mirror of
https://github.com/slackhq/nebula.git
synced 2026-11-16 07:46:52 +01:00
Compare commits
23
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
556b457389 | ||
|
|
5bba4de3c6 | ||
|
|
06035252b2 | ||
|
|
1fb77bdda5 | ||
|
|
556747d9b7 | ||
|
|
8f34981764 | ||
|
|
4d6f757d1f | ||
|
|
94372f4f51 | ||
|
|
42280f04ce | ||
|
|
11c1917387 | ||
|
|
5ba7aff4b3 | ||
|
|
8a2de9b1b5 | ||
|
|
a8c52f5c13 | ||
|
|
13af8dd6f3 | ||
|
|
5bb0622192 | ||
|
|
f3705cbb67 | ||
|
|
94be454f4f | ||
|
|
5f26c1a7e5 | ||
|
|
d47c8c71f1 | ||
|
|
f57539a163 | ||
|
|
8d8ae0f744 | ||
|
|
f9b0853476 | ||
|
|
d5062dacf0 |
No files matched your search
@@ -7,6 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
|
||||
|
||||
## [Unreleased]
|
||||
|
||||
### Fixed
|
||||
|
||||
- `create-tunnel -address <ip:port>` panicked on a nil RemoteList and, over the ssh debug server, took nebula
|
||||
down with it. A hostinfo returned by StartHandshake had no RemoteList until the handshake loop's first
|
||||
pass attached one, and `create-tunnel -address` set the remote before that. The list is now attached when
|
||||
the handshake starts, so a pending hostinfo never exists without one.
|
||||
|
||||
## [1.11.1] - 2026-08-21
|
||||
|
||||
See the [v1.11.1](https://github.com/slackhq/nebula/milestone/30?closed=1) milestone for a complete list of changes.
|
||||
|
||||
@@ -174,10 +174,8 @@ func (cm *connectionManager) doTrafficCheck(localIndex uint32, p, nb, out []byte
|
||||
|
||||
switch decision {
|
||||
case deleteTunnel:
|
||||
if cm.hostMap.DeleteHostInfo(hostinfo) {
|
||||
// Only clearing the lighthouse cache if this is the last hostinfo for this vpn ip in the hostmap
|
||||
cm.intf.lightHouse.DeleteVpnAddrs(hostinfo.vpnAddrs)
|
||||
}
|
||||
// Local delete only, clearing the lighthouse cache if this was the last hostinfo for the vpn addr
|
||||
cm.intf.closeTunnel(hostinfo)
|
||||
|
||||
case closeTunnel:
|
||||
cm.intf.sendCloseTunnel(hostinfo)
|
||||
|
||||
@@ -7,6 +7,7 @@ import (
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/cert"
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/overlay/overlaytest"
|
||||
@@ -21,6 +22,8 @@ func newTestLighthouse() *LightHouse {
|
||||
l: test.NewLogger(),
|
||||
addrMap: map[netip.Addr]*RemoteList{},
|
||||
queryChan: make(chan netip.Addr, 10),
|
||||
// Nothing drains queryChan here, so an 11th query is dropped.
|
||||
queryDropped: metrics.NilCounter{},
|
||||
}
|
||||
lighthouses := []netip.Addr{}
|
||||
staticList := map[netip.Addr]struct{}{}
|
||||
|
||||
+12
-3
@@ -103,14 +103,23 @@ func (cs *ConnectionState) Decrypt(l *slog.Logger, messageCounter uint64, packet
|
||||
if !result {
|
||||
return nil, ErrAlreadySeen
|
||||
}
|
||||
out, err := cs.decryptPacket(messageCounter, packet, nb)
|
||||
return cs.acceptDecrypted(l, messageCounter, out, err)
|
||||
}
|
||||
|
||||
out, err := cs.dKey.DecryptDanger(packet[header.Len:header.Len], packet[:header.Len], packet[header.Len:], messageCounter, nb)
|
||||
// decryptPacket authenticates and decrypts packet in place, leaving the replay window to the caller.
|
||||
func (cs *ConnectionState) decryptPacket(messageCounter uint64, packet, nb []byte) ([]byte, error) {
|
||||
return cs.dKey.DecryptDanger(packet[header.Len:header.Len], packet[:header.Len], packet[header.Len:], messageCounter, nb)
|
||||
}
|
||||
|
||||
// acceptDecrypted finishes a decryptPacket result: an authenticated packet is recorded in the replay window, and
|
||||
// rejected if its counter was recorded since it was checked.
|
||||
func (cs *ConnectionState) acceptDecrypted(l *slog.Logger, messageCounter uint64, out []byte, err error) ([]byte, error) {
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
cs.decryptLock.Lock()
|
||||
result = cs.window.Update(l, messageCounter)
|
||||
result := cs.window.Update(l, messageCounter)
|
||||
cs.decryptLock.Unlock()
|
||||
if !result {
|
||||
return nil, ErrAlreadySeen
|
||||
|
||||
+131
-3
@@ -2,6 +2,8 @@ package nebula
|
||||
|
||||
import (
|
||||
"net/netip"
|
||||
"strconv"
|
||||
"sync"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
@@ -11,6 +13,7 @@ import (
|
||||
ct "github.com/slackhq/nebula/cert_test"
|
||||
"github.com/slackhq/nebula/handshake"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/overlay"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
"github.com/stretchr/testify/assert"
|
||||
@@ -20,11 +23,20 @@ import (
|
||||
// runTestHandshake runs a complete IX handshake between two freshly-built
|
||||
// peers and returns the initiator and responder Results. Used to produce
|
||||
// real cipher states for tests that need to exercise post-handshake glue.
|
||||
func runTestHandshake(t *testing.T) (initR, respR *handshake.Result) {
|
||||
func runTestHandshake(t testing.TB) (initR, respR *handshake.Result) {
|
||||
t.Helper()
|
||||
return runTestHandshakeWith(t, noise.CipherChaChaPoly)
|
||||
}
|
||||
|
||||
// runTestHandshakeWith is runTestHandshake with the data plane cipher picked by the caller.
|
||||
func runTestHandshakeWith(t testing.TB, cipher noise.CipherFunc) (initR, respR *handshake.Result) {
|
||||
t.Helper()
|
||||
|
||||
// Valid around now rather than the cert_test default, which is fixed at process start and lapses a minute
|
||||
// later, inside a long -count run
|
||||
now := time.Now()
|
||||
ca, _, caKey, _ := ct.NewTestCaCert(
|
||||
cert.Version2, cert.Curve_CURVE25519, time.Time{}, time.Time{}, nil, nil, nil,
|
||||
cert.Version2, cert.Curve_CURVE25519, now.Add(-time.Minute), now.Add(time.Hour), nil, nil, nil,
|
||||
)
|
||||
caPool := ct.NewTestCAPool(ca)
|
||||
|
||||
@@ -37,7 +49,7 @@ func runTestHandshake(t *testing.T) (initR, respR *handshake.Result) {
|
||||
require.NoError(t, err)
|
||||
hsBytes, err := c.MarshalForHandshakes()
|
||||
require.NoError(t, err)
|
||||
ncs := noise.NewCipherSuite(noise.DH25519, noise.CipherChaChaPoly, noise.HashSHA256)
|
||||
ncs := noise.NewCipherSuite(noise.DH25519, cipher, noise.HashSHA256)
|
||||
cred := handshake.NewCredential(c, hsBytes, priv, ncs)
|
||||
return func(v cert.Version) *handshake.Credential {
|
||||
if v == cert.Version2 {
|
||||
@@ -147,6 +159,122 @@ func TestSendNoMetricsCloseTunnelKeepsRebindEpoch(t *testing.T) {
|
||||
assert.True(t, hostinfo.markOut(1), "a CloseTunnel send must not consume the rebind epoch")
|
||||
}
|
||||
|
||||
// capturingConn is a udp.Conn that records every WriteTo.
|
||||
type capturingConn struct {
|
||||
udp.NoopConn
|
||||
writes [][]byte
|
||||
addrs []netip.AddrPort
|
||||
}
|
||||
|
||||
func (c *capturingConn) WriteTo(b []byte, addr netip.AddrPort) error {
|
||||
c.writes = append(c.writes, append([]byte(nil), b...))
|
||||
c.addrs = append(c.addrs, addr)
|
||||
return nil
|
||||
}
|
||||
|
||||
// TestSendNoMetricsViaRelay sends through a relay from the mtu sized buffer the cached packet flush uses. Payloads up
|
||||
// to overlay.MaxMTU go out intact, anything bigger is dropped rather than outgrowing the buffer, which once panicked.
|
||||
func TestSendNoMetricsViaRelay(t *testing.T) {
|
||||
tests := []struct {
|
||||
n int
|
||||
sent bool
|
||||
}{
|
||||
{0, true},
|
||||
{1300, true},
|
||||
{overlay.MaxMTU, true},
|
||||
{overlay.MaxMTU + 1, false}, // too big for the relay's tag
|
||||
{mtu - 47, false}, // too big for the inner tag, this panicked
|
||||
{9000, false}, // a full packet at tun.mtu 9000
|
||||
}
|
||||
for _, tt := range tests {
|
||||
t.Run(strconv.Itoa(tt.n), func(t *testing.T) {
|
||||
// One tunnel to the target, carried end to end, and one to the relay that wraps it
|
||||
targetInit, targetResp := runTestHandshake(t)
|
||||
relayInit, relayResp := runTestHandshake(t)
|
||||
ci, err := newConnectionStateFromResult(targetInit)
|
||||
require.NoError(t, err)
|
||||
relayCI, err := newConnectionStateFromResult(relayInit)
|
||||
require.NoError(t, err)
|
||||
|
||||
targetAddr := netip.MustParseAddr("10.0.0.2")
|
||||
relayAddr := netip.MustParseAddr("10.0.0.3")
|
||||
relayRemote := netip.MustParseAddrPort("192.0.2.1:4242")
|
||||
|
||||
relay := &Relay{Type: TerminalType, State: Established, LocalIndex: 5, RemoteIndex: 6, PeerAddr: targetAddr}
|
||||
relayHI := &HostInfo{
|
||||
vpnAddrs: []netip.Addr{relayAddr},
|
||||
ConnectionState: relayCI,
|
||||
relayState: RelayState{
|
||||
relayForByAddr: map[netip.Addr]*Relay{targetAddr: relay},
|
||||
relayForByIdx: map[uint32]*Relay{relay.LocalIndex: relay},
|
||||
},
|
||||
}
|
||||
relayHI.remote.Store(&relayRemote)
|
||||
|
||||
// No remote of its own, so the send has to go via the relay
|
||||
hostinfo := &HostInfo{
|
||||
vpnAddrs: []netip.Addr{targetAddr},
|
||||
ConnectionState: ci,
|
||||
remoteIndexId: 7,
|
||||
relayState: RelayState{
|
||||
relays: []netip.Addr{relayAddr},
|
||||
relayForByAddr: map[netip.Addr]*Relay{},
|
||||
relayForByIdx: map[uint32]*Relay{},
|
||||
},
|
||||
}
|
||||
|
||||
l := test.NewLogger()
|
||||
hm := newHostMap(l)
|
||||
hm.Hosts[relayAddr] = relayHI
|
||||
conn := &capturingConn{}
|
||||
f := &Interface{
|
||||
l: l,
|
||||
hostMap: hm,
|
||||
messageMetrics: &MessageMetrics{txExhausted: metrics.NewCounter()},
|
||||
writers: []udp.Conn{conn},
|
||||
}
|
||||
f.connectionManager = &connectionManager{intf: f, relayUsed: map[uint32]struct{}{}, relayUsedLock: &sync.RWMutex{}}
|
||||
|
||||
payload := make([]byte, tt.n)
|
||||
for i := range payload {
|
||||
payload[i] = byte(i)
|
||||
}
|
||||
counter := ci.messageCounter.Load()
|
||||
f.sendNoMetrics(header.Message, 0, ci, hostinfo, netip.AddrPort{}, payload, make([]byte, 12), make([]byte, mtu), 0)
|
||||
|
||||
if !tt.sent {
|
||||
assert.Empty(t, conn.writes)
|
||||
assert.Equal(t, counter, ci.messageCounter.Load(), "a dropped packet must not spend a counter")
|
||||
return
|
||||
}
|
||||
require.Len(t, conn.writes, 1)
|
||||
assert.Equal(t, relayRemote, conn.addrs[0])
|
||||
pkt := conn.writes[0]
|
||||
|
||||
// The relay authenticates the whole packet under its own header
|
||||
outer := &header.H{}
|
||||
require.NoError(t, outer.Parse(pkt))
|
||||
assert.Equal(t, header.MessageRelay, outer.Subtype)
|
||||
assert.Equal(t, relay.RemoteIndex, outer.RemoteIndex)
|
||||
relayPeer, err := newConnectionStateFromResult(relayResp)
|
||||
require.NoError(t, err)
|
||||
require.NoError(t, relayPeer.VerifyRelay(l, outer.MessageCounter, pkt, make([]byte, 12)))
|
||||
|
||||
// Inside it, the target decrypts the end to end packet back to the payload
|
||||
inner := pkt[header.Len : len(pkt)-relayPeer.dKey.Overhead()]
|
||||
h := &header.H{}
|
||||
require.NoError(t, h.Parse(inner))
|
||||
assert.Equal(t, header.Message, h.Type)
|
||||
assert.Equal(t, hostinfo.remoteIndexId, h.RemoteIndex)
|
||||
targetPeer, err := newConnectionStateFromResult(targetResp)
|
||||
require.NoError(t, err)
|
||||
got, err := targetPeer.Decrypt(l, h.MessageCounter, inner, make([]byte, 12))
|
||||
require.NoError(t, err)
|
||||
assert.Equal(t, payload, got)
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
func TestNewConnectionStateFromResult(t *testing.T) {
|
||||
initR, respR := runTestHandshake(t)
|
||||
|
||||
|
||||
@@ -144,10 +144,10 @@ type fakeConn struct {
|
||||
rebinds int
|
||||
}
|
||||
|
||||
func (c *fakeConn) Rebind() error { c.rebinds++; return nil }
|
||||
func (c *fakeConn) LocalAddr() (netip.AddrPort, error) { return netip.AddrPort{}, nil }
|
||||
func (c *fakeConn) ListenOut(_ udp.EncReader, _ func()) error { return nil }
|
||||
func (c *fakeConn) WriteTo(_ []byte, _ netip.AddrPort) error { return nil }
|
||||
func (c *fakeConn) Rebind() error { c.rebinds++; return nil }
|
||||
func (c *fakeConn) LocalAddr() (netip.AddrPort, error) { return netip.AddrPort{}, nil }
|
||||
func (c *fakeConn) ListenOut(_ udp.EncReader, _ func([][]byte) [][]byte) error { return nil }
|
||||
func (c *fakeConn) WriteTo(_ []byte, _ netip.AddrPort) error { return nil }
|
||||
func (c *fakeConn) WriteBatch(bufs [][]byte, _ []netip.AddrPort) (int, error) {
|
||||
return len(bufs), nil
|
||||
}
|
||||
|
||||
@@ -5,9 +5,11 @@ package e2e
|
||||
import (
|
||||
"fmt"
|
||||
"net/netip"
|
||||
"runtime"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/cert"
|
||||
"github.com/slackhq/nebula/cert_test"
|
||||
"github.com/slackhq/nebula/e2e/router"
|
||||
@@ -570,3 +572,29 @@ func TestCloseTunnelAuthenticated(t *testing.T) {
|
||||
myControl.Stop()
|
||||
theirControl.Stop()
|
||||
}
|
||||
|
||||
// TestStagedTunnel completes a handshake and traffic both ways with staged receive. Every received packet,
|
||||
// handshakes included, uses the receive pipeline on Linux. Other platforms must do the same over the inline fallback.
|
||||
func TestStagedTunnel(t *testing.T) {
|
||||
t.Parallel()
|
||||
received := metrics.GetOrRegisterCounter("flow.rx.inline", nil)
|
||||
before := received.Count()
|
||||
staged := m{"tun": m{"staged": m{"workers": 2}}}
|
||||
ca, _, caKey, _ := cert_test.NewTestCaCert(cert.Version1, cert.Curve_CURVE25519, time.Now(), time.Now().Add(10*time.Minute), nil, nil, []string{})
|
||||
myControl, myVpnIpNet, myUdpAddr, _ := newSimpleServer(cert.Version1, ca, caKey, "me", "10.128.0.1/24", staged)
|
||||
theirControl, theirVpnIpNet, theirUdpAddr, _ := newSimpleServer(cert.Version1, ca, caKey, "them", "10.128.0.2/24", staged)
|
||||
myControl.InjectLightHouseAddr(theirVpnIpNet[0].Addr(), theirUdpAddr)
|
||||
theirControl.InjectLightHouseAddr(myVpnIpNet[0].Addr(), myUdpAddr)
|
||||
myControl.Start()
|
||||
theirControl.Start()
|
||||
r := router.NewR(t, myControl, theirControl)
|
||||
defer r.RenderFlow()
|
||||
assertTunnel(t, myVpnIpNet[0].Addr(), theirVpnIpNet[0].Addr(), myControl, theirControl, r)
|
||||
myControl.Stop()
|
||||
theirControl.Stop()
|
||||
if runtime.GOOS != "linux" {
|
||||
assert.Zero(t, received.Count()-before)
|
||||
return
|
||||
}
|
||||
assert.GreaterOrEqual(t, received.Count()-before, int64(2))
|
||||
}
|
||||
+16
-7
@@ -135,12 +135,12 @@ listen:
|
||||
# superpacket, so the receive scratch is batch * 64KiB per listening socket (~4MiB per routine at the
|
||||
# default of 64). Lower this to trade peak per-syscall throughput for memory on constrained hosts.
|
||||
#batch: 64
|
||||
# Configure socket buffers for the udp side (outside), leave unset to use the system defaults. Values will be doubled by the kernel
|
||||
# Linux only. Configure socket buffers for the udp side (outside), leave unset to use the system defaults. Values will be doubled by the kernel
|
||||
# Default is net.core.rmem_default and net.core.wmem_default (/proc/sys/net/core/rmem_default and /proc/sys/net/core/rmem_default)
|
||||
# Maximum is limited by memory in the system, SO_RCVBUFFORCE and SO_SNDBUFFORCE is used to avoid having to raise the system wide
|
||||
# max, net.core.rmem_max and net.core.wmem_max
|
||||
#read_buffer: 10485760
|
||||
#write_buffer: 10485760
|
||||
read_buffer: 3412512
|
||||
write_buffer: 3412512
|
||||
|
||||
# On Windows only
|
||||
# When true, Nebula installs a WFP (Windows Filtering Platform) PERMIT filter scoped to UDP at the listener port.
|
||||
@@ -262,14 +262,23 @@ tun:
|
||||
drop_local_broadcast: false
|
||||
# Toggles forwarding of multicast packets
|
||||
drop_multicast: false
|
||||
# Sets the transmit queue length, if you notice lots of transmit drops on the tun it may help to raise this number. Default is 500
|
||||
tx_queue: 500
|
||||
# Sets the transmit queue length, if you notice lots of transmit drops on the tun it may help to raise this number. Default is 5000
|
||||
tx_queue: 5000
|
||||
# Default MTU for every packet, safe setting is (and the default) 1300 for internet based traffic
|
||||
# Anything above 8937, here or on a route below, is capped to 8937 with a warning
|
||||
mtu: 1300
|
||||
|
||||
# the use_offloads setting controls if Nebula will attempt to enable GSO and GRO for the tun device. Linux only, not reloadable.
|
||||
#use_offloads: false
|
||||
|
||||
# Linux only. staged.workers enables staged receive: each routine hands its received batches to this many decrypt
|
||||
# workers, then an ordered writer commits them in arrival order through the same replay, roaming and firewall
|
||||
# handling as inline receive. 0, the default, keeps receive inline. Batches below 256 packets on an idle listener
|
||||
# stay on the listener. Needs no support from peers. Costs about 24 MiB of buffers per routine at listen.batch 64
|
||||
# with GRO. Not reloadable.
|
||||
#staged:
|
||||
# workers: 0
|
||||
|
||||
# Linux only. pin_threads pins each tun reader/encrypt OS thread to a single CPU. This keeps every goroutine's
|
||||
# batched sends flowing through one XPS-selected NIC TX ring, so packets within a flow stay ordered on the wire
|
||||
# instead of being sprayed across multiple TX rings and reordered. Not reloadable. Coerced to false if routines <= 1.
|
||||
@@ -410,8 +419,8 @@ logging:
|
||||
#try_interval: 100ms
|
||||
#retries: 20
|
||||
|
||||
# query_buffer is the size of the buffer channel for querying lighthouses
|
||||
#query_buffer: 64
|
||||
# query_buffer is the size of the buffer channel for querying lighthouses, drops are recorded in the lighthouse.query.dropped metric.
|
||||
#query_buffer: 1024
|
||||
|
||||
# trigger_buffer is the size of the buffer channel for quickly sending handshakes
|
||||
# after receiving the response for lighthouse queries
|
||||
|
||||
@@ -0,0 +1,40 @@
|
||||
package nebula
|
||||
|
||||
import (
|
||||
"log/slog"
|
||||
"runtime"
|
||||
|
||||
"github.com/slackhq/nebula/config"
|
||||
)
|
||||
|
||||
// Staged receive (tun.staged.workers) moves decryption off each udp listener onto worker goroutines, behind which an
|
||||
// ordered writer commits every batch in arrival order. It reuses the inline packet handlers: staging changes ownership
|
||||
// and scheduling, not policy. Each routine has its own workers and writer, so order is per routine. Full queues block
|
||||
// the listener. A read error drains the batches in flight, while shutdown waits for the workers and then drops what
|
||||
// is left. Each routine keeps six batches; at listen.batch=64 with GRO their buffers take about 24 MiB per routine.
|
||||
//
|
||||
// Metrics are registered only for active pipelines. flow.rx.batches counts batches handed to the writer and
|
||||
// flow.rx.inline those an idle listener handled itself; their rates show whether staging is used.
|
||||
|
||||
// stagedConfig is tun.staged. workers is per routine; zero keeps receive inline.
|
||||
type stagedConfig struct {
|
||||
workers int
|
||||
// inlineBelow is the packet count below which an idle listener skips the worker handoff
|
||||
inlineBelow int
|
||||
}
|
||||
|
||||
// parseStagedConfig enables staging on Linux only; conns that cannot hand off batches also stay inline.
|
||||
func parseStagedConfig(c *config.C, l *slog.Logger) stagedConfig {
|
||||
sc := stagedConfig{workers: c.GetInt("tun.staged.workers", 0)}
|
||||
if sc.workers <= 0 {
|
||||
return stagedConfig{}
|
||||
}
|
||||
if runtime.GOOS != "linux" {
|
||||
l.Warn("tun.staged is only supported on Linux; packet processing stays inline")
|
||||
return stagedConfig{}
|
||||
}
|
||||
// 256 is the measured crossover
|
||||
sc.inlineBelow = 256
|
||||
l.Info("Staged receive processing enabled", "workers", sc.workers)
|
||||
return sc
|
||||
}
|
||||
+303
@@ -0,0 +1,303 @@
|
||||
package nebula
|
||||
|
||||
import (
|
||||
"net/netip"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
)
|
||||
|
||||
// handoffConn is a udp conn that receives in batches of a fixed shape, so staged receive can have replacement buffers
|
||||
// ready for every batch it keeps.
|
||||
type handoffConn interface {
|
||||
ListenOut(udp.EncReader, func([][]byte) [][]byte) error
|
||||
HandoffShape() (n, size int)
|
||||
}
|
||||
|
||||
const stagedRxSlots = 6
|
||||
const rxChunkMin = 16
|
||||
|
||||
// rxItem is one received packet. A data packet for an established tunnel has ci set; out and err are written by
|
||||
// whoever decrypts it.
|
||||
type rxItem struct {
|
||||
packet []byte
|
||||
ci *ConnectionState
|
||||
counter uint64
|
||||
via netip.AddrPort
|
||||
out []byte
|
||||
err error
|
||||
}
|
||||
|
||||
// rxChunk is a contiguous run of a batch's items for one worker to decrypt.
|
||||
type rxChunk struct {
|
||||
items []rxItem
|
||||
done chan struct{}
|
||||
}
|
||||
|
||||
// decryptRxItems decrypts the data packets among items in place, skipping the replays checkReplays marked. Recording
|
||||
// the counters is left to readOutsidePackets on the writer, which also handles the packets without a tunnel.
|
||||
func decryptRxItems(items []rxItem, nb []byte) {
|
||||
for i := range items {
|
||||
it := &items[i]
|
||||
if it.ci == nil || it.err != nil {
|
||||
continue
|
||||
}
|
||||
it.out, it.err = it.ci.decryptPacket(it.counter, it.packet, nb)
|
||||
}
|
||||
}
|
||||
|
||||
// rxBatch is one recvmmsg batch moving through the staged receive pipeline.
|
||||
type rxBatch struct {
|
||||
// items is every packet of the batch in arrival order. Data packets are decrypted by a worker; anything else is
|
||||
// handled whole by the writer, in its place.
|
||||
items []rxItem
|
||||
// bufs is the set of recvmmsg buffers the items point into, owned by the batch from handoff until it is reused.
|
||||
bufs [][]byte
|
||||
// chunks are the decrypt jobs, nChunks of them in use.
|
||||
chunks []rxChunk
|
||||
nChunks int
|
||||
}
|
||||
|
||||
func (b *rxBatch) reset() {
|
||||
clear(b.items)
|
||||
b.items = b.items[:0]
|
||||
for i := range b.nChunks {
|
||||
b.chunks[i].items = nil
|
||||
}
|
||||
b.nChunks = 0
|
||||
}
|
||||
|
||||
// stagedRx is the staged receive pipeline of one udp listener. deliver and handoff run on the listener goroutine.
|
||||
type stagedRx struct {
|
||||
f *Interface
|
||||
q int
|
||||
conn handoffConn
|
||||
// jobs feeds the decrypt workers. The listener is the only producer and closes it once the writer has drained.
|
||||
jobs chan *rxChunk
|
||||
workers sync.WaitGroup
|
||||
// h and hostmapCache are the listener's own. wrxc is the commit context: the writer's, or the listener's while
|
||||
// nothing is in flight.
|
||||
h *header.H
|
||||
hostmapCache map[uint32]*HostInfo
|
||||
wrxc *rxContext
|
||||
|
||||
// cur is the batch the listener is filling. The others wait in free, or travel to the writer on ordered and come
|
||||
// back through free; inFlight counts those, and the writer drops it only once it is done with a batch.
|
||||
cur *rxBatch
|
||||
free chan *rxBatch
|
||||
ordered chan *rxBatch
|
||||
inFlight atomic.Int32
|
||||
writerDone chan struct{}
|
||||
|
||||
metricBatches metrics.Counter
|
||||
metricInline metrics.Counter
|
||||
}
|
||||
|
||||
func newStagedRx(f *Interface, q int, wrxc *rxContext, conn handoffConn) *stagedRx {
|
||||
p := &stagedRx{
|
||||
f: f,
|
||||
q: q,
|
||||
conn: conn,
|
||||
h: &header.H{},
|
||||
hostmapCache: map[uint32]*HostInfo{},
|
||||
wrxc: wrxc,
|
||||
free: make(chan *rxBatch, stagedRxSlots),
|
||||
ordered: make(chan *rxBatch, stagedRxSlots),
|
||||
writerDone: make(chan struct{}),
|
||||
metricBatches: metrics.GetOrRegisterCounter("flow.rx.batches", nil),
|
||||
metricInline: metrics.GetOrRegisterCounter("flow.rx.inline", nil),
|
||||
}
|
||||
n := f.staged.workers
|
||||
p.jobs = make(chan *rxChunk, stagedRxSlots*2*n)
|
||||
for range n {
|
||||
p.workers.Go(p.decrypt)
|
||||
}
|
||||
bufs, size := conn.HandoffShape()
|
||||
for i := range stagedRxSlots {
|
||||
// Two chunks per worker, so the writer can start committing while the workers are still on the rest
|
||||
b := &rxBatch{items: make([]rxItem, 0, 256), chunks: make([]rxChunk, 2*n)}
|
||||
for j := range b.chunks {
|
||||
b.chunks[j].done = make(chan struct{}, 1)
|
||||
}
|
||||
if i == 0 {
|
||||
// The first batch reads into the conn's own buffers; every other one brings its own
|
||||
p.cur = b
|
||||
continue
|
||||
}
|
||||
b.bufs = make([][]byte, bufs)
|
||||
for j := range b.bufs {
|
||||
b.bufs[j] = make([]byte, size)
|
||||
}
|
||||
p.free <- b
|
||||
}
|
||||
go p.writer()
|
||||
return p
|
||||
}
|
||||
|
||||
// decrypt is a worker: it decrypts the chunks it is handed and signals each one done.
|
||||
func (p *stagedRx) decrypt() {
|
||||
nb := make([]byte, 12)
|
||||
for c := range p.jobs {
|
||||
decryptRxItems(c.items, nb)
|
||||
c.done <- struct{}{}
|
||||
}
|
||||
}
|
||||
|
||||
// listen runs the conn's receive loop through the pipeline until it fails, then drains the pipeline. The conn hands
|
||||
// its buffers over, so a batch moves between stages without a copy.
|
||||
func (p *stagedRx) listen() error {
|
||||
defer p.close()
|
||||
return p.conn.ListenOut(p.deliver, p.handoff)
|
||||
}
|
||||
|
||||
// deliver is the listener's per-packet work: classify the packet and add it to the current batch. Nothing here
|
||||
// decrypts or touches the tun.
