go-ethereum/p2p/server.go
Felix Lange 6fb810adaa p2p: throttle all discovery lookups
Lookup calls would spin out of control when network connectivity was
lost. The throttling that was in place only took effect when the table
returned zero results, which doesn't happen very often.

The new throttling should not have a negative impact when the host is
online. Lookups against the network take some time and dials for all
results must complete or hit the cache before a new one is started. This
usually takes longer than four seconds, leaving online lookups
unaffected.

Fixes #1296
2015-06-22 01:07:58 +02:00

654 lines
18 KiB
Go

package p2p
import (
"crypto/ecdsa"
"errors"
"fmt"
"net"
"sync"
"time"
"github.com/ethereum/go-ethereum/logger"
"github.com/ethereum/go-ethereum/logger/glog"
"github.com/ethereum/go-ethereum/p2p/discover"
"github.com/ethereum/go-ethereum/p2p/nat"
)
const (
defaultDialTimeout = 15 * time.Second
refreshPeersInterval = 30 * time.Second
staticPeerCheckInterval = 15 * time.Second
// Maximum number of concurrently handshaking inbound connections.
maxAcceptConns = 50
// Maximum number of concurrently dialing outbound connections.
maxActiveDialTasks = 16
// Maximum time allowed for reading a complete message.
// This is effectively the amount of time a connection can be idle.
frameReadTimeout = 30 * time.Second
// Maximum amount of time allowed for writing a complete message.
frameWriteTimeout = 20 * time.Second
)
var errServerStopped = errors.New("server stopped")
var srvjslog = logger.NewJsonLogger()
// Server manages all peer connections.
//
// The fields of Server are used as configuration parameters.
// You should set them before starting the Server. Fields may not be
// modified while the server is running.
type Server struct {
// This field must be set to a valid secp256k1 private key.
PrivateKey *ecdsa.PrivateKey
// MaxPeers is the maximum number of peers that can be
// connected. It must be greater than zero.
MaxPeers int
// MaxPendingPeers is the maximum number of peers that can be pending in the
// handshake phase, counted separately for inbound and outbound connections.
// Zero defaults to preset values.
MaxPendingPeers int
// Discovery specifies whether the peer discovery mechanism should be started
// or not. Disabling is usually useful for protocol debugging (manual topology).
Discovery bool
// Name sets the node name of this server.
// Use common.MakeName to create a name that follows existing conventions.
Name string
// Bootstrap nodes are used to establish connectivity
// with the rest of the network.
BootstrapNodes []*discover.Node
// Static nodes are used as pre-configured connections which are always
// maintained and re-connected on disconnects.
StaticNodes []*discover.Node
// Trusted nodes are used as pre-configured connections which are always
// allowed to connect, even above the peer limit.
TrustedNodes []*discover.Node
// NodeDatabase is the path to the database containing the previously seen
// live nodes in the network.
NodeDatabase string
// Protocols should contain the protocols supported
// by the server. Matching protocols are launched for
// each peer.
Protocols []Protocol
// If ListenAddr is set to a non-nil address, the server
// will listen for incoming connections.
//
// If the port is zero, the operating system will pick a port. The
// ListenAddr field will be updated with the actual address when
// the server is started.
ListenAddr string
// If set to a non-nil value, the given NAT port mapper
// is used to make the listening port available to the
// Internet.
NAT nat.Interface
// If Dialer is set to a non-nil value, the given Dialer
// is used to dial outbound peer connections.
Dialer *net.Dialer
// If NoDial is true, the server will not dial any peers.
