I’m questioning how nodes uncover different nodes
Nodes study different nodes by means of:
- DNS seeds (which run crawlers that preserve databases of community nodes)
- From their friends (nodes gossip IP addresses of different nodes to one another)
- From their very own database of nodes (which is fed by the above two mechanisms)
- From a hardcoded checklist built-in to the shopper, as a last fall again.
and miners
Miners are simply nodes like all others; there’s nothing particular about them on this regard.
principally, who to broadcast obtained transactions and blocks to.
To make clear:
- Nodes make connections to different nodes; that is what must be found. Each node makes a couple of random connections to different nodes, to not everybody within the community.
- Nodes broadcast transactions and blocks (and IP addresses of different nodes) to those friends, to not everybody.
Finally everybody will hear about each block, and most transactions, as a result of everyone seems to be linked to 1 one other.
I am assuming that when a brand new node or miner needs to be found, it broadcasts to all identified nodes informing them that its a brand new node or miner.
It simply rumors its IP tackle to its friends. These friends will put the node’s IP tackle of their peer database, and gossip it additional to their friends. Over time, increasingly nodes will study it.
Then the node places the node or miner on an inventory of IP addresses, to broadcast transactions and blocks accordingly.
Nodes solely ever talk with their very own friends (sometimes 8-10 outgoing connections, up to a couple 100 incoming ones), however they’ll preserve databases of IP addresses with 10s of hundreds of nodes. Most of these won’t ever be linked with, however could also be rumoured to others.