|
||||
func (p *stagedRx) deliver(via netip.AddrPort, packet []byte) {
|
||||
it := rxItem{packet: packet, via: via}
|
||||
p.classify(&it)
|
||||
p.cur.items = append(p.cur.items, it)
|
||||
}
|
||||
|
||||
// classify fills in the tunnel for a plain data packet of an established, non-relayed tunnel, leaving anything else
|
||||
// for the writer to handle whole. It is speculative: readOutsidePackets resolves the tunnel again on the writer and
|
||||
// uses the worker's result only when it is still the same one.
|
||||
func (p *stagedRx) classify(it *rxItem) {
|
||||
f := p.f
|
||||
h := p.h
|
||||
packet := it.packet
|
||||
if h.Parse(packet) != nil || h.Version != header.Version || h.Type != header.Message || h.Subtype != header.MessageNone {
|
||||
return
|
||||
}
|
||||
if f.myVpnNetworksTable.Contains(it.via.Addr()) {
|
||||
return
|
||||
}
|
||||
hostinfo := f.hostMap.QueryIndexCached(h.RemoteIndex, p.hostmapCache)
|
||||
if hostinfo == nil || hostinfo.ConnectionState == nil {
|
||||
return
|
||||
}
|
||||
ci := hostinfo.ConnectionState
|
||||
if len(packet) < header.Len+ci.dKey.Overhead() {
|
||||
return
|
||||
}
|
||||
it.ci, it.counter = ci, h.MessageCounter
|
||||
}
|
||||
|
||||
// handoff ends a batch: the batch takes bufs, and the conn reads the next batch into the buffers of the next free
|
||||
// batch.
|
||||
func (p *stagedRx) handoff(bufs [][]byte) [][]byte {
|
||||
p.cur.bufs = bufs
|
||||
p.endBatch()
|
||||
return p.cur.bufs
|
||||
}
|
||||
|
||||
// endBatch hands the batch to the workers and the ordered writer, then takes a free batch.
|
||||
func (p *stagedRx) endBatch() {
|
||||
clear(p.hostmapCache)
|
||||
b := p.cur
|
||||
n := len(b.items)
|
||||
if n == 0 {
|
||||
b.reset()
|
||||
return
|
||||
}
|
||||
p.checkReplays(b.items)
|
||||
|
||||
if p.inFlight.Load() == 0 && n < p.f.staged.inlineBelow {
|
||||
// The writer is idle with nothing in flight, so the listener may commit with its context and flush the tun
|
||||
p.metricInline.Inc(1)
|
||||
p.processInline(b, p.wrxc)
|
||||
b.reset()
|
||||
return
|
||||
}
|
||||
|
||||
parts := max(1, min(len(b.chunks), n/rxChunkMin))
|
||||
size := (n + parts - 1) / parts
|
||||
for off := 0; off < n; off += size {
|
||||
c := &b.chunks[b.nChunks]
|
||||
c.items = b.items[off:min(off+size, n)]
|
||||
b.nChunks++
|
||||
}
|
||||
// The jobs go out before the batch does: once the writer has it, it may finish and reset it at any moment
|
||||
for i := range b.nChunks {
|
||||
p.jobs <- &b.chunks[i]
|
||||
}
|
||||
p.metricBatches.Inc(1)
|
||||
p.inFlight.Add(1)
|
||||
p.ordered <- b
|
||||
// With every batch in flight this waits for the writer, the socket buffer absorbs the backlog meanwhile
|
||||
p.cur = <-p.free
|
||||
}
|
||||
|
||||
// checkReplays marks the replays among items before anything decrypts them, as Decrypt checks before it decrypts,
|
||||
// taking each tunnel's lock once per run of its packets rather than once per packet. The window only moves forward
|
||||
// until the writer records these counters, so a stale check can only let a replay through to acceptDecrypted, which
|
||||
// rejects it.
|
||||
func (p *stagedRx) checkReplays(items []rxItem) {
|
||||
for i := 0; i < len(items); {
|
||||
ci := items[i].ci
|
||||
if ci == nil {
|
||||
i++
|
||||
continue
|
||||
}
|
||||
ci.decryptLock.Lock()
|
||||
for ; i < len(items) && items[i].ci == ci; i++ {
|
||||
if !ci.window.Check(p.f.l, items[i].counter) {
|
||||
items[i].err = ErrAlreadySeen
|
||||
}
|
||||
}
|
||||
ci.decryptLock.Unlock()
|
||||
}
|
||||
}
|
||||
|
||||
// processInline decrypts and commits a batch on the calling goroutine, then flushes the tun.
|
||||
func (p *stagedRx) processInline(b *rxBatch, rxc *rxContext) {
|
||||
decryptRxItems(b.items, rxc.nb)
|
||||
p.commit(b.items, rxc)
|
||||
p.f.flushOutside(rxc)
|
||||
}
|
||||
|
||||
// commit runs every item through readOutsidePackets in arrival order, handing it the worker's result for a data
|
||||
// packet, exactly as inline receive handles them.
|
||||
func (p *stagedRx) commit(items []rxItem, rxc *rxContext) {
|
||||
for i := range items {
|
||||
it := &items[i]
|
||||
if it.ci != nil {
|
||||
rxc.decrypted = it
|
||||
}
|
||||
p.f.readOutsidePackets(ViaSender{UdpAddr: it.via}, it.packet, rxc)
|
||||
rxc.decrypted = nil
|
||||
}
|
||||
}
|
||||
|
||||
// writer is the ordered receive stage: it commits each batch chunk by chunk as the workers finish them, so the commit
|
||||
// overlaps the workers still on later chunks, then flushes the tun and returns the batch.
|
||||
func (p *stagedRx) writer() {
|
||||
defer close(p.writerDone)
|
||||
for b := range p.ordered {
|
||||
// Once the tun is going away a write would only log errors, so only wait out the workers
|
||||
closed := p.f.closed.Load()
|
||||
for i := range b.nChunks {
|
||||
c := &b.chunks[i]
|
||||
<-c.done
|
||||
if !closed {
|
||||
p.commit(c.items, p.wrxc)
|
||||
}
|
||||
}
|
||||
if !closed {
|
||||
p.f.flushOutside(p.wrxc)
|
||||
}
|
||||
b.reset()
|
||||
p.inFlight.Add(-1)
|
||||
p.free <- b
|
||||
}
|
||||
}
|
||||
|
||||
// close stops the writer once it has committed everything in flight, in order, then the workers.
|
||||
func (p *stagedRx) close() {
|
||||
close(p.ordered)
|
||||
<-p.writerDone
|
||||
close(p.jobs)
|
||||
p.workers.Wait()
|
||||
}
|
||||
+219
@@ -0,0 +1,219 @@
|
||||
//go:build linux && !android && !e2e_testing
|
||||
|
||||
package nebula
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"net/netip"
|
||||
"sync/atomic"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/flynn/noise"
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/noiseutil"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
"github.com/stretchr/testify/require"
|
||||
"go.uber.org/goleak"
|
||||
)
|
||||
|
||||
// TestFlowStagedRXOrderAndReplay feeds every packet behind a forgery of it, from another address, and a replay of
|
||||
// it, then a test packet that jumps the window and the whole stream again, with shuffled workers and a slow tun,
|
||||
// through small and large batches: the tun sees each segment once, in order, the window marks them all, the forgeries
|
||||
// neither consume a counter nor roam, and the jump does not get the earlier data packets of its batch rejected.
|
||||
func TestFlowStagedRXOrderAndReplay(t *testing.T) {
|
||||
defer goleak.VerifyNone(t, goleak.IgnoreCurrent())
|
||||
for _, cipher := range []noise.CipherFunc{noiseutil.CipherAESGCM, noise.CipherChaChaPoly} {
|
||||
for _, batchLen := range []int{8, 1000} {
|
||||
t.Run(fmt.Sprintf("%s/batch=%d", cipher.CipherName(), batchLen), func(t *testing.T) {
|
||||
sender, receiver := newFlowTestPair(t, cipher, stagedOn(4))
|
||||
delayCrypto(receiver, func(n uint64) {
|
||||
if n%32 == 0 {
|
||||
time.Sleep(time.Millisecond)
|
||||
}
|
||||
})
|
||||
receiver.tun.delay = 100 * time.Microsecond
|
||||
batches, want := flowTestBatches(30, 4, 1, 10)
|
||||
wire := wireFor(t, sender, batches)
|
||||
var feed [][]byte
|
||||
for i, p := range wire {
|
||||
// Corrupt the tag or the payload
|
||||
bad := append([]byte(nil), p...)
|
||||
bad[[]int{len(p) - 1, header.Len + 3}[i%2]] ^= 1
|
||||
feed = append(feed, bad, p, p)
|
||||
}
|
||||
// The jump lands in a batch with fresh data before it, at batch=1000
|
||||
sender.hostinfo.ConnectionState.messageCounter.Add(2 * ReplayWindow)
|
||||
feed = append(append(feed, sendControl(sender, header.Test, header.TestReply)), wire...)
|
||||
forged := func(i int) netip.AddrPort {
|
||||
if i < 3*len(wire) && i%3 == 0 {
|
||||
return netip.MustParseAddrPort("198.51.100.9:4242")
|
||||
}
|
||||
return flowTestUnder
|
||||
}
|
||||
rx0 := rxBatches.Count()
|
||||
conn := &rxConn{wire: feed, batchLen: batchLen, from: forged}
|
||||
receiver.receive(conn)
|
||||
require.Positive(t, rxBatches.Count()-rx0)
|
||||
require.Equal(t, (len(feed)+batchLen-1)/batchLen, conn.swaps, "every batch is handed off")
|
||||
require.Equal(t, want, receiver.tun.segments(t), "every packet exactly once, in order")
|
||||
require.Equal(t, flowTestUnder, receiver.hostinfo.GetRemote())
|
||||
for _, p := range wire {
|
||||
h := &header.H{}
|
||||
require.NoError(t, h.Parse(p))
|
||||
require.False(t, receiver.hostinfo.ConnectionState.window.Check(receiver.f.l, h.MessageCounter), "counter %d must be marked seen", h.MessageCounter)
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// TestFlowStagedRXClose: a CloseTunnel in the middle of a batch is handled in its place, by the writer. The data after
|
||||
// it in the batch is still delivered from the batch's hostmap cache, as inline receive does.
|
||||
func TestFlowStagedRXClose(t *testing.T) {
|
||||
defer goleak.VerifyNone(t, goleak.IgnoreCurrent())
|
||||
sender, receiver := newFlowTestPair(t, noise.CipherChaChaPoly, stagedOn(2))
|
||||
batches, want := flowTestBatches(4, 4, 1, 8)
|
||||
wire := wireFor(t, sender, batches)
|
||||
half := len(wire) / 2
|
||||
feed := append(append(append([][]byte{}, wire[:half]...), sendControl(sender, header.CloseTunnel, 0)), wire[half:]...)
|
||||
receiver.receive(&rxConn{wire: feed, batchLen: len(feed)})
|
||||
require.Equal(t, want, receiver.tun.segments(t))
|
||||
require.Nil(t, receiver.f.hostMap.QueryIndex(receiver.hostinfo.localIndexId))
|
||||
}
|
||||
|
||||
// sendControl returns the wire packet of a control message from sender.
|
||||
func sendControl(sender *flowTestPeer, typ header.MessageType, st header.MessageSubType) []byte {
|
||||
sender.f.sendNoMetrics(typ, st, sender.hostinfo.ConnectionState, sender.hostinfo, flowTestUnder, nil, make([]byte, 12), make([]byte, mtu), 0)
|
||||
wire := sender.conn.captured()
|
||||
sender.conn.pkts, sender.conn.addrs = nil, nil
|
||||
return wire[len(wire)-1]
|
||||
}
|
||||
|
||||
// receivePaths is inline, staged and, with the threshold, adaptive receive.
|
||||
func receivePaths() []stagedConfig {
|
||||
return []stagedConfig{{}, stagedOn(2), adaptive(2)}
|
||||
}
|
||||
|
||||
// TestFlowStagedRXRecvError feeds a RecvError that deletes the tunnel, then a known replay and new data for it, in one
|
||||
// batch. Every receive path must drop the data through its own view of the hostmap, without touching the replay window,
|
||||
// and answer each packet with a RecvError, as inline receive does, without decrypting the replay again.
|
||||
func TestFlowStagedRXRecvError(t *testing.T) {
|
||||
for _, sc := range receivePaths() {
|
||||
t.Run(stagedName(sc), func(t *testing.T) {
|
||||
sender, receiver := newFlowTestPair(t, noise.CipherChaChaPoly, sc)
|
||||
receiver.f.messageMetrics = newMessageMetrics()
|
||||
receiver.f.acceptRecvErrorConfig = recvErrorAlways
|
||||
receiver.f.sendRecvErrorConfig = recvErrorAlways
|
||||
batches, want := flowTestBatches(1, 1, 1, 33)
|
||||
wire := wireFor(t, sender, batches)
|
||||
rxc := newRxContext(receiver.f, 0)
|
||||
receiver.f.readOutsidePackets(ViaSender{UdpAddr: flowTestUnder}, append([]byte(nil), wire[0]...), rxc)
|
||||
receiver.f.flushOutside(rxc)
|
||||
h := &header.H{}
|
||||
require.NoError(t, h.Parse(wire[0]))
|
||||
replay := h.MessageCounter
|
||||
var decrypts atomic.Int64
|
||||
delayCrypto(receiver, func(n uint64) {
|
||||
if n == replay {
|
||||
decrypts.Add(1)
|
||||
}
|
||||
})
|
||||
ctl := header.Encode(make([]byte, header.Len), header.Version, header.RecvError, 0, receiver.hostinfo.remoteIndexId, 0)
|
||||
conn := &rxConn{wire: append([][]byte{ctl}, wire...), batchLen: len(wire) + 1}
|
||||
receiver.receive(conn)
|
||||
require.Nil(t, receiver.f.hostMap.QueryIndex(receiver.hostinfo.localIndexId))
|
||||
require.Equal(t, want[:1], receiver.tun.segments(t))
|
||||
require.Zero(t, decrypts.Load(), "known replays must not be decrypted again")
|
||||
replies := conn.captured()
|
||||
require.Len(t, replies, len(wire))
|
||||
for i, reply := range replies {
|
||||
require.NoError(t, h.Parse(reply))
|
||||
require.Equal(t, header.RecvError, h.Type)
|
||||
require.NoError(t, h.Parse(wire[i]))
|
||||
require.True(t, i == 0 || receiver.hostinfo.ConnectionState.window.Check(nil, h.MessageCounter), "dropped data must not update replay state")
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
// TestFlowStagedRXReplacedTunnel classifies a data packet, then a close and a new handshake replace its tunnel at the
|
||||
// same index before the writer commits it, as when the listener runs ahead: it must be dropped untouched.
|
||||
func TestFlowStagedRXReplacedTunnel(t *testing.T) {
|
||||
sender, receiver := newFlowTestPair(t, noise.CipherChaChaPoly, stagedOn(2))
|
||||
receiver.f.messageMetrics = newMessageMetrics()
|
||||
batches, _ := flowTestBatches(1, 1, 1, 1)
|
||||
wire := wireFor(t, sender, batches)
|
||||
closePacket := sendControl(sender, header.CloseTunnel, 0)
|
||||
p := &stagedRx{f: receiver.f, h: &header.H{}, hostmapCache: map[uint32]*HostInfo{}}
|
||||
item := rxItem{packet: wire[0], via: flowTestUnder}
|
||||
p.classify(&item)
|
||||
require.NotNil(t, item.ci)
|
||||
rxc := newRxContext(receiver.f, 0)
|
||||
p.processInline(&rxBatch{items: []rxItem{{packet: closePacket, via: flowTestUnder}}}, rxc)
|
||||
require.Nil(t, receiver.f.hostMap.QueryIndex(receiver.hostinfo.localIndexId))
|
||||
_, replacement := newFlowTestPair(t, noise.CipherChaChaPoly, stagedOn(2))
|
||||
receiver.f.hostMap.Lock()
|
||||
receiver.f.hostMap.unlockedAddHostInfo(replacement.hostinfo, receiver.f)
|
||||
receiver.f.hostMap.Unlock()
|
||||
p.processInline(&rxBatch{items: []rxItem{item}}, rxc)
|
||||
require.Empty(t, receiver.tun.writes)
|
||||
require.True(t, item.ci.window.Check(nil, item.counter))
|
||||
require.True(t, replacement.hostinfo.ConnectionState.window.Check(nil, item.counter))
|
||||
}
|
||||
|
||||
// TestFlowStagedRXRealSocket exercises buffer handoff through the real Linux UDP receive loop.
|
||||
func TestFlowStagedRXRealSocket(t *testing.T) {
|
||||
defer goleak.VerifyNone(t, goleak.IgnoreCurrent())
|
||||
sender, receiver := newFlowTestPair(t, noise.CipherChaChaPoly, stagedOn(3))
|
||||
batches, want := flowTestBatches(60, 4, 1, 8)
|
||||
wire := wireFor(t, sender, batches)
|
||||
|
||||
l := receiver.f.l
|
||||
conn, err := udp.NewListener(l, udp.Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0"), Batch: 16, Offloads: true})
|
||||
require.NoError(t, err)
|
||||
addr, err := conn.LocalAddr()
|
||||
require.NoError(t, err)
|
||||
out, err := udp.NewListener(l, udp.Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0"), Batch: 16, Offloads: true})
|
||||
require.NoError(t, err)
|
||||
defer out.Close()
|
||||
// The sender's address is the tunnel's remote, so no roaming
|
||||
from, err := out.LocalAddr()
|
||||
require.NoError(t, err)
|
||||
receiver.hostinfo.remote.Store(&from)
|
||||
|
||||
done := make(chan struct{})
|
||||
go func() { receiver.receive(conn); close(done) }()
|
||||
// Send in bursts small enough for the default socket buffer
|
||||
for off := 0; off < len(wire); off += 32 {
|
||||
end := min(off+32, len(wire))
|
||||
addrs := make([]netip.AddrPort, end-off)
|
||||
for i := range addrs {
|
||||
addrs[i] = addr
|
||||
}
|
||||
_, err := out.WriteBatch(wire[off:end], addrs)
|
||||
require.NoError(t, err)
|
||||
require.Eventually(t, func() bool { return len(receiver.tun.segments(t)) >= end }, 5*time.Second, 100*time.Microsecond)
|
||||
}
|
||||
receiver.f.closed.Store(true)
|
||||
require.NoError(t, conn.Close())
|
||||
<-done
|
||||
require.Equal(t, want, receiver.tun.segments(t))
|
||||
}
|
||||
|
||||
// TestFlowStagedRXInlineBelow: small idle batches stay on the listener, larger batches still use the ordered writer.
|
||||
func TestFlowStagedRXInlineBelow(t *testing.T) {
|
||||
defer goleak.VerifyNone(t, goleak.IgnoreCurrent())
|
||||
sc := adaptive(3)
|
||||
sc.inlineBelow = 16
|
||||
sender, receiver := newFlowTestPair(t, noise.CipherChaChaPoly, sc)
|
||||
batches, want := flowTestBatches(40, 4, 1, 16)
|
||||
wire := wireFor(t, sender, batches)
|
||||
inline := metrics.GetOrRegisterCounter("flow.rx.inline", nil)
|
||||
inline0, rx0 := inline.Count(), rxBatches.Count()
|
||||
receiver.receive(&rxConn{wire: wire, sizes: []int{8, 8, 8, 128, 128, 8, 8, 128, 4, 128}})
|
||||
require.Equal(t, want, receiver.tun.segments(t))
|
||||
require.Positive(t, inline.Count()-inline0)
|
||||
require.Positive(t, rxBatches.Count()-rx0)
|
||||
}
|
||||
+489
@@ -0,0 +1,489 @@
|
||||
//go:build linux && !android && !e2e_testing
|
||||
|
||||
package nebula
|
||||
|
||||
import (
|
||||
"encoding/binary"
|
||||
"io"
|
||||
"math"
|
||||
"net/netip"
|
||||
"os"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/flynn/noise"
|
||||
"github.com/gaissmai/bart"
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/cert"
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/firewall"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/iputil"
|
||||
"github.com/slackhq/nebula/noiseutil"
|
||||
"github.com/slackhq/nebula/overlay/batch"
|
||||
"github.com/slackhq/nebula/overlay/tio"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
"github.com/stretchr/testify/require"
|
||||
"go.uber.org/goleak"
|
||||
)
|
||||
|
||||
// flowTestMSS is the segment payload size of the synthetic TSO superpackets.
|
||||
const flowTestMSS = 1200
|
||||
|
||||
var (
|
||||
flowTestLocal = netip.MustParseAddr("10.0.0.1")
|
||||
flowTestRemote = netip.MustParseAddr("10.0.0.2")
|
||||
flowTestUnder = netip.MustParseAddrPort("192.0.2.2:4242")
|
||||
)
|
||||
|
||||
// flowTCPSuperpacket builds an IPv4/TCP TSO superpacket from local to remote carrying segs segments. Segment i's
|
||||
// payload starts with the flow id, the superpacket seq and i, so a receiver can check what it got and in which order.
|
||||
func flowTCPSuperpacket(flow uint16, seq uint32, segs int) tio.Packet {
|
||||
const hdr = 40
|
||||
pay := segs * flowTestMSS
|
||||
b := make([]byte, hdr+pay)
|
||||
b[0] = 0x45
|
||||
binary.BigEndian.PutUint16(b[2:4], uint16(min(hdr+pay, 65535)))
|
||||
b[8] = 64
|
||||
b[9] = iputil.IPProtocolTCP
|
||||
copy(b[12:16], flowTestLocal.AsSlice())
|
||||
copy(b[16:20], flowTestRemote.AsSlice())
|
||||
binary.BigEndian.PutUint16(b[20:22], 10000+flow)
|
||||
binary.BigEndian.PutUint16(b[22:24], 80)
|
||||
binary.BigEndian.PutUint32(b[24:28], seq*uint32(pay))
|
||||
b[32] = 0x50
|
||||
b[33] = 0x10 // ACK
|
||||
binary.BigEndian.PutUint16(b[34:36], 65535)
|
||||
for i := range segs {
|
||||
p := b[hdr+i*flowTestMSS:]
|
||||
binary.BigEndian.PutUint16(p[0:2], flow)
|
||||
binary.BigEndian.PutUint32(p[2:6], seq)
|
||||
binary.BigEndian.PutUint16(p[6:8], uint16(i))
|
||||
}
|
||||
if segs == 1 {
|
||||
return tio.Packet{Bytes: b}
|
||||
}
|
||||
return tio.Packet{Bytes: b, GSO: tio.GSOInfo{Size: flowTestMSS, HdrLen: hdr, CsumStart: 20, Proto: tio.GSOProtoTCP}}
|
||||
}
|
||||
|
||||
// decodedSeg identifies one segment by the marker flowTCPSuperpacket put in its payload.
|
||||
type decodedSeg struct {
|
||||
flow uint16
|
||||
seq uint32
|
||||
idx uint16
|
||||
}
|
||||
|
||||
func decodeSeg(p []byte) decodedSeg {
|
||||
return decodedSeg{flow: binary.BigEndian.Uint16(p[0:2]), seq: binary.BigEndian.Uint32(p[2:6]), idx: binary.BigEndian.Uint16(p[6:8])}
|
||||
}
|
||||
|
||||
// flowTestBatches builds n reads of per superpackets of segs segments, across flows interleaved flows, and the
|
||||
// segments they carry in order.
|
||||
func flowTestBatches(n, per, flows, segs int) ([][]tio.Packet, []decodedSeg) {
|
||||
var batches [][]tio.Packet
|
||||
var want []decodedSeg
|
||||
seq := make([]uint32, flows)
|
||||
for range n {
|
||||
var b []tio.Packet
|
||||
for j := range per {
|
||||
flow := uint16(j % flows)
|
||||
b = append(b, flowTCPSuperpacket(flow, seq[flow], segs))
|
||||
for s := range segs {
|
||||
want = append(want, decodedSeg{flow: flow, seq: seq[flow], idx: uint16(s)})
|
||||
}
|
||||
seq[flow]++
|
||||
}
|
||||
batches = append(batches, b)
|
||||
}
|
||||
return batches, want
|
||||
}
|
||||
|
||||
// flowTestPeer is one end of a tunnel: the interface, the hostinfo for the other end, and its capture points.
|
||||
type flowTestPeer struct {
|
||||
f *Interface
|
||||
hostinfo *HostInfo
|
||||
conn *flowCaptureConn
|
||||
tun *flowTunCapture
|
||||
}
|
||||
|
||||
// newFlowTestPair builds two interfaces joined by one real tunnel (keys from an actual handshake). sender sends from
|
||||
// flowTestLocal to flowTestRemote, receiver is the far side.
|
||||
func newFlowTestPair(t testing.TB, cipher noise.CipherFunc, sc stagedConfig) (sender, receiver *flowTestPeer) {
|
||||
t.Helper()
|
||||
initR, respR := runTestHandshakeWith(t, cipher)
|
||||
sci, err := newConnectionStateFromResult(initR)
|
||||
require.NoError(t, err)
|
||||
rci, err := newConnectionStateFromResult(respR)
|
||||
require.NoError(t, err)
|
||||
sender = newFlowTestPeer(t, flowTestLocal, flowTestRemote, sci, 100, 200)
|
||||
receiver = newFlowTestPeer(t, flowTestRemote, flowTestLocal, rci, 200, 100)
|
||||
sender.f.staged, receiver.f.staged = sc, sc
|
||||
return sender, receiver
|
||||
}
|
||||
|
||||
func newFlowTestPeer(t testing.TB, me, them netip.Addr, ci *ConnectionState, localIdx, remoteIdx uint32) *flowTestPeer {
|
||||
l := test.NewLogger()
|
||||
hm := newHostMap(l)
|
||||
hm.preferredRanges.Store(&[]netip.Prefix{})
|
||||
|
||||
myNets, myAddrs := &bart.Lite{}, &bart.Lite{}
|
||||
myNets.Insert(netip.PrefixFrom(me, 24))
|
||||
myAddrs.Insert(netip.PrefixFrom(me, 32))
|
||||
myCert := &dummyCert{version: cert.Version1, networks: []netip.Prefix{netip.PrefixFrom(me, 24)}}
|
||||
theirCert := &dummyCert{version: cert.Version1, networks: []netip.Prefix{netip.PrefixFrom(them, 24)}, name: "peer"}
|
||||
ci.peerCert = &cert.CachedCertificate{Certificate: theirCert, InvertedGroups: map[string]struct{}{}}
|
||||
|
||||
fw := NewFirewall(l, time.Minute, time.Minute, time.Minute, myCert)
|
||||
require.NoError(t, fw.AddRule(false, firewall.ProtoAny, 0, 0, []string{"any"}, "", "", "", "", ""))
|
||||
require.NoError(t, fw.AddRule(true, firewall.ProtoAny, 0, 0, []string{"any"}, "", "", "", "", ""))
|
||||
|
||||
conn, tun := &flowCaptureConn{}, &flowTunCapture{}
|
||||
lh := newTestLighthouse()
|
||||
lh.remoteAllowList.Store(&RemoteAllowList{})
|
||||
|
||||
f := &Interface{
|
||||
hostMap: hm,
|
||||
outside: conn,
|
||||
firewall: fw,
|
||||
lightHouse: lh,
|
||||
pki: &PKI{},
|
||||
myVpnNetworksTable: myNets,
|
||||
myVpnAddrsTable: myAddrs,
|
||||
routines: 1,
|
||||
writers: []udp.Conn{conn},
|
||||
messageMetrics: &MessageMetrics{txExhausted: metrics.NewCounter()},
|
||||
metricTxDropped: metrics.NilCounter{},
|
||||
l: l,
|
||||
}
|
||||
f.ctx = t.Context()
|
||||
f.pki.caPool.Store(cert.NewCAPool())
|
||||
// Never promote or requery, those would put extra test packets on the wire
|
||||
f.tryPromoteEvery.Store(math.MaxUint32)
|
||||
f.reQueryEvery.Store(math.MaxUint32)
|
||||
f.batchers = []*batch.MultiCoalescer{batch.NewMultiCoalescer(tun, l)}
|
||||
f.connectionManager = &connectionManager{
|
||||
intf: f,
|
||||
relayUsed: map[uint32]struct{}{},
|
||||
relayUsedLock: &sync.RWMutex{},
|
||||
trafficTimer: NewLockingTimerWheel[uint32](time.Second, time.Minute),
|
||||
}
|
||||
f.handshakeManager = NewHandshakeManager(l, hm, lh, conn, defaultHandshakeConfig)
|
||||
f.handshakeManager.f = f
|
||||
|
||||
hostinfo := &HostInfo{
|
||||
ConnectionState: ci,
|
||||
vpnAddrs: []netip.Addr{them},
|
||||
localIndexId: localIdx,
|
||||
remoteIndexId: remoteIdx,
|
||||
relayState: RelayState{relayForByAddr: map[netip.Addr]*Relay{}, relayForByIdx: map[uint32]*Relay{}},
|
||||
}
|
||||
hostinfo.remote.Store(&flowTestUnder)
|
||||
hostinfo.buildNetworks(myNets, theirCert)
|
||||
hm.unlockedAddHostInfo(hostinfo, f)
|
||||
|
||||
return &flowTestPeer{f: f, hostinfo: hostinfo, conn: conn, tun: tun}
|
||||
}
|
||||
|
||||
// flowCaptureConn records every packet sent, in order, as copies.
