aboutsummaryrefslogtreecommitdiffstats
path: root/p2p
diff options
context:
space:
mode:
authorFelix Lange <fjl@twurst.com>2017-02-24 16:58:04 +0800
committerFelix Lange <fjl@twurst.com>2017-02-28 17:20:29 +0800
commit96ae35e2ac8c360781407d7294081aabdcbb3652 (patch)
tree378cc7aae99127a4e85a721e5cd319ced87e103a /p2p
parent35e8308bf76df11c3c1e8d11fac782814583ea5c (diff)
downloadgo-tangerine-96ae35e2ac8c360781407d7294081aabdcbb3652.tar
go-tangerine-96ae35e2ac8c360781407d7294081aabdcbb3652.tar.gz
go-tangerine-96ae35e2ac8c360781407d7294081aabdcbb3652.tar.bz2
go-tangerine-96ae35e2ac8c360781407d7294081aabdcbb3652.tar.lz
go-tangerine-96ae35e2ac8c360781407d7294081aabdcbb3652.tar.xz
go-tangerine-96ae35e2ac8c360781407d7294081aabdcbb3652.tar.zst
go-tangerine-96ae35e2ac8c360781407d7294081aabdcbb3652.zip
p2p, p2p/discover, p2p/nat: rework logging using context keys
Diffstat (limited to 'p2p')
-rw-r--r--p2p/dial.go13
-rw-r--r--p2p/discover/database.go8
-rw-r--r--p2p/discover/node.go5
-rw-r--r--p2p/discover/ntp.go13
-rw-r--r--p2p/discover/table.go14
-rw-r--r--p2p/discover/udp.go46
-rw-r--r--p2p/nat/nat.go11
-rw-r--r--p2p/peer.go43
-rw-r--r--p2p/peer_error.go34
-rw-r--r--p2p/peer_test.go20
-rw-r--r--p2p/server.go115
11 files changed, 171 insertions, 151 deletions
diff --git a/p2p/dial.go b/p2p/dial.go
index 65180e029..bb3befab2 100644
--- a/p2p/dial.go
+++ b/p2p/dial.go
@@ -133,7 +133,7 @@ func (s *dialstate) newTasks(nRunning int, peers map[discover.NodeID]*Peer, now
var newtasks []task
addDial := func(flag connFlag, n *discover.Node) bool {
if err := s.checkDial(n, peers); err != nil {
- log.Debug(fmt.Sprintf("skipping dial candidate %x@%v:%d: %v", n.ID[:8], n.IP, n.TCP, err))
+ log.Trace("Skipping dial candidate", "id", n.ID, "addr", &net.TCPAddr{IP: n.IP, Port: int(n.TCP)}, "err", err)
return false
}
s.dialing[n.ID] = flag
@@ -162,7 +162,7 @@ func (s *dialstate) newTasks(nRunning int, peers map[discover.NodeID]*Peer, now
err := s.checkDial(t.dest, peers)
switch err {
case errNotWhitelisted, errSelf:
- log.Debug(fmt.Sprintf("removing static dial candidate %x@%v:%d: %v", t.dest.ID[:8], t.dest.IP, t.dest.TCP, err))
+ log.Warn("Removing static dial candidate", "id", t.dest.ID, "addr", &net.TCPAddr{IP: t.dest.IP, Port: int(t.dest.TCP)}, "err", err)
delete(s.static, t.dest.ID)
case nil:
s.dialing[id] = t.flags
@@ -266,7 +266,7 @@ func (t *dialTask) Do(srv *Server) {
// The backoff delay resets when the node is found.
func (t *dialTask) resolve(srv *Server) bool {
if srv.ntab == nil {
- log.Debug(fmt.Sprintf("can't resolve node %x: discovery is disabled", t.dest.ID[:6]))
+ log.Debug("Can't resolve node", "id", t.dest.ID, "err", "discovery is disabled")
return false
}
if t.resolveDelay == 0 {
@@ -282,23 +282,22 @@ func (t *dialTask) resolve(srv *Server) bool {
if t.resolveDelay > maxResolveDelay {
t.resolveDelay = maxResolveDelay
}
- log.Debug(fmt.Sprintf("resolving node %x failed (new delay: %v)", t.dest.ID[:6], t.resolveDelay))
+ log.Debug("Resolving node failed", "id", t.dest.ID, "newdelay", t.resolveDelay)
return false
}
// The node was found.
t.resolveDelay = initialResolveDelay
t.dest = resolved
- log.Debug(fmt.Sprintf("resolved node %x: %v:%d", t.dest.ID[:6], t.dest.IP, t.dest.TCP))
+ log.Debug("Resolved node", "id", t.dest.ID, "addr", &net.TCPAddr{IP: t.dest.IP, Port: int(t.dest.TCP)})
return true
}
// dial performs the actual connection attempt.
