How do peer-to-peer networks exchange data?

Question

P2P

Good time, dear Toaster.
I understand theoretically how peer-to-peer decentralized networks work. One rank and all. The data is not stored on a single server, but is transferred from client to client.
Here's what I don't understand:
Let's say there are 4 people. For the convenience of A, B, C and D.
Client A wants to send a message to client D, the central server, which would receive the message, look at the addressee and send it to him. Theoretically, a message can go from A, to B, then to C, and finally to D, but how? How are messages transmitted in p2p networks? Where exactly is the message sent as a result and where does this address come from? How does a p2p program find out client B's address, how does client B get C's address, and so on?
I understand that the question is very basic, but in the abundance and hype of blockchain and other technologies, it is now difficult to pull out the technical details. And I just can’t imagine how messages can be transmitted in an anonymous decentralized network without sorting through ip addresses, and how, as a result, a relatively short let to the addressee is generally built, if, in theory, the recipient’s ip should not know individual points, otherwise where is anonymity.
Tell me where to look and how it generally happens in practice, I will be very grateful,
Regards.

Answer 1

something about bitcoin could be found, or about a torrent
, all nodes transmit ip and info to each other (synchronize) through the very first, the very first (there are several) - sewn into the client

Answer 2

People mixed up in a bunch, horses: "p2p", "blockchain", "anonymity" ...
Firstly, the blockchain has a very mediocre relation to p2p.
Secondly, with anonymity, any pure p2p will quickly be polluted by corporations and copyists, as happened with eMule and other p2p "pioneers". There must be some trust in at least some nodes. For torrents, the load of trust is carried by trackers, for Bitcoin there are several supernodes registered in the client, etc.
And thirdly, in p2p, Internet nodes will never find each other if they do not know each other's IP, or do not know through whom to look for each other - i.e. do not have at least some entry points.
For example, several dozen supernodes are registered in Tox clients. Clients at startup connect to supernodes and through them they view other nodes and "enter" the DHT. Further, requests are sent to these nodes, and responses are received from them.
As for traffic routing and data integrity, you can read another article of mine .