|
||||
type flowCaptureConn struct {
|
||||
udp.NoopConn
|
||||
mu sync.Mutex
|
||||
pkts [][]byte
|
||||
addrs []netip.AddrPort
|
||||
// delay, when set, slows every WriteBatch down so batches pile up in flight
|
||||
delay time.Duration
|
||||
}
|
||||
|
||||
func (c *flowCaptureConn) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, error) {
|
||||
time.Sleep(c.delay)
|
||||
c.mu.Lock()
|
||||
defer c.mu.Unlock()
|
||||
for i, b := range bufs {
|
||||
c.pkts = append(c.pkts, append([]byte(nil), b...))
|
||||
c.addrs = append(c.addrs, addrs[i])
|
||||
}
|
||||
return len(bufs), nil
|
||||
}
|
||||
|
||||
func (c *flowCaptureConn) WriteTo(b []byte, addr netip.AddrPort) error {
|
||||
_, err := c.WriteBatch([][]byte{b}, []netip.AddrPort{addr})
|
||||
return err
|
||||
}
|
||||
|
||||
func (c *flowCaptureConn) captured() [][]byte {
|
||||
c.mu.Lock()
|
||||
defer c.mu.Unlock()
|
||||
return append([][]byte(nil), c.pkts...)
|
||||
}
|
||||
|
||||
// flowTunCapture is a tun queue that records every plaintext write.
|
||||
type flowTunCapture struct {
|
||||
mu sync.Mutex
|
||||
writes [][]byte
|
||||
// delay, when set, slows every write down
|
||||
delay time.Duration
|
||||
}
|
||||
|
||||
func (c *flowTunCapture) Read() ([]tio.Packet, error) { return nil, io.EOF }
|
||||
func (c *flowTunCapture) Close() error { return nil }
|
||||
func (c *flowTunCapture) Write(b []byte) (int, error) {
|
||||
time.Sleep(c.delay)
|
||||
c.mu.Lock()
|
||||
defer c.mu.Unlock()
|
||||
c.writes = append(c.writes, append([]byte(nil), b...))
|
||||
return len(b), nil
|
||||
}
|
||||
|
||||
// segments decodes what reached the tun. The coalescer may have merged segments back into superpackets, so this
|
||||
// walks each payload in MSS steps.
|
||||
func (c *flowTunCapture) segments(t testing.TB) []decodedSeg {
|
||||
t.Helper()
|
||||
c.mu.Lock()
|
||||
defer c.mu.Unlock()
|
||||
var out []decodedSeg
|
||||
for _, w := range c.writes {
|
||||
pay := w[40:]
|
||||
require.Zero(t, len(pay)%flowTestMSS, "tun write payload %d bytes", len(pay))
|
||||
for off := 0; off < len(pay); off += flowTestMSS {
|
||||
out = append(out, decodeSeg(pay[off:]))
|
||||
}
|
||||
}
|
||||
return out
|
||||
}
|
||||
|
||||
// decryptWire checks every wire packet decrypts on the receiver, in order, with strictly increasing counters, and
|
||||
// returns what each one carried. It uses its own replay window and copies, so the wire can still be fed to the
|
||||
// receiver afterwards.
|
||||
func decryptWire(t testing.TB, receiver *flowTestPeer, wire [][]byte) []decodedSeg {
|
||||
t.Helper()
|
||||
peer := receiver.hostinfo.ConnectionState
|
||||
ci := &ConnectionState{dKey: peer.dKey, window: NewBits(peer.window.length)}
|
||||
nb := make([]byte, 12)
|
||||
var last uint64
|
||||
out := make([]decodedSeg, 0, len(wire))
|
||||
for i, pkt := range wire {
|
||||
h := &header.H{}
|
||||
require.NoError(t, h.Parse(pkt))
|
||||
require.Equal(t, header.Message, h.Type)
|
||||
require.Equal(t, receiver.hostinfo.localIndexId, h.RemoteIndex)
|
||||
require.Greater(t, h.MessageCounter, last, "wire packet %d: counter %d after %d", i, h.MessageCounter, last)
|
||||
last = h.MessageCounter
|
||||
plain, err := ci.Decrypt(receiver.f.l, h.MessageCounter, append([]byte(nil), pkt...), nb)
|
||||
require.NoError(t, err, "wire packet %d", i)
|
||||
out = append(out, decodeSeg(plain[40:]))
|
||||
}
|
||||
return out
|
||||
}
|
||||
|
||||
// flowBatchQueue is a tun queue that hands out scripted Read batches, then fails with io.EOF. Every batch is copied
|
||||
// into the current receive buffer, so a test sees the buffer ownership the Offload queue has.
|
||||
type flowBatchQueue struct {
|
||||
batches [][]tio.Packet
|
||||
rxBuf []byte
|
||||
out []tio.Packet
|
||||
closed chan struct{}
|
||||
// reads counts the batches handed out
|
||||
reads atomic.Int64
|
||||
}
|
||||
|
||||
func newFlowBatchQueue(batches [][]tio.Packet) *flowBatchQueue {
|
||||
return &flowBatchQueue{batches: batches, rxBuf: make([]byte, 8*65536), closed: make(chan struct{})}
|
||||
}
|
||||
|
||||
func (q *flowBatchQueue) Read() ([]tio.Packet, error) {
|
||||
select {
|
||||
case <-q.closed:
|
||||
return nil, os.ErrClosed
|
||||
default:
|
||||
}
|
||||
n := int(q.reads.Load())
|
||||
if n == len(q.batches) {
|
||||
return nil, io.EOF
|
||||
}
|
||||
q.out = q.out[:0]
|
||||
off := 0
|
||||
for _, p := range q.batches[n] {
|
||||
l := copy(q.rxBuf[off:], p.Bytes)
|
||||
q.out = append(q.out, tio.Packet{Bytes: q.rxBuf[off : off+l : off+l], GSO: p.GSO})
|
||||
off += l
|
||||
}
|
||||
q.reads.Add(1)
|
||||
return q.out, nil
|
||||
}
|
||||
|
||||
func (q *flowBatchQueue) Write(b []byte) (int, error) { return len(b), nil }
|
||||
|
||||
func (q *flowBatchQueue) Close() error { close(q.closed); return nil }
|
||||
|
||||
// delayedCipher changes scheduling through the real cipher interface, only in tests.
|
||||
type delayedCipher struct {
|
||||
noiseutil.CipherState
|
||||
before func(uint64)
|
||||
}
|
||||
|
||||
func (c delayedCipher) EncryptDanger(out, ad, plain []byte, n uint64, nb []byte) ([]byte, error) {
|
||||
c.before(n)
|
||||
return c.CipherState.EncryptDanger(out, ad, plain, n, nb)
|
||||
}
|
||||
|
||||
func (c delayedCipher) DecryptDanger(out, ad, ciphertext []byte, n uint64, nb []byte) ([]byte, error) {
|
||||
c.before(n)
|
||||
return c.CipherState.DecryptDanger(out, ad, ciphertext, n, nb)
|
||||
}
|
||||
|
||||
func delayCrypto(peer *flowTestPeer, before func(uint64)) {
|
||||
ci := peer.hostinfo.ConnectionState
|
||||
ci.eKey = delayedCipher{ci.eKey, before}
|
||||
ci.dKey = delayedCipher{ci.dKey, before}
|
||||
}
|
||||
|
||||
// stagedOn enables staging with n workers.
|
||||
func stagedOn(n int) stagedConfig {
|
||||
return stagedConfig{workers: n}
|
||||
}
|
||||
|
||||
// adaptive is staging with the inline threshold parseStagedConfig sets.
|
||||
func adaptive(n int) stagedConfig {
|
||||
return stagedConfig{workers: n, inlineBelow: 256}
|
||||
}
|
||||
|
||||
func stagedName(sc stagedConfig) string {
|
||||
switch {
|
||||
case sc.workers == 0:
|
||||
return "inline"
|
||||
case sc.inlineBelow > 0:
|
||||
return "adaptive"
|
||||
default:
|
||||
return "staged"
|
||||
}
|
||||
}
|
||||
|
||||
// TestFlowStagedFallback: a conn that cannot hand off its batches keeps receive on the inline path.
|
||||
func TestFlowStagedFallback(t *testing.T) {
|
||||
defer goleak.VerifyNone(t, goleak.IgnoreCurrent())
|
||||
sender, receiver := newFlowTestPair(t, noise.CipherChaChaPoly, stagedOn(3))
|
||||
batches, want := flowTestBatches(10, 2, 1, 4)
|
||||
wire := wireFor(t, sender, batches)
|
||||
require.Equal(t, want, decryptWire(t, receiver, wire))
|
||||
rx0 := rxBatches.Count()
|
||||
receiver.receive(plainConn{&rxConn{wire: wire, batchLen: 16}})
|
||||
require.Equal(t, want, receiver.tun.segments(t))
|
||||
require.Equal(t, rx0, rxBatches.Count())
|
||||
}
|
||||
|
||||
// wireFor produces the wire packets the sender emits for batches.
|
||||
func wireFor(t testing.TB, sender *flowTestPeer, batches [][]tio.Packet) [][]byte {
|
||||
t.Helper()
|
||||
sender.f.listenIn(newFlowBatchQueue(batches), 0)
|
||||
wire := sender.conn.captured()
|
||||
sender.conn.pkts, sender.conn.addrs = nil, nil
|
||||
return wire
|
||||
}
|
||||
|
||||
// rxConn plays wire back to listenOut through the buffer handoff, in batches of batchLen or of the scripted sizes
|
||||
// (the last one repeating), and captures what is sent back. Every buffer handed back is scribbled over before it is
|
||||
// read into, so a stage still using one shows corruption. from, when set, picks each packet's source address by its
|
||||
// index in wire.
|
||||
type rxConn struct {
|
||||
flowCaptureConn
|
||||
wire [][]byte
|
||||
batchLen int
|
||||
sizes []int
|
||||
from func(i int) netip.AddrPort
|
||||
swaps int
|
||||
}
|
||||
|
||||
func (c *rxConn) HandoffShape() (n, size int) {
|
||||
n = c.batchLen
|
||||
for _, s := range c.sizes {
|
||||
n = max(n, s)
|
||||
}
|
||||
return n, udp.MTU
|
||||
}
|
||||
|
||||
func (c *rxConn) ListenOut(r udp.EncReader, handoff func(bufs [][]byte) [][]byte) error {
|
||||
n, _ := c.HandoffShape()
|
||||
bufs := make([][]byte, n)
|
||||
for i := range bufs {
|
||||
bufs[i] = make([]byte, udp.MTU)
|
||||
}
|
||||
for off, k := 0, 0; off < len(c.wire); k++ {
|
||||
size := c.batchLen
|
||||
if len(c.sizes) > 0 {
|
||||
size = c.sizes[min(k, len(c.sizes)-1)]
|
||||
}
|
||||
end := min(off+size, len(c.wire))
|
||||
for i, p := range c.wire[off:end] {
|
||||
n := copy(bufs[i], p)
|
||||
from := flowTestUnder
|
||||
if c.from != nil {
|
||||
from = c.from(off + i)
|
||||
}
|
||||
r(from, bufs[i][:n:n])
|
||||
}
|
||||
next := handoff(bufs)
|
||||
if len(next) != len(bufs) {
|
||||
panic("handoff returned the wrong number of buffers")
|
||||
}
|
||||
bufs = next
|
||||
for _, b := range bufs {
|
||||
for j := range b[:udp.MTU] {
|
||||
b[j] = 0x5a
|
||||
}
|
||||
}
|
||||
c.swaps++
|
||||
off = end
|
||||
}
|
||||
return os.ErrClosed
|
||||
}
|
||||
|
||||
// plainConn hides HandoffShape, so receive stays inline.
|
||||
type plainConn struct{ udp.Conn }
|
||||
|
||||
var rxBatches = metrics.GetOrRegisterCounter("flow.rx.batches", nil)
|
||||
|
||||
// receive runs the peer's listenOut over conn until conn runs out.
|
||||
func (p *flowTestPeer) receive(conn udp.Conn) {
|
||||
p.f.writers = []udp.Conn{conn}
|
||||
p.f.outside = conn
|
||||
p.f.listenOut(0)
|
||||
}
|
||||
|
||||
// TestFlowStagedFirewallParity: both receive paths apply the same inbound policy. An authenticated inbound denial
|
||||
// still consumes its replay counter, exactly as inline Decrypt does.
|
||||
func TestFlowStagedFirewallParity(t *testing.T) {
|
||||
for _, sc := range []stagedConfig{{}, stagedOn(2)} {
|
||||
t.Run(stagedName(sc), func(t *testing.T) {
|
||||
batches, _ := flowTestBatches(4, 3, 1, 8)
|
||||
sender, receiver := newFlowTestPair(t, noise.CipherChaChaPoly, sc)
|
||||
wire := wireFor(t, sender, batches)
|
||||
receiver.f.firewall.InRules = newFirewallTable()
|
||||
receiver.receive(&rxConn{wire: wire, batchLen: 64})
|
||||
require.Empty(t, receiver.tun.writes)
|
||||
for _, pkt := range wire {
|
||||
h := &header.H{}
|
||||
require.NoError(t, h.Parse(pkt))
|
||||
require.False(t, receiver.hostinfo.ConnectionState.window.Check(receiver.f.l, h.MessageCounter))
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
// TestFlowStagedAESGCM runs the parsed configuration with the real AES cipher. Receive decryption has no nonce
|
||||
// ordering requirement, so it stays staged in BoringCrypto and FIPS builds too.
|
||||
func TestFlowStagedAESGCM(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
c := config.NewC(l)
|
||||
c.Settings = map[string]any{"tun": map[string]any{"staged": map[string]any{"workers": 2}}}
|
||||
sender, receiver := newFlowTestPair(t, noiseutil.CipherAESGCM, parseStagedConfig(c, l))
|
||||
require.Equal(t, 2, receiver.f.staged.workers)
|
||||
batches, want := flowTestBatches(4, 8, 1, 45)
|
||||
wire := wireFor(t, sender, batches)
|
||||
rx0 := rxBatches.Count()
|
||||
receiver.receive(&rxConn{wire: wire, batchLen: 512})
|
||||
require.Greater(t, rxBatches.Count(), rx0)
|
||||
require.Equal(t, want, receiver.tun.segments(t))
|
||||
}
|
||||
+18
-5
@@ -31,6 +31,7 @@ package nebula
|
||||
|
||||
import (
|
||||
"bufio"
|
||||
"context"
|
||||
"fmt"
|
||||
"net"
|
||||
"strconv"
|
||||
@@ -51,17 +52,29 @@ type graphiteConfigExport struct {
|
||||
Percentiles []float64 // Percentiles to export from timers and histograms
|
||||
}
|
||||
|
||||
// graphiteTimeout bounds one submission, the connect and the write together. An export is a few KB, a host that can't
|
||||
// take that in this long isn't going to.
|
||||
const graphiteTimeout = 30 * time.Second
|
||||
|
||||
// graphiteOnce performs a single submission to Graphite, returning a non-nil
|
||||
// error on failed connections.
|
||||
func graphiteOnce(c graphiteConfigExport) error {
|
||||
// error if it could not connect or write all of it within timeout. It gives up
|
||||
// as soon as ctx is done.
|
||||
func graphiteOnce(ctx context.Context, c graphiteConfigExport, timeout time.Duration) error {
|
||||
now := time.Now().Unix()
|
||||
du := float64(c.DurationUnit)
|
||||
flushSeconds := float64(c.FlushInterval) / float64(time.Second)
|
||||
conn, err := net.DialTCP("tcp", nil, c.Addr)
|
||||
deadline := time.Now().Add(timeout)
|
||||
d := net.Dialer{Deadline: deadline}
|
||||
conn, err := d.DialContext(ctx, "tcp", c.Addr.String())
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
defer conn.Close()
|
||||
if err := conn.SetDeadline(deadline); err != nil {
|
||||
return err
|
||||
}
|
||||
stop := context.AfterFunc(ctx, func() { _ = conn.Close() })
|
||||
defer stop()
|
||||
w := bufio.NewWriter(conn)
|
||||
c.Registry.Each(func(name string, i any) {
|
||||
switch metric := i.(type) {
|
||||
@@ -111,7 +124,7 @@ func graphiteOnce(c graphiteConfigExport) error {
|
||||
fmt.Fprintf(w, "%s.%s.fifteen-minute %.2f %d\n", c.Prefix, name, t.Rate15(), now)
|
||||
fmt.Fprintf(w, "%s.%s.mean-rate %.2f %d\n", c.Prefix, name, t.RateMean(), now)
|
||||
}
|
||||
w.Flush()
|
||||
})
|
||||
return nil
|
||||
// A failed write sticks, so this reports the first one
|
||||
return w.Flush()
|
||||
}
|
||||
@@ -0,0 +1,315 @@
|
||||
package nebula
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"context"
|
||||
"errors"
|
||||
"fmt"
|
||||
"io"
|
||||
"net"
|
||||
"strings"
|
||||
"sync"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
// graphiteHost accepts connections and hands each one to the test
|
||||
type graphiteHost struct {
|
||||
ln net.Listener
|
||||
conns chan net.Conn
|
||||
|
||||
mu sync.Mutex
|
||||
closed bool
|
||||
open []net.Conn
|
||||
}
|
||||
|
||||
// newGraphiteHost accepts connections for the test. A non-zero window shrinks each connection's receive buffer, so a
|
||||
// sender that is not read from blocks after a little whatever the host's TCP tuning. Only for tests that need a stall,
|
||||
// a small window makes reading very slow on Linux.
|
||||
func newGraphiteHost(t *testing.T, window int) *graphiteHost {
|
||||
t.Helper()
|
||||
ln, err := net.Listen("tcp", "127.0.0.1:0")
|
||||
require.NoError(t, err)
|
||||
h := &graphiteHost{ln: ln, conns: make(chan net.Conn, 8)}
|
||||
done := make(chan struct{})
|
||||
t.Cleanup(func() {
|
||||
_ = ln.Close()
|
||||
h.mu.Lock()
|
||||
h.closed = true
|
||||
for _, c := range h.open {
|
||||
_ = c.Close()
|
||||
}
|
||||
h.mu.Unlock()
|
||||
close(done)
|
||||
})
|
||||
go func() {
|
||||
for {
|
||||
c, err := ln.Accept()
|
||||
if err != nil {
|
||||
return
|
||||
}
|
||||
h.mu.Lock()
|
||||
if h.closed {
|
||||
h.mu.Unlock()
|
||||
_ = c.Close()
|
||||
return
|
||||
}
|
||||
h.open = append(h.open, c)
|
||||
h.mu.Unlock()
|
||||
if tc, ok := c.(*net.TCPConn); ok && window > 0 {
|
||||
_ = tc.SetReadBuffer(window)
|
||||
}
|
||||
select {
|
||||
case h.conns <- c:
|
||||
case <-done:
|
||||
return
|
||||
}
|
||||
}
|
||||
}()
|
||||
return h
|
||||
}
|
||||
|
||||
func (h *graphiteHost) addr(t *testing.T) *net.TCPAddr {
|
||||
a, err := net.ResolveTCPAddr("tcp", h.ln.Addr().String())
|
||||
require.NoError(t, err)
|
||||
return a
|
||||
}
|
||||
|
||||
// readExport reads what one send wrote, bounded so a send that never finishes fails the test rather than hanging it
|
||||
func readExport(t *testing.T, c net.Conn) []byte {
|
||||
t.Helper()
|
||||
require.NoError(t, c.SetReadDeadline(time.Now().Add(10*time.Second)))
|
||||
b, err := io.ReadAll(c)
|
||||
require.NoError(t, err)
|
||||
return b
|
||||
}
|
||||
|
||||
func (h *graphiteHost) next(t *testing.T, wait time.Duration) net.Conn {
|
||||
t.Helper()
|
||||
select {
|
||||
case c := <-h.conns:
|
||||
return c
|
||||
case <-time.After(wait):
|
||||
t.Fatal("no connection")
|
||||
return nil
|
||||
}
|
||||
}
|
||||
|
||||
// lockedBuffer is a log destination a test can read while the code under test writes to it from another goroutine
|
||||
type lockedBuffer struct {
|
||||
mu sync.Mutex
|
||||
buf bytes.Buffer
|
||||
}
|
||||
|
||||
func (b *lockedBuffer) Write(p []byte) (int, error) {
|
||||
b.mu.Lock()
|
||||
defer b.mu.Unlock()
|
||||
return b.buf.Write(p)
|
||||
}
|
||||
|
||||
func (b *lockedBuffer) String() string {
|
||||
b.mu.Lock()
|
||||
defer b.mu.Unlock()
|
||||
return b.buf.String()
|
||||
}
|
||||
|
||||
func testGraphiteConfig(addr *net.TCPAddr, r metrics.Registry) graphiteConfigExport {
|
||||
return graphiteConfigExport{Addr: addr, Registry: r, FlushInterval: time.Second, DurationUnit: time.Nanosecond, Prefix: "nebula"}
|
||||
}
|
||||
|
||||
// Enough metrics that one export overflows the socket buffers on any platform, so a send to a host that stops
|
||||
// reading blocks
|
||||
func bigGraphiteRegistry(r metrics.Registry, t *testing.T) {
|
||||
for i := range 200000 {
|
||||
name := fmt.Sprintf("big.gauge.%d", i)
|
||||
metrics.GetOrRegisterGauge(name, r).Update(int64(i))
|
||||
t.Cleanup(func() { r.Unregister(name) })
|
||||
}
|
||||
}
|
||||
|
||||
func TestGraphiteOnce_sends(t *testing.T) {
|
||||
r := metrics.NewRegistry()
|
||||
metrics.GetOrRegisterGauge("g", r).Update(7)
|
||||
metrics.GetOrRegisterCounter("c", r).Inc(3)
|
||||
h := newGraphiteHost(t, 0)
|
||||
|
||||
require.NoError(t, graphiteOnce(t.Context(), testGraphiteConfig(h.addr(t), r), time.Minute))
|
||||
b := readExport(t, h.next(t, 5*time.Second))
|
||||
assert.Contains(t, string(b), "nebula.g.value 7 ")
|
||||
assert.Contains(t, string(b), "nebula.c.count 3 ")
|
||||
assert.Contains(t, string(b), "nebula.c.count_ps 3.00 ")
|
||||
}
|
||||
|
||||
// A host that accepts and never reads is given up on, and the failed write is reported
|
||||
func TestGraphiteOnce_stalledHostTimesOut(t *testing.T) {
|
||||
r := metrics.NewRegistry()
|
||||
bigGraphiteRegistry(r, t)
|
||||
h := newGraphiteHost(t, 4096)
|
||||
done := make(chan error, 1)
|
||||
cfg := testGraphiteConfig(h.addr(t), r)
|
||||
go func() { done <- graphiteOnce(t.Context(), cfg, 300*time.Millisecond) }()
|
||||
h.next(t, 5*time.Second) // accepted, never read
|
||||
|
||||
select {
|
||||
case err := <-done:
|
||||
require.Error(t, err)
|
||||
case <-time.After(10 * time.Second):
|
||||
t.Fatal("the send did not give up on a host that stopped reading")
|
||||
}
|
||||
}
|
||||
|
||||
// The timeout covers the whole send, a host that keeps reading but too slowly is still given up on
|
||||
func TestGraphiteOnce_timeoutCoversTheWholeSend(t *testing.T) {
|
||||
r := metrics.NewRegistry()
|
||||
bigGraphiteRegistry(r, t)
|
||||
h := newGraphiteHost(t, 0)
|
||||
done := make(chan error, 1)
|
||||
start := time.Now()
|
||||
cfg := testGraphiteConfig(h.addr(t), r)
|
||||
go func() { done <- graphiteOnce(t.Context(), cfg, 500*time.Millisecond) }()
|
||||
c := h.next(t, 5*time.Second)
|
||||
stopReading := make(chan struct{})
|
||||
defer close(stopReading)
|
||||
// 32KB every 10ms keeps every write well inside the timeout, the whole export takes seconds
|
||||
go func() {
|
||||
buf := make([]byte, 32<<10)
|
||||
for {
|
||||
select {
|
||||
case <-stopReading:
|
||||
return
|
||||
case <-time.After(10 * time.Millisecond):
|
||||
}
|
||||
if _, err := c.Read(buf); err != nil {
|
||||
return
|
||||
}
|
||||
}
|
||||
}()
|
||||
|
||||
select {
|
||||
case err := <-done:
|
||||
require.Error(t, err)
|
||||
assert.Less(t, time.Since(start), 5*time.Second)
|
||||
case <-time.After(10 * time.Second):
|
||||
t.Fatal("a host that kept reading slowly held the send past its timeout")
|
||||
}
|
||||
}
|
||||
|
||||
func TestGraphiteOnce_cancelledCtxDoesNotConnect(t *testing.T) {
|
||||
h := newGraphiteHost(t, 0)
|
||||
ctx, cancel := context.WithCancel(t.Context())
|
||||
cancel()
|
||||
require.Error(t, graphiteOnce(ctx, testGraphiteConfig(h.addr(t), metrics.NewRegistry()), time.Minute))
|
||||
select {
|
||||
case <-h.conns:
|
||||
t.Fatal("connected on a cancelled ctx")
|
||||
case <-time.After(200 * time.Millisecond):
|
||||
}
|
||||
}
|
||||
|
||||
// A stop or reload abandons a stuck send rather than waiting out its timeout
|
||||
func TestGraphiteOnce_cancelAbandonsSend(t *testing.T) {
|
||||
r := metrics.NewRegistry()
|
||||
bigGraphiteRegistry(r, t)
|
||||
h := newGraphiteHost(t, 4096)
|
||||
ctx, cancel := context.WithCancel(t.Context())
|
||||
done := make(chan error, 1)
|
||||
cfg := testGraphiteConfig(h.addr(t), r)
|
||||
go func() { done <- graphiteOnce(ctx, cfg, time.Minute) }()
|
||||
h.next(t, 5*time.Second)
|
||||
select {
|
||||
case err := <-done:
|
||||
t.Fatalf("the send should be stuck on a host that is not reading, it ended with %v", err)
|
||||
case <-time.After(200 * time.Millisecond):
|
||||
}
|
||||
|
||||
cancel()
|
||||
select {
|
||||
case err := <-done:
|
||||
require.Error(t, err)
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("the send is still stuck after its ctx was cancelled")
|
||||
}
|
||||
}
|
||||
|
||||
// buildRuntime hands graphite the configured prefix and interval
|
||||
func TestStatsServer_graphiteExportUsesConfig(t *testing.T) {
|
||||
const name = "graphite.wiring.test"
|
||||
metrics.GetOrRegisterCounter(name, nil).Inc(8)
|
||||
t.Cleanup(func() { metrics.DefaultRegistry.Unregister(name) })
|
||||
h := newGraphiteHost(t, 0)
|
||||
s, c := newTestStatsServer(t)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "4s", "prefix": "pfx", "host": h.ln.Addr().String()})
|
||||
cfg, err := loadStatsConfig(c)
|
||||
require.NoError(t, err)
|
||||
|
||||
s.runMu.Lock()
|
||||
fns, _ := s.buildRuntime(t.Context(), cfg)
|
||||
s.runMu.Unlock()
|
||||
for _, fn := range fns {
|
||||
fn()
|
||||
}
|
||||
b := readExport(t, h.next(t, 5*time.Second))
|
||||
assert.Contains(t, string(b), "pfx."+name+".count 8 ")
|
||||
assert.Contains(t, string(b), "pfx."+name+".count_ps 2.00 ")
|
||||
}
|
||||
|
||||
func TestStatsServer_graphiteFailureIsLogged(t *testing.T) {
|
||||
logs := &lockedBuffer{}
|
||||
s, c := newTestStatsServer(t)
|
||||
s.l = test.NewLoggerWithOutput(logs)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "1h", "host": "127.0.0.1:" + freeTCPPort(t)})
|
||||
cfg, err := loadStatsConfig(c)
|
||||
require.NoError(t, err)
|
||||
|
||||
s.runMu.Lock()
|
||||
fns, _ := s.buildRuntime(t.Context(), cfg)
|
||||
s.runMu.Unlock()
|
||||
for _, fn := range fns {
|
||||
fn()
|
||||
}
|
||||
assert.Contains(t, logs.String(), "Graphite export failed")
|
||||
}
|
||||
|
||||
// Stop abandons the running export, so it goes through Start rather than handing buildRuntime a ctx, and a stop is
|
||||
// not logged as a failure
|
||||
func TestStatsServer_stopAbandonsGraphiteSend(t *testing.T) {
|
||||
bigGraphiteRegistry(metrics.DefaultRegistry, t)
|
||||
h := newGraphiteHost(t, 4096)
|
||||
logs := &lockedBuffer{}
|
||||
s, c := newTestStatsServer(t)
|
||||
s.l = test.NewLoggerWithOutput(logs)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "50ms", "host": h.ln.Addr().String()})
|
||||
require.NoError(t, s.reload(c, true))
|
||||
|
||||
started := make(chan struct{})
|
||||
go func() {
|
||||
s.Start()
|
||||
close(started)
|
||||
}()
|
||||
stuck := h.next(t, 30*time.Second)
|
||||
s.Stop()
|
||||
select {
|
||||
case <-started:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("Start did not return after Stop")
|
||||
}
|
||||
|
||||
// A closed sender resets the connection once it sees this, one still sending just buffers it. Either way the
|
||||
// read below doesn't have to drain the export through the small window to find out
|
||||
_, err := stuck.Write([]byte("x"))
|
||||
require.NoError(t, err)
|
||||
require.NoError(t, stuck.SetReadDeadline(time.Now().Add(10*time.Second)))
|
||||
n, err := io.Copy(io.Discard, stuck)
|
||||
var ne net.Error
|
||||
if errors.As(err, &ne) && ne.Timeout() {
|
||||
t.Fatalf("the send kept the connection after Stop, read %d bytes", n)
|
||||
}
|
||||
assert.Less(t, n, int64(1<<20), "the send kept going after Stop")
|
||||
// The capture loop logs on its own goroutine after Start has returned, give it the chance
|
||||
assert.Never(t, func() bool { return strings.Contains(logs.String(), "Graphite export failed") }, 500*time.Millisecond, 10*time.Millisecond, "a stop is not a failure")
|
||||
}
|
||||
+9
-12
@@ -54,7 +54,8 @@ type HandshakeConfig struct {
|
||||
}
|
||||
|
||||
type HandshakeManager struct {
|
||||
// Mutex for interacting with the vpnIps and indexes maps
|
||||
// Mutex for interacting with the vpnIps and indexes maps.