func (t *dialTask) dial(srv *Server, dest *discover.Node) bool {
addr := &net.TCPAddr{IP: dest.IP, Port: int(dest.TCP)}
- log.Debug(fmt.Sprintf("dial tcp %v (%x)", addr, dest.ID[:6]))
fd, err := srv.Dialer.Dial("tcp", addr.String())
if err != nil {
- log.Trace(fmt.Sprintf("%v", err))
+ log.Trace("Dial error", "task", t, "err", err)
return false
}
mfd := newMeteredConn(fd, false)
diff --git a/p2p/discover/database.go b/p2p/discover/database.go
index a8b32d31e..7206a63c6 100644
--- a/p2p/discover/database.go
+++ b/p2p/discover/database.go
@@ -23,7 +23,6 @@ import (
"bytes"
"crypto/rand"
"encoding/binary"
- "fmt"
"os"
"sync"
"time"
@@ -180,12 +179,11 @@ func (db *nodeDB) storeInt64(key []byte, n int64) error {
func (db *nodeDB) node(id NodeID) *Node {
blob, err := db.lvl.Get(makeKey(id, nodeDBDiscoverRoot), nil)
if err != nil {
- log.Trace(fmt.Sprintf("failed to retrieve node %v: %v", id, err))
return nil
}
node := new(Node)
if err := rlp.DecodeBytes(blob, node); err != nil {
- log.Warn(fmt.Sprintf("failed to decode node RLP: %v", err))
+ log.Error("Failed to decode node RLP", "err", err)
return nil
}
node.sha = crypto.Keccak256Hash(node.ID[:])
@@ -233,7 +231,7 @@ func (db *nodeDB) expirer() {
select {
case <-tick:
if err := db.expireNodes(); err != nil {
- log.Error(fmt.Sprintf("Failed to expire nodedb items: %v", err))
+ log.Error("Failed to expire nodedb items", "err", err)
}
case <-db.quit:
@@ -352,7 +350,7 @@ func nextNode(it iterator.Iterator) *Node {
}
var n Node
if err := rlp.DecodeBytes(it.Value(), &n); err != nil {
- log.Warn(fmt.Sprintf("invalid node %x: %v", id, err))
+ log.Warn("Failed to decode node RLP", "id", id, "err", err)
continue
}
return &n
diff --git a/p2p/discover/node.go b/p2p/discover/node.go
index f0262762e..6a7ab814e 100644
--- a/p2p/discover/node.go
+++ b/p2p/discover/node.go
@@ -221,6 +221,11 @@ func (n NodeID) GoString() string {
return fmt.Sprintf("discover.HexID(\"%x\")", n[:])
}
+// TerminalString returns a shortened hex string for terminal logging.
+func (n NodeID) TerminalString() string {
+ return hex.EncodeToString(n[:8])
+}
+
// HexID converts a hex string to a NodeID.
// The string may be prefixed with 0x.
func HexID(in string) (NodeID, error) {
diff --git a/p2p/discover/ntp.go b/p2p/discover/ntp.go
index df67e1c5b..1bb52399f 100644
--- a/p2p/discover/ntp.go
+++ b/p2p/discover/ntp.go
@@ -23,7 +23,6 @@ import (
"fmt"
"net"
"sort"
- "strings"
"time"
"github.com/ethereum/go-ethereum/log"
@@ -50,16 +49,10 @@ func checkClockDrift() {
return
}
if drift < -driftThreshold || drift > driftThreshold {
- warning := fmt.Sprintf("System clock seems off by %v, which can prevent network connectivity", drift)
- howtofix := fmt.Sprintf("Please enable network time synchronisation in system settings")
- separator := strings.Repeat("-", len(warning))
-
- log.Warn(fmt.Sprint(separator))
- log.Warn(fmt.Sprint(warning))
- log.Warn(fmt.Sprint(howtofix))
- log.Warn(fmt.Sprint(separator))
+ log.Warn(fmt.Sprintf("System clock seems off by %v, which can prevent network connectivity", drift))
+ log.Warn("Please enable network time synchronisation in system settings.")
} else {
- log.Debug(fmt.Sprintf("Sanity NTP check reported %v drift, all ok", drift))
+ log.Debug("NTP sanity check done", "drift", drift)
}
}
diff --git a/p2p/discover/table.go b/p2p/discover/table.go
index 03392b563..2f5a26c34 100644
--- a/p2p/discover/table.go
+++ b/p2p/discover/table.go
@@ -277,10 +277,10 @@ func (tab *Table) lookup(targetID NodeID, refreshIfEmpty bool) []*Node {
// Bump the failure counter to detect and evacuate non-bonded entries
fails := tab.db.findFails(n.ID) + 1
tab.db.updateFindFails(n.ID, fails)
- log.Trace(fmt.Sprintf("Bumping failures for %x: %d", n.ID[:8], fails))
+ log.Trace("Bumping findnode failure counter", "id", n.ID, "failcount", fails)
if fails >= maxFindnodeFailures {
- log.Trace(fmt.Sprintf("Evacuating node %x: %d findnode failures", n.ID[:8], fails))
+ log.Trace("Too many findnode failures, dropping", "id", n.ID, "failcount", fails)
tab.delete(n)
}
}
@@ -385,13 +385,11 @@ func (tab *Table) doRefresh(done chan struct{}) {
seeds = tab.bondall(append(seeds, tab.nursery...))
