Thomas Eizinger
Thomas Eizinger
> It's probably a good idea to filter out 0.0.0.0 advertisements. My point was towards this in that we never advertise `0.0.0.0` because we translate them to actual addresses of...
Filtering the addresses sounds good to me. We need to do it as late as possible, _after_ we gathered the addresses from the behaviours. We should also add a log...
Adding a new variant will be a breaking change so I'd rather not do that for such a corner case. I think filtering is fine. Most users will never hit...
We just need to change these addresses to something else then! :)
> would work on mDNS and DHT simultaneously, you can assign me this as well Make them two PRs please! :)
You are welcome to work on this too!
What is the actual error that is returned here? I think to correctly fix #4460, we should catch this case explicitly and **not** return on the kernel to fail dialing...
> Attacker continuously tells target a new (fake) external address of the target. Target sends N updates each time where N is the number of connected peers. I.e. an amplification...
> If I'm reading it correctly, it only has effect if particular peer ID is not yet in k-buckets, however it if it already there, insertion configuration is ignored: @mxinden...
> > I am taking away that this only controls the initial insert > > This might indeed be the intention when it was introduced, however the way I see...