|
||||
// Lock order is hostmap, then this lock, then the lighthouse, then a RemoteList.
|
||||
sync.RWMutex
|
||||
|
||||
vpnIps map[netip.Addr]*HandshakeHostInfo
|
||||
@@ -259,14 +260,6 @@ func (hm *HandshakeManager) handleOutbound(vpnIp netip.Addr, lighthouseTriggered
|
||||
"style": header.SubTypeName(header.Handshake, hh.machine.Subtype()),
|
||||
}
|
||||
|
||||
// Get a remotes object if we don't already have one.
|
||||
// This is mainly to protect us as this should never be the case
|
||||
// NB ^ This comment doesn't jive. It's how the thing gets initialized.
|
||||
// It's the common path. Should it update every time, in case a future LH query/queries give us more info?
|
||||
if hostinfo.remotes == nil {
|
||||
hostinfo.remotes = hm.lightHouse.QueryCache([]netip.Addr{vpnIp})
|
||||
}
|
||||
|
||||
remotes := hostinfo.remotes.CopyAddrs(hm.mainHostMap.GetPreferredRanges())
|
||||
remotesHaveChanged := !slices.Equal(remotes, hh.lastRemotes)
|
||||
|
||||
@@ -371,6 +364,9 @@ func (hm *HandshakeManager) StartHandshake(vpnAddr netip.Addr, cacheCb func(*Han
|
||||
hostinfo := &HostInfo{
|
||||
vpnAddrs: []netip.Addr{vpnAddr},
|
||||
HandshakePacket: make(map[uint8][]byte, 0),
|
||||
// Attached here so a pending hostinfo never exists without one: SetRemote dereferences
|
||||
// it, and create-tunnel -address can reach SetRemote before the handshake loop runs.
|
||||
remotes: hm.lightHouse.QueryCache([]netip.Addr{vpnAddr}),
|
||||
relayState: RelayState{
|
||||
relays: nil,
|
||||
relayForByAddr: map[netip.Addr]*Relay{},
|
||||
@@ -946,13 +942,14 @@ func (hm *HandshakeManager) continueHandshake(via ViaSender, hh *HandshakeHostIn
|
||||
|
||||
hm.DeleteHostInfo(hostinfo)
|
||||
hm.StartHandshake(hostinfo.vpnAddrs[0], func(newHH *HandshakeHostInfo) {
|
||||
newHH.hostinfo.remotes = hostinfo.remotes
|
||||
newHH.hostinfo.remotes.BlockRemote(via)
|
||||
newHH.packetStore = hh.packetStore
|
||||
hh.packetStore = []*cachedPacket{}
|
||||
hostinfo.vpnAddrs = vpnAddrs
|
||||
f.sendCloseTunnel(hostinfo)
|
||||
})
|
||||
// sendCloseTunnel may take the hostmap lock, so it must not run under the handshake
|
||||
// manager lock that StartHandshake holds around the callback.
|
||||
hostinfo.vpnAddrs = vpnAddrs
|
||||
f.sendCloseTunnel(hostinfo)
|
||||
return
|
||||
}
|
||||
|
||||
|
||||
@@ -11,6 +11,7 @@ import (
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
func Test_NewHandshakeManagerVpnIp(t *testing.T) {
|
||||
@@ -42,8 +43,6 @@ func Test_NewHandshakeManagerVpnIp(t *testing.T) {
|
||||
i2 := blah.StartHandshake(ip, nil)
|
||||
assert.Same(t, i, i2)
|
||||
|
||||
i.remotes = NewRemoteList([]netip.Addr{}, nil)
|
||||
|
||||
// Adding something to pending should not affect the main hostmap
|
||||
assert.Empty(t, mainHM.Hosts)
|
||||
|
||||
@@ -235,3 +234,70 @@ func TestHandleIncomingDispatch(t *testing.T) {
|
||||
assert.Empty(t, hm.indexes, "orphan stage-2 must not create state")
|
||||
})
|
||||
}
|
||||
|
||||
// Test_PendingHandshakeHasRemotes covers create-tunnel -address: a pending hostinfo must already
|
||||
// carry its RemoteList when StartHandshake runs its callback, so SetRemote there, before the
|
||||
// handshake loop has run, is safe and the address becomes one the handshake is sent to.
|
||||
func Test_PendingHandshakeHasRemotes(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
mainHM := newHostMap(l)
|
||||
preferredRanges := []netip.Prefix{}
|
||||
mainHM.preferredRanges.Store(&preferredRanges)
|
||||
|
||||
hm := NewHandshakeManager(l, mainHM, newTestLighthouse(), &udp.NoopConn{}, defaultHandshakeConfig)
|
||||
|
||||
vpnIp := netip.MustParseAddr("172.1.1.2")
|
||||
remote := netip.MustParseAddrPort("192.168.1.9:4242")
|
||||
|
||||
hi := hm.StartHandshake(vpnIp, func(hh *HandshakeHostInfo) {
|
||||
require.NotNil(t, hh.hostinfo.remotes, "a pending handshake must never be without its RemoteList")
|
||||
hh.hostinfo.SetRemote(remote)
|
||||
})
|
||||
assert.Contains(t, hi.remotes.CopyAddrs(preferredRanges), remote, "the handshake must be sent to the supplied remote")
|
||||
assert.Equal(t, remote, hi.GetRemote())
|
||||
|
||||
// A RemoteList keeps one learned address per host, so a second remote replaces the
|
||||
// first — the same behaviour change-remote has on an established tunnel.
|
||||
other := netip.MustParseAddrPort("192.168.1.10:4242")
|
||||
hi.SetRemote(other)
|
||||
assert.Equal(t, []netip.AddrPort{other}, hi.remotes.CopyAddrs(preferredRanges))
|
||||
assert.Equal(t, other, hi.GetRemote())
|
||||
}
|
||||
|
||||
// newRemotesRaceFixture builds a handshake manager with an established tunnel to vpnAddr whose
|
||||
// RemoteList is the lighthouse's cached list, the state every teardown-during-handshake race starts from.
|
||||
func newRemotesRaceFixture(t *testing.T, vpnAddr netip.Addr) (*HandshakeManager, *Interface, *HostInfo) {
|
||||
t.Helper()
|
||||
l := test.NewLogger()
|
||||
mainHM := newHostMap(l)
|
||||
preferredRanges := []netip.Prefix{}
|
||||
mainHM.preferredRanges.Store(&preferredRanges)
|
||||
|
||||
lh := newTestLighthouse()
|
||||
hm := NewHandshakeManager(l, mainHM, lh, &udp.NoopConn{}, defaultHandshakeConfig)
|
||||
f := &Interface{hostMap: mainHM, lightHouse: lh, handshakeManager: hm, l: l}
|
||||
hm.f = f
|
||||
|
||||
established := &HostInfo{
|
||||
vpnAddrs: []netip.Addr{vpnAddr},
|
||||
localIndexId: 1,
|
||||
remoteIndexId: 101,
|
||||
remotes: lh.QueryCache([]netip.Addr{vpnAddr}),
|
||||
HandshakePacket: map[uint8][]byte{},
|
||||
}
|
||||
mainHM.Lock()
|
||||
mainHM.unlockedAddHostInfo(established, f)
|
||||
mainHM.Unlock()
|
||||
|
||||
return hm, f, established
|
||||
}
|
||||
|
||||
// With no handshake pending, tearing down the last tunnel still evicts the list.
|
||||
func Test_CloseTunnelEvictsWithoutPendingHandshake(t *testing.T) {
|
||||
vpnAddr := netip.MustParseAddr("172.1.1.2")
|
||||
_, f, established := newRemotesRaceFixture(t, vpnAddr)
|
||||
|
||||
f.closeTunnel(established)
|
||||
assert.NotContains(t, f.hostMap.Hosts, vpnAddr)
|
||||
assert.NotContains(t, f.lightHouse.addrMap, vpnAddr)
|
||||
}
|
||||
@@ -24,6 +24,11 @@ type m = map[string]any
|
||||
const (
|
||||
Version uint8 = 1
|
||||
Len = 16
|
||||
|
||||
// MaxOverhead is the most a payload grows on its way to the underlay.
|
||||
// A relayed packet has two headers and two AEAD tags, one for the inner tunnel and another for the relay tunnel.
|
||||
// Every supported cipher (today) has a 16 byte tag.
|
||||
MaxOverhead = 2 * (Len + 16)
|
||||
)
|
||||
|
||||
type MessageType uint8
|
||||
|
||||
@@ -10,6 +10,7 @@ import (
|
||||
"github.com/slackhq/nebula/firewall"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/iputil"
|
||||
"github.com/slackhq/nebula/logging"
|
||||
"github.com/slackhq/nebula/noiseutil"
|
||||
"github.com/slackhq/nebula/overlay/batch"
|
||||
"github.com/slackhq/nebula/overlay/tio"
|
||||
@@ -198,8 +199,8 @@ func (f *Interface) sendInsideMessage(hostinfo *HostInfo, pkt tio.Packet, nb []b
|
||||
}
|
||||
|
||||
err = tio.SegmentSuperpacket(pkt, func(seg []byte) error {
|
||||
//relay header + header + plaintext + AEAD tag (16 bytes for both AES-GCM and ChaCha20-Poly1305) + relay tag
|
||||
scratch := sendBatch.Reserve(header.Len + header.Len + len(seg) + 16 + 16)
|
||||
//relay header + header + plaintext + AEAD tag + relay tag
|
||||
scratch := sendBatch.Reserve(len(seg) + header.MaxOverhead)
|
||||
|
||||
innerPacket := f.sendInsideEncrypt(hostinfo, ci, seg, scratch[header.Len:], nb)
|
||||
if innerPacket == nil {
|
||||
@@ -517,7 +518,9 @@ func (f *Interface) SendVia(via *HostInfo, relay *Relay, ad, nb, out []byte, noc
|
||||
|
||||
err = f.writers[q].WriteTo(toSend, via.GetRemote())
|
||||
if err != nil {
|
||||
via.logger(f.l).Info("Failed to WriteTo in sendVia", "error", err)
|
||||
if l, ok := logging.RateLimited(f.l, f.udpRelayTxErrLimiter); ok {
|
||||
via.logger(l).Info("Failed to WriteTo in sendVia", "error", err)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -526,16 +529,26 @@ func (f *Interface) sendNoMetrics(t header.MessageType, st header.MessageSubType
|
||||
return
|
||||
}
|
||||
useRelay := !remote.IsValid() && !hostinfo.GetRemote().IsValid()
|
||||
fullOut := out
|
||||
|
||||
// The encrypt appends to out in place and a relayed packet is then wrapped in place,
|
||||
// so out has to fit the whole packet up front or the encrypt reallocates out from under SendVia.
|
||||
// We cap the MTU of tun devices and routes we create ourselves, so only a host that set a bigger MTU than we allow gets here.
|
||||
need := header.Len + len(p) + ci.eKey.Overhead()
|
||||
if useRelay {
|
||||
if len(out) < header.Len {
|
||||
// out always has a capacity of mtu, but not always a length greater than the header.Len.
|
||||
// Grow it to make sure the next operation works.
|
||||
out = out[:header.Len]
|
||||
// The relay tunnel runs the same cipher, and adds its own header and tag
|
||||
need += header.Len + ci.eKey.Overhead()
|
||||
}
|
||||
if cap(out) < need {
|
||||
if l, ok := logging.RateLimited(f.l, f.txOversizeLimiter); ok {
|
||||
hostinfo.logger(l).Error("Dropping outbound packet, out buffer not large enough", "outCap", cap(out), "payloadLen", len(p), "relayed", useRelay)
|
||||
}
|
||||
return
|
||||
}
|
||||
|
||||
fullOut := out
|
||||
if useRelay {
|
||||
// Save a header's worth of data at the front of the 'out' buffer.
|
||||
out = out[header.Len:]
|
||||
out = out[header.Len:header.Len]
|
||||
}
|
||||
|
||||
if noiseutil.EncryptLockNeeded {
|
||||
@@ -582,18 +595,22 @@ func (f *Interface) sendNoMetrics(t header.MessageType, st header.MessageSubType
|
||||
if remote.IsValid() {
|
||||
err = f.writers[q].WriteTo(out, remote)
|
||||
if err != nil {
|
||||
hostinfo.logger(f.l).Error("Failed to write outgoing packet",
|
||||
"error", err,
|
||||
"udpAddr", remote,
|
||||
)
|
||||
if l, ok := logging.RateLimited(f.l, f.udpTxErrLimiter); ok {
|
||||
hostinfo.logger(l).Error("Failed to write outgoing packet",
|
||||
"error", err,
|
||||
"udpAddr", remote,
|
||||
)
|
||||
}
|
||||
}
|
||||
} else if hr := hostinfo.GetRemote(); hr.IsValid() {
|
||||
err = f.writers[q].WriteTo(out, hr)
|
||||
if err != nil {
|
||||
hostinfo.logger(f.l).Error("Failed to write outgoing packet",
|
||||
"error", err,
|
||||
"udpAddr", hr,
|
||||
)
|
||||
if l, ok := logging.RateLimited(f.l, f.udpTxErrLimiter); ok {
|
||||
hostinfo.logger(l).Error("Failed to write outgoing packet",
|
||||
"error", err,
|
||||
"udpAddr", hr,
|
||||
)
|
||||
}
|
||||
}
|
||||
} else {
|
||||
// Try to send via a relay
|
||||
|
||||
+42
-24
@@ -24,6 +24,7 @@ import (
|
||||
"github.com/slackhq/nebula/overlay"
|
||||
"github.com/slackhq/nebula/overlay/batch"
|
||||
"github.com/slackhq/nebula/overlay/tio"
|
||||
"github.com/slackhq/nebula/ratelimit"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
)
|
||||
|
||||
@@ -64,6 +65,7 @@ type InterfaceConfig struct {
|
||||
// goroutine's sendmmsg on one XPS-selected NIC TX ring so per-flow
|
||||
// packets stay ordered on the wire.
|
||||
PinThreads bool
|
||||
staged stagedConfig
|
||||
|
||||
l *slog.Logger
|
||||
}
|
||||
@@ -98,6 +100,7 @@ type Interface struct {
|
||||
// a CPU at all (tun.pin_threads, default true). When false, threads are
|
||||
// left free to migrate as on stock nebula.
|
||||
pinThreads bool
|
||||
staged stagedConfig
|
||||
relayManager *relayManager
|
||||
|
||||
tryPromoteEvery atomic.Uint32
|
||||
@@ -124,6 +127,10 @@ type Interface struct {
|
||||
batchers []*batch.MultiCoalescer
|
||||
wg sync.WaitGroup
|
||||
|
||||
udpTxErrLimiter *ratelimit.Limiter // udp direct transmit error log limiter
|
||||
udpRelayTxErrLimiter *ratelimit.Limiter // udp relay transmit error log limiter
|
||||
txOversizeLimiter *ratelimit.Limiter // oversized outbound packet drop log limiter
|
||||
|
||||
// fatalErr holds the first unexpected reader error that caused shutdown.
|
||||
// nil means "no fatal error" (yet)
|
||||
fatalErr atomic.Pointer[error]
|
||||
@@ -198,6 +205,19 @@ func NewInterface(ctx context.Context, c *InterfaceConfig) (*Interface, error) {
|
||||
return nil, errors.New("no connection manager")
|
||||
}
|
||||
|
||||
udpTxErrLimiter, err := ratelimit.New(1, time.Second)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
udpRelayTxErrLimiter, err := ratelimit.New(1, time.Second)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
txOversizeLimiter, err := ratelimit.New(1, time.Second)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
if c.routines <= 1 {
|
||||
c.PinThreads = false //pinning is not useful unless there's more than one tun reader
|
||||
}
|
||||
@@ -230,6 +250,10 @@ func NewInterface(ctx context.Context, c *InterfaceConfig) (*Interface, error) {
|
||||
conntrackCacheTimeout: c.ConntrackCacheTimeout,
|
||||
cpuAffinity: c.CpuAffinity,
|
||||
pinThreads: c.PinThreads,
|
||||
staged: c.staged,
|
||||
udpTxErrLimiter: udpTxErrLimiter,
|
||||
udpRelayTxErrLimiter: udpRelayTxErrLimiter,
|
||||
txOversizeLimiter: txOversizeLimiter,
|
||||
|
||||
metricHandshakes: metrics.GetOrRegisterHistogram("handshakes", nil, metrics.NewExpDecaySample(1028, 0.015)),
|
||||
metricTxDropped: metrics.GetOrRegisterCounter("udp.tx.dropped", nil),
|
||||
@@ -363,6 +387,8 @@ type rxContext struct {
|
||||
hostmapCache map[uint32]*HostInfo
|
||||
lhh *LightHouseHandler
|
||||
ctCache *firewall.ConntrackCacheTicker
|
||||
// decrypted is staged receive's result for the packet being handled, decrypted ahead of time on a worker
|
||||
decrypted *rxItem
|
||||
}
|
||||
|
||||
func newRxContext(f *Interface, q int) *rxContext {
|
||||
@@ -392,14 +418,21 @@ func (f *Interface) listenOut(i int) {
|
||||
f.readOutsidePackets(ViaSender{UdpAddr: fromUdpAddr}, payload, rxc)
|
||||
}
|
||||
|
||||
flusher := func() {
|
||||
if err := f.batchers[i].Flush(); err != nil {
|
||||
f.l.Error("Failed to flush tun coalescer", "error", err)
|
||||
}
|
||||
clear(rxc.hostmapCache)
|
||||
flusher := func(bufs [][]byte) [][]byte {
|
||||
f.flushOutside(rxc)
|
||||
return bufs
|
||||
}
|
||||
|
||||
err := li.ListenOut(listener, flusher)
|
||||
var err error
|
||||
if hc, ok := li.(handoffConn); ok && f.staged.workers > 0 {
|
||||
// Every stage on its own goroutine, see flow_rx.go. rxc becomes the ordered writer's context.
|
||||
err = newStagedRx(f, i, rxc, hc).listen()
|
||||
} else {
|
||||
if f.staged.workers > 0 {
|
||||
f.l.Info("udp conn cannot hand off receive batches, receive stays inline", "reader", i)
|
||||
}
|
||||
err = li.ListenOut(listener, flusher)
|
||||
}
|
||||
|
||||
// An error after teardown began is shutdown noise, the closed flag covers resources
|
||||
// Close releases itself and the cancelled ctx covers ones torn down by their owners
|
||||
@@ -624,17 +657,15 @@ func (f *Interface) reloadMisc(c *config.C) {
|
||||
}
|
||||
}
|
||||
|
||||
func (f *Interface) emitStats(ctx context.Context, i time.Duration) {
|
||||
ticker := time.NewTicker(i)
|
||||
defer ticker.Stop()
|
||||
|
||||
// statsEmitter updates the interface's gauges, the stats server runs it on its capture loop.
|
||||
func (f *Interface) statsEmitter() func() {
|
||||
udpStats := udp.NewUDPStatsEmitter(f.writers)
|
||||
|
||||
certExpirationGauge := metrics.GetOrRegisterGauge("certificate.ttl_seconds", nil)
|
||||
certInitiatingVersion := metrics.GetOrRegisterGauge("certificate.initiating_version", nil)
|
||||
certMaxVersion := metrics.GetOrRegisterGauge("certificate.max_version", nil)
|
||||
|
||||
emit := func() {
|
||||
return func() {
|
||||
f.firewall.EmitStats()
|
||||
f.handshakeManager.EmitStats()
|
||||
udpStats()
|
||||
@@ -651,19 +682,6 @@ func (f *Interface) emitStats(ctx context.Context, i time.Duration) {
|
||||
certMaxVersion.Update(int64(certState.v1Cert.Version()))
|
||||
}
|
||||
}
|
||||
|
||||
// Prime gauges so a Prometheus scrape that lands before the first tick
|
||||
// sees real values instead of the zero defaults (issue #907).
|
||||
emit()
|
||||
|
||||
for {
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
return
|
||||
case <-ticker.C:
|
||||
emit()
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func (f *Interface) GetHostInfo(vpnIp netip.Addr) *HostInfo {
|
||||
|
||||
+5
-14
@@ -3,7 +3,6 @@
|
||||
package nebula
|
||||
|
||||
import (
|
||||
"context"
|
||||
"net/netip"
|
||||
"testing"
|
||||
"time"
|
||||
@@ -18,11 +17,8 @@ import (
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
// Test_emitStats_primesGauges covers issue #907: a Prometheus scrape that
|
||||
// landed before the first ticker fire used to read 0 for the cert gauges.
|
||||
// emitStats now primes the gauges before entering the ticker loop. We assert
|
||||
// the gauge is zero before the first call and non-zero after.
|
||||
func Test_emitStats_primesGauges(t *testing.T) {
|
||||
// Priming before the first tick (issue #907) is Start's job, see TestStatsServer_Start_primes.
|
||||
func Test_statsEmitter_setsGauges(t *testing.T) {
|
||||
defer metrics.DefaultRegistry.UnregisterAll()
|
||||
|
||||
l := test.NewLogger()
|
||||
@@ -56,17 +52,12 @@ func Test_emitStats_primesGauges(t *testing.T) {
|
||||
ifce.pki.cs.Store(cs)
|
||||
|
||||
ttlGauge := metrics.GetOrRegisterGauge("certificate.ttl_seconds", nil)
|
||||
require.Zero(t, ttlGauge.Value(), "gauge should be zero before emitStats runs")
|
||||
require.Zero(t, ttlGauge.Value(), "gauge should be zero before the emitter runs")
|
||||
|
||||
// Pre-cancel the context so emitStats returns after priming the gauges
|
||||
// without ever reading from ticker.C. The one hour interval is just a
|
||||
// belt-and-suspenders, the test does not expect the ticker to fire.
|
||||
ctx, cancel := context.WithCancel(context.Background())
|
||||
cancel()
|
||||
ifce.emitStats(ctx, time.Hour)
|
||||
ifce.statsEmitter()()
|
||||
|
||||
ttl := ttlGauge.Value()
|
||||
assert.Positive(t, ttl, "ttl gauge should be primed by emitStats before its first tick")
|
||||
assert.Positive(t, ttl, "ttl gauge should be set by one run of the emitter")
|
||||
assert.LessOrEqual(t, ttl, int64(3600))
|
||||
assert.Equal(t, int64(cert.Version1), metrics.GetOrRegisterGauge("certificate.initiating_version", nil).Value())
|
||||
assert.Equal(t, int64(cert.Version1), metrics.GetOrRegisterGauge("certificate.max_version", nil).Value())
|
||||
|
||||
+28
-5
@@ -15,6 +15,7 @@ import (
|
||||
"time"
|
||||
|
||||
"github.com/gaissmai/bart"
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/cert"
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/header"
|
||||
@@ -81,6 +82,8 @@ type LightHouse struct {
|
||||
|
||||
updateTrigger chan struct{}
|
||||
queryChan chan netip.Addr
|
||||
// queryDropped counts queries QueryServer dropped because queryChan was full.
|
||||
queryDropped metrics.Counter
|
||||
|
||||
calculatedRemotes atomic.Pointer[bart.Table[[]*calculatedRemote]] // Maps VpnAddr to []*calculatedRemote
|
||||
|
||||
@@ -116,7 +119,8 @@ func NewLightHouseFromConfig(ctx context.Context, l *slog.Logger, c *config.C, c
|
||||
nebulaPort: nebulaPort,
|
||||
punchy: p,
|
||||
updateTrigger: make(chan struct{}, 1),
|
||||
queryChan: make(chan netip.Addr, c.GetUint32("handshakes.query_buffer", 64)),
|
||||
queryChan: make(chan netip.Addr, c.GetUint32("handshakes.query_buffer", 1024)),
|
||||
queryDropped: metrics.GetOrRegisterCounter("lighthouse.query.dropped", nil),
|
||||
l: l,
|
||||
}
|
||||
h.localAddrsFn = func(al *LocalAllowList) []netip.Addr {
|
||||
@@ -355,7 +359,13 @@ func (lh *LightHouse) reload(c *config.C, initial bool) error {
|
||||
}
|
||||
}
|
||||
|
||||
if initial || c.HasChanged("relay.relays") {
|
||||
// am_relay decides whether relay.relays applies to us at all
|
||||
if initial || c.HasChanged("relay.relays") || c.HasChanged("relay.am_relay") {
|
||||
var before []netip.Addr
|
||||
if !initial {
|
||||
before = lh.GetRelaysForMe()
|
||||
}
|
||||
|
||||
switch c.GetBool("relay.am_relay", false) {
|
||||
case true:
|
||||
// Relays aren't allowed to specify other relays
|
||||
@@ -380,6 +390,11 @@ func (lh *LightHouse) reload(c *config.C, initial bool) error {
|
||||
}
|
||||
lh.relaysForMe.Store(&relaysForMe)
|
||||
}
|
||||
|
||||
// Tell the lighthouses now rather than at the next interval, but only when what we advertise changed
|
||||
if !initial && !slices.Equal(before, lh.GetRelaysForMe()) {
|
||||
lh.TriggerUpdate()
|
||||
}
|
||||
}
|
||||
|
||||
return nil
|
||||
@@ -512,14 +527,19 @@ func (lh *LightHouse) Query(vpnAddr netip.Addr) *RemoteList {
|
||||
return nil
|
||||
}
|
||||
|
||||
// QueryServer is asynchronous so no reply should be expected
|
||||
// QueryServer is asynchronous and non-blocking, no reply should be expected.
|
||||
// If the query channel is full, the query is dropped.
|
||||
func (lh *LightHouse) QueryServer(vpnAddr netip.Addr) {
|
||||
// Don't put lighthouse addrs in the query channel because we can't query lighthouses about lighthouses
|
||||
if lh.amLighthouse || lh.IsLighthouseAddr(vpnAddr) {
|
||||
return
|
||||
}
|
||||
|
||||
lh.queryChan <- vpnAddr
|
||||
select {
|
||||
case lh.queryChan <- vpnAddr:
|
||||
default:
|
||||
lh.queryDropped.Inc(1)
|
||||
}
|
||||
}
|
||||
|
||||
func (lh *LightHouse) QueryCache(vpnAddrs []netip.Addr) *RemoteList {
|
||||
@@ -1235,10 +1255,13 @@ func (lhh *LightHouseHandler) handleHostQuery(n *NebulaMeta, fromVpnAddrs []neti
|
||||
// sendHostPunchNotification signals the other side to punch some zero byte udp packets
|
||||
func (lhh *LightHouseHandler) sendHostPunchNotification(n *NebulaMeta, fromVpnAddrs []netip.Addr, punchNotifDest netip.Addr, w EncWriter) {
|
||||
whereToPunch := fromVpnAddrs[0]
|
||||
// Resolve the target before taking the querier's RemoteList lock below. GetHostInfo takes the
|
||||
// hostmap lock, and the hostmap is taken before the lighthouse, which is taken before a
|
||||
// RemoteList, so looking it up inside the callback would deadlock.
|
||||
targetHI := lhh.lh.ifce.GetHostInfo(punchNotifDest)
|
||||
found, ln, err := lhh.lh.queryAndPrepMessage(whereToPunch, func(c *cache) (int, error) {
|
||||
n = lhh.resetMeta()
|
||||
n.Type = NebulaMeta_HostPunchNotification
|
||||
targetHI := lhh.lh.ifce.GetHostInfo(punchNotifDest)
|
||||
var useVersion cert.Version
|
||||
if targetHI == nil {
|
||||
useVersion = lhh.lh.ifce.GetCertState().initiatingVersion
|
||||
|
||||
@@ -6,13 +6,16 @@ import (
|
||||
"errors"
|
||||
"fmt"
|
||||
"net/netip"
|
||||
"sync"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/gaissmai/bart"
|
||||
"github.com/slackhq/nebula/cert"
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
"go.yaml.in/yaml/v3"
|
||||
@@ -795,3 +798,202 @@ func TestLightHouse_logLocalAddrsErr(t *testing.T) {
|
||||
lh.logLocalAddrsErr(errors.New("something else"))
|
||||
assert.Contains(t, out.String(), "level=WARN")
|
||||
}
|
||||
|
||||
// heldEncWriter holds the query worker in its first GetHostInfo until release closes.
|
||||
type heldEncWriter struct {
|
||||
mockEncWriter
|
||||
once sync.Once
|
||||
held chan struct{}
|
||||
release chan struct{}
|
||||
}
|
||||
|
||||
func (w *heldEncWriter) GetHostInfo(netip.Addr) *HostInfo {
|
||||
w.once.Do(func() { close(w.held) })
|
||||
<-w.release
|
||||
return nil
|
||||
}
|
||||
|
||||
// QueryServer never waits on the query worker, since the packet readers call it and nothing drains the channel at
|
||||
// shutdown; a query that finds the channel full is dropped and counted. The worker is held on its first query so the
|
||||
// channel fills.