if len(seeds) == 0 {
- log.Debug(fmt.Sprintf("no seed nodes found"))
+ log.Debug("No discv4 seed nodes found")
}
for _, n := range seeds {
- log.Debug("", "msg", log.Lazy{Fn: func() string {
- age := time.Since(tab.db.lastPong(n.ID))
- return fmt.Sprintf("seed node (age %v): %v", age, n)
- }})
+ age := log.Lazy{Fn: func() time.Duration { return time.Since(tab.db.lastPong(n.ID)) }}
+ log.Trace("Found seed node in database", "id", n.ID, "addr", n.addr(), "age", age)
}
tab.mutex.Lock()
tab.stuff(seeds)
@@ -470,7 +468,7 @@ func (tab *Table) bond(pinged bool, id NodeID, addr *net.UDPAddr, tcpPort uint16
var result error
age := time.Since(tab.db.lastPong(id))
if node == nil || fails > 0 || age > nodeDBNodeExpiration {
- log.Trace(fmt.Sprintf("Bonding %x: known=%t, fails=%d age=%v", id[:8], node != nil, fails, age))
+ log.Trace("Starting bonding ping/pong", "id", id, "known", node != nil, "failcount", fails, "age", age)
tab.bondmu.Lock()
w := tab.bonding[id]
diff --git a/p2p/discover/udp.go b/p2p/discover/udp.go
index eafc3f394..93545e7d5 100644
--- a/p2p/discover/udp.go
+++ b/p2p/discover/udp.go
@@ -147,6 +147,7 @@ func nodeToRPC(n *Node) rpcNode {
type packet interface {
handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte) error
+ name() string
}
type conn interface {
@@ -223,7 +224,7 @@ func ListenUDP(priv *ecdsa.PrivateKey, laddr string, natm nat.Interface, nodeDBP
if err != nil {
return nil, err
}
- log.Info(fmt.Sprint("Listening,", tab.self))
+ log.Debug("UDP listener up", "self", tab.self)
return tab, nil
}
@@ -269,7 +270,7 @@ func (t *udp) close() {
func (t *udp) ping(toid NodeID, toaddr *net.UDPAddr) error {
// TODO: maybe check for ReplyTo field in callback to measure RTT
errc := t.pending(toid, pongPacket, func(interface{}) bool { return true })
- t.send(toaddr, pingPacket, ping{
+ t.send(toaddr, pingPacket, &ping{
Version: Version,
From: t.ourEndpoint,
To: makeEndpoint(toaddr, 0), // TODO: maybe use known TCP port from DB
@@ -293,14 +294,14 @@ func (t *udp) findnode(toid NodeID, toaddr *net.UDPAddr, target NodeID) ([]*Node
nreceived++
n, err := t.nodeFromRPC(toaddr, rn)
if err != nil {
- log.Trace(fmt.Sprintf("invalid neighbor node (%v) from %v: %v", rn.IP, toaddr, err))
+ log.Trace("Invalid neighbor node received", "ip", rn.IP, "addr", toaddr, "err", err)
continue
}
nodes = append(nodes, n)
}
return nreceived >= bucketSize
})
- t.send(toaddr, findnodePacket, findnode{
+ t.send(toaddr, findnodePacket, &findnode{
Target: target,
Expiration: uint64(time.Now().Add(expiration).Unix()),
})
@@ -458,15 +459,13 @@ func init() {
}
}
-func (t *udp) send(toaddr *net.UDPAddr, ptype byte, req interface{}) error {
+func (t *udp) send(toaddr *net.UDPAddr, ptype byte, req packet) error {
packet, err := encodePacket(t.priv, ptype, req)
if err != nil {
return err
}
- log.Trace(fmt.Sprintf(">>> %v %T", toaddr, req))
- if _, err = t.conn.WriteToUDP(packet, toaddr); err != nil {
- log.Trace(fmt.Sprint("UDP send failed:", err))
- }
+ _, err = t.conn.WriteToUDP(packet, toaddr)
+ log.Trace(">> "+req.name(), "addr", toaddr, "err", err)
return err
}
@@ -475,13 +474,13 @@ func encodePacket(priv *ecdsa.PrivateKey, ptype byte, req interface{}) ([]byte,
b.Write(headSpace)
b.WriteByte(ptype)
if err := rlp.Encode(b, req); err != nil {
- log.Error(fmt.Sprint("error encoding packet:", err))
+ log.Error("Can't encode discv4 packet", "err", err)
return nil, err
}
packet := b.Bytes()
sig, err := crypto.Sign(crypto.Keccak256(packet[headSize:]), priv)
if err != nil {
- log.Error(fmt.Sprint("could not sign packet:", err))
+ log.Error("Can't sign discv4 packet", "err", err)
return nil, err
}
copy(packet[macSize:], sig)
@@ -503,11 +502,11 @@ func (t *udp) readLoop() {
nbytes, from, err := t.conn.ReadFromUDP(buf)
if netutil.IsTemporaryError(err) {
// Ignore temporary read errors.
- log.Debug(fmt.Sprintf("Temporary read error: %v", err))
+ log.Debug("Temporary UDP read error", "err", err)
continue
} else if err != nil {
// Shut down the loop for permament errors.
