Initial PeerPool mockup

This commit is contained in:
Yuriy Glukhov 2018-04-19 20:28:44 +03:00
parent 732b7941c4
commit d567911c14
1 changed files with 173 additions and 0 deletions

173
ethp2p/peer_pool.nim Normal file
View File

@ -0,0 +1,173 @@
import logging, tables, asyncdispatch, times, random
import eth_keys
import discovery, rlpx, kademlia
type
PeerPool* = ref object
keyPair: KeyPair
networkId: int
minPeers: int
discovery: DiscoveryProtocol
lastLookupTime: float
connectedNodes: Table[Node, Peer]
running: bool
AsyncChainDb = ref object # TODO: This should be defined elsewhere
# class PeerPool:
# PeerPool attempts to keep connections to at least min_peers on the given network.
const
lookupInterval = 5
connectLoopSleepMs = 2000
proc newPeerPool*(chainDb: AsyncChainDb, networkId: int, keyPair: KeyPair,
discovery: DiscoveryProtocol, minPeers = 10): PeerPool =
result.new()
result.keyPair = keyPair
result.minPeers = minPeers
result.networkId = networkId
result.discovery = discovery
result.connectedNodes = initTable[Node, Peer]()
template ensureFuture(f: untyped) = asyncCheck f
proc nodesToConnect(p: PeerPool): seq[Node] {.inline.} =
p.discovery.randomNodes(p.minPeers)
# def subscribe(self, subscriber: PeerPoolSubscriber) -> None:
# self._subscribers.append(subscriber)
# for peer in self.connected_nodes.values():
# subscriber.register_peer(peer)
# def unsubscribe(self, subscriber: PeerPoolSubscriber) -> None:
# if subscriber in self._subscribers:
# self._subscribers.remove(subscriber)
proc stopAllPeers(p: PeerPool) {.async.} =
info "Stopping all peers ..."
# TODO: ...
# await asyncio.gather(
# *[peer.stop() for peer in self.connected_nodes.values()])
# async def stop(self) -> None:
# self.cancel_token.trigger()
# await self.stop_all_peers()
proc connect(p: PeerPool, remote: Node): Future[Peer] {.async.} =
## Connect to the given remote and return a Peer instance when successful.
## Returns nil if the remote is unreachable, times out or is useless.
if remote in p.connectedNodes:
debug "Skipping ", remote, "; already connected to it"
return nil
result = await rlpxConnect(p.keyPair, remote)
# expected_exceptions = (
# UnreachablePeer, TimeoutError, PeerConnectionLost, HandshakeFailure)
# try:
# self.logger.debug("Connecting to %s...", remote)
# peer = await wait_with_token(
# handshake(remote, self.privkey, self.peer_class, self.chaindb, self.network_id),
# token=self.cancel_token,
# timeout=HANDSHAKE_TIMEOUT)
# return peer
# except OperationCancelled:
# # Pass it on to instruct our main loop to stop.
# raise
# except expected_exceptions as e:
# self.logger.debug("Could not complete handshake with %s: %s", remote, repr(e))
# except Exception:
# self.logger.exception("Unexpected error during auth/p2p handshake with %s", remote)
# return None
proc lookupRandomNode(p: PeerPool) {.async.} =
# This method runs in the background, so we must catch OperationCancelled here otherwise
# asyncio will warn that its exception was never retrieved.
try:
discard await p.discovery.lookupRandom()
except: # OperationCancelled
discard
p.lastLookupTime = epochTime()
proc getRandomBootnode(p: PeerPool): seq[Node] = @[p.discovery.bootstrapNodes.rand()]
proc peerFinished(p: PeerPool, peer: Peer) =
## Remove the given peer from our list of connected nodes.
## This is passed as a callback to be called when a peer finishes.
p.connectedNodes.del(peer.remote)
proc run(p: Peer, completionHandler: proc() = nil) {.async.} =
# TODO: This is a stub that should be implemented in rlpx.nim
await sleepAsync(20000) # sleep 20 sec
if not completionHandler.isNil: completionHandler()
proc connectToNodes(p: PeerPool, nodes: seq[Node]) {.async.} =
for node in nodes:
# TODO: Consider changing connect() to raise an exception instead of returning None,
# as discussed in
# https://github.com/ethereum/py-evm/pull/139#discussion_r152067425
let peer = await p.connect(node)
if not peer.isNil:
info "Successfully connected to ", peer
ensureFuture peer.run() do():
p.peerFinished(peer)
p.connectedNodes[peer.remote] = peer
# for subscriber in self._subscribers:
# subscriber.register_peer(peer)
if p.connectedNodes.len >= p.minPeers:
return
proc maybeConnectToMorePeers(p: PeerPool) {.async.} =
## Connect to more peers if we're not yet connected to at least self.minPeers.
if p.connectedNodes.len >= p.minPeers:
debug "Already connected to enough peers: ", p.connectedNodes, "; sleeping"
return
if p.lastLookupTime + lookupInterval < epochTime():
ensureFuture p.lookupRandomNode()
await p.connectToNodes(p.nodesToConnect())
# In some cases (e.g ROPSTEN or private testnets), the discovery table might be full of
# bad peers so if we can't connect to any peers we try a random bootstrap node as well.
if p.connectedNodes.len == 0:
await p.connectToNodes(p.getRandomBootnode())
proc run(p: PeerPool) {.async.} =
info "Running PeerPool..."
p.running = true
while p.running:
var dropConnections = false
try:
await p.maybeConnectToMorePeers()
except:
# Most unexpected errors should be transient, so we log and restart from scratch.
error "Unexpected error, restarting"
dropConnections = true
if dropConnections:
await p.stopAllPeers()
await sleepAsync(connectLoopSleepMs)
proc start*(p: PeerPool) =
if not p.running:
asyncCheck p.run()
# @property
# def peers(self) -> List[BasePeer]:
# peers = list(self.connected_nodes.values())
# # Shuffle the list of peers so that dumb callsites are less likely to send all requests to
# # a single peer even if they always pick the first one from the list.
# random.shuffle(peers)
# return peers
# async def get_random_peer(self) -> BasePeer:
# while not self.peers:
# self.logger.debug("No connected peers, sleeping a bit")
# await asyncio.sleep(0.5)
# return random.choice(self.peers)