Steven Allen
Steven Allen
Hm. Actually, I think we're talking about multiple timeouts here. I agree the one @magik6k referenced should probably be shorter in many cases. However, that was really intended to be...
Interesting... We should also be able to use relays without killing the DHT too much but the problem there is: 1. Picking the relay. 2. Not killing the relay. 3....
For tracking, the related open PRs are: * libp2p/go-libp2p-swarm#77 -- The modifications we need in the swarm for the DialPeer/NewStream timeout. Also fixes the per-address timeouts. * libp2p/go-libp2p-net#26 -- the...
(@bigs, you may be interested in this discussion)
Anyone have any thoughts on this?
Given that @magik6k has signed off on https://github.com/libp2p/go-libp2p-net/pull/26, I'm going to move forward on that part. I think we'll want it regardless.
Yeah, sorry. I just wanted to consolidate so we can see the big picture.
Note: we went with this approach for relays because it was easy, not because it was the "best" way. I'm still open to doing it here, but this is kind...
> That's why I suggested an opt-in. If an application doesn't change anything, it won't get access to a 0-RTT enabled stream, so the same strong security guarantees continue to...
> Unfortunately, TLS session tickets are only sent from the server to the client. Yes, but we can generate and send one back in a side-channel (new stream). It shouldn't...