- log.Debug(fmt.Sprintf("Read error: %v", err))
+ log.Debug("UDP read error", "err", err)
return
}
t.handlePacket(from, buf[:nbytes])
@@ -517,14 +516,11 @@ func (t *udp) readLoop() {
func (t *udp) handlePacket(from *net.UDPAddr, buf []byte) error {
packet, fromID, hash, err := decodePacket(buf)
if err != nil {
- log.Debug(fmt.Sprintf("Bad packet from %v: %v", from, err))
+ log.Debug("Bad discv4 packet", "addr", from, "err", err)
return err
}
- status := "ok"
- if err = packet.handle(t, from, fromID, hash); err != nil {
- status = err.Error()
- }
- log.Trace(fmt.Sprintf("<<< %v %T: %s", from, packet, status))
+ err = packet.handle(t, from, fromID, hash)
+ log.Trace("<< "+packet.name(), "addr", from, "err", err)
return err
}
@@ -563,7 +559,7 @@ func (req *ping) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte) er
if expired(req.Expiration) {
return errExpired
}
- t.send(from, pongPacket, pong{
+ t.send(from, pongPacket, &pong{
To: makeEndpoint(from, req.From.TCP),
ReplyTok: mac,
Expiration: uint64(time.Now().Add(expiration).Unix()),
@@ -575,6 +571,8 @@ func (req *ping) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte) er
return nil
}
+func (req *ping) name() string { return "PING/v4" }
+
func (req *pong) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte) error {
if expired(req.Expiration) {
return errExpired
@@ -585,6 +583,8 @@ func (req *pong) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte) er
return nil
}
+func (req *pong) name() string { return "PONG/v4" }
+
func (req *findnode) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte) error {
if expired(req.Expiration) {
return errExpired
@@ -613,13 +613,15 @@ func (req *findnode) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte
}
p.Nodes = append(p.Nodes, nodeToRPC(n))
if len(p.Nodes) == maxNeighbors || i == len(closest)-1 {
- t.send(from, neighborsPacket, p)
+ t.send(from, neighborsPacket, &p)
p.Nodes = p.Nodes[:0]
}
}
return nil
}
+func (req *findnode) name() string { return "FINDNODE/v4" }
+
func (req *neighbors) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byte) error {
if expired(req.Expiration) {
return errExpired
@@ -630,6 +632,8 @@ func (req *neighbors) handle(t *udp, from *net.UDPAddr, fromID NodeID, mac []byt
return nil
}
+func (req *neighbors) name() string { return "NEIGHBORS/v4" }
+
func expired(ts uint64) bool {
return time.Unix(int64(ts), 0).Before(time.Now())
}
diff --git a/p2p/nat/nat.go b/p2p/nat/nat.go
index e5883cf98..a254648c6 100644
--- a/p2p/nat/nat.go
+++ b/p2p/nat/nat.go
@@ -98,16 +98,17 @@ const (
// Map adds a port mapping on m and keeps it alive until c is closed.
// This function is typically invoked in its own goroutine.
func Map(m Interface, c chan struct{}, protocol string, extport, intport int, name string) {
+ log := log.New("proto", protocol, "extport", extport, "intport", intport, "interface", m)
refresh := time.NewTimer(mapUpdateInterval)
defer func() {
refresh.Stop()
- log.Debug(fmt.Sprintf("deleting port mapping: %s %d -> %d (%s) using %s", protocol, extport, intport, name, m))
+ log.Debug("Deleting port mapping")
m.DeleteMapping(protocol, extport, intport)
}()
if err := m.AddMapping(protocol, extport, intport, name, mapTimeout); err != nil {
- log.Debug(fmt.Sprintf("network port %s:%d could not be mapped: %v", protocol, intport, err))
+ log.Debug("Couldn't add port mapping", "err", err)
} else {
- log.Info(fmt.Sprintf("mapped network port %s:%d -> %d (%s) using %s", protocol, extport, intport, name, m))
+ log.Info("Mapped network port")
}
for {
select {
@@ -116,9 +117,9 @@ func Map(m Interface, c chan struct{}, protocol string, extport, intport int, na
return
}
case <-refresh.C:
- log.Trace(fmt.Sprintf("refresh port mapping %s:%d -> %d (%s) using %s", protocol, extport, intport, name, m))
+ log.Trace("Refreshing port mapping")
if err := m.AddMapping(protocol, extport, intport, name, mapTimeout); err != nil {
- log.Debug(fmt.Sprintf("network port %s:%d could not be mapped: %v", protocol, intport, err))
+ log.Debug("Couldn't add port mapping", "err", err)
}
refresh.Reset(mapUpdateInterval)
}
diff --git a/p2p/peer.go b/p2p/peer.go
index 5d09927a5..a9c20189a 100644
--- a/p2p/peer.go
+++ b/p2p/peer.go
@@ -17,7 +17,6 @@
package p2p
import (
- "errors"
"fmt"
"io"
"net"
@@ -25,6 +24,7 @@ import (
"sync"
"time"
+ "github.com/ethereum/go-ethereum/common/mclock"
"github.com/ethereum/go-ethereum/log"
"github.com/ethereum/go-ethereum/p2p/discover"
"github.com/ethereum/go-ethereum/rlp"
@@ -64,6 +64,8 @@ type protoHandshake struct {
type Peer struct {
rw *conn
running map[string]*protoRW
+ log log.Logger
+ created mclock.AbsTime
wg sync.WaitGroup
protoErr chan error
@@ -125,20 +127,25 @@ func newPeer(conn *conn, protocols []Protocol) *Peer {
p := &Peer{
rw: conn,
running: protomap,
+ created: mclock.Now(),
disc: make(chan DiscReason),
protoErr: make(chan error, len(protomap)+1), // protocols + pingLoop
closed: make(chan struct{}),
+ log: log.New("id", conn.id, "conn", conn.flags),
}
return p
}
-func (p *Peer) run() DiscReason {
+func (p *Peer) Log() log.Logger {
+ return p.log
+}
+
+func (p *Peer) run() (remoteRequested bool, err error) {
var (
writeStart = make(chan struct{}, 1)
writeErr = make(chan error, 1)
readErr = make(chan error, 1)
- reason DiscReason
- requested bool
+ reason DiscReason // sent to the peer
)
p.wg.Add(2)
go p.readLoop(readErr)
@@ -152,31 +159,26 @@ func (p *Peer) run() DiscReason {
loop:
for {
select {
- case err := <-writeErr:
+ case err = <-writeErr:
// A write finished. Allow the next write to start if
// there was no error.