|
||||
func TestLighthouse_QueryServerDoesNotBlock(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
cs := testCertState(netip.MustParsePrefix("10.128.0.1/16"))
|
||||
lh1 := "10.128.0.2"
|
||||
c := config.NewC(l)
|
||||
c.Settings["lighthouse"] = map[string]any{"hosts": []any{lh1}}
|
||||
c.Settings["static_host_map"] = map[string]any{lh1: []any{"1.1.1.1:4242"}}
|
||||
c.Settings["handshakes"] = map[string]any{"query_buffer": 1}
|
||||
lh, err := NewLightHouseFromConfig(t.Context(), l, c, cs, nil, nil)
|
||||
require.NoError(t, err)
|
||||
w := &heldEncWriter{held: make(chan struct{}), release: make(chan struct{})}
|
||||
lh.ifce = w
|
||||
t.Cleanup(func() { close(w.release) })
|
||||
|
||||
lh.QueryServer(netip.MustParseAddr("10.128.0.10"))
|
||||
select {
|
||||
case <-w.held:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("the query worker never took the first query")
|
||||
}
|
||||
dropped := lh.queryDropped.Count()
|
||||
done := make(chan struct{})
|
||||
go func() {
|
||||
lh.QueryServer(netip.MustParseAddr("10.128.0.11"))
|
||||
lh.QueryServer(netip.MustParseAddr("10.128.0.12"))
|
||||
close(done)
|
||||
}()
|
||||
select {
|
||||
case <-done:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("QueryServer blocked on a full query channel")
|
||||
}
|
||||
assert.EqualValues(t, 1, lh.queryDropped.Count()-dropped)
|
||||
}
|
||||
|
||||
// hostmapEncWriter answers GetHostInfo from a real hostmap, the way Interface does.
|
||||
type hostmapEncWriter struct {
|
||||
mockEncWriter
|
||||
hm *HostMap
|
||||
}
|
||||
|
||||
func (w *hostmapEncWriter) GetHostInfo(a netip.Addr) *HostInfo { return w.hm.QueryVpnAddr(a) }
|
||||
func (w *hostmapEncWriter) GetCertState() *CertState {
|
||||
return &CertState{initiatingVersion: cert.Version1}
|
||||
}
|
||||
|
||||
// A lighthouse runs four paths that each hold one lock while waiting for the next: CheckAndComplete
|
||||
// holds the hostmap and takes the handshake manager, StartHandshake holds the handshake manager and
|
||||
// takes the lighthouse, a host update holds the lighthouse and takes the sender's RemoteList, and a
|
||||
// punch notification holds the querier's RemoteList while it resolves the target. If that last step
|
||||
// reaches the hostmap under the RemoteList lock, the four deadlock. Drive them all at once.
|
||||
func TestLighthouse_PunchNotificationDoesNotCloseLockCycle(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
myVpnNet := netip.MustParsePrefix("10.128.0.1/24")
|
||||
|
||||
c := config.NewC(l)
|
||||
c.Settings["lighthouse"] = map[string]any{"am_lighthouse": true}
|
||||
c.Settings["listen"] = map[string]any{"port": 4242}
|
||||
lh, err := NewLightHouseFromConfig(t.Context(), l, c, testCertState(myVpnNet), nil, nil)
|
||||
require.NoError(t, err)
|
||||
|
||||
mainHM := newHostMap(l)
|
||||
pr := []netip.Prefix{}
|
||||
mainHM.preferredRanges.Store(&pr)
|
||||
lh.ifce = &hostmapEncWriter{hm: mainHM}
|
||||
|
||||
hm := NewHandshakeManager(l, mainHM, lh, &udp.NoopConn{}, defaultHandshakeConfig)
|
||||
f := &Interface{hostMap: mainHM, lightHouse: lh, handshakeManager: hm, l: l}
|
||||
hm.f = f
|
||||
|
||||
peer := netip.MustParseAddr("10.128.0.2")
|
||||
peerUdp := netip.MustParseAddrPort("10.0.0.2:4242")
|
||||
target := netip.MustParseAddr("10.128.0.3")
|
||||
targetUdp := netip.MustParseAddrPort("10.0.0.3:4242")
|
||||
outbound := netip.MustParseAddr("10.128.0.4")
|
||||
inbound := netip.MustParseAddr("10.128.0.5")
|
||||
|
||||
seed := lh.NewRequestHandler()
|
||||
newLHHostUpdate(peerUdp, peer, []netip.AddrPort{peerUdp}, seed)
|
||||
newLHHostUpdate(targetUdp, target, []netip.AddrPort{targetUdp}, seed)
|
||||
|
||||
stop := make(chan struct{})
|
||||
var wg sync.WaitGroup
|
||||
loop := func(body func()) {
|
||||
defer wg.Done()
|
||||
for {
|
||||
select {
|
||||
case <-stop:
|
||||
return
|
||||
default:
|
||||
body()
|
||||
}
|
||||
}
|
||||
}
|
||||
wg.Add(4)
|
||||
|
||||
// Outbound handshakes keep starting and being dropped.
|
||||
go loop(func() {
|
||||
hm.DeleteHostInfo(hm.StartHandshake(outbound, nil))
|
||||
})
|
||||
|
||||
// Inbound handshakes keep completing and being torn down.
|
||||
var idx uint32
|
||||
go loop(func() {
|
||||
idx++
|
||||
hi := &HostInfo{
|
||||
vpnAddrs: []netip.Addr{inbound},
|
||||
localIndexId: idx,
|
||||
remoteIndexId: idx,
|
||||
remotes: lh.QueryCache([]netip.Addr{inbound}),
|
||||
HandshakePacket: map[uint8][]byte{handshakePacketStage0: {byte(idx)}},
|
||||
}
|
||||
if _, err := hm.CheckAndComplete(hi, handshakePacketStage0, f); err == nil {
|
||||
mainHM.DeleteHostInfo(hi)
|
||||
}
|
||||
})
|
||||
|
||||
// The peer keeps sending host updates.
|
||||
updates := lh.NewRequestHandler()
|
||||
go loop(func() {
|
||||
newLHHostUpdate(peerUdp, peer, []netip.AddrPort{peerUdp}, updates)
|
||||
})
|
||||
|
||||
// The same peer keeps querying for the target, which ends in a punch notification.
|
||||
queries := lh.NewRequestHandler()
|
||||
go loop(func() {
|
||||
sendLHHostRequest(peerUdp, peer, target, queries, nil)
|
||||
})
|
||||
|
||||
time.Sleep(time.Second)
|
||||
close(stop)
|
||||
done := make(chan struct{})
|
||||
go func() { wg.Wait(); close(done) }()
|
||||
select {
|
||||
case <-done:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("deadlock: the lighthouse paths never finished after stop was signalled")
|
||||
}
|
||||
}
|
||||
|
||||
func TestLighthouse_reloadRelaysTriggersUpdate(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
c := config.NewC(l)
|
||||
base := "lighthouse:\n am_lighthouse: true\nlisten:\n port: 4242\n"
|
||||
require.NoError(t, c.LoadString(base+"relay:\n relays: [10.128.0.5]\n"))
|
||||
|
||||
lh, err := NewLightHouseFromConfig(t.Context(), l, c, testCertState(netip.MustParsePrefix("10.128.0.1/24")), nil, nil)
|
||||
require.NoError(t, err)
|
||||
assert.Empty(t, lh.updateTrigger)
|
||||
|
||||
require.NoError(t, c.ReloadConfigString(base+"relay:\n relays: [10.128.0.6]\n"))
|
||||
assert.Equal(t, []netip.Addr{netip.MustParseAddr("10.128.0.6")}, lh.GetRelaysForMe())
|
||||
require.Len(t, lh.updateTrigger, 1)
|
||||
<-lh.updateTrigger
|
||||
|
||||
// The config changed but what we advertise did not, a bad entry is skipped
|
||||
require.NoError(t, c.ReloadConfigString(base+"relay:\n relays: [10.128.0.6, nope]\n"))
|
||||
assert.Equal(t, []netip.Addr{netip.MustParseAddr("10.128.0.6")}, lh.GetRelaysForMe())
|
||||
assert.Empty(t, lh.updateTrigger)
|
||||
}
|
||||
|
||||
func TestLighthouse_reloadAmRelayAlone(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
c := config.NewC(l)
|
||||
base := "lighthouse:\n am_lighthouse: true\nlisten:\n port: 4242\nrelay:\n relays: [10.128.0.5]\n"
|
||||
require.NoError(t, c.LoadString(base+" am_relay: false\n"))
|
||||
|
||||
lh, err := NewLightHouseFromConfig(t.Context(), l, c, testCertState(netip.MustParsePrefix("10.128.0.1/24")), nil, nil)
|
||||
require.NoError(t, err)
|
||||
assert.Equal(t, []netip.Addr{netip.MustParseAddr("10.128.0.5")}, lh.GetRelaysForMe())
|
||||
|
||||
// Relays aren't allowed to have relays, flipping am_relay alone must drop them and tell the lighthouses
|
||||
require.NoError(t, c.ReloadConfigString(base+" am_relay: true\n"))
|
||||
assert.Empty(t, lh.GetRelaysForMe())
|
||||
require.Len(t, lh.updateTrigger, 1)
|
||||
<-lh.updateTrigger
|
||||
|
||||
require.NoError(t, c.ReloadConfigString(base+" am_relay: false\n"))
|
||||
assert.Equal(t, []netip.Addr{netip.MustParseAddr("10.128.0.5")}, lh.GetRelaysForMe())
|
||||
require.Len(t, lh.updateTrigger, 1)
|
||||
}
|
||||
@@ -0,0 +1,31 @@
|
||||
package logging
|
||||
|
||||
import "log/slog"
|
||||
|
||||
// RateLimiter is what RateLimited needs from a rate limiter, such as a *ratelimit.Limiter.
|
||||
// Allow reports whether this call may proceed, and how many calls were denied since the last one that did.
|
||||
type RateLimiter interface {
|
||||
Allow() (denied uint64, ok bool)
|
||||
}
|
||||
|
||||
// RateLimited checks lim before anything is built.
|
||||
// When allowed it returns l, with a "suppressed" count attached if lim denied calls since its last allowed one.
|
||||
// A nil lim allows everything.
|
||||
//
|
||||
// if l, ok := logging.RateLimited(f.l, f.udpTxErrLimiter); ok {
|
||||
// hostinfo.logger(l).Error("Failed to write outgoing packet", "error", err)
|
||||
// }
|
||||
func RateLimited(l *slog.Logger, lim RateLimiter) (*slog.Logger, bool) {
|
||||
if lim == nil {
|
||||
return l, true
|
||||
}
|
||||
|
||||
denied, ok := lim.Allow()
|
||||
if !ok {
|
||||
return nil, false
|
||||
}
|
||||
if denied > 0 {
|
||||
return l.With("suppressed", denied), true
|
||||
}
|
||||
return l, true
|
||||
}
|
||||
@@ -0,0 +1,106 @@
|
||||
package logging
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"errors"
|
||||
"io"
|
||||
"log/slog"
|
||||
"net/netip"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/slackhq/nebula/ratelimit"
|
||||
)
|
||||
|
||||
type fakeLimiter struct {
|
||||
denied uint64
|
||||
ok bool
|
||||
}
|
||||
|
||||
func (f fakeLimiter) Allow() (uint64, bool) { return f.denied, f.ok }
|
||||
|
||||
func TestRateLimited(t *testing.T) {
|
||||
var buf bytes.Buffer
|
||||
base := NewLogger(&buf)
|
||||
|
||||
if l, ok := RateLimited(base, fakeLimiter{ok: false}); ok || l != nil {
|
||||
t.Fatalf("denied call: ok=%v logger=%v", ok, l)
|
||||
}
|
||||
|
||||
if l, ok := RateLimited(base, fakeLimiter{ok: true}); !ok || l != base {
|
||||
t.Fatal("nothing denied should hand back the same logger")
|
||||
}
|
||||
|
||||
if l, ok := RateLimited(base, nil); !ok || l != base {
|
||||
t.Fatal("a nil limiter should allow with the same logger")
|
||||
}
|
||||
|
||||
l, ok := RateLimited(base, fakeLimiter{denied: 7, ok: true})
|
||||
if !ok {
|
||||
t.Fatal("allowed call denied")
|
||||
}
|
||||
l.Info("hello", "k", "v")
|
||||
if !strings.Contains(buf.String(), "suppressed=7") {
|
||||
t.Fatalf("want suppressed=7, got %q", buf.String())
|
||||
}
|
||||
}
|
||||
|
||||
// The benchmarks compare a log line through slog directly against the same line behind RateLimited: logged with
|
||||
// nothing denied, denied, and logged after denials.
|
||||
|
||||
var (
|
||||
benchErr = errors.New("sendto: no buffer space available")
|
||||
benchAddr = netip.MustParseAddrPort("192.0.2.1:4242")
|
||||
)
|
||||
|
||||
func benchLogger() *slog.Logger {
|
||||
return NewLogger(io.Discard).With("localIndex", 1234, "remoteIndex", 5678)
|
||||
}
|
||||
|
||||
func BenchmarkRateLimitedBaseline(b *testing.B) {
|
||||
lg := benchLogger()
|
||||
b.ReportAllocs()
|
||||
for i := 0; i < b.N; i++ {
|
||||
lg.Error("Failed to write outgoing packet", "error", benchErr, "udpAddr", benchAddr)
|
||||
}
|
||||
}
|
||||
|
||||
func BenchmarkRateLimitedAllowed(b *testing.B) {
|
||||
lg, lim := benchLogger(), mustLimiter(b, 1, 0)
|
||||
b.ReportAllocs()
|
||||
for i := 0; i < b.N; i++ {
|
||||
if l, ok := RateLimited(lg, lim); ok {
|
||||
l.Error("Failed to write outgoing packet", "error", benchErr, "udpAddr", benchAddr)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func BenchmarkRateLimitedDenied(b *testing.B) {
|
||||
lg, lim := benchLogger(), mustLimiter(b, 1, time.Hour)
|
||||
lim.Allow()
|
||||
b.ReportAllocs()
|
||||
for i := 0; i < b.N; i++ {
|
||||
if l, ok := RateLimited(lg, lim); ok {
|
||||
l.Error("Failed to write outgoing packet", "error", benchErr, "udpAddr", benchAddr)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func BenchmarkRateLimitedAfterDenied(b *testing.B) {
|
||||
lg, lim := benchLogger(), fakeLimiter{denied: 5, ok: true}
|
||||
b.ReportAllocs()
|
||||
for i := 0; i < b.N; i++ {
|
||||
if l, ok := RateLimited(lg, lim); ok {
|
||||
l.Error("Failed to write outgoing packet", "error", benchErr, "udpAddr", benchAddr)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func mustLimiter(b *testing.B, burst int, every time.Duration) *ratelimit.Limiter {
|
||||
l, err := ratelimit.New(burst, every)
|
||||
if err != nil {
|
||||
b.Fatal(err)
|
||||
}
|
||||
return l
|
||||
}
|
||||
@@ -244,6 +244,7 @@ func Main(c *config.C, configTest bool, buildVersion string, l *slog.Logger, dev
|
||||
|
||||
pinThreads := c.GetBool("tun.pin_threads", true)
|
||||
cpuAffinity := parseCpuAffinity(c, l, routines)
|
||||
staged := parseStagedConfig(c, l)
|
||||
if pinThreads && routines > 1 && len(cpuAffinity) == 0 && !configTest {
|
||||
// The operator didn't choose pin CPUs, so pick a default set that
|
||||
// prefers performance cores and doesn't stack co-located instances
|
||||
@@ -296,10 +297,12 @@ func Main(c *config.C, configTest bool, buildVersion string, l *slog.Logger, dev
|
||||
ConntrackCacheTimeout: conntrackCacheTimeout,
|
||||
CpuAffinity: cpuAffinity,
|
||||
PinThreads: pinThreads,
|
||||
staged: staged,
|
||||
l: l,
|
||||
}
|
||||
|
||||
var ifce *Interface
|
||||
var emitStats func()
|
||||
if !configTest {
|
||||
ifce, err = NewInterface(ctx, ifConfig)
|
||||
if err != nil {
|
||||
@@ -318,9 +321,12 @@ func Main(c *config.C, configTest bool, buildVersion string, l *slog.Logger, dev
|
||||
go handshakeManager.Run(ctx)
|
||||
|
||||
punchy.Start(ctx, ifce, hostMap, lightHouse)
|
||||
|
||||
// On the stats server's loop, so the gauges follow the interval it accepted
|
||||
emitStats = ifce.statsEmitter()
|
||||
}
|
||||
|
||||
stats, err := newStatsServerFromConfig(ctx, l, c, buildVersion, configTest)
|
||||
stats, err := newStatsServerFromConfig(ctx, l, c, buildVersion, configTest, emitStats)
|
||||
if err != nil {
|
||||
return nil, util.ContextualizeIfNeeded("Failed to start stats emitter", err)
|
||||
}
|
||||
@@ -329,8 +335,6 @@ func Main(c *config.C, configTest bool, buildVersion string, l *slog.Logger, dev
|
||||
return nil, nil
|
||||
}
|
||||
|
||||
go ifce.emitStats(ctx, c.GetDuration("stats.interval", time.Second*10))
|
||||
|
||||
attachCommands(l, c, ssh, ifce)
|
||||
|
||||
networkChanges := udp.NewNetworkChangeMonitor(ctx, l, c)
|
||||
|
||||
@@ -5,6 +5,7 @@ import (
|
||||
"net/netip"
|
||||
"os"
|
||||
"path/filepath"
|
||||
"runtime"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
@@ -123,3 +124,23 @@ firewall:
|
||||
require.NoError(t, err)
|
||||
require.Nil(t, ctrl, "a config test hands back nothing to stop, so it must stop itself")
|
||||
}
|
||||
|
||||
func TestParseStagedConfig(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
staged := func(workers any) stagedConfig {
|
||||
c := config.NewC(l)
|
||||
if workers != nil {
|
||||
c.Settings["tun"] = map[string]any{"staged": map[string]any{"workers": workers}}
|
||||
}
|
||||
return parseStagedConfig(c, l)
|
||||
}
|
||||
for _, workers := range []any{nil, -1, 0} {
|
||||
require.Equal(t, stagedConfig{}, staged(workers))
|
||||
}
|
||||
if runtime.GOOS != "linux" {
|
||||
require.Equal(t, stagedConfig{}, staged(2))
|
||||
return
|
||||
}
|
||||
want := stagedConfig{workers: 2, inlineBelow: 256}
|
||||
require.Equal(t, want, staged(2))
|
||||
}
|
||||
+25
-8
@@ -129,7 +129,13 @@ func (f *Interface) readOutsidePackets(via ViaSender, packet []byte, rxc *rxCont
|
||||
return
|
||||
}
|
||||
|
||||
out, err := hostinfo.ConnectionState.Decrypt(f.l, h.MessageCounter, packet, rxc.nb)
|
||||
var out []byte
|
||||
if pre := rxc.decrypted; pre != nil && pre.ci == hostinfo.ConnectionState {
|
||||
// Staged receive already checked and decrypted this packet on a worker, only the replay window is left
|
||||
out, err = pre.ci.acceptDecrypted(f.l, h.MessageCounter, pre.out, pre.err)
|
||||
} else {
|
||||
out, err = hostinfo.ConnectionState.Decrypt(f.l, h.MessageCounter, packet, rxc.nb)
|
||||
}
|
||||
if err != nil {
|
||||
if f.l.Enabled(context.Background(), slog.LevelDebug) {
|
||||
hostinfo.logger(f.l).Debug("Failed to decrypt packet", "error", err, "from", via, "header", h)
|
||||
@@ -160,9 +166,7 @@ func (f *Interface) readOutsidePackets(via ViaSender, packet []byte, rxc *rxCont
|
||||
case header.TestReply:
|
||||
// No-op, useful for the Roaming and connectionManager side-effects above
|
||||
case header.TestRequest:
|
||||
const maxCipherOverhead = 16 //todo we use this too often, needs a real importable const
|
||||
const maxOverhead = header.Len + header.Len + maxCipherOverhead + maxCipherOverhead
|
||||
if maxOverhead+len(out) > len(rxc.scratch) {
|
||||
if header.MaxOverhead+len(out) > len(rxc.scratch) {
|
||||
// A reply that cannot fit in scratch is dropped no matter the log level.
|
||||
if f.l.Enabled(context.Background(), slog.LevelDebug) {
|
||||
hostinfo.logger(f.l).Debug("dropping oversized test request", "payloadLen", len(out), "from", via)
|
||||
@@ -263,11 +267,16 @@ func (f *Interface) handleOutsideRelayPacket(hostinfo *HostInfo, via ViaSender,
|
||||
|
||||
// closeTunnel closes a tunnel locally, it does not send a closeTunnel packet to the remote
|
||||
func (f *Interface) closeTunnel(hostInfo *HostInfo) {
|
||||
final := f.hostMap.DeleteHostInfo(hostInfo)
|
||||
if final {
|
||||
// We no longer have any tunnels with this vpn addr, clear learned lighthouse state to lower memory usage
|
||||
f.lightHouse.DeleteVpnAddrs(hostInfo.vpnAddrs)
|
||||
// Hold the hostmap lock across both the delete and the lighthouse eviction,
|
||||
// to avoid a race with handshake completion.
|
||||
f.hostMap.Lock()
|
||||
defer f.hostMap.Unlock()
|
||||
if !f.hostMap.unlockedDeleteHostInfo(hostInfo) {
|
||||
return
|
||||
}
|
||||
|
||||
// We no longer have any tunnels with this vpn addr, clear learned lighthouse state to lower memory usage
|
||||
f.lightHouse.DeleteVpnAddrs(hostInfo.vpnAddrs)
|
||||
}
|
||||
|
||||
// sendCloseTunnel is a helper function to send a proper close tunnel packet to a remote
|
||||
@@ -470,6 +479,14 @@ func parseV4(data []byte, incoming bool, fp *firewall.ParsedPacket) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
// flushOutside ends a receive batch: it writes out the tun coalescer and drops the batch's hostmap cache.
|
||||
func (f *Interface) flushOutside(rxc *rxContext) {
|
||||
if err := f.batchers[rxc.q].Flush(); err != nil {
|
||||
f.l.Error("Failed to flush tun coalescer", "error", err)
|
||||
}
|
||||
clear(rxc.hostmapCache)
|
||||
}
|
||||
|
||||
func (f *Interface) handleOutsideMessagePacket(hostinfo *HostInfo, messageCounter uint64, out []byte, rxc *rxContext) {
|
||||
err := newPacket(out, true, rxc.fwPacket)
|
||||
if err != nil {
|
||||
|
||||
@@ -0,0 +1,57 @@
|
||||
//go:build !e2e_testing
|
||||
|
||||
package overlay
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"log/slog"
|
||||
|
||||
"golang.org/x/sys/windows"
|
||||
"golang.org/x/sys/windows/registry"
|
||||
)
|
||||
|
||||
var tcpipInterfaceKeys = []string{
|
||||
`SYSTEM\CurrentControlSet\Services\Tcpip\Parameters\Interfaces`,
|
||||
`SYSTEM\CurrentControlSet\Services\Tcpip6\Parameters\Interfaces`,
|
||||
}
|
||||
|
||||
// disableDNSRegistration prevents publishing overlay addresses in the host's
|
||||
// DNS records, where clients outside the overlay cannot reach them.
|
||||
// Failures are non-fatal so registry access cannot prevent tunnel creation.
|
||||
func disableDNSRegistration(l *slog.Logger, guid windows.GUID) {
|
||||
id := guid.String()
|
||||
for _, base := range tcpipInterfaceKeys {
|
||||
path := base + `\` + id
|
||||
if err := setDNSRegistrationValues(path); err != nil {
|
||||
l.Warn("Failed to disable DNS registration on the tun adapter", "error", err, "key", path)
|
||||
continue
|
||||
}
|
||||
l.Debug("Disabled DNS registration on the tun adapter", "key", path)
|
||||
}
|
||||
}
|
||||
|
||||
func setDNSRegistrationValues(path string) error {
|
||||
// The stack may not have created the adapter's interface key yet.
|
||||
k, _, err := registry.CreateKey(registry.LOCAL_MACHINE, path, registry.SET_VALUE)
|
||||
if err != nil {
|
||||
return fmt.Errorf("open HKLM\\%s: %w", path, err)
|
||||
}
|
||||
defer k.Close()
|
||||
|
||||
// Neither RegistrationEnabled nor DisableDynamicUpdate alone reliably
|
||||
// prevents registration, so all three settings are required.