NoDial bool
// Hooks for testing. These are useful because we can inhibit
// the whole protocol stack.
newTransport func(net.Conn) transport
newPeerHook func(*Peer)
lock sync.Mutex // protects running
running bool
ntab discoverTable
listener net.Listener
ourHandshake *protoHandshake
lastLookup time.Time
// These are for Peers, PeerCount (and nothing else).
peerOp chan peerOpFunc
peerOpDone chan struct{}
quit chan struct{}
addstatic chan *discover.Node
posthandshake chan *conn
addpeer chan *conn
delpeer chan *Peer
loopWG sync.WaitGroup // loop, listenLoop
}
type peerOpFunc func(map[discover.NodeID]*Peer)
type connFlag int
const (
dynDialedConn connFlag = 1 << iota
staticDialedConn
inboundConn
trustedConn
)
// conn wraps a network connection with information gathered
// during the two handshakes.
type conn struct {
fd net.Conn
transport
flags connFlag
cont chan error // The run loop uses cont to signal errors to setupConn.
id discover.NodeID // valid after the encryption handshake
caps []Cap // valid after the protocol handshake
name string // valid after the protocol handshake
}
type transport interface {
// The two handshakes.
doEncHandshake(prv *ecdsa.PrivateKey, dialDest *discover.Node) (discover.NodeID, error)
doProtoHandshake(our *protoHandshake) (*protoHandshake, error)
// The MsgReadWriter can only be used after the encryption
// handshake has completed. The code uses conn.id to track this
// by setting it to a non-nil value after the encryption handshake.
MsgReadWriter
// transports must provide Close because we use MsgPipe in some of
// the tests. Closing the actual network connection doesn't do
// anything in those tests because NsgPipe doesn't use it.
close(err error)
}
func (c *conn) String() string {
s := c.flags.String() + " conn"
if (c.id != discover.NodeID{}) {
s += fmt.Sprintf(" %x", c.id[:8])
}
s += " " + c.fd.RemoteAddr().String()
return s
}
func (f connFlag) String() string {
s := ""
if f&trustedConn != 0 {
s += " trusted"
}
if f&dynDialedConn != 0 {
s += " dyn dial"
}
if f&staticDialedConn != 0 {
s += " static dial"
}
if f&inboundConn != 0 {
s += " inbound"
}
if s != "" {
s = s[1:]
}
return s
}
func (c *conn) is(f connFlag) bool {
return c.flags&f != 0
}
// Peers returns all connected peers.
func (srv *Server) Peers() []*Peer {
var ps []*Peer
select {
// Note: We'd love to put this function into a variable but
// that seems to cause a weird compiler error in some
// environments.
case srv.peerOp <- func(peers map[discover.NodeID]*Peer) {
for _, p := range peers {
ps = append(ps, p)
}
}:
<-srv.peerOpDone
case <-srv.quit:
}
return ps
}
// PeerCount returns the number of connected peers.
func (srv *Server) PeerCount() int {
var count int
select {
case srv.peerOp <- func(ps map[discover.NodeID]*Peer) { count = len(ps) }:
<-srv.peerOpDone
case <-srv.quit:
}
return count
}
// AddPeer connects to the given node and maintains the connection until the
// server is shut down. If the connection fails for any reason, the server will
// attempt to reconnect the peer.
func (srv *Server) AddPeer(node *discover.Node) {
select {
case srv.addstatic <- node:
case <-srv.quit:
}
}
// Self returns the local node's endpoint information.
func (srv *Server) Self() *discover.Node {
srv.lock.Lock()
defer srv.lock.Unlock()
// If the server's not running, return an empty node
if !srv.running {
return &discover.Node{IP: net.ParseIP("0.0.0.0")}
}
// If the node is running but discovery is off, manually assemble the node infos
if srv.ntab == nil {
// Inbound connections disabled, use zero address
if srv.listener == nil {
return &discover.Node{IP: net.ParseIP("0.0.0.0"), ID: discover.PubkeyID(&srv.PrivateKey.PublicKey)}
}
// Otherwise inject the listener address too
addr := srv.listener.Addr().(*net.TCPAddr)
return &discover.Node{
ID: discover.PubkeyID(&srv.PrivateKey.PublicKey),
IP: addr.IP,
TCP: uint16(addr.Port),
}
}
// Otherwise return the live node infos
return srv.ntab.Self()
}
// Stop terminates the server and all active peer connections.
// It blocks until all active connections have been closed.
func (srv *Server) Stop() {
srv.lock.Lock()
defer srv.lock.Unlock()
if !srv.running {
return
}
srv.running = false
if srv.listener != nil {
// this unblocks listener Accept
srv.listener.Close()
}
close(srv.quit)
srv.loopWG.Wait()
}
// Start starts running the server.