if err != nil {
- log.Trace(fmt.Sprintf("%v: write error: %v", p, err))
reason = DiscNetworkError
break loop
}
writeStart <- struct{}{}
- case err := <-readErr:
+ case err = <-readErr:
if r, ok := err.(DiscReason); ok {
- log.Debug(fmt.Sprintf("%v: remote requested disconnect: %v", p, r))
- requested = true
+ remoteRequested = true
reason = r
} else {
- log.Trace(fmt.Sprintf("%v: read error: %v", p, err))
reason = DiscNetworkError
}
break loop
- case err := <-p.protoErr:
+ case err = <-p.protoErr:
reason = discReasonForError(err)
- log.Debug(fmt.Sprintf("%v: protocol error: %v (%v)", p, err, reason))
break loop
- case reason = <-p.disc:
- log.Debug(fmt.Sprintf("%v: locally requested disconnect: %v", p, reason))
+ case err = <-p.disc:
break loop
}
}
@@ -184,10 +186,7 @@ loop:
close(p.closed)
p.rw.close(reason)
p.wg.Wait()
- if requested {
- reason = DiscRequested
- }
- return reason
+ return remoteRequested, err
}
func (p *Peer) pingLoop() {
@@ -297,14 +296,14 @@ func (p *Peer) startProtocols(writeStart <-chan struct{}, writeErr chan<- error)
proto.closed = p.closed
proto.wstart = writeStart
proto.werr = writeErr
- log.Trace(fmt.Sprintf("%v: Starting protocol %s/%d", p, proto.Name, proto.Version))
+ p.log.Trace(fmt.Sprintf("Starting protocol %s/%d", proto.Name, proto.Version))
go func() {
err := proto.Run(p, proto)
if err == nil {
- log.Trace(fmt.Sprintf("%v: Protocol %s/%d returned", p, proto.Name, proto.Version))
- err = errors.New("protocol returned")
+ p.log.Trace(fmt.Sprintf("Protocol %s/%d returned", proto.Name, proto.Version))
+ err = errProtocolReturned
} else if err != io.EOF {
- log.Trace(fmt.Sprintf("%v: Protocol %s/%d error: %v", p, proto.Name, proto.Version, err))
+ p.log.Trace(fmt.Sprintf("Protocol %s/%d failed", proto.Name, proto.Version), "err", err)
}
p.protoErr <- err
p.wg.Done()
diff --git a/p2p/peer_error.go b/p2p/peer_error.go
index 62c7b665d..a1cddb707 100644
--- a/p2p/peer_error.go
+++ b/p2p/peer_error.go
@@ -17,6 +17,7 @@
package p2p
import (
+ "errors"
"fmt"
)
@@ -51,6 +52,8 @@ func (self *peerError) Error() string {
return self.message
}
+var errProtocolReturned = errors.New("protocol returned")
+
type DiscReason uint
const (
@@ -70,24 +73,24 @@ const (
)
var discReasonToString = [...]string{
- DiscRequested: "Disconnect requested",
- DiscNetworkError: "Network error",
- DiscProtocolError: "Breach of protocol",
- DiscUselessPeer: "Useless peer",
- DiscTooManyPeers: "Too many peers",
- DiscAlreadyConnected: "Already connected",
- DiscIncompatibleVersion: "Incompatible P2P protocol version",
- DiscInvalidIdentity: "Invalid node identity",
- DiscQuitting: "Client quitting",
- DiscUnexpectedIdentity: "Unexpected identity",
- DiscSelf: "Connected to self",
- DiscReadTimeout: "Read timeout",
- DiscSubprotocolError: "Subprotocol error",
+ DiscRequested: "disconnect requested",
+ DiscNetworkError: "network error",
+ DiscProtocolError: "breach of protocol",
+ DiscUselessPeer: "useless peer",
+ DiscTooManyPeers: "too many peers",
+ DiscAlreadyConnected: "already connected",
+ DiscIncompatibleVersion: "incompatible p2p protocol version",
+ DiscInvalidIdentity: "invalid node identity",
+ DiscQuitting: "client quitting",
+ DiscUnexpectedIdentity: "unexpected identity",
+ DiscSelf: "connected to self",
+ DiscReadTimeout: "read timeout",
+ DiscSubprotocolError: "subprotocol error",
}
func (d DiscReason) String() string {
if len(discReasonToString) < int(d) {
- return fmt.Sprintf("Unknown Reason(%d)", d)
+ return fmt.Sprintf("unknown disconnect reason %d", d)
}
return discReasonToString[d]
}
@@ -100,6 +103,9 @@ func discReasonForError(err error) DiscReason {
if reason, ok := err.(DiscReason); ok {
return reason
