NetworkBridge: validator (authorities) discovery api (#1699)

* stupid, but it compiles * redo * cleanup * add ValidatorDiscovery to msgs * sketch network bridge code * ConnectToAuthorities instead of validators * more stuff * cleanup * more stuff * complete ConnectToAuthoritiesState * Update node/network/bridge/src/lib.rs Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com> * Collator protocol subsystem (#1659) * WIP * The initial implementation of the collator side. * Improve comments * Multiple collation requests * Add more tests and comments to validator side * Add comments, remove dead code * Apply suggestions from code review Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com> * Fix build after suggested changes * Also connect to the next validator group * Remove a Future impl and move TimeoutExt to util * Minor nits * Fix build * Change FetchCollations back to FetchCollation * Try this * Final fixes * Fix build Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com> * handle multiple in-flight connection requests * handle cancelled requests * Update node/core/runtime-api/src/lib.rs Co-authored-by: Bernhard Schuster <bernhard@ahoi.io> * redo it again * more stuff * redo it again * update comments * workaround Future is not Send * fix trailing spaces * clarify comments * bridge: fix compilation in tests * update more comments * small fixes * port collator protocol to new validator discovery api * collator tests compile * collator tests pass * do not revoke a request when the stream receiver is closed * make revoking opt-in * fix is_fulfilled * handle request revokation in collator * tests * wait for validator connections asyncronously * fix compilation * relabel my todos * apply Fedor's patch * resolve reconnection TODO * resolve revoking TODO * resolve channel capacity TODO * resolve peer cloning TODO * resolve peer disconnected TODO * resolve PeerSet TODO * wip tests * more tests * resolve Arc TODO * rename pending to non_revoked * one more test * extract utility function into util crate * fix compilation in tests * Apply suggestions from code review Co-authored-by: Fedor Sakharov <fedor.sakharov@gmail.com> * revert pin_project removal * fix while let loop * Revert "revert pin_project removal" This reverts commit ae7f529d8de982ef66c3007dd1ff74c6ddce80d2. * fix compilation * Update node/subsystem/src/messages.rs * docs on pub items * guide updates * remove a TODO * small guide update * fix a typo * link to the issue * validator discovery: on_request docs Co-authored-by: Peter Goodspeed-Niklaus <coriolinus@users.noreply.github.com> Co-authored-by: Fedor Sakharov <fedor.sakharov@gmail.com> Co-authored-by: Bernhard Schuster <bernhard@ahoi.io>
2026-06-13 18:41:05 +00:00 · 2020-10-06 13:34:57 +02:00
parent 96f6b5ae2d
commit ca89e3edbe
20 changed files with 1102 additions and 124 deletions
@@ -57,6 +57,8 @@ use std::{
 };
 use streamunordered::{StreamUnordered, StreamYield};

+pub mod validator_discovery;
+
 /// These reexports are required so that external crates can use the `delegated_subsystem` macro properly.
 pub mod reexports {
 	pub use sp_core::traits::SpawnNamed;
@@ -0,0 +1,157 @@
+// Copyright 2020 Parity Technologies (UK) Ltd.
+// This file is part of Polkadot.
+
+// Polkadot is free software: you can redistribute it and/or modify
+// it under the terms of the GNU General Public License as published by
+// the Free Software Foundation, either version 3 of the License, or
+// (at your option) any later version.
+
+// Polkadot is distributed in the hope that it will be useful,
+// but WITHOUT ANY WARRANTY; without even the implied warranty of
+// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+// GNU General Public License for more details.
+
+// You should have received a copy of the GNU General Public License
+// along with Polkadot.  If not, see <http://www.gnu.org/licenses/>.
+
+//! Utility function to make it easier to connect to validators.
+
+use std::collections::HashMap;
+use std::pin::Pin;
+
+use futures::{
+	channel::{mpsc, oneshot},
+	task::{Poll, self},
+	stream,
+};
+
+use polkadot_node_subsystem::{
+	errors::RuntimeApiError, SubsystemError,
+	messages::{AllMessages, RuntimeApiMessage, RuntimeApiRequest, NetworkBridgeMessage},
+	SubsystemContext,
+};
+use polkadot_primitives::v1::{Hash, ValidatorId, AuthorityDiscoveryId};
+use sc_network::PeerId;
+
+/// Error when making a request to connect to validators.