// Servers can not be re-used after stopping.
func (srv *Server) Start() (err error) {
srv.lock.Lock()
defer srv.lock.Unlock()
if srv.running {
return errors.New("server already running")
}
srv.running = true
glog.V(logger.Info).Infoln("Starting Server")
// static fields
if srv.PrivateKey == nil {
return fmt.Errorf("Server.PrivateKey must be set to a non-nil key")
}
if srv.newTransport == nil {
srv.newTransport = newRLPX
}
if srv.Dialer == nil {
srv.Dialer = &net.Dialer{Timeout: defaultDialTimeout}
}
srv.quit = make(chan struct{})
srv.addpeer = make(chan *conn)
srv.delpeer = make(chan *Peer)
srv.posthandshake = make(chan *conn)
srv.addstatic = make(chan *discover.Node)
srv.peerOp = make(chan peerOpFunc)
srv.peerOpDone = make(chan struct{})
// node table
if srv.Discovery {
ntab, err := discover.ListenUDP(srv.PrivateKey, srv.ListenAddr, srv.NAT, srv.NodeDatabase)
if err != nil {
return err
}
srv.ntab = ntab
}
dynPeers := srv.MaxPeers / 2
if !srv.Discovery {
dynPeers = 0
}
dialer := newDialState(srv.StaticNodes, srv.ntab, dynPeers)
// handshake
srv.ourHandshake = &protoHandshake{Version: baseProtocolVersion, Name: srv.Name, ID: discover.PubkeyID(&srv.PrivateKey.PublicKey)}
for _, p := range srv.Protocols {
srv.ourHandshake.Caps = append(srv.ourHandshake.Caps, p.cap())
}
// listen/dial
if srv.ListenAddr != "" {
if err := srv.startListening(); err != nil {
return err
}
}
if srv.NoDial && srv.ListenAddr == "" {
glog.V(logger.Warn).Infoln("I will be kind-of useless, neither dialing nor listening.")
}
srv.loopWG.Add(1)
go srv.run(dialer)
srv.running = true
return nil
}
func (srv *Server) startListening() error {
// Launch the TCP listener.
listener, err := net.Listen("tcp", srv.ListenAddr)
if err != nil {
return err
}
laddr := listener.Addr().(*net.TCPAddr)
srv.ListenAddr = laddr.String()
srv.listener = listener
srv.loopWG.Add(1)
go srv.listenLoop()
// Map the TCP listening port if NAT is configured.
if !laddr.IP.IsLoopback() && srv.NAT != nil {
srv.loopWG.Add(1)
go func() {
nat.Map(srv.NAT, srv.quit, "tcp", laddr.Port, laddr.Port, "ethereum p2p")
srv.loopWG.Done()
}()
}
return nil
}
type dialer interface {
newTasks(running int, peers map[discover.NodeID]*Peer, now time.Time) []task
taskDone(task, time.Time)
addStatic(*discover.Node)
}
func (srv *Server) run(dialstate dialer) {
defer srv.loopWG.Done()
var (
peers = make(map[discover.NodeID]*Peer)
trusted = make(map[discover.NodeID]bool, len(srv.TrustedNodes))
tasks []task
pendingTasks []task
taskdone = make(chan task, maxActiveDialTasks)
)
// Put trusted nodes into a map to speed up checks.
// Trusted peers are loaded on startup and cannot be
// modified while the server is running.
for _, n := range srv.TrustedNodes {
trusted[n.ID] = true
}
// Some task list helpers.
delTask := func(t task) {
for i := range tasks {
if tasks[i] == t {
tasks = append(tasks[:i], tasks[i+1:]...)
break
}
}
}
scheduleTasks := func(new []task) {
pt := append(pendingTasks, new...)
start := maxActiveDialTasks - len(tasks)
if len(pt) < start {
start = len(pt)
}
if start > 0 {
tasks = append(tasks, pt[:start]...)