}
+ if err == errProtocolReturned {
+ return DiscQuitting
+ }
peerError, ok := err.(*peerError)
if ok {
switch peerError.code {
diff --git a/p2p/peer_test.go b/p2p/peer_test.go
index f44300b15..a3e1c74fd 100644
--- a/p2p/peer_test.go
+++ b/p2p/peer_test.go
@@ -43,7 +43,7 @@ var discard = Protocol{
},
}
-func testPeer(protos []Protocol) (func(), *conn, *Peer, <-chan DiscReason) {
+func testPeer(protos []Protocol) (func(), *conn, *Peer, <-chan error) {
fd1, fd2 := net.Pipe()
c1 := &conn{fd: fd1, transport: newTestTransport(randomID(), fd1)}
c2 := &conn{fd: fd2, transport: newTestTransport(randomID(), fd2)}
@@ -53,15 +53,17 @@ func testPeer(protos []Protocol) (func(), *conn, *Peer, <-chan DiscReason) {
}
peer := newPeer(c1, protos)
- errc := make(chan DiscReason, 1)
- go func() { errc <- peer.run() }()
+ errc := make(chan error, 1)
+ go func() {
+ _, err := peer.run()
+ errc <- err
+ }()
closer := func() { c2.close(errors.New("close func called")) }
return closer, c2, peer, errc
}
func TestPeerProtoReadMsg(t *testing.T) {
- done := make(chan struct{})
proto := Protocol{
Name: "a",
Length: 5,
@@ -75,7 +77,6 @@ func TestPeerProtoReadMsg(t *testing.T) {
if err := ExpectMsg(rw, 4, []uint{3}); err != nil {
t.Error(err)
}
- close(done)
return nil
},
}
@@ -88,9 +89,10 @@ func TestPeerProtoReadMsg(t *testing.T) {
Send(rw, baseProtocolLength+4, []uint{3})
select {
- case <-done:
case err := <-errc:
- t.Errorf("peer returned: %v", err)
+ if err != errProtocolReturned {
+ t.Errorf("peer returned error: %v", err)
+ }
case <-time.After(2 * time.Second):
t.Errorf("receive timeout")
}
@@ -137,8 +139,8 @@ func TestPeerDisconnect(t *testing.T) {
}
select {
case reason := <-disc:
- if reason != DiscRequested {
- t.Errorf("run returned wrong reason: got %v, want %v", reason, DiscRequested)
+ if reason != DiscQuitting {
+ t.Errorf("run returned wrong reason: got %v, want %v", reason, DiscQuitting)
}
case <-time.After(500 * time.Millisecond):
t.Error("peer did not return")
diff --git a/p2p/server.go b/p2p/server.go
index 9f1478a41..48b4e8be3 100644
--- a/p2p/server.go
+++ b/p2p/server.go
@@ -25,6 +25,8 @@ import (
"sync"
"time"
+ "github.com/ethereum/go-ethereum/common"
+ "github.com/ethereum/go-ethereum/common/mclock"
"github.com/ethereum/go-ethereum/log"
"github.com/ethereum/go-ethereum/p2p/discover"
"github.com/ethereum/go-ethereum/p2p/discv5"
@@ -162,12 +164,18 @@ type Server struct {
removestatic chan *discover.Node
posthandshake chan *conn
addpeer chan *conn
- delpeer chan *Peer
+ delpeer chan peerDrop
loopWG sync.WaitGroup // loop, listenLoop
}
type peerOpFunc func(map[discover.NodeID]*Peer)
+type peerDrop struct {
+ *Peer
+ err error
+ requested bool // true if signaled by the peer
+}
+
type connFlag int
const (
@@ -204,9 +212,9 @@ type transport interface {
}
func (c *conn) String() string {
- s := c.flags.String() + " conn"
+ s := c.flags.String()
if (c.id != discover.NodeID{}) {
- s += fmt.Sprintf(" %x", c.id[:8])
+ s += " " + c.id.String()
}
s += " " + c.fd.RemoteAddr().String()
return s
@@ -215,16 +223,16 @@ func (c *conn) String() string {
func (f connFlag) String() string {
s := ""
if f&trustedConn != 0 {
- s += " trusted"
+ s += "-trusted"
}
if f&dynDialedConn != 0 {
- s += " dyn dial"
+ s += "-dyndial"
}
if f&staticDialedConn != 0 {
- s += " static dial"
+ s += "-staticdial"
}
if f&inboundConn != 0 {
- s += " inbound"
+ s += "-inbound"
}
if s != "" {
s = s[1:]
@@ -288,26 +296,30 @@ 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 srv.makeSelf(srv.listener, srv.ntab)
+}
+
+func (srv *Server) makeSelf(listener net.Listener, ntab discoverTable) *discover.Node {
+ // If the server's not running, return an empty node.