+#[derive(Debug, derive_more::From)]
+pub enum Error {
+	/// Attempted to send or receive on a oneshot channel which had been canceled
+	#[from]
+	Oneshot(oneshot::Canceled),
+	/// A subsystem error.
+	#[from]
+	Subsystem(SubsystemError),
+	/// An error in the Runtime API.
+	#[from]
+	RuntimeApi(RuntimeApiError),
+}
+
+/// Utility function to make it easier to connect to validators.
+pub async fn connect_to_validators<Context: SubsystemContext>(
+	ctx: &mut Context,
+	relay_parent: Hash,
+	validators: Vec<ValidatorId>,
+) -> Result<ConnectionRequest, Error> {
+	// ValidatorId -> AuthorityDiscoveryId
+	let (tx, rx) = oneshot::channel();
+
+	ctx.send_message(AllMessages::RuntimeApi(
+		RuntimeApiMessage::Request(
+			relay_parent,
+			RuntimeApiRequest::ValidatorDiscovery(validators.clone(), tx),
+		)
+	)).await?;
+
+	let maybe_authorities = rx.await??;
+	let authorities: Vec<_> = maybe_authorities.iter()
+		.cloned()
+		.filter_map(|id| id)
+		.collect();
+
+	let validator_map = validators.into_iter()
+		.zip(maybe_authorities.into_iter())
+		.filter_map(|(k, v)| v.map(|v| (v, k)))
+		.collect::<HashMap<AuthorityDiscoveryId, ValidatorId>>();
+
+	let (connections, revoke) = connect_to_authorities(ctx, authorities).await?;
+
+	Ok(ConnectionRequest {
+		validator_map,
+		connections,
+		revoke,
+	})
+}
+
+async fn connect_to_authorities<Context: SubsystemContext>(
+	ctx: &mut Context,
+	validator_ids: Vec<AuthorityDiscoveryId>,
+) -> Result<(mpsc::Receiver<(AuthorityDiscoveryId, PeerId)>, oneshot::Sender<()>), Error> {
+	const PEERS_CAPACITY: usize = 8;
+
+	let (revoke_tx, revoke) = oneshot::channel();
+	let (connected, connected_rx) = mpsc::channel(PEERS_CAPACITY);
+
+	ctx.send_message(AllMessages::NetworkBridge(
+		NetworkBridgeMessage::ConnectToValidators {
+			validator_ids,
+			connected,
+			revoke,
+		}
+	)).await?;
+
+	Ok((connected_rx, revoke_tx))
+}
+
+/// A pending connection request to validators.
+/// This struct implements `Stream` to allow for asynchronous
+/// discovery of validator addresses.
+///
+/// NOTE: you should call `revoke` on this struct
+/// when you're no longer interested in the requested validators.
+#[must_use = "dropping a request will result in its immediate revokation"]
+pub struct ConnectionRequest {
+	validator_map: HashMap<AuthorityDiscoveryId, ValidatorId>,
+	#[must_use = "streams do nothing unless polled"]
+	connections: mpsc::Receiver<(AuthorityDiscoveryId, PeerId)>,
+	#[must_use = "a request should be revoked at some point"]
+	revoke: oneshot::Sender<()>,
+}
+
+impl stream::Stream for ConnectionRequest {
+	type Item = (ValidatorId, PeerId);
+
+	fn poll_next(mut self: Pin<&mut Self>, cx: &mut task::Context) -> Poll<Option<Self::Item>> {
+		if self.validator_map.is_empty() {
+			return Poll::Ready(None);
+		}
+		match Pin::new(&mut self.connections).poll_next(cx) {
+			Poll::Ready(Some((id, peer_id))) => {
+				if let Some(validator_id) = self.validator_map.remove(&id) {
+					return Poll::Ready(Some((validator_id, peer_id)));
+				} else {
+					// unknown authority_id
+					// should be unreachable
+				}
+			}
+			_ => {},
+		}
+		Poll::Pending
+	}
+}
+
+impl ConnectionRequest {
+	/// By revoking the request the caller allows the network to
+	/// free some peer slots thus freeing the resources.
+	/// It doesn't necessarily lead to peers disconnection though.
+	/// The revokation is enacted on in the next connection request.
+	///
+	/// This can be done either by calling this function or dropping the request.
+	pub fn revoke(self) {
+		if let Err(_) = self.revoke.send(()) {
+			log::warn!(
+				"Failed to revoke a validator connection request",
+			);
+		}
+	}
+}