for _, t := range pt[:start] {
t := t
glog.V(logger.Detail).Infoln("new task:", t)
go func() { t.Do(srv); taskdone <- t }()
}
copy(pt, pt[start:])
pendingTasks = pt[:len(pt)-start]
}
}
running:
for {
// Query the dialer for new tasks and launch them.
now := time.Now()
nt := dialstate.newTasks(len(pendingTasks)+len(tasks), peers, now)
scheduleTasks(nt)
select {
case <-srv.quit:
// The server was stopped. Run the cleanup logic.
glog.V(logger.Detail).Infoln("<-quit: spinning down")
break running
case n := <-srv.addstatic:
// This channel is used by AddPeer to add to the
// ephemeral static peer list. Add it to the dialer,
// it will keep the node connected.
glog.V(logger.Detail).Infoln("<-addstatic:", n)
dialstate.addStatic(n)
case op := <-srv.peerOp:
// This channel is used by Peers and PeerCount.
op(peers)
srv.peerOpDone <- struct{}{}
case t := <-taskdone:
// A task got done. Tell dialstate about it so it
// can update its state and remove it from the active
// tasks list.
glog.V(logger.Detail).Infoln("<-taskdone:", t)
dialstate.taskDone(t, now)
delTask(t)
case c := <-srv.posthandshake:
// A connection has passed the encryption handshake so
// the remote identity is known (but hasn't been verified yet).
if trusted[c.id] {
// Ensure that the trusted flag is set before checking against MaxPeers.
c.flags |= trustedConn
}
glog.V(logger.Detail).Infoln("<-posthandshake:", c)
// TODO: track in-progress inbound node IDs (pre-Peer) to avoid dialing them.
c.cont <- srv.encHandshakeChecks(peers, c)
case c := <-srv.addpeer:
// At this point the connection is past the protocol handshake.
// Its capabilities are known and the remote identity is verified.
glog.V(logger.Detail).Infoln("<-addpeer:", c)
err := srv.protoHandshakeChecks(peers, c)
if err != nil {
glog.V(logger.Detail).Infof("Not adding %v as peer: %v", c, err)
} else {
// The handshakes are done and it passed all checks.
p := newPeer(c, srv.Protocols)
peers[c.id] = p
go srv.runPeer(p)
}
// The dialer logic relies on the assumption that
// dial tasks complete after the peer has been added or
// discarded. Unblock the task last.
c.cont <- err
case p := <-srv.delpeer:
// A peer disconnected.
glog.V(logger.Detail).Infoln("<-delpeer:", p)
delete(peers, p.ID())
}
}
// Terminate discovery. If there is a running lookup it will terminate soon.
if srv.ntab != nil {
srv.ntab.Close()
}
// Disconnect all peers.
for _, p := range peers {
p.Disconnect(DiscQuitting)
}
// Wait for peers to shut down. Pending connections and tasks are
// not handled here and will terminate soon-ish because srv.quit
// is closed.
glog.V(logger.Detail).Infof("ignoring %d pending tasks at spindown", len(tasks))
for len(peers) > 0 {
p := <-srv.delpeer
glog.V(logger.Detail).Infoln("<-delpeer (spindown):", p)
delete(peers, p.ID())
}
}
func (srv *Server) protoHandshakeChecks(peers map[discover.NodeID]*Peer, c *conn) error {
// Drop connections with no matching protocols.
if len(srv.Protocols) > 0 && countMatchingProtocols(srv.Protocols, c.caps) == 0 {
return DiscUselessPeer
}
// Repeat the encryption handshake checks because the
// peer set might have changed between the handshakes.
return srv.encHandshakeChecks(peers, c)
}
func (srv *Server) encHandshakeChecks(peers map[discover.NodeID]*Peer, c *conn) error {
switch {
case !c.is(trustedConn|staticDialedConn) && len(peers) >= srv.MaxPeers:
return DiscTooManyPeers
case peers[c.id] != nil:
return DiscAlreadyConnected
case c.id == srv.Self().ID:
return DiscSelf
default:
return nil
}
}
// listenLoop runs in its own goroutine and accepts
// inbound connections.
func (srv *Server) listenLoop() {
defer srv.loopWG.Done()
glog.V(logger.Info).Infoln("Listening on", srv.listener.Addr())
// This channel acts as a semaphore limiting
// active inbound connections that are lingering pre-handshake.