+ // If the node is running but discovery is off, manually assemble the node infos.
+ if ntab == nil {
+ // Inbound connections disabled, use zero address.
+ if 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)
+ addr := 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()
+ // Otherwise return the discovery node.
+ return ntab.Self()
}
// Stop terminates the server and all active peer connections.
@@ -336,7 +348,7 @@ func (srv *Server) Start() (err error) {
return errors.New("server already running")
}
srv.running = true
- log.Info(fmt.Sprint("Starting Server"))
+ log.Info("Starting P2P networking")
// static fields
if srv.PrivateKey == nil {
@@ -350,7 +362,7 @@ func (srv *Server) Start() (err error) {
}
srv.quit = make(chan struct{})
srv.addpeer = make(chan *conn)
- srv.delpeer = make(chan *Peer)
+ srv.delpeer = make(chan peerDrop)
srv.posthandshake = make(chan *conn)
srv.addstatic = make(chan *discover.Node)
srv.removestatic = make(chan *discover.Node)
@@ -398,7 +410,7 @@ func (srv *Server) Start() (err error) {
}
}
if srv.NoDial && srv.ListenAddr == "" {
- log.Warn(fmt.Sprint("I will be kind-of useless, neither dialing nor listening."))
+ log.Warn("P2P server will be useless, neither dialing nor listening")
}
srv.loopWG.Add(1)
@@ -466,7 +478,7 @@ func (srv *Server) run(dialstate dialer) {
i := 0
for ; len(runningTasks) < maxActiveDialTasks && i < len(ts); i++ {
t := ts[i]
- log.Trace(fmt.Sprint("new task:", t))
+ log.Trace("New dial task", "task", t)
go func() { t.Do(srv); taskdone <- t }()
runningTasks = append(runningTasks, t)
}
@@ -489,19 +501,18 @@ running:
select {
case <-srv.quit:
// The server was stopped. Run the cleanup logic.
- log.Trace(fmt.Sprint("<-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.
- log.Trace(fmt.Sprint("<-addstatic:", n))
+ log.Debug("Adding static node", "node", n)
dialstate.addStatic(n)
case n := <-srv.removestatic:
// This channel is used by RemovePeer to send a
// disconnect request to a peer and begin the
// stop keeping the node connected
- log.Trace(fmt.Sprint("<-removestatic:", n))
+ log.Debug("Removing static node", "node", n)
dialstate.removeStatic(n)
if p, ok := peers[n.ID]; ok {
p.Disconnect(DiscRequested)
@@ -514,7 +525,7 @@ running:
// A task got done. Tell dialstate about it so it
// can update its state and remove it from the active
// tasks list.
- log.Trace(fmt.Sprint("<-taskdone:", t))
+ log.Trace("Dial task done", "task", t)
dialstate.taskDone(t, time.Now())
delTask(t)
case c := <-srv.posthandshake:
@@ -524,19 +535,17 @@ running:
// Ensure that the trusted flag is set before checking against MaxPeers.
c.flags |= trustedConn
}
- log.Trace(fmt.Sprint("<-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.
- log.Trace(fmt.Sprint("<-addpeer:", c))
err := srv.protoHandshakeChecks(peers, c)
- if err != nil {
- log.Trace(fmt.Sprintf("Not adding %v as peer: %v", c, err))
- } else {
+ if err == nil {
// The handshakes are done and it passed all checks.
p := newPeer(c, srv.Protocols)
+ name := truncateName(c.name)
+ log.Debug("Adding p2p peer", "id", c.id, "name", name, "addr", c.fd.RemoteAddr(), "peers", len(peers)+1)
peers[c.id] = p
go srv.runPeer(p)
}
@@ -544,13 +553,16 @@ running:
// dial tasks complete after the peer has been added or
// discarded. Unblock the task last.
c.cont <- err
- case p := <-srv.delpeer:
+ case pd := <-srv.delpeer:
// A peer disconnected.
- log.Trace(fmt.Sprint("<-delpeer:", p))
- delete(peers, p.ID())
+ d := common.PrettyDuration(mclock.Now() - pd.created)
+ pd.log.Debug("Removing p2p peer", "duration", d, "peers", len(peers)-1, "req", pd.requested, "err", pd.err)
+ delete(peers, pd.ID())
}
}
+ log.Trace("P2P networking is spinning down")
+
// Terminate discovery. If there is a running lookup it will terminate soon.
if srv.ntab != nil {
srv.ntab.Close()
@@ -565,10 +577,9 @@ running:
// Wait for peers to shut down. Pending connections and tasks are
// not handled here and will terminate soon-ish because srv.quit
// is closed.