|
||||
values := []struct {
|
||||
name string
|
||||
data uint32
|
||||
}{
|
||||
{"RegistrationEnabled", 0},
|
||||
{"DisableDynamicUpdate", 1},
|
||||
{"MaxNumberOfAddressesToRegister", 0},
|
||||
}
|
||||
for _, v := range values {
|
||||
if err := k.SetDWordValue(v.name, v.data); err != nil {
|
||||
return fmt.Errorf("set %s: %w", v.name, err)
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
+28
-4
@@ -50,7 +50,13 @@ func (r Route) String() string {
|
||||
|
||||
func makeRouteTree(l *slog.Logger, routes []Route, allowMTU bool) (*bart.Table[routing.Gateways], error) {
|
||||
routeTree := new(bart.Table[routing.Gateways])
|
||||
seen := make(map[netip.Prefix]bool, len(routes))
|
||||
for _, r := range routes {
|
||||
if seen[r.Cidr] {
|
||||
l.Warn("route is listed more than once, only the last entry's via is used", "route", r)
|
||||
}
|
||||
seen[r.Cidr] = true
|
||||
|
||||
if !allowMTU && r.MTU > 0 {
|
||||
l.Warn("route MTU is not supported on this platform",
|
||||
"goos", runtime.GOOS,
|
||||
@@ -98,7 +104,11 @@ func parseRoutes(c *config.C, networks []netip.Prefix) ([]Route, error) {
|
||||
|
||||
mtu, ok := rMtu.(int)
|
||||
if !ok {
|
||||
mtu, err = strconv.Atoi(rMtu.(string))
|
||||
s, ok := rMtu.(string)
|
||||
if !ok {
|
||||
return nil, fmt.Errorf("entry %v.mtu in tun.routes is not an integer: %v", i+1, rMtu)
|
||||
}
|
||||
mtu, err = strconv.Atoi(s)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("entry %v.mtu in tun.routes is not an integer: %v", i+1, err)
|
||||
}
|
||||
@@ -174,7 +184,11 @@ func parseUnsafeRoutes(c *config.C, networks []netip.Prefix) ([]Route, error) {
|
||||
if rMtu, ok := m["mtu"]; ok {
|
||||
mtu, ok = rMtu.(int)
|
||||
if !ok {
|
||||
mtu, err = strconv.Atoi(rMtu.(string))
|
||||
s, ok := rMtu.(string)
|
||||
if !ok {
|
||||
return nil, fmt.Errorf("entry %v.mtu in tun.unsafe_routes is not an integer: %v", i+1, rMtu)
|
||||
}
|
||||
mtu, err = strconv.Atoi(s)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("entry %v.mtu in tun.unsafe_routes is not an integer: %v", i+1, err)
|
||||
}
|
||||
@@ -192,10 +206,15 @@ func parseUnsafeRoutes(c *config.C, networks []netip.Prefix) ([]Route, error) {
|
||||
|
||||
metric, ok := rMetric.(int)
|
||||
if !ok {
|
||||
_, err = strconv.ParseInt(rMetric.(string), 10, 32)
|
||||
s, ok := rMetric.(string)
|
||||
if !ok {
|
||||
return nil, fmt.Errorf("entry %v.metric in tun.unsafe_routes is not an integer: %v", i+1, rMetric)
|
||||
}
|
||||
n, err := strconv.ParseInt(s, 10, 32)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("entry %v.metric in tun.unsafe_routes is not an integer: %v", i+1, err)
|
||||
}
|
||||
metric = int(n)
|
||||
}
|
||||
|
||||
if metric < 0 || metric > math.MaxInt32 {
|
||||
@@ -248,10 +267,15 @@ func parseUnsafeRoutes(c *config.C, networks []netip.Prefix) ([]Route, error) {
|
||||
|
||||
gatewayWeight, ok := rGatewayWeight.(int)
|
||||
if !ok {
|
||||
_, err = strconv.ParseInt(rGatewayWeight.(string), 10, 32)
|
||||
s, ok := rGatewayWeight.(string)
|
||||
if !ok {
|
||||
return nil, fmt.Errorf("entry .weight in tun.unsafe_routes[%v].via[%v] is not an integer", i+1, ig+1)
|
||||
}
|
||||
w, err := strconv.ParseInt(s, 10, 32)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("entry .weight in tun.unsafe_routes[%v].via[%v] is not an integer", i+1, ig+1)
|
||||
}
|
||||
gatewayWeight = int(w)
|
||||
}
|
||||
|
||||
if gatewayWeight < 1 || gatewayWeight > math.MaxInt32 {
|
||||
|
||||
@@ -1,13 +1,17 @@
|
||||
package overlay
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"fmt"
|
||||
"net/netip"
|
||||
"strings"
|
||||
"testing"
|
||||
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/routing"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
@@ -402,3 +406,115 @@ func Test_makeMultipathUnsafeRouteTree(t *testing.T) {
|
||||
routing.CalculateBucketsForGateways(expectedGateways)
|
||||
assert.ElementsMatch(t, expectedGateways, r)
|
||||
}
|
||||
|
||||
func Test_getMTU(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
c := config.NewC(l)
|
||||
|
||||
// unset
|
||||
assert.Equal(t, DefaultMTU, getMTU(l, c))
|
||||
|
||||
// the largest a relayed packet can carry in one underlay buffer
|
||||
require.Equal(t, udp.MTU, MaxMTU+header.MaxOverhead)
|
||||
c.Settings["tun"] = map[string]any{"mtu": MaxMTU}
|
||||
assert.Equal(t, MaxMTU, getMTU(l, c))
|
||||
|
||||
// too big is capped
|
||||
c.Settings["tun"] = map[string]any{"mtu": 9000}
|
||||
assert.Equal(t, MaxMTU, getMTU(l, c))
|
||||
}
|
||||
|
||||
func Test_getAllRoutesFromConfigCapsMTU(t *testing.T) {
|
||||
l := test.NewLogger()
|
||||
c := config.NewC(l)
|
||||
c.Settings["tun"] = map[string]any{
|
||||
"routes": []any{
|
||||
map[string]any{"mtu": "9000", "route": "10.0.0.0/29"},
|
||||
map[string]any{"mtu": "1300", "route": "10.0.0.8/29"},
|
||||
},
|
||||
"unsafe_routes": []any{
|
||||
map[string]any{"via": "127.0.0.1", "mtu": "9000", "route": "1.0.0.0/29"},
|
||||
},
|
||||
}
|
||||
|
||||
changed, routes, err := getAllRoutesFromConfig(l, c, []netip.Prefix{netip.MustParsePrefix("10.0.0.0/24")}, true)
|
||||
require.NoError(t, err)
|
||||
assert.True(t, changed)
|
||||
|
||||
mtus := map[string]int{}
|
||||
for _, r := range routes {
|
||||
mtus[r.Cidr.String()] = r.MTU
|
||||
}
|
||||
assert.Equal(t, map[string]int{"10.0.0.0/29": MaxMTU, "10.0.0.8/29": 1300, "1.0.0.0/29": MaxMTU}, mtus)
|
||||
}
|
||||
|
||||
// The route tree keeps one entry per CIDR, so a CIDR listed twice only routes by the last one. Say so instead of
|
||||
// letting the earlier via, or on Windows the earlier metric, quietly lose
|
||||
func Test_makeRouteTree_WarnsOnDuplicate(t *testing.T) {
|
||||
buf := &bytes.Buffer{}
|
||||
l := test.NewLoggerWithOutput(buf)
|
||||
via := func(s string) routing.Gateways {
|
||||
return routing.Gateways{routing.NewGateway(netip.MustParseAddr(s), 1)}
|
||||
}
|
||||
cidr := netip.MustParsePrefix("10.251.1.0/24")
|
||||
|
||||
_, err := makeRouteTree(l, []Route{
|
||||
{Cidr: cidr, Via: via("10.250.0.2"), Install: true, Metric: 100},
|
||||
{Cidr: netip.MustParsePrefix("10.251.2.0/24"), Via: via("10.250.0.2"), Install: true},
|
||||
}, false)
|
||||
require.NoError(t, err)
|
||||
assert.Empty(t, buf.String())
|
||||
|
||||
tree, err := makeRouteTree(l, []Route{
|
||||
{Cidr: cidr, Via: via("10.250.0.2"), Install: true, Metric: 100},
|
||||
{Cidr: cidr, Via: via("10.250.0.3"), Install: true, Metric: 200},
|
||||
}, false)
|
||||
require.NoError(t, err)
|
||||
assert.Equal(t, 1, strings.Count(buf.String(), "route is listed more than once, only the last entry's via is used"))
|
||||
gw, ok := tree.Lookup(netip.MustParseAddr("10.251.1.1"))
|
||||
require.True(t, ok)
|
||||
assert.Equal(t, netip.MustParseAddr("10.250.0.3"), gw[0].Addr(), "the last entry wins")
|
||||
}
|
||||
|
||||
// A quoted metric or weight is a number written as a string, it means that number. Anything else is an error, not a
|
||||
// panic
|
||||
func Test_parseUnsafeRoutes_QuotedNumbers(t *testing.T) {
|
||||
c := config.NewC(test.NewLogger())
|
||||
n := netip.MustParsePrefix("10.0.0.1/24")
|
||||
route := func(metric, weight any) {
|
||||
c.Settings["tun"] = map[string]any{"unsafe_routes": []any{map[string]any{
|
||||
"route": "1.0.0.0/8",
|
||||
"metric": metric,
|
||||
"via": []any{map[string]any{"gateway": "10.0.0.2", "weight": weight}},
|
||||
}}}
|
||||
}
|
||||
|
||||
route("100", "5")
|
||||
routes, err := parseUnsafeRoutes(c, []netip.Prefix{n})
|
||||
require.NoError(t, err)
|
||||
require.Len(t, routes, 1)
|
||||
assert.Equal(t, 100, routes[0].Metric)
|
||||
assert.Equal(t, "{addr: 10.0.0.2, weight: 5}", routes[0].Via[0].String())
|
||||
|
||||
route(1.5, 1)
|
||||
_, err = parseUnsafeRoutes(c, []netip.Prefix{n})
|
||||
require.EqualError(t, err, "entry 1.metric in tun.unsafe_routes is not an integer: 1.5")
|
||||
|
||||
route(1, true)
|
||||
_, err = parseUnsafeRoutes(c, []netip.Prefix{n})
|
||||
require.EqualError(t, err, "entry .weight in tun.unsafe_routes[1].via[1] is not an integer")
|
||||
}
|
||||
|
||||
// An mtu that is neither a number nor a quoted one is an error, not a panic
|
||||
func Test_parseRoutes_MTUNotANumber(t *testing.T) {
|
||||
c := config.NewC(test.NewLogger())
|
||||
n := netip.MustParsePrefix("10.0.0.1/24")
|
||||
|
||||
c.Settings["tun"] = map[string]any{"routes": []any{map[string]any{"route": "1.0.0.0/8", "mtu": 1500.5}}}
|
||||
_, err := parseRoutes(c, []netip.Prefix{n})
|
||||
require.EqualError(t, err, "entry 1.mtu in tun.routes is not an integer: 1500.5")
|
||||
|
||||
c.Settings["tun"] = map[string]any{"unsafe_routes": []any{map[string]any{"route": "1.0.0.0/8", "via": "10.0.0.2", "mtu": 1500.5}}}
|
||||
_, err = parseUnsafeRoutes(c, []netip.Prefix{n})
|
||||
require.EqualError(t, err, "entry 1.mtu in tun.unsafe_routes is not an integer: 1500.5")
|
||||
}
|
||||
+24
-2
@@ -7,11 +7,27 @@ import (
|
||||
"net/netip"
|
||||
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/header"
|
||||
"github.com/slackhq/nebula/udp"
|
||||
"github.com/slackhq/nebula/util"
|
||||
)
|
||||
|
||||
const DefaultMTU = 1300
|
||||
|
||||
// MaxMTU is the largest tun.mtu, or route mtu, the underlay can carry.
|
||||
// Every underlay buffer is udp.MTU bytes, and a relayed packet adds header.MaxOverhead to its plaintext.
|
||||
const MaxMTU = udp.MTU - header.MaxOverhead
|
||||
|
||||
// getMTU reads tun.mtu, capping a value too big for the underlay to carry.
|
||||
func getMTU(l *slog.Logger, c *config.C) int {
|
||||
mtu := c.GetInt("tun.mtu", DefaultMTU)
|
||||
if mtu > MaxMTU {
|
||||
l.Warn("tun.mtu is too big for the underlay to carry, capping it", "mtu", mtu, "maxMTU", MaxMTU)
|
||||
return MaxMTU
|
||||
}
|
||||
return mtu
|
||||
}
|
||||
|
||||
type NameError struct {
|
||||
Name string
|
||||
Underlying error
|
||||
@@ -27,7 +43,7 @@ type DeviceFactory func(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix,
|
||||
func NewDeviceFromConfig(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, routines int) (Device, error) {
|
||||
switch {
|
||||
case c.GetBool("tun.disabled", false):
|
||||
tun := newDisabledTun(vpnNetworks, c.GetInt("tun.tx_queue", 500), c.GetBool("stats.message_metrics", false), l)
|
||||
tun := newDisabledTun(vpnNetworks, c.GetInt("tun.tx_queue", 5000), c.GetBool("stats.message_metrics", false), l)
|
||||
return tun, nil
|
||||
|
||||
default:
|
||||
@@ -41,7 +57,7 @@ func NewFdDeviceFromConfig(fd *int) DeviceFactory {
|
||||
}
|
||||
}
|
||||
|
||||
func getAllRoutesFromConfig(c *config.C, vpnNetworks []netip.Prefix, initial bool) (bool, []Route, error) {
|
||||
func getAllRoutesFromConfig(l *slog.Logger, c *config.C, vpnNetworks []netip.Prefix, initial bool) (bool, []Route, error) {
|
||||
if !initial && !c.HasChanged("tun.routes") && !c.HasChanged("tun.unsafe_routes") {
|
||||
return false, nil, nil
|
||||
}
|
||||
@@ -57,6 +73,12 @@ func getAllRoutesFromConfig(c *config.C, vpnNetworks []netip.Prefix, initial boo
|
||||
}
|
||||
|
||||
routes = append(routes, unsafeRoutes...)
|
||||
for i, r := range routes {
|
||||
if r.MTU > MaxMTU {
|
||||
l.Warn("route mtu is too big for the underlay to carry, capping it", "route", r.Cidr, "mtu", r.MTU, "maxMTU", MaxMTU)
|
||||
routes[i].MTU = MaxMTU
|
||||
}
|
||||
}
|
||||
return true, routes, nil
|
||||
}
|
||||
|
||||
|
||||
@@ -68,7 +68,7 @@ func (t *tun) Activate() error {
|
||||
}
|
||||
|
||||
func (t *tun) reload(c *config.C, initial bool) error {
|
||||
change, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
change, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
+101
-17
@@ -6,8 +6,10 @@ import (
|
||||
"errors"
|
||||
"fmt"
|
||||
"log/slog"
|
||||
"math/bits"
|
||||
"net/netip"
|
||||
"os"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"syscall"
|
||||
"unsafe"
|
||||
@@ -28,7 +30,7 @@ type tun struct {
|
||||
DefaultMTU int
|
||||
Routes atomic.Pointer[[]Route]
|
||||
routeTree atomic.Pointer[bart.Table[routing.Gateways]]
|
||||
linkAddr *netroute.LinkAddr
|
||||
linkAddr atomic.Pointer[netroute.LinkAddr]
|
||||
// hostOwned means the fd arrived from the OS, which has already configured addressing, mtu
|
||||
// and routes for it. NEPacketTunnelProvider on darwin does this.
|
||||
hostOwned bool
|
||||
@@ -126,7 +128,7 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*t
|
||||
f: os.NewFile(uintptr(fd), ""),
|
||||
Device: name,
|
||||
vpnNetworks: vpnNetworks,
|
||||
DefaultMTU: c.GetInt("tun.mtu", DefaultMTU),
|
||||
DefaultMTU: getMTU(l, c),
|
||||
l: l,
|
||||
}
|
||||
|
||||
@@ -167,7 +169,7 @@ func newTunFromFd(c *config.C, l *slog.Logger, deviceFd int, vpnNetworks []netip
|
||||
f: file,
|
||||
Device: utunNameFromFd(deviceFd),
|
||||
vpnNetworks: vpnNetworks,
|
||||
DefaultMTU: c.GetInt("tun.mtu", DefaultMTU),
|
||||
DefaultMTU: getMTU(l, c),
|
||||
hostOwned: true,
|
||||
l: l,
|
||||
}
|
||||
@@ -243,7 +245,7 @@ func (t *tun) Activate() error {
|
||||
if linkAddr == nil {
|
||||
return fmt.Errorf("unable to discover link_addr for tun interface")
|
||||
}
|
||||
t.linkAddr = linkAddr
|
||||
t.linkAddr.Store(linkAddr)
|
||||
|
||||
for _, network := range t.vpnNetworks {
|
||||
if network.Addr().Is4() {
|
||||
@@ -269,6 +271,7 @@ func (t *tun) Activate() error {
|
||||
return t.addRoutes(false)
|
||||
}
|
||||
|
||||
// activate4 assigns our vpn address as a /32 and routes the vpn network to the tun explicitly.
|
||||
func (t *tun) activate4(network netip.Prefix) error {
|
||||
s, err := unix.Socket(
|
||||
unix.AF_INET,
|
||||
@@ -295,22 +298,32 @@ func (t *tun) activate4(network netip.Prefix) error {
|
||||
MaskAddr: unix.RawSockaddrInet4{
|
||||
Len: unix.SizeofSockaddrInet4,
|
||||
Family: unix.AF_INET,
|
||||
Addr: prefixToMask(network).As4(),
|
||||
Addr: [4]byte{255, 255, 255, 255}, //always make the address a /32
|
||||
},
|
||||
}
|
||||
|
||||
// On Darwin, VPNs don't edit the routing table directly.
|
||||
// A network extension hands its routes to the system, and configd installs them.
|
||||
// configd replaces any route already holding the same prefix, including ours.
|
||||
// As an example, Tailscale with an exit node treats any interface prefix wider than one address as a LAN.
|
||||
// With LAN access off, it routes each LAN into Tailscale's utun.
|
||||
// With LAN access on, it excludes each LAN from its tunnel.
|
||||
// configd routes an excluded prefix like its own routes would, ignoring ours, so it goes out the default route.
|
||||
// Either way our route is replaced, and traffic for the vpn network no longer reaches nebula.
|
||||
// A single address isn't a LAN to Tailscale, so our route is left alone.
|
||||
// We route the vpn network to the tun explicitly instead.
|
||||
|
||||
if err := ioctl(uintptr(s), unix.SIOCAIFADDR, uintptr(unsafe.Pointer(&ifr))); err != nil {
|
||||
return fmt.Errorf("failed to set tun v4 address: %s", err)
|
||||
}
|
||||
|
||||
err = addRoute(network, t.linkAddr)
|
||||
if err != nil {
|
||||
return err
|
||||
if err := addRoute(network, t.linkAddr.Load()); err != nil {
|
||||
return fmt.Errorf("failed to add route for %v: %w", network.Masked(), err)
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
// activate6 is activate4 for IPv6
|
||||
func (t *tun) activate6(network netip.Prefix) error {
|
||||
s, err := unix.Socket(
|
||||
unix.AF_INET6,
|
||||
@@ -332,25 +345,29 @@ func (t *tun) activate6(network netip.Prefix) error {
|
||||
PrefixMask: unix.RawSockaddrInet6{
|
||||
Len: unix.SizeofSockaddrInet6,
|
||||
Family: unix.AF_INET6,
|
||||
Addr: prefixToMask(network).As16(),
|
||||
// Always set the netmask to be a /128. See activate4 for why we do this.
|
||||
Addr: [16]byte{255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255},
|
||||
},
|
||||
Lifetime: addrLifetime{
|
||||
// never expires
|
||||
Vltime: 0xffffffff,
|
||||
Pltime: 0xffffffff,
|
||||
},
|
||||
Flags: _IN6_IFF_NODAD,
|
||||
Flags: _IN6_IFF_NODAD, // No DAD, a utun has no neighbors.
|
||||
}
|
||||
|
||||
if err := ioctl(uintptr(s), _SIOCAIFADDR_IN6, uintptr(unsafe.Pointer(&ifr))); err != nil {
|
||||
return fmt.Errorf("failed to set tun address: %s", err)
|
||||
}
|
||||
|
||||
if err := addRoute(network, t.linkAddr.Load()); err != nil {
|
||||
return fmt.Errorf("failed to add route for %v: %w", network.Masked(), err)
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
func (t *tun) reload(c *config.C, initial bool) error {
|
||||
change, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
change, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
@@ -394,6 +411,56 @@ func (t *tun) RoutesFor(ip netip.Addr) routing.Gateways {
|
||||
return routing.Gateways{}
|
||||
}
|
||||
|
||||
func fetchRoutes() ([]netroute.Message, error) {
|
||||
rib, err := netroute.FetchRIB(unix.AF_UNSPEC, unix.NET_RT_DUMP, 0)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return netroute.ParseRIB(unix.NET_RT_DUMP, rib)
|
||||
}
|
||||
|
||||
// routeInterfaceIn returns the index of the interface prefix is routed through, or 0 if msgs has no route for exactly
|
||||
// that prefix. Routes scoped to an interface are skipped, they don't stop us adding ours.
|
||||
func routeInterfaceIn(msgs []netroute.Message, prefix netip.Prefix) int {
|
||||
for _, msg := range msgs {
|
||||
m, ok := msg.(*netroute.RouteMessage)
|
||||
if !ok || len(m.Addrs) <= unix.RTAX_NETMASK || m.Flags&unix.RTF_IFSCOPE != 0 {
|
||||
continue
|
||||
}
|
||||
if p, ok := routePrefix(m.Addrs[unix.RTAX_DST], m.Addrs[unix.RTAX_NETMASK]); ok && p == prefix.Masked() {
|
||||
return m.Index
|
||||
}
|
||||
}
|
||||
return 0
|
||||
}
|
||||
|
||||
// routePrefix returns the prefix a route covers. Host routes have no netmask.
|
||||
func routePrefix(dst, mask netroute.Addr) (netip.Prefix, bool) {
|
||||
switch d := dst.(type) {
|
||||
case *netroute.Inet4Addr:
|
||||
bits := 32
|
||||
if m, ok := mask.(*netroute.Inet4Addr); ok {
|
||||
bits = maskBits(m.IP[:])
|
||||
}
|
||||
return netip.PrefixFrom(netip.AddrFrom4(d.IP), bits), true
|
||||
case *netroute.Inet6Addr:
|
||||
bits := 128
|
||||
if m, ok := mask.(*netroute.Inet6Addr); ok {
|
||||
bits = maskBits(m.IP[:])
|
||||
}
|
||||
return netip.PrefixFrom(netip.AddrFrom16(d.IP), bits), true
|
||||
}
|
||||
return netip.Prefix{}, false
|
||||
}
|
||||
|
||||
func maskBits(mask []byte) int {
|
||||
n := 0
|
||||
for _, b := range mask {
|
||||
n += bits.OnesCount8(b)
|
||||
}
|
||||
return n
|
||||
}
|
||||
|
||||
// Get the LinkAddr for the interface of the given name
|
||||
// Is there an easier way to fetch this when we create the interface?
|
||||
// Maybe SIOCGIFINDEX? but this doesn't appear to exist in the darwin headers.
|
||||
@@ -428,7 +495,14 @@ func (t *tun) addRoutes(logErrors bool) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
// Activate hasn't run yet, it will install the routes
|
||||
linkAddr := t.linkAddr.Load()
|
||||
if linkAddr == nil {
|
||||
return nil
|
||||
}
|
||||
|
||||
routes := *t.Routes.Load()
|
||||
systemRoutes := sync.OnceValues(fetchRoutes)
|
||||
|
||||
for _, r := range routes {
|
||||
if len(r.Via) == 0 || !r.Install {
|
||||
@@ -436,10 +510,16 @@ func (t *tun) addRoutes(logErrors bool) error {
|
||||
continue
|
||||
}
|
||||
|
||||
err := addRoute(r.Cidr, t.linkAddr)
|
||||
err := addRoute(r.Cidr, linkAddr)
|
||||
if err != nil {
|
||||
if errors.Is(err, unix.EEXIST) {
|
||||
t.l.Warn("unable to add unsafe_route, identical route already exists", "route", r.Cidr)
|
||||
// Every route we added already exists on a reload, only warn about someone else's
|
||||
msgs, err := systemRoutes()
|
||||
if err != nil {
|
||||
t.l.Warn("unable to add unsafe_route, a route already exists but its interface could not be determined", "route", r.Cidr, "error", err)
|
||||
} else if routeInterfaceIn(msgs, r.Cidr) != linkAddr.Index {
|
||||
t.l.Warn("unable to add unsafe_route, the destination is already routed through another interface", "route", r.Cidr)
|
||||
}
|
||||
} else {
|
||||
retErr := util.NewContextualError("Failed to add route", map[string]any{"route": r}, err)
|
||||
if logErrors {
|
||||
@@ -450,6 +530,8 @@ func (t *tun) addRoutes(logErrors bool) error {
|
||||
}
|
||||
} else {
|
||||
t.l.Info("Added route", "route", r)
|
||||
// A cidr listed twice would otherwise find our own route missing from an older fetch
|
||||
systemRoutes = sync.OnceValues(fetchRoutes)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -457,16 +539,18 @@ func (t *tun) addRoutes(logErrors bool) error {
|
||||
}
|
||||
|
||||
func (t *tun) removeRoutes(routes []Route) error {
|
||||
if t.hostOwned {
|
||||
linkAddr := t.linkAddr.Load()
|
||||
if t.hostOwned || linkAddr == nil {
|
||||
return nil
|
||||
}
|
||||
|
||||
for _, r := range routes {
|
||||
if !r.Install {
|
||||
if len(r.Via) == 0 || !r.Install {
|
||||
// addRoutes never installs these
|
||||
continue
|
||||
}
|
||||
|
||||
err := delRoute(r.Cidr, t.linkAddr)
|
||||
err := delRoute(r.Cidr, linkAddr)
|
||||
if err != nil {
|
||||
t.l.Error("Failed to remove route", "error", err, "route", r)
|
||||
} else {
|
||||
|
||||
@@ -0,0 +1,103 @@
|
||||
//go:build darwin && !ios && !e2e_testing
|
||||
|
||||
package overlay
|
||||
|
||||
import (
|
||||
"net"
|
||||
"net/netip"
|
||||
"os"
|
||||
"os/exec"
|
||||
"strings"
|
||||
"testing"
|
||||
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/routing"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
netroute "golang.org/x/net/route"
|
||||
"golang.org/x/sys/unix"
|
||||
)
|
||||
|
||||
// Every mac routes loopback through lo0
|
||||
func TestRouteInterfaceIn(t *testing.T) {
|
||||
msgs, err := fetchRoutes()
|
||||
require.NoError(t, err)
|
||||
routeInterface := func(s string) int { return routeInterfaceIn(msgs, netip.MustParsePrefix(s)) }
|
||||
|
||||
lo0, err := net.InterfaceByName("lo0")
|
||||
require.NoError(t, err)
|
||||
|
||||
assert.Equal(t, lo0.Index, routeInterface("127.0.0.0/8"))
|
||||
assert.Equal(t, lo0.Index, routeInterface("127.0.0.1/32"), "a host route has no netmask")
|
||||
assert.Equal(t, lo0.Index, routeInterface("::1/128"))
|
||||
assert.Equal(t, 0, routeInterface("127.0.0.0/16"), "only an exact prefix counts")
|
||||
|
||||
// Most macs also list the default route scoped to their other interfaces
|
||||
out, err := exec.Command("route", "-n", "get", "default").Output()
|
||||
require.NoError(t, err)
|
||||
var name string
|
||||
for _, line := range strings.Split(string(out), "\n") {
|
||||
if v, ok := strings.CutPrefix(strings.TrimSpace(line), "interface: "); ok {
|
||||
name = v
|
||||
}
|
||||
}
|
||||
if name == "" {
|
||||
t.Skip("no default route")
|
||||
}
|
||||
def, err := net.InterfaceByName(name)
|
||||
require.NoError(t, err)
|
||||
assert.Equal(t, def.Index, routeInterface("0.0.0.0/0"))
|
||||
}
|
||||
|
||||
func TestRouteInterfaceIn_SkipsScopedCopies(t *testing.T) {
|
||||
route := func(index, flags int) netroute.Message {
|
||||
addrs := make([]netroute.Addr, unix.RTAX_MAX)
|
||||
addrs[unix.RTAX_DST] = &netroute.Inet4Addr{IP: [4]byte{192, 168, 1, 0}}
|
||||
addrs[unix.RTAX_NETMASK] = &netroute.Inet4Addr{IP: [4]byte{255, 255, 255, 0}}
|
||||
return &netroute.RouteMessage{Index: index, Flags: flags, Addrs: addrs}
|
||||
}
|
||||
p := netip.MustParsePrefix("192.168.1.0/24")
|
||||
assert.Equal(t, 7, routeInterfaceIn([]netroute.Message{route(4, unix.RTF_UP|unix.RTF_IFSCOPE), route(7, unix.RTF_UP)}, p))
|
||||
assert.Equal(t, 0, routeInterfaceIn([]netroute.Message{route(4, unix.RTF_UP|unix.RTF_IFSCOPE)}, p))
|
||||
}
|
||||
|
||||
// A reload can happen before Activate
|
||||
func TestTunRoutesBeforeActivate(t *testing.T) {
|
||||
tn := &tun{l: test.NewLogger()}
|
||||
routes := []Route{{
|
||||
Cidr: netip.MustParsePrefix("10.253.1.0/24"),
|
||||
Via: routing.Gateways{routing.NewGateway(netip.MustParseAddr("10.250.0.2"), 1)},
|
||||
Install: true,
|
||||
}}
|
||||
tn.Routes.Store(&routes)
|
||||
|
||||
assert.NoError(t, tn.addRoutes(true))
|
||||
assert.NoError(t, tn.removeRoutes(routes))
|
||||
}
|
||||
|
||||
// A tun.routes entry with no via is never installed, so removing it must not delete a route for the same cidr that
|
||||
// belongs to someone else
|
||||
func TestTunReloadLeavesRoutesItNeverInstalled(t *testing.T) {
|
||||
if os.Geteuid() != 0 {
|
||||
t.Skip("creating a utun and adding routes needs root")
|
||||
}
|
||||
lo0, err := net.InterfaceByName("lo0")
|
||||
require.NoError(t, err)
|
||||
cidr := "10.250.0.128/25"
|
||||
require.NoError(t, exec.Command("route", "-q", "-n", "add", "-net", cidr, "-interface", "lo0").Run())
|
||||
t.Cleanup(func() { _ = exec.Command("route", "-q", "-n", "delete", "-net", cidr).Run() })
|
||||
|
||||
l := test.NewLogger()
|
||||
c := config.NewC(l)
|
||||
require.NoError(t, c.LoadString("tun:\n routes:\n - route: "+cidr+"\n mtu: 1300\n"))
|
||||
tn, err := newTun(c, l, []netip.Prefix{netip.MustParsePrefix("10.250.0.1/24")}, false)
|
||||
require.NoError(t, err)
|
||||
t.Cleanup(func() { _ = tn.Close() })
|
||||
require.NoError(t, tn.Activate())
|
||||
|
||||
require.NoError(t, c.ReloadConfigString("tun: {}\n"))
|
||||
msgs, err := fetchRoutes()
|
||||
require.NoError(t, err)
|
||||
assert.Equal(t, lo0.Index, routeInterfaceIn(msgs, netip.MustParsePrefix(cidr)))
|
||||
}
|
||||
@@ -370,7 +370,7 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*t
|
||||
t := &tun{
|
||||
Device: deviceName,
|
||||
vpnNetworks: vpnNetworks,
|
||||
MTU: c.GetInt("tun.mtu", DefaultMTU),
|
||||
MTU: getMTU(l, c),
|
||||
l: l,
|
||||
fd: fd,
|
||||
shutdownR: shutdownR,
|
||||
@@ -510,7 +510,7 @@ func (t *tun) setMTU() error {
|
||||
}
|
||||
|
||||
func (t *tun) reload(c *config.C, initial bool) error {
|
||||
change, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
change, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
@@ -589,7 +589,8 @@ func (t *tun) addRoutes(logErrors bool) error {
|
||||
|
||||
func (t *tun) removeRoutes(routes []Route) error {
|
||||
for _, r := range routes {
|
||||
if !r.Install {
|
||||
if len(r.Via) == 0 || !r.Install {
|
||||
// addRoutes never installs these
|
||||
continue
|
||||
}
|
||||
|
||||
|
||||
+1
-1
@@ -68,7 +68,7 @@ func (t *tun) Activate() error {
|
||||
}
|
||||
|
||||
func (t *tun) reload(c *config.C, initial bool) error {
|
||||
change, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
change, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
+13
-4
@@ -221,7 +221,7 @@ func newTunGeneric(c *config.C, l *slog.Logger, fd int, vnetHdr bool, offloadFla
|
||||
vnetHdr: vnetHdr,
|
||||
offloadFlags: offloadFlags,
|
||||
vpnNetworks: vpnNetworks,
|
||||
TXQueueLen: c.GetInt("tun.tx_queue", 500),
|
||||
TXQueueLen: c.GetInt("tun.tx_queue", 5000),
|
||||
useSystemRoutes: c.GetBool("tun.use_system_route_table", false),
|
||||
useSystemRoutesBufferSize: c.GetInt("tun.use_system_route_table_buffer_size", 0),
|
||||
routesFromSystem: map[netip.Prefix]routing.Gateways{},
|
||||
@@ -244,7 +244,7 @@ func newTunGeneric(c *config.C, l *slog.Logger, fd int, vnetHdr bool, offloadFla
|
||||
}
|
||||
|
||||
func (t *tun) reload(c *config.C, initial bool) error {
|
||||
routeChange, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
routeChange, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
@@ -253,6 +253,15 @@ func (t *tun) reload(c *config.C, initial bool) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
// Only the mtu changed. The routes are the same, parse them again so the ones without their own mtu pick up the
|
||||
// new default. Without this an mtu change would remove them all
|
||||
if !routeChange {
|
||||
_, routes, err = getAllRoutesFromConfig(t.l, c, t.vpnNetworks, true)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
}
|
||||
|
||||
routeTree, err := makeRouteTree(t.l, routes, true)
|
||||
if err != nil {
|
||||
return err
|
||||
@@ -267,14 +276,14 @@ func (t *tun) reload(c *config.C, initial bool) error {
|
||||
|
||||
oldDefaultMTU := t.DefaultMTU
|
||||
oldMaxMTU := t.MaxMTU
|
||||
newDefaultMTU := c.GetInt("tun.mtu", DefaultMTU)
|
||||
newDefaultMTU := getMTU(t.l, c)
|
||||
newMaxMTU := newDefaultMTU
|
||||
for i, r := range routes {
|
||||
if r.MTU == 0 {
|
||||
routes[i].MTU = newDefaultMTU
|
||||
}
|
||||
|
||||
if r.MTU > t.MaxMTU {
|
||||
if r.MTU > newMaxMTU {
|
||||
newMaxMTU = r.MTU
|
||||
}
|
||||
}
|
||||
|
||||
@@ -99,7 +99,7 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*t
|
||||
fd: fd,
|
||||
Device: deviceName,
|
||||
vpnNetworks: vpnNetworks,
|
||||
MTU: c.GetInt("tun.mtu", DefaultMTU),
|
||||
MTU: getMTU(l, c),
|
||||
l: l,
|
||||
}
|
||||
|
||||
@@ -340,7 +340,7 @@ func (t *tun) doIoctlByName(ctl uintptr, value uint32) error {
|
||||
}
|
||||
|
||||
func (t *tun) reload(c *config.C, initial bool) error {
|
||||
change, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
change, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
@@ -420,7 +420,8 @@ func (t *tun) addRoutes(logErrors bool) error {
|
||||
|
||||
func (t *tun) removeRoutes(routes []Route) error {
|
||||
for _, r := range routes {
|
||||
if !r.Install {
|
||||
if len(r.Via) == 0 || !r.Install {
|
||||
// addRoutes never installs these
|
||||
continue
|
||||
}
|
||||
|
||||
|
||||
@@ -90,7 +90,7 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*t
|
||||
fd: fd,
|
||||
Device: deviceName,
|
||||
vpnNetworks: vpnNetworks,
|
||||
MTU: c.GetInt("tun.mtu", DefaultMTU),
|
||||
MTU: getMTU(l, c),
|
||||
l: l,
|
||||
}
|
||||
|
||||
@@ -319,7 +319,7 @@ func (t *tun) doIoctlByName(ctl uintptr, value uint32) error {
|
||||
}
|
||||
|
||||
func (t *tun) reload(c *config.C, initial bool) error {
|
||||
change, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
change, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
@@ -399,7 +399,8 @@ func (t *tun) addRoutes(logErrors bool) error {
|
||||
|
||||
func (t *tun) removeRoutes(routes []Route) error {
|
||||
for _, r := range routes {
|
||||
if !r.Install {
|
||||
if len(r.Via) == 0 || !r.Install {
|
||||
// addRoutes never installs these
|
||||
continue
|
||||
}
|
||||
|
||||
|
||||
@@ -0,0 +1,56 @@
|
||||
//go:build !e2e_testing
|
||||
|
||||
package overlay
|
||||
|
||||
import (
|
||||
"net/netip"
|
||||
"testing"
|
||||
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/routing"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
)
|
||||
|
||||
// Only tun.mtu changes. The system calls fail without a real device, what matters is the routes nebula keeps
|
||||
func TestReloadMTUOnlyKeepsRoutes(t *testing.T) {
|
||||
routes := " unsafe_routes:\n - route: 192.168.100.0/24\n via: 10.1.0.1\n"
|
||||
c := config.NewC(test.NewLogger())
|
||||
require.NoError(t, c.LoadString("tun:\n mtu: 1300\n"+routes))
|
||||
|
||||
tn := &tun{
|
||||
l: test.NewLogger(),
|
||||
vpnNetworks: []netip.Prefix{netip.MustParsePrefix("10.1.0.2/24")},
|
||||
ioctlFd: ^uintptr(0),
|
||||
routesFromSystem: map[netip.Prefix]routing.Gateways{},
|
||||
}
|
||||
require.NoError(t, tn.reload(c, true))
|
||||
|
||||
require.NoError(t, c.ReloadConfigString("tun:\n mtu: 1400\n"+routes))
|
||||
_ = tn.reload(c, false)
|
||||
|
||||
got := *tn.Routes.Load()
|
||||
require.Len(t, got, 1)
|
||||
assert.Equal(t, netip.MustParsePrefix("192.168.100.0/24"), got[0].Cidr)
|
||||
assert.Equal(t, 1400, got[0].MTU)
|
||||
assert.Equal(t, 1400, tn.DefaultMTU)
|
||||
}
|
||||
|
||||
// The device mtu has to fit the largest route mtu, wherever it sits in the list
|
||||
func TestReloadMaxMTUIsTheLargestRoute(t *testing.T) {
|
||||
c := config.NewC(test.NewLogger())
|
||||
require.NoError(t, c.LoadString("tun:\n mtu: 1300\n unsafe_routes:\n"+
|
||||
" - route: 192.168.100.0/24\n via: 10.1.0.1\n mtu: 1500\n"+
|
||||
" - route: 192.168.101.0/24\n via: 10.1.0.1\n mtu: 1400\n"))
|
||||
|
||||
tn := &tun{
|
||||
l: test.NewLogger(),
|
||||
vpnNetworks: []netip.Prefix{netip.MustParsePrefix("10.1.0.2/24")},
|
||||
ioctlFd: ^uintptr(0),
|
||||
routesFromSystem: map[netip.Prefix]routing.Gateways{},
|
||||
}
|
||||
require.NoError(t, tn.reload(c, true))
|
||||
assert.Equal(t, 1500, tn.MaxMTU)
|
||||
assert.Equal(t, 1300, tn.DefaultMTU)
|
||||
}
|
||||
@@ -31,7 +31,7 @@ type TestTun struct {
|
||||
}
|
||||
|
||||
func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*TestTun, error) {
|
||||
_, routes, err := getAllRoutesFromConfig(c, vpnNetworks, true)
|
||||
_, routes, err := getAllRoutesFromConfig(l, c, vpnNetworks, true)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
+80
-6
@@ -4,6 +4,7 @@ package overlay
|
||||
|
||||
import (
|
||||
"crypto"
|
||||
"errors"
|
||||
"fmt"
|
||||
"log/slog"
|
||||
"net/netip"
|
||||
@@ -75,7 +76,7 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*w
|
||||
t := &winTun{
|
||||
Device: deviceName,
|
||||
vpnNetworks: vpnNetworks,
|
||||
MTU: c.GetInt("tun.mtu", DefaultMTU),
|
||||
MTU: getMTU(l, c),
|
||||
guid: *guid,
|
||||
networkCategory: cat,
|
||||
setCategory: setCat,
|
||||
@@ -104,6 +105,9 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*w
|
||||
}
|
||||
t.tun = tunDevice.(*wintun.NativeTun)
|
||||
|
||||
// Disable registration before Activate assigns addresses and triggers it.