// If all slots are taken, no further connections are accepted.
tokens := maxAcceptConns
if srv.MaxPendingPeers > 0 {
tokens = srv.MaxPendingPeers
}
slots := make(chan struct{}, tokens)
for i := 0; i < tokens; i++ {
slots <- struct{}{}
}
for {
<-slots
fd, err := srv.listener.Accept()
if err != nil {
return
}
glog.V(logger.Debug).Infof("Accepted conn %v\n", fd.RemoteAddr())
go func() {
srv.setupConn(fd, inboundConn, nil)
slots <- struct{}{}
}()
}
}
// setupConn runs the handshakes and attempts to add the connection
// as a peer. It returns when the connection has been added as a peer
// or the handshakes have failed.
func (srv *Server) setupConn(fd net.Conn, flags connFlag, dialDest *discover.Node) {
// Prevent leftover pending conns from entering the handshake.
srv.lock.Lock()
running := srv.running
srv.lock.Unlock()
c := &conn{fd: fd, transport: srv.newTransport(fd), flags: flags, cont: make(chan error)}
if !running {
c.close(errServerStopped)
return
}
// Run the encryption handshake.
var err error
if c.id, err = c.doEncHandshake(srv.PrivateKey, dialDest); err != nil {
glog.V(logger.Debug).Infof("%v faild enc handshake: %v", c, err)
c.close(err)
return
}
// For dialed connections, check that the remote public key matches.
if dialDest != nil && c.id != dialDest.ID {
c.close(DiscUnexpectedIdentity)
glog.V(logger.Debug).Infof("%v dialed identity mismatch, want %x", c, dialDest.ID[:8])
return
}
if err := srv.checkpoint(c, srv.posthandshake); err != nil {
glog.V(logger.Debug).Infof("%v failed checkpoint posthandshake: %v", c, err)
c.close(err)
return
}
// Run the protocol handshake
phs, err := c.doProtoHandshake(srv.ourHandshake)
if err != nil {
glog.V(logger.Debug).Infof("%v failed proto handshake: %v", c, err)
c.close(err)
return
}
if phs.ID != c.id {
glog.V(logger.Debug).Infof("%v wrong proto handshake identity: %x", c, phs.ID[:8])
c.close(DiscUnexpectedIdentity)
return
}
c.caps, c.name = phs.Caps, phs.Name
if err := srv.checkpoint(c, srv.addpeer); err != nil {
glog.V(logger.Debug).Infof("%v failed checkpoint addpeer: %v", c, err)
c.close(err)
return
}
// If the checks completed successfully, runPeer has now been
// launched by run.
}
// checkpoint sends the conn to run, which performs the
// post-handshake checks for the stage (posthandshake, addpeer).
func (srv *Server) checkpoint(c *conn, stage chan<- *conn) error {
select {
case stage <- c:
case <-srv.quit:
return errServerStopped
}
select {
case err := <-c.cont:
return err
case <-srv.quit:
return errServerStopped
}
}
// runPeer runs in its own goroutine for each peer.
// it waits until the Peer logic returns and removes
// the peer.
func (srv *Server) runPeer(p *Peer) {
glog.V(logger.Debug).Infof("Added %v\n", p)
srvjslog.LogJson(&logger.P2PConnected{
RemoteId: p.ID().String(),
RemoteAddress: p.RemoteAddr().String(),
RemoteVersionString: p.Name(),
NumConnections: srv.PeerCount(),
})
if srv.newPeerHook != nil {
srv.newPeerHook(p)
}
discreason := p.run()
// Note: run waits for existing peers to be sent on srv.delpeer
// before returning, so this send should not select on srv.quit.
srv.delpeer <- p
glog.V(logger.Debug).Infof("Removed %v (%v)\n", p, discreason)
srvjslog.LogJson(&logger.P2PDisconnected{
RemoteId: p.ID().String(),
NumConnections: srv.PeerCount(),
})
}