Darius Clark
Darius Clark
> An alternative would be to entirely remove auto-closing and instead emit an event notifying the user that the connection is idle. Would that perhaps be useful? That actually doesnt...
> 1, the connection that being openned (via get_closest_peer or put/get record ops or by gossipsub) remains alive and not get `auto closed` , even with `idel_connection_timeout` got setup via...
You can disable that in your config
@manucorporat any documents on that?
I honestly thought tokio-redis was more of a PoC on how tokio works
> I manually hooked it up to the swarm :) Glad to see this implemented in the protocol. However there's the question whether the peer wants to participate in the...
As I think about it, we could still add the peer to the routing table since in client mode, they wont be able to participate in the DHT until it...
> IIUC the goal here is to update the address of a peer that is already a `kad` routing table, not to add new peers to it. Possible, though based...
My question would be, is this actually needed? Not against adding `RSA_PKCS1_1024_8192_SHA256_FOR_LEGACY_USE_ONLY` though but at the same time Im not sure if it really fits to validate weaker keys. Further...
If this is to be automatic, I wonder if we should check the external addresses for any public addresses that isnt a relay circuit (either reported by upnp, autonat or...