|
||||
disableDNSRegistration(l, *guid)
|
||||
|
||||
c.RegisterReloadCallback(func(c *config.C) {
|
||||
err := t.reload(c, false)
|
||||
if err != nil {
|
||||
@@ -115,7 +119,7 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*w
|
||||
}
|
||||
|
||||
func (t *winTun) reload(c *config.C, initial bool) error {
|
||||
change, routes, err := getAllRoutesFromConfig(c, t.vpnNetworks, initial)
|
||||
change, routes, err := getAllRoutesFromConfig(t.l, c, t.vpnNetworks, initial)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
@@ -135,7 +139,7 @@ func (t *winTun) reload(c *config.C, initial bool) error {
|
||||
|
||||
if !initial {
|
||||
// Remove first, if the system removes a wanted route hopefully it will be re-added next
|
||||
err := t.removeRoutes(findRemovedRoutes(routes, *oldRoutes))
|
||||
err := t.removeRoutes(findRemovedSystemRoutes(routes, *oldRoutes))
|
||||
if err != nil {
|
||||
util.LogWithContextIfNeeded("Failed to remove routes", err, t.l)
|
||||
}
|
||||
@@ -184,17 +188,27 @@ func (t *winTun) addRoutes(logErrors bool) error {
|
||||
foundDefault4 := false
|
||||
carriesV6 := slices.ContainsFunc(t.vpnNetworks, func(p netip.Prefix) bool { return p.Addr().Is6() })
|
||||
|
||||
for _, r := range routes {
|
||||
// A CIDR listed twice routes by its last entry, install only that one or each reload would flip between them. The
|
||||
// same rule as systemRoutes
|
||||
last := make(map[netip.Prefix]int, len(routes))
|
||||
for i, r := range routes {
|
||||
last[r.Cidr] = i
|
||||
}
|
||||
|
||||
for i, r := range routes {
|
||||
if len(r.Via) == 0 || !r.Install {
|
||||
// We don't allow route MTUs so only install routes with a via
|
||||
continue
|
||||
}
|
||||
if last[r.Cidr] != i {
|
||||
continue
|
||||
}
|
||||
|
||||
// A v6 unsafe_route is legal under a v4-only cert; uninstalled ones put nothing on the adapter.
|
||||
carriesV6 = carriesV6 || r.Cidr.Addr().Is6()
|
||||
|
||||
// Add our unsafe route as an on-link route to the nebula tun device.
|
||||
err := luid.AddRoute(r.Cidr, unspecifiedNextHop(r.Cidr), uint32(r.Metric))
|
||||
result, err := addRoute(luid, r)
|
||||
if err != nil {
|
||||
retErr := util.NewContextualError("Failed to add route", map[string]any{"route": r}, err)
|
||||
if logErrors {
|
||||
@@ -204,7 +218,12 @@ func (t *winTun) addRoutes(logErrors bool) error {
|
||||
return retErr
|
||||
}
|
||||
} else {
|
||||
t.l.Info("Added route", "route", r)
|
||||
switch result {
|
||||
case routeAdded:
|
||||
t.l.Info("Added route", "route", r)
|
||||
case routeMetricUpdated:
|
||||
t.l.Info("Updated route metric", "route", r)
|
||||
}
|
||||
}
|
||||
|
||||
if !foundDefault4 {
|
||||
@@ -316,6 +335,61 @@ func (t *winTun) Close() error {
|
||||
return t.tun.Close()
|
||||
}
|
||||
|
||||
// systemRoutes maps each CIDR to the route that puts it in the system table. A CIDR's last entry decides, including
|
||||
// one that says not to install it
|
||||
func systemRoutes(routes []Route) map[netip.Prefix]Route {
|
||||
out := make(map[netip.Prefix]Route, len(routes))
|
||||
for _, r := range routes {
|
||||
if len(r.Via) > 0 && r.Install {
|
||||
out[r.Cidr] = r
|
||||
} else {
|
||||
delete(out, r.Cidr)
|
||||
}
|
||||
}
|
||||
return out
|
||||
}
|
||||
|
||||
// findRemovedSystemRoutes is the routes to delete from the system table. Windows deletes by destination and next hop,
|
||||
// so only a CIDR nothing installs anymore goes, a changed metric is set on the route in place by addRoute
|
||||
func findRemovedSystemRoutes(newRoutes, oldRoutes []Route) []Route {
|
||||
keep := systemRoutes(newRoutes)
|
||||
var removed []Route
|
||||
for _, r := range systemRoutes(oldRoutes) {
|
||||
if _, ok := keep[r.Cidr]; !ok {
|
||||
removed = append(removed, r)
|
||||
}
|
||||
}
|
||||
return removed
|
||||
}
|
||||
|
||||
type routeResult int
|
||||
|
||||
const (
|
||||
routeAdded routeResult = iota
|
||||
routeMetricUpdated
|
||||
routeAlreadyInstalled
|
||||
)
|
||||
|
||||
// addRoute installs r on the interface. A reload re-adds every route to catch any the system dropped, and Windows
|
||||
// refuses one it has with the same destination and next hop whatever its metric. One that is there with another
|
||||
// metric, ours from before a metric change or someone else's, is brought to ours in place
|
||||
func addRoute(luid winipcfg.LUID, r Route) (routeResult, error) {
|
||||
nextHop := unspecifiedNextHop(r.Cidr)
|
||||
err := luid.AddRoute(r.Cidr, nextHop, uint32(r.Metric))
|
||||
if !errors.Is(err, windows.ERROR_OBJECT_ALREADY_EXISTS) {
|
||||
return routeAdded, err
|
||||
}
|
||||
row, err := luid.Route(r.Cidr, nextHop)
|
||||
if err != nil {
|
||||
return routeAlreadyInstalled, err
|
||||
}
|
||||
if row.Metric != uint32(r.Metric) {
|
||||
row.Metric = uint32(r.Metric)
|
||||
return routeMetricUpdated, row.Set()
|
||||
}
|
||||
return routeAlreadyInstalled, nil
|
||||
}
|
||||
|
||||
func unspecifiedNextHop(p netip.Prefix) netip.Addr {
|
||||
if p.Addr().Is4() {
|
||||
return netip.IPv4Unspecified()
|
||||
|
||||
@@ -0,0 +1,168 @@
|
||||
//go:build windows && !e2e_testing
|
||||
|
||||
package overlay
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"fmt"
|
||||
"net/netip"
|
||||
"strings"
|
||||
"testing"
|
||||
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
"golang.org/x/sys/windows"
|
||||
"golang.zx2c4.com/wireguard/windows/tunnel/winipcfg"
|
||||
)
|
||||
|
||||
type routeTest struct {
|
||||
c *config.C
|
||||
tun *winTun
|
||||
luid winipcfg.LUID
|
||||
log *bytes.Buffer
|
||||
}
|
||||
|
||||
// newRouteTest makes a wintun device of its own, before installing tun.unsafe_routes from cfg
|
||||
func newRouteTest(t *testing.T, cfg string) *routeTest {
|
||||
t.Helper()
|
||||
if !windows.GetCurrentProcessToken().IsElevated() {
|
||||
t.Skip("creating a wintun device needs an elevated process")
|
||||
}
|
||||
if err := checkWinTunExists(); err != nil {
|
||||
t.Skipf("wintun.dll is not where nebula looks for it: %v", err)
|
||||
}
|
||||
rt := &routeTest{log: &bytes.Buffer{}}
|
||||
l := test.NewLoggerWithOutput(rt.log)
|
||||
rt.c = config.NewC(l)
|
||||
require.NoError(t, rt.c.LoadString(cfg))
|
||||
var err error
|
||||
rt.tun, err = newTun(rt.c, l, []netip.Prefix{netip.MustParsePrefix("10.250.0.1/24")}, false)
|
||||
require.NoError(t, err)
|
||||
t.Cleanup(func() { _ = rt.tun.Close() })
|
||||
rt.luid = winipcfg.LUID(rt.tun.tun.LUID())
|
||||
return rt
|
||||
}
|
||||
|
||||
func (rt *routeTest) metric(t *testing.T, cidr string) uint32 {
|
||||
t.Helper()
|
||||
p := netip.MustParsePrefix(cidr)
|
||||
row, err := rt.luid.Route(p, unspecifiedNextHop(p))
|
||||
require.NoError(t, err)
|
||||
return row.Metric
|
||||
}
|
||||
|
||||
func unsafeRoutes(routes ...string) string {
|
||||
var b strings.Builder
|
||||
b.WriteString("tun:\n dev: nebroutetest\n unsafe_routes:\n")
|
||||
// cidr[@metric][!], the ! for install: false
|
||||
for _, r := range routes {
|
||||
r, noInstall := strings.CutSuffix(r, "!")
|
||||
cidr, metric, _ := strings.Cut(r, "@")
|
||||
fmt.Fprintf(&b, " - route: %s\n via: 10.250.0.2\n", cidr)
|
||||
if metric != "" {
|
||||
fmt.Fprintf(&b, " metric: %s\n", metric)
|
||||
}
|
||||
if noInstall {
|
||||
b.WriteString(" install: false\n")
|
||||
}
|
||||
}
|
||||
return b.String()
|
||||
}
|
||||
|
||||
// A reload that adds a route re-adds the ones already installed, Windows refusing those as already there isn't a failure
|
||||
func TestWinTun_ReloadKeepsInstalledRoutes(t *testing.T) {
|
||||
forFamilies(t, func(t *testing.T, a, b string) {
|
||||
rt := newRouteTest(t, unsafeRoutes(a))
|
||||
require.NoError(t, rt.tun.Activate())
|
||||
|
||||
require.NoError(t, rt.c.ReloadConfigString(unsafeRoutes(a, b)))
|
||||
assert.NotContains(t, rt.log.String(), "Failed to add route")
|
||||
rt.metric(t, a)
|
||||
rt.metric(t, b)
|
||||
})
|
||||
}
|
||||
|
||||
// Windows calls a route the same by destination and next hop, so a changed metric is set on the installed route in
|
||||
// place rather than deleted and added, which would leave a moment with no route
|
||||
func TestWinTun_ReloadMetricChangeApplies(t *testing.T) {
|
||||
forFamilies(t, func(t *testing.T, a, b string) {
|
||||
rt := newRouteTest(t, unsafeRoutes(a+"@100"))
|
||||
require.NoError(t, rt.tun.Activate())
|
||||
require.Equal(t, uint32(100), rt.metric(t, a))
|
||||
|
||||
require.NoError(t, rt.c.ReloadConfigString(unsafeRoutes(a+"@200")))
|
||||
assert.NotContains(t, rt.log.String(), "Failed to add route")
|
||||
assert.NotContains(t, rt.log.String(), "Removed route")
|
||||
assert.Equal(t, uint32(200), rt.metric(t, a))
|
||||
})
|
||||
}
|
||||
|
||||
// A route someone else put on the device with another metric is brought to ours, at startup too
|
||||
func TestWinTun_ForeignRouteGetsOurMetric(t *testing.T) {
|
||||
forFamilies(t, func(t *testing.T, a, b string) {
|
||||
rt := newRouteTest(t, unsafeRoutes(a+"@100"))
|
||||
p := netip.MustParsePrefix(a)
|
||||
require.NoError(t, rt.luid.AddRoute(p, unspecifiedNextHop(p), 50))
|
||||
|
||||
require.NoError(t, rt.tun.Activate())
|
||||
assert.Equal(t, uint32(100), rt.metric(t, a))
|
||||
assert.Contains(t, rt.log.String(), "level=INFO msg=\"Updated route metric\"")
|
||||
})
|
||||
}
|
||||
|
||||
// A CIDR listed twice is installed once, by its last entry, and reloads leave it there instead of flipping between the
|
||||
// two metrics
|
||||
func TestWinTun_DuplicateCidrInstallsTheLastEntry(t *testing.T) {
|
||||
forFamilies(t, func(t *testing.T, a, b string) {
|
||||
rt := newRouteTest(t, unsafeRoutes(a+"@100", a+"@200"))
|
||||
require.NoError(t, rt.tun.Activate())
|
||||
assert.Equal(t, uint32(200), rt.metric(t, a))
|
||||
|
||||
require.NoError(t, rt.c.ReloadConfigString(unsafeRoutes(a+"@100", a+"@200", b)))
|
||||
assert.Equal(t, uint32(200), rt.metric(t, a))
|
||||
assert.NotContains(t, rt.log.String(), "Updated route metric")
|
||||
assert.NotContains(t, rt.log.String(), "Failed to add route")
|
||||
})
|
||||
}
|
||||
|
||||
// Dropping the duplicate that wasn't installed leaves the installed route alone. Windows deletes by destination and
|
||||
// next hop, deleting the dropped one would take the live route with it
|
||||
func TestWinTun_DroppingAnUninstalledDuplicateKeepsTheRoute(t *testing.T) {
|
||||
forFamilies(t, func(t *testing.T, a, b string) {
|
||||
rt := newRouteTest(t, unsafeRoutes(a+"@100", a+"@200"))
|
||||
require.NoError(t, rt.tun.Activate())
|
||||
|
||||
require.NoError(t, rt.c.ReloadConfigString(unsafeRoutes(a+"@200")))
|
||||
assert.NotContains(t, rt.log.String(), "Removed route")
|
||||
assert.Equal(t, uint32(200), rt.metric(t, a))
|
||||
})
|
||||
}
|
||||
|
||||
// A reload ends where a fresh start with the same config would, a CIDR whose last entry says not to install it is
|
||||
// removed
|
||||
func TestWinTun_ReloadMatchesAFreshStart(t *testing.T) {
|
||||
forFamilies(t, func(t *testing.T, a, b string) {
|
||||
cidr := netip.MustParsePrefix(a)
|
||||
routes := unsafeRoutes(a+"@100", a+"@100!")
|
||||
|
||||
fresh := newRouteTest(t, routes)
|
||||
require.NoError(t, fresh.tun.Activate())
|
||||
_, err := fresh.luid.Route(cidr, unspecifiedNextHop(cidr))
|
||||
require.Error(t, err, "a fresh start installed it")
|
||||
require.NoError(t, fresh.tun.Close())
|
||||
|
||||
rt := newRouteTest(t, unsafeRoutes(a+"@100"))
|
||||
require.NoError(t, rt.tun.Activate())
|
||||
require.NoError(t, rt.c.ReloadConfigString(routes))
|
||||
_, err = rt.luid.Route(cidr, unspecifiedNextHop(cidr))
|
||||
assert.Error(t, err, "the reload left it installed")
|
||||
})
|
||||
}
|
||||
|
||||
// forFamilies runs f with two unsafe_routes CIDRs of each address family, the tun's own network stays IPv4
|
||||
func forFamilies(t *testing.T, f func(t *testing.T, a, b string)) {
|
||||
t.Run("ipv4", func(t *testing.T) { f(t, "10.251.1.0/24", "10.251.2.0/24") })
|
||||
t.Run("ipv6", func(t *testing.T) { f(t, "fd00:251:1::/64", "fd00:251:2::/64") })
|
||||
}
|
||||
@@ -0,0 +1,91 @@
|
||||
package ratelimit
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"math"
|
||||
"sync/atomic"
|
||||
"time"
|
||||
)
|
||||
|
||||
// Limiter is a token bucket: `burst` calls are allowed at once, then one more per `every`.
|
||||
// It counts the calls it denies and hands that count to the next allowed call.
|
||||
//
|
||||
// The budget belongs to the Limiter instance.
|
||||
// Every Allow that returns true spends from it, whatever the caller then does.
|
||||
//
|
||||
// Time follows Go's monotonic clock.
|
||||
//
|
||||
// Create one Limiter for each thing you want to limit,
|
||||
// sharing one between unrelated events lets the noisy one starve the quiet one.
|
||||
//
|
||||
// Check Allow before doing the work being limited, so a denied call stays cheap:
|
||||
//
|
||||
// if denied, ok := probeLimiter.Allow(); ok {
|
||||
// sendProbe()
|
||||
// probesDenied.Inc(int64(denied))
|
||||
// }
|
||||
//
|
||||
// For log lines use logging.RateLimited, which attaches the denied count to the logger as "suppressed".
|
||||
type Limiter struct {
|
||||
every int64
|
||||
slack int64
|
||||
tat atomic.Int64
|
||||
denied atomic.Uint64
|
||||
}
|
||||
|
||||
// maxSpan bounds `burst` * `every` to half of int64, leaving the other half for process uptime so the bucket
|
||||
// math can't overflow.
|
||||
const maxSpan = math.MaxInt64 / 2
|
||||
|
||||
// New returns a Limiter allowing `burst` calls at once and one more per `every`.
|
||||
// A `burst` under 1 is 1, and an `every` of 0 or less never limits.
|
||||
// It returns an error if `burst` * `every` is over about 146 years.
|
||||
func New(burst int, every time.Duration) (*Limiter, error) {
|
||||
l := &Limiter{}
|
||||
if every > 0 {
|
||||
burst = max(burst, 1)
|
||||
if int64(burst) > maxSpan/int64(every) {
|
||||
return nil, fmt.Errorf("ratelimit: burst %d * every %v is too long", burst, every)
|
||||
}
|
||||
l.every = int64(every)
|
||||
l.slack = int64(burst-1) * l.every
|
||||
}
|
||||
return l, nil
|
||||
}
|
||||
|
||||
var monoStart = time.Now()
|
||||
|
||||
func monoNow() int64 { return int64(time.Since(monoStart)) }
|
||||
|
||||
// Allow reports whether this call is allowed.
|
||||
// If a token is available it spends it and returns the calls denied since the last allowed one, and true.
|
||||
// Otherwise it returns 0 and false, and counts the denial.
|
||||
// A nil Limiter allows everything.
|
||||
func (l *Limiter) Allow() (uint64, bool) {
|
||||
if l == nil {
|
||||
return 0, true
|
||||
}
|
||||
return l.allow(monoNow())
|
||||
}
|
||||
|
||||
// allow is Allow at a given monotonic time.
|
||||
func (l *Limiter) allow(now int64) (uint64, bool) {
|
||||
// Concurrent Allow calls read the clock before racing for the bucket, so one can see a time another already spent
|
||||
// past and be denied. An unlimited Limiter must never deny, so it skips the bucket.
|
||||
if l.every == 0 {
|
||||
return 0, true
|
||||
}
|
||||
|
||||
for {
|
||||
tat := l.tat.Load()
|
||||
next := max(tat, now)
|
||||
if next-now > l.slack {
|
||||
l.denied.Add(1)
|
||||
return 0, false
|
||||
}
|
||||
|
||||
if l.tat.CompareAndSwap(tat, next+l.every) {
|
||||
return l.denied.Swap(0), true
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,264 @@
|
||||
package ratelimit
|
||||
|
||||
import (
|
||||
"math"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"testing"
|
||||
"time"
|
||||
)
|
||||
|
||||
func mustLimiter(t testing.TB, burst int, every time.Duration) *Limiter {
|
||||
t.Helper()
|
||||
l, err := New(burst, every)
|
||||
if err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
return l
|
||||
}
|
||||
|
||||
func TestLimiterBurstAndRefill(t *testing.T) {
|
||||
l, now := mustLimiter(t, 3, time.Second), int64(0)
|
||||
|
||||
for i := 0; i < 3; i++ {
|
||||
if n, ok := l.allow(now); !ok || n != 0 {
|
||||
t.Fatalf("burst line %d: ok=%v denied=%d", i, ok, n)
|
||||
}
|
||||
}
|
||||
for i := 0; i < 5; i++ {
|
||||
if _, ok := l.allow(now); ok {
|
||||
t.Fatalf("line %d past the burst logged", i)
|
||||
}
|
||||
}
|
||||
|
||||
now += int64(time.Second)
|
||||
if n, ok := l.allow(now); !ok || n != 5 {
|
||||
t.Fatalf("after one refill: ok=%v denied=%d, want true 5", ok, n)
|
||||
}
|
||||
if _, ok := l.allow(now); ok {
|
||||
t.Fatal("one refill earned more than one line")
|
||||
}
|
||||
|
||||
// A long quiet spell refills to burst, not past it, and the count reported above does not carry over.
|
||||
now += int64(time.Hour)
|
||||
for i, want := range []uint64{1, 0, 0} {
|
||||
if n, ok := l.allow(now); !ok || n != want {
|
||||
t.Fatalf("refilled line %d: ok=%v denied=%d, want true %d", i, ok, n, want)
|
||||
}
|
||||
}
|
||||
if _, ok := l.allow(now); ok {
|
||||
t.Fatal("bucket refilled past burst")
|
||||
}
|
||||
}
|
||||
|
||||
func TestLimiterBounds(t *testing.T) {
|
||||
for _, every := range []time.Duration{0, -time.Second} {
|
||||
l := mustLimiter(t, 1, every)
|
||||
for i := 0; i < 100; i++ {
|
||||
if _, ok := l.allow(0); !ok {
|
||||
t.Fatalf("every=%v limited line %d", every, i)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
l := mustLimiter(t, 0, time.Second)
|
||||
if _, ok := l.allow(0); !ok {
|
||||
t.Fatal("burst 0 should allow one line")
|
||||
}
|
||||
if _, ok := l.allow(0); ok {
|
||||
t.Fatal("burst 0 should allow only one line")
|
||||
}
|
||||
}
|
||||
|
||||
func TestLimiterOverflow(t *testing.T) {
|
||||
for _, c := range []struct {
|
||||
burst int
|
||||
every time.Duration
|
||||
}{
|
||||
{math.MaxInt, time.Second},
|
||||
{1, math.MaxInt64},
|
||||
{2, maxSpan/2 + 1},
|
||||
} {
|
||||
if _, err := New(c.burst, c.every); err == nil {
|
||||
t.Fatalf("burst %d every %d should not fit", c.burst, c.every)
|
||||
}
|
||||
}
|
||||
|
||||
l := mustLimiter(t, 2, maxSpan/2)
|
||||
for i := 0; i < 2; i++ {
|
||||
if _, ok := l.allow(0); !ok {
|
||||
t.Fatalf("the largest span should allow its burst, denied call %d", i)
|
||||
}
|
||||
}
|
||||
if _, ok := l.allow(0); ok {
|
||||
t.Fatal("the largest span allowed past its burst")
|
||||
}
|
||||
}
|
||||
|
||||
// Concurrent callers read the clock at slightly different times, which must not limit a Limiter that never limits.