- log.Trace(fmt.Sprintf("ignoring %d pending tasks at spindown", len(runningTasks)))
for len(peers) > 0 {
p := <-srv.delpeer
- log.Trace(fmt.Sprint("<-delpeer (spindown):", p))
+ p.log.Trace("<-delpeer (spindown)", "remainingTasks", len(runningTasks))
delete(peers, p.ID())
}
}
@@ -604,7 +615,7 @@ type tempError interface {
// inbound connections.
func (srv *Server) listenLoop() {
defer srv.loopWG.Done()
- log.Info(fmt.Sprint("Listening on", srv.listener.Addr()))
+ log.Info("RLPx listener up", "self", srv.makeSelf(srv.listener, srv.ntab))
// This channel acts as a semaphore limiting
// active inbound connections that are lingering pre-handshake.
@@ -629,10 +640,10 @@ func (srv *Server) listenLoop() {
for {
fd, err = srv.listener.Accept()
if tempErr, ok := err.(tempError); ok && tempErr.Temporary() {
- log.Debug(fmt.Sprintf("Temporary read error: %v", err))
+ log.Debug("Temporary read error", "err", err)
continue
} else if err != nil {
- log.Debug(fmt.Sprintf("Read error: %v", err))
+ log.Debug("Read error", "err", err)
return
}
break
@@ -641,7 +652,7 @@ func (srv *Server) listenLoop() {
// Reject connections that do not match NetRestrict.
if srv.NetRestrict != nil {
if tcp, ok := fd.RemoteAddr().(*net.TCPAddr); ok && !srv.NetRestrict.Contains(tcp.IP) {
- log.Debug(fmt.Sprintf("Rejected conn %v because it is not whitelisted in NetRestrict", fd.RemoteAddr()))
+ log.Debug("Rejected conn (not whitelisted in NetRestrict)", "addr", fd.RemoteAddr())
fd.Close()
slots <- struct{}{}
continue
@@ -649,7 +660,7 @@ func (srv *Server) listenLoop() {
}
fd = newMeteredConn(fd, true)
- log.Debug(fmt.Sprintf("Accepted conn %v", fd.RemoteAddr()))
+ log.Trace("Accepted connection", "addr", fd.RemoteAddr())
// Spawn the handler. It will give the slot back when the connection
// has been established.
@@ -676,36 +687,37 @@ func (srv *Server) setupConn(fd net.Conn, flags connFlag, dialDest *discover.Nod
// Run the encryption handshake.
var err error
if c.id, err = c.doEncHandshake(srv.PrivateKey, dialDest); err != nil {
- log.Debug(fmt.Sprintf("%v faild enc handshake: %v", c, err))
+ log.Trace("Failed RLPx handshake", "addr", c.fd.RemoteAddr(), "conn", c.flags, "err", err)
c.close(err)
return
}
+ clog := log.New("id", c.id, "addr", c.fd.RemoteAddr(), "conn", c.flags)
// For dialed connections, check that the remote public key matches.
if dialDest != nil && c.id != dialDest.ID {
c.close(DiscUnexpectedIdentity)
- log.Debug(fmt.Sprintf("%v dialed identity mismatch, want %x", c, dialDest.ID[:8]))
+ clog.Trace("Dialed identity mismatch", "want", c, dialDest.ID)
return
}
if err := srv.checkpoint(c, srv.posthandshake); err != nil {
- log.Debug(fmt.Sprintf("%v failed checkpoint posthandshake: %v", c, err))
+ clog.Trace("Rejected peer before protocol handshake", "err", err)
c.close(err)
return
}
// Run the protocol handshake
phs, err := c.doProtoHandshake(srv.ourHandshake)
if err != nil {
- log.Debug(fmt.Sprintf("%v failed proto handshake: %v", c, err))
+ clog.Trace("Failed proto handshake", "err", err)
c.close(err)
return
}
if phs.ID != c.id {
- log.Debug(fmt.Sprintf("%v wrong proto handshake identity: %x", c, phs.ID[:8]))
+ clog.Trace("Wrong devp2p handshake identity", "err", phs.ID)
c.close(DiscUnexpectedIdentity)
return
}
c.caps, c.name = phs.Caps, phs.Name
if err := srv.checkpoint(c, srv.addpeer); err != nil {
- log.Debug(fmt.Sprintf("%v failed checkpoint addpeer: %v", c, err))
+ clog.Trace("Rejected peer", "err", err)
c.close(err)
return
}
@@ -713,6 +725,13 @@ func (srv *Server) setupConn(fd net.Conn, flags connFlag, dialDest *discover.Nod
// launched by run.
}
+func truncateName(s string) string {
+ if len(s) > 20 {
+ return s[:20] + "..."
+ }
+ return s
+}
+
// 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 {
@@ -733,17 +752,13 @@ func (srv *Server) checkpoint(c *conn, stage chan<- *conn) error {
// it waits until the Peer logic returns and removes
// the peer.
func (srv *Server) runPeer(p *Peer) {
- log.Debug(fmt.Sprintf("Added %v", p))
-
if srv.newPeerHook != nil {
srv.newPeerHook(p)
}
- discreason := p.run()
+ remoteRequested, err := 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
-
- log.Debug(fmt.Sprintf("Removed %v (%v)", p, discreason))
+ srv.delpeer <- peerDrop{p, err, remoteRequested}
}
// NodeInfo represents a short summary of the information known about the host.