure thing! But frankly speaking, I'm...
@andreasbros I kicked CI which is currently blocked on you being a new contributor. To make iteration smoother, perhaps find yourself a typo / small rewording or some other small...
> Do we want to check to determine if a peer has been previously discovered so we dont return the event for the same peer and address or should we...
Thanks for the ping! The way we handle `0.0.0.0` is that we watch the local system for interface changes and listen on the specific interfaces as they come up, see...