|
||||
func TestLimiterUnlimitedConcurrent(t *testing.T) {
|
||||
l := mustLimiter(t, 1, 0)
|
||||
|
||||
var denied atomic.Int64
|
||||
var wg sync.WaitGroup
|
||||
for i := 0; i < 16; i++ {
|
||||
wg.Add(1)
|
||||
go func() {
|
||||
defer wg.Done()
|
||||
for j := 0; j < 10000; j++ {
|
||||
if _, ok := l.Allow(); !ok {
|
||||
denied.Add(1)
|
||||
}
|
||||
}
|
||||
}()
|
||||
}
|
||||
wg.Wait()
|
||||
|
||||
if got := denied.Load(); got != 0 {
|
||||
t.Fatalf("denied %d lines", got)
|
||||
}
|
||||
}
|
||||
|
||||
func TestLimiterNil(t *testing.T) {
|
||||
var l *Limiter
|
||||
for i := 0; i < 10; i++ {
|
||||
if _, ok := l.Allow(); !ok {
|
||||
t.Fatal("nil limiter limited")
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func TestLimiterConcurrent(t *testing.T) {
|
||||
l := mustLimiter(t, 10, time.Hour)
|
||||
|
||||
var allowed atomic.Int64
|
||||
var reported atomic.Uint64
|
||||
var wg sync.WaitGroup
|
||||
for i := 0; i < 16; i++ {
|
||||
wg.Add(1)
|
||||
go func() {
|
||||
defer wg.Done()
|
||||
for j := 0; j < 1000; j++ {
|
||||
if n, ok := l.allow(0); ok {
|
||||
allowed.Add(1)
|
||||
reported.Add(n)
|
||||
}
|
||||
}
|
||||
}()
|
||||
}
|
||||
wg.Wait()
|
||||
|
||||
if got := allowed.Load(); got != 10 {
|
||||
t.Fatalf("allowed %d lines, want the burst of 10", got)
|
||||
}
|
||||
// An allowed call that lands after some denials reports them, so the rest are still pending.
|
||||
if got := reported.Load() + l.denied.Load(); got != 16*1000-10 {
|
||||
t.Fatalf("denied %d, want %d", got, 16*1000-10)
|
||||
}
|
||||
}
|
||||
|
||||
func BenchmarkAllowAllowed(b *testing.B) {
|
||||
l := mustLimiter(b, 1, time.Nanosecond)
|
||||
b.ReportAllocs()
|
||||
for i := 0; i < b.N; i++ {
|
||||
l.tat.Store(0)
|
||||
l.Allow()
|
||||
}
|
||||
}
|
||||
|
||||
func BenchmarkAllowDenied(b *testing.B) {
|
||||
l := mustLimiter(b, 1, time.Hour)
|
||||
l.Allow()
|
||||
b.ReportAllocs()
|
||||
for i := 0; i < b.N; i++ {
|
||||
l.Allow()
|
||||
}
|
||||
}
|
||||
|
||||
func BenchmarkAllowDeniedParallel(b *testing.B) {
|
||||
l := mustLimiter(b, 1, time.Hour)
|
||||
l.Allow()
|
||||
b.ReportAllocs()
|
||||
b.RunParallel(func(pb *testing.PB) {
|
||||
for pb.Next() {
|
||||
l.Allow()
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
// FuzzLimiter checks the bucket against its exact definition over arbitrary call timings.
|
||||
// A call is denied if and only if some allowed call a would leave too many calls for the time since a:
|
||||
// (allowed calls since a, plus this one) * every > burst * every + (now - a).
|
||||
// Every denied call must also be reported exactly once, by a later allowed call or still pending.
|
||||
func FuzzLimiter(f *testing.F) {
|
||||
f.Add(uint8(3), uint16(1000), []byte{0, 0, 0, 0, 16, 0, 0, 200, 1, 1, 1, 15, 17, 0})
|
||||
f.Add(uint8(1), uint16(1), []byte{0, 16, 15, 16, 0, 255, 0, 0})
|
||||
f.Fuzz(func(t *testing.T, burst uint8, everyUs uint16, steps []byte) {
|
||||
if len(steps) > 512 {
|
||||
steps = steps[:512]
|
||||
}
|
||||
b := int64(burst%16) + 1
|
||||
every := int64(everyUs%1000+1) * int64(time.Microsecond)
|
||||
l := mustLimiter(t, int(b), time.Duration(every))
|
||||
|
||||
var now int64
|
||||
var allowedAt []int64
|
||||
var denied, reported uint64
|
||||
for _, s := range steps {
|
||||
// Each step advances by a sixteenth of every, so timings land on and around the refill.
|
||||
now += int64(s) * every / 16
|
||||
|
||||
over := false
|
||||
for i, a := range allowedAt {
|
||||
if int64(len(allowedAt)-i+1)*every > b*every+now-a {
|
||||
over = true
|
||||
break
|
||||
}
|
||||
}
|
||||
|
||||
n, ok := l.allow(now)
|
||||
if ok == over {
|
||||
t.Fatalf("at %d with %d allowed before: ok=%v, the definition says %v", now, len(allowedAt), ok, !over)
|
||||
}
|
||||
if !ok {
|
||||
denied++
|
||||
continue
|
||||
}
|
||||
reported += n
|
||||
allowedAt = append(allowedAt, now)
|
||||
}
|
||||
|
||||
if got := reported + l.denied.Load(); got != denied {
|
||||
t.Fatalf("reported %d denied calls, made %d", got, denied)
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
// Concurrent callers on the real clock must account for every denied call exactly once.
|
||||
func TestLimiterDeniedConservedConcurrent(t *testing.T) {
|
||||
l := mustLimiter(t, 4, time.Microsecond)
|
||||
|
||||
var denied, reported atomic.Uint64
|
||||
var wg sync.WaitGroup
|
||||
for i := 0; i < 16; i++ {
|
||||
wg.Add(1)
|
||||
go func() {
|
||||
defer wg.Done()
|
||||
for j := 0; j < 20000; j++ {
|
||||
if n, ok := l.Allow(); ok {
|
||||
reported.Add(n)
|
||||
} else {
|
||||
denied.Add(1)
|
||||
}
|
||||
}
|
||||
}()
|
||||
}
|
||||
wg.Wait()
|
||||
|
||||
if denied.Load() == 0 {
|
||||
t.Fatal("nothing was denied, the test proves nothing")
|
||||
}
|
||||
if got := reported.Load() + l.denied.Load(); got != denied.Load() {
|
||||
t.Fatalf("reported %d denied calls, made %d", got, denied.Load())
|
||||
}
|
||||
}
|
||||
@@ -60,6 +60,20 @@ type sshDeviceInfoFlags struct {
|
||||
func wireSSHReload(l *slog.Logger, ssh *sshd.SSHServer, c *config.C) {
|
||||
c.RegisterReloadCallback(func(c *config.C) {
|
||||
if c.GetBool("sshd.enabled", false) {
|
||||
// Restarting the server drops every open session, the one that asked for this reload included. Only restart
|
||||
// when it isn't running or its config changed. The host key file can change under the same config, new
|
||||
// handshakes use the new key without a restart
|
||||
if ssh.Running() && !c.HasChanged("sshd") {
|
||||
key, err := loadSSHHostKey(c)
|
||||
if err == nil {
|
||||
err = ssh.SetHostKey(key)
|
||||
}
|
||||
if err != nil {
|
||||
l.Error("Failed to reconfigure the sshd", "error", err)
|
||||
ssh.Stop()
|
||||
}
|
||||
return
|
||||
}
|
||||
sshRun, err := configSSH(l, ssh, c)
|
||||
if err != nil {
|
||||
l.Error("Failed to reconfigure the sshd", "error", err)
|
||||
@@ -74,6 +88,22 @@ func wireSSHReload(l *slog.Logger, ssh *sshd.SSHServer, c *config.C) {
|
||||
})
|
||||
}
|
||||
|
||||
// loadSSHHostKey returns sshd.host_key, read from its file when it names one.
|
||||
func loadSSHHostKey(c *config.C) ([]byte, error) {
|
||||
hostKeyPathOrKey := c.GetString("sshd.host_key", "")
|
||||
if hostKeyPathOrKey == "" {
|
||||
return nil, fmt.Errorf("sshd.host_key must be provided")
|
||||
}
|
||||
if strings.Contains(hostKeyPathOrKey, "-----BEGIN") {
|
||||
return []byte(hostKeyPathOrKey), nil
|
||||
}
|
||||
b, err := os.ReadFile(hostKeyPathOrKey)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("error while loading sshd.host_key file: %s", err)
|
||||
}
|
||||
return b, nil
|
||||
}
|
||||
|
||||
// configSSH reads the ssh info out of the passed-in Config and
|
||||
// updates the passed-in SSHServer. On success, it returns a function
|
||||
// that callers may invoke to run the configured ssh server. On
|
||||
@@ -92,19 +122,9 @@ func configSSH(l *slog.Logger, ssh *sshd.SSHServer, c *config.C) (func(), error)
|
||||
return nil, fmt.Errorf("sshd.listen can not use port 22")
|
||||
}
|
||||
|
||||
hostKeyPathOrKey := c.GetString("sshd.host_key", "")
|
||||
if hostKeyPathOrKey == "" {
|
||||
return nil, fmt.Errorf("sshd.host_key must be provided")
|
||||
}
|
||||
|
||||
var hostKeyBytes []byte
|
||||
if strings.Contains(hostKeyPathOrKey, "-----BEGIN") {
|
||||
hostKeyBytes = []byte(hostKeyPathOrKey)
|
||||
} else {
|
||||
hostKeyBytes, err = os.ReadFile(hostKeyPathOrKey)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("error while loading sshd.host_key file: %s", err)
|
||||
}
|
||||
hostKeyBytes, err := loadSSHHostKey(c)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
err = ssh.SetHostKey(hostKeyBytes)
|
||||
@@ -684,10 +704,14 @@ func sshCreateTunnel(ifce *Interface, fs any, a []string, w sshd.StringWriter) e
|
||||
}
|
||||
}
|
||||
|
||||
hostInfo = ifce.handshakeManager.StartHandshake(vpnAddr, nil)
|
||||
if addr.IsValid() {
|
||||
hostInfo.SetRemote(addr)
|
||||
}
|
||||
// Set the remote from the callback, which StartHandshake runs under the handshake manager lock
|
||||
// before the hostinfo reaches the handshake loop or the static-host trigger, so the address is
|
||||
// in the RemoteList before the first handshake is sent.
|
||||
ifce.handshakeManager.StartHandshake(vpnAddr, func(hh *HandshakeHostInfo) {
|
||||
if addr.IsValid() {
|
||||
hh.hostinfo.SetRemote(addr)
|
||||
}
|
||||
})
|
||||
|
||||
return w.WriteLine("Created")
|
||||
}
|
||||
|
||||
+48
-13
@@ -8,14 +8,18 @@ import (
|
||||
"log/slog"
|
||||
"net"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
|
||||
"github.com/armon/go-radix"
|
||||
"golang.org/x/crypto/ssh"
|
||||
)
|
||||
|
||||
type SSHServer struct {
|
||||
config *ssh.ServerConfig
|
||||
l *slog.Logger
|
||||
// config is replaced whole when the host key changes, ssh.ServerConfig.AddHostKey only replaces a key of the same
|
||||
// algorithm and appends any other
|
||||
config atomic.Pointer[ssh.ServerConfig]
|
||||
authenticate func(ssh.ConnMetadata, ssh.PublicKey) (*ssh.Permissions, error)
|
||||
l *slog.Logger
|
||||
|
||||
certChecker *ssh.CertChecker
|
||||
|
||||
@@ -28,7 +32,12 @@ type SSHServer struct {
|
||||
// List of available commands
|
||||
helpCommand *Command
|
||||
commands *radix.Tree
|
||||
listener net.Listener
|
||||
listener atomic.Pointer[net.Listener]
|
||||
|
||||
// runs counts the Runs that are listening. A reload's Run can start before the one it replaced has returned
|
||||
runs atomic.Int32
|
||||
// addr is where the newest Run listens, nil when none is
|
||||
addr atomic.Pointer[net.Addr]
|
||||
|
||||
// ctx parents per-Run contexts. Cancelling it (e.g. via Control.Stop) tears the server down even
|
||||
// across reloads, since each Run derives a fresh child rather than reusing this one directly.
|
||||
@@ -85,10 +94,8 @@ func NewSSHServer(ctx context.Context, l *slog.Logger) (*SSHServer, error) {
|
||||
},
|
||||
}
|
||||
|
||||
s.config = &ssh.ServerConfig{
|
||||
PublicKeyCallback: cc.Authenticate,
|
||||
ServerVersion: fmt.Sprintf("SSH-2.0-Nebula???"),
|
||||
}
|
||||
s.authenticate = cc.Authenticate
|
||||
s.config.Store(s.newConfig())
|
||||
|
||||
s.RegisterCommand(&Command{
|
||||
Name: "help",
|
||||
@@ -107,10 +114,32 @@ func (s *SSHServer) SetHostKey(hostPrivateKey []byte) error {
|
||||
return fmt.Errorf("failed to parse private key: %s", err)
|
||||
}
|
||||
|
||||
s.config.AddHostKey(private)
|
||||
config := s.newConfig()
|
||||
config.AddHostKey(private)
|
||||
s.config.Store(config)
|
||||
return nil
|
||||
}
|
||||
|
||||
func (s *SSHServer) newConfig() *ssh.ServerConfig {
|
||||
return &ssh.ServerConfig{
|
||||
PublicKeyCallback: s.authenticate,
|
||||
ServerVersion: fmt.Sprintf("SSH-2.0-Nebula???"),
|
||||
}
|
||||
}
|
||||
|
||||
// Addr is the address the newest Run listens on, nil when none is.
|
||||
func (s *SSHServer) Addr() net.Addr {
|
||||
if a := s.addr.Load(); a != nil {
|
||||
return *a
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// Running is true while a Run is listening.
|
||||
func (s *SSHServer) Running() bool {
|
||||
return s.runs.Load() > 0
|
||||
}
|
||||
|
||||
func (s *SSHServer) ClearTrustedCAs() {
|
||||
s.authLock.Lock()
|
||||
s.trustedCAs = []ssh.PublicKey{}
|
||||
@@ -181,10 +210,16 @@ func (s *SSHServer) Run(addr string) error {
|
||||
// this run owns. They start equal but a fast reload may overwrite s.listener with the next run's
|
||||
// listener before this run's watcher fires, so each run must close its own listener via the local
|
||||
// reference.
|
||||
s.listener = listener
|
||||
s.listener.Store(&listener)
|
||||
s.runs.Add(1)
|
||||
bound := listener.Addr()
|
||||
s.addr.Store(&bound)
|
||||
|
||||
runCtx, cancel := context.WithCancel(s.ctx)
|
||||
defer cancel()
|
||||
// Deferred after cancel so they run first, a run on its way out no longer counts as listening
|
||||
defer s.runs.Add(-1)
|
||||
defer s.addr.CompareAndSwap(&bound, nil)
|
||||
|
||||
// Close the listener when this run's context is cancelled. That can come from the parent
|
||||
// (Control.Stop), from Run returning normally (defer cancel above), or transitively when a sibling
|
||||
@@ -197,7 +232,7 @@ func (s *SSHServer) Run(addr string) error {
|
||||
}
|
||||
}()
|
||||
|
||||
s.l.Info("SSH server is listening", "sshListener", addr)
|
||||
s.l.Info("SSH server is listening", "sshListener", bound)
|
||||
|
||||
// Run loops until there is an error
|
||||
s.run(runCtx, listener)
|
||||
@@ -226,7 +261,7 @@ func (s *SSHServer) run(ctx context.Context, listener net.Listener) {
|
||||
<-sessionContext.Done()
|
||||
c.Close()
|
||||
}()
|
||||
conn, chans, reqs, err := ssh.NewServerConn(c, s.config)
|
||||
conn, chans, reqs, err := ssh.NewServerConn(c, s.config.Load())
|
||||
fp := ""
|
||||
if conn != nil {
|
||||
fp = conn.Permissions.Extensions["fp"]
|
||||
@@ -264,8 +299,8 @@ func (s *SSHServer) run(ctx context.Context, listener net.Listener) {
|
||||
}
|
||||
|
||||
func (s *SSHServer) Stop() {
|
||||
if s.listener != nil {
|
||||
if err := s.listener.Close(); err != nil {
|
||||
if l := s.listener.Load(); l != nil {
|
||||
if err := (*l).Close(); err != nil && !errors.Is(err, net.ErrClosed) {
|
||||
s.l.Warn("Failed to close the sshd listener", "error", err)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -0,0 +1,196 @@
|
||||
package nebula
|
||||
|
||||
import (
|
||||
"crypto"
|
||||
"crypto/ed25519"
|
||||
"crypto/rand"
|
||||
"crypto/rsa"
|
||||
"encoding/pem"
|
||||
"fmt"
|
||||
"net"
|
||||
"os"
|
||||
"path/filepath"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/slackhq/nebula/sshd"
|
||||
"github.com/slackhq/nebula/test"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
"golang.org/x/crypto/ssh"
|
||||
)
|
||||
|
||||
type sshdReloadHarness struct {
|
||||
c *config.C
|
||||
server *sshd.SSHServer
|
||||
keyFile string
|
||||
signer ssh.Signer
|
||||
extra string
|
||||
}
|
||||
|
||||
func newSSHDReloadHarness(t *testing.T) *sshdReloadHarness {
|
||||
t.Helper()
|
||||
l := test.NewLogger()
|
||||
h := &sshdReloadHarness{keyFile: filepath.Join(t.TempDir(), "host_key")}
|
||||
h.writeHostKey(t, newEd25519Key(t))
|
||||
|
||||
_, userKey, err := ed25519.GenerateKey(rand.Reader)
|
||||
require.NoError(t, err)
|
||||
h.signer, err = ssh.NewSignerFromKey(userKey)
|
||||
require.NoError(t, err)
|
||||
|
||||
h.c = config.NewC(l)
|
||||
require.NoError(t, h.c.LoadString(h.config()))
|
||||
h.server, err = sshd.NewSSHServer(t.Context(), l)
|
||||
require.NoError(t, err)
|
||||
wireSSHReload(l, h.server, h.c)
|
||||
run, err := configSSH(l, h.server, h.c)
|
||||
require.NoError(t, err)
|
||||
go run()
|
||||
t.Cleanup(h.server.Stop)
|
||||
require.Eventually(t, h.server.Running, 5*time.Second, 10*time.Millisecond)
|
||||
return h
|
||||
}
|
||||
|
||||
func newEd25519Key(t *testing.T) crypto.Signer {
|
||||
_, key, err := ed25519.GenerateKey(rand.Reader)
|
||||
require.NoError(t, err)
|
||||
return key
|
||||
}
|
||||
|
||||
// writeHostKey replaces the host key file and returns the public key a client should now see
|
||||
func (h *sshdReloadHarness) writeHostKey(t *testing.T, key crypto.Signer) ssh.PublicKey {
|
||||
t.Helper()
|
||||
block, err := ssh.MarshalPrivateKey(key, "")
|
||||
require.NoError(t, err)
|
||||
require.NoError(t, os.WriteFile(h.keyFile, pem.EncodeToMemory(block), 0o600))
|
||||
pub, err := ssh.NewPublicKey(key.Public())
|
||||
require.NoError(t, err)
|
||||
return pub
|
||||
}
|
||||
|
||||
func (h *sshdReloadHarness) config() string {
|
||||
return fmt.Sprintf("sshd:\n enabled: true\n listen: 127.0.0.1:0\n host_key: %s\n authorized_users:\n - user: test\n keys: [%q]\n%s",
|
||||
h.keyFile, string(ssh.MarshalAuthorizedKey(h.signer.PublicKey())), h.extra)
|
||||
}
|
||||
|
||||
// dial connects to wherever the server listens now, a restart on :0 moves it. hostKeyAlgorithms limits which host keys
|
||||
// the client accepts, none means any
|
||||
func (h *sshdReloadHarness) dial(hostKeyAlgorithms ...string) (*ssh.Client, ssh.PublicKey, error) {
|
||||
addr := h.server.Addr()
|
||||
if addr == nil {
|
||||
return nil, nil, fmt.Errorf("not listening")
|
||||
}
|
||||
var hostKey ssh.PublicKey
|
||||
c, err := ssh.Dial("tcp", addr.String(), &ssh.ClientConfig{
|
||||
User: "test",
|
||||
Auth: []ssh.AuthMethod{ssh.PublicKeys(h.signer)},
|
||||
HostKeyCallback: func(_ string, _ net.Addr, key ssh.PublicKey) error {
|
||||
hostKey = key
|
||||
return nil
|
||||
},
|
||||
HostKeyAlgorithms: hostKeyAlgorithms,
|
||||
Timeout: time.Second,
|
||||
})
|
||||
return c, hostKey, err
|
||||
}
|
||||
|
||||
func (h *sshdReloadHarness) connect(t *testing.T) (*ssh.Client, ssh.PublicKey) {
|
||||
t.Helper()
|
||||
var client *ssh.Client
|
||||
var hostKey ssh.PublicKey
|
||||
require.Eventually(t, func() bool {
|
||||
var err error
|
||||
client, hostKey, err = h.dial()
|
||||
return err == nil
|
||||
}, 5*time.Second, 20*time.Millisecond)
|
||||
t.Cleanup(func() { _ = client.Close() })
|
||||
return client, hostKey
|
||||
}
|
||||
|
||||
// dropped is true when the server closed the client's connection within wait
|
||||
func dropped(client *ssh.Client, wait time.Duration) bool {
|
||||
done := make(chan struct{})
|
||||
go func() {
|
||||
_ = client.Wait()
|
||||
close(done)
|
||||
}()
|
||||
select {
|
||||
case <-done:
|
||||
return true
|
||||
case <-time.After(wait):
|
||||
return false
|
||||
}
|
||||
}
|
||||
|
||||
// Restarting sshd drops every open session, the one that asked for the reload included. A reload that touches nothing
|
||||
// sshd reads leaves it alone
|
||||
func TestSSHDReload_UnchangedKeepsSessions(t *testing.T) {
|
||||
h := newSSHDReloadHarness(t)
|
||||
client, _ := h.connect(t)
|
||||
|
||||
require.NoError(t, h.c.ReloadConfigString(h.config()+"punchy:\n punch: true\n"))
|
||||
assert.False(t, dropped(client, 500*time.Millisecond), "an unchanged sshd was restarted")
|
||||
}
|
||||
|
||||
func TestSSHDReload_ChangedConfigRestarts(t *testing.T) {
|
||||
h := newSSHDReloadHarness(t)
|
||||
client, _ := h.connect(t)
|
||||
|
||||
h.extra = " trusted_cas: []\n"
|
||||
require.NoError(t, h.c.ReloadConfigString(h.config()))
|
||||
assert.True(t, dropped(client, 5*time.Second), "a changed sshd config was not applied")
|
||||
require.Eventually(t, h.server.Running, 5*time.Second, 10*time.Millisecond)
|
||||
h.connect(t)
|
||||
}
|
||||
|
||||
// The config can stay the same while the host key file it names changes. New connections get the new key, open
|
||||
// sessions are left alone
|
||||
func TestSSHDReload_ChangedHostKeyFile(t *testing.T) {
|
||||
h := newSSHDReloadHarness(t)
|
||||
client, _ := h.connect(t)
|
||||
|
||||
want := h.writeHostKey(t, newEd25519Key(t))
|
||||
require.NoError(t, h.c.ReloadConfigString(h.config()))
|
||||
_, got := h.connect(t)
|
||||
assert.Equal(t, want.Marshal(), got.Marshal(), "a new host key was not applied")
|
||||
assert.False(t, dropped(client, 500*time.Millisecond), "a new host key restarted sshd")
|
||||
}
|
||||
|
||||
// A host key of another algorithm replaces the old one rather than being offered next to it
|
||||
func TestSSHDReload_HostKeyAlgorithmChange(t *testing.T) {
|
||||
h := newSSHDReloadHarness(t)
|
||||
h.connect(t)
|
||||
|
||||
rsaKey, err := rsa.GenerateKey(rand.Reader, 2048)
|
||||
require.NoError(t, err)
|
||||
want := h.writeHostKey(t, rsaKey)
|
||||
require.NoError(t, h.c.ReloadConfigString(h.config()))
|
||||
|
||||
_, got := h.connect(t)
|
||||
assert.Equal(t, want.Marshal(), got.Marshal())
|
||||
_, _, err = h.dial(ssh.KeyAlgoED25519)
|
||||
assert.Error(t, err, "the old ed25519 host key is still offered")
|
||||
}
|
||||
|
||||
// A server that isn't running, say its port was busy, is tried again on every reload as before
|
||||
func TestSSHDReload_NotRunningRestarts(t *testing.T) {
|
||||
h := newSSHDReloadHarness(t)
|
||||
h.server.Stop()
|
||||
require.Eventually(t, func() bool { return !h.server.Running() }, 5*time.Second, 10*time.Millisecond)
|
||||
|
||||
require.NoError(t, h.c.ReloadConfigString(h.config()))
|
||||
require.Eventually(t, h.server.Running, 5*time.Second, 10*time.Millisecond)
|
||||
h.connect(t)
|
||||
}
|
||||
|
||||
// A host key that no longer loads stops sshd, the same as a config that doesn't
|
||||
func TestSSHDReload_BadHostKeyFileStops(t *testing.T) {
|
||||
h := newSSHDReloadHarness(t)
|
||||
h.connect(t)
|
||||
|
||||
require.NoError(t, os.WriteFile(h.keyFile, []byte("not a key"), 0o600))
|
||||
require.NoError(t, h.c.ReloadConfigString(h.config()))
|
||||
require.Eventually(t, func() bool { return !h.server.Running() }, 5*time.Second, 10*time.Millisecond)
|
||||
}
|
||||
@@ -38,6 +38,11 @@ type statsServer struct {
|
||||
runMu sync.Mutex
|
||||
runCfg *statsConfig
|
||||
run *statsRuntime // non-nil while a runtime is live
|
||||
// Closed when the last runtime's capture loop has returned. Stop waits on it so nothing captures once it returns,
|
||||
// and a reload's new runtime never overlaps the old one, go-metrics keeps its GC capture state in unguarded globals
|
||||
loopDone chan struct{}
|
||||
// See newStatsServerFromConfig, never changes after it
|
||||
emitStats func()
|
||||
}
|
||||
|
||||
// statsRuntime is the live state owned by a single Start invocation. Start
|
||||
@@ -82,12 +87,16 @@ type promConfig struct {
|
||||
//
|
||||
// Start is safe to call unconditionally: it no-ops when stats are disabled.
|
||||
// The returned pointer is always non-nil, even on error.
|
||||
func newStatsServerFromConfig(ctx context.Context, l *slog.Logger, c *config.C, buildVersion string, configTest bool) (*statsServer, error) {
|
||||
//
|
||||
// emitStats, if set, updates metrics on every capture, ahead of the exporters so each export sees what it just set. It
|
||||
// is fixed here, before the reload callback exists, so every runtime has it.
|
||||
func newStatsServerFromConfig(ctx context.Context, l *slog.Logger, c *config.C, buildVersion string, configTest bool, emitStats func()) (*statsServer, error) {
|
||||
s := &statsServer{
|
||||
l: l,
|
||||
ctx: ctx,
|
||||
buildVersion: buildVersion,
|
||||
configTest: configTest,
|
||||
emitStats: emitStats,
|
||||
}
|
||||
|
||||
c.RegisterReloadCallback(func(c *config.C) {
|
||||
@@ -120,7 +129,6 @@ func (s *statsServer) reload(c *config.C, initial bool) error {
|
||||
s.runMu.Lock()
|
||||
sameCfg := s.runCfg != nil && *s.runCfg == newCfg
|
||||
s.runCfg = &newCfg
|
||||
running := s.run != nil
|
||||
s.runMu.Unlock()
|
||||
|
||||
s.enabled.Store(enabled)
|
||||
@@ -129,9 +137,8 @@ func (s *statsServer) reload(c *config.C, initial bool) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
if running {
|
||||
s.Stop()
|
||||
}
|
||||
// Even with nothing running, a runtime whose listener failed can still be finishing its last pass
|
||||
s.Stop()
|
||||
if enabled && !s.configTest {
|
||||
go s.Start()
|
||||
}
|
||||
@@ -153,13 +160,19 @@ func (s *statsServer) Start() {
|
||||
return
|
||||
}
|
||||
cfg := *s.runCfg
|
||||
captureFns, listener := s.buildRuntime(cfg)
|
||||
runCtx, cancel := context.WithCancel(s.ctx)
|
||||
captureFns, listener := s.buildRuntime(runCtx, cfg)
|
||||
rt := &statsRuntime{cancel: cancel, listener: listener}
|
||||
s.run = rt
|
||||
done := make(chan struct{})
|
||||
s.loopDone = done
|
||||
s.runMu.Unlock()
|
||||
|
||||
go captureStatsLoop(runCtx, cfg.interval, captureFns)
|
||||
// Prime before serving, so a scrape that lands before the first tick sees real values instead of zeros (issue #907)
|
||||
for _, fn := range captureFns {
|
||||
fn()
|
||||
}
|
||||
go captureStatsLoop(runCtx, cfg.interval, captureFns, done)
|
||||
|
||||
cleanExit := true
|
||||
if listener == nil {
|
||||
@@ -218,26 +231,32 @@ func (s *statsServer) serveListener(listener *http.Server) bool {
|
||||
|
||||
// Stop tears down the active runtime, if any. Idempotent.
|
||||
func (s *statsServer) Stop() {
|
||||
// Held until the last pass is done, a Start that slipped in after s.run cleared would capture alongside it
|
||||
s.runMu.Lock()
|
||||
defer s.runMu.Unlock()
|
||||
rt := s.run
|
||||
s.run = nil
|
||||
s.runMu.Unlock()
|
||||
if rt == nil {
|
||||
return
|
||||
}
|
||||
rt.cancel()
|
||||
if rt.listener != nil {
|
||||
shutdownCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
|
||||
if err := rt.listener.Shutdown(shutdownCtx); err != nil {
|
||||
s.l.Warn("Failed to shut down prometheus stats listener", "error", err)
|
||||
if rt != nil {
|
||||
rt.cancel()
|
||||
if rt.listener != nil {
|
||||
shutdownCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
|
||||
if err := rt.listener.Shutdown(shutdownCtx); err != nil {
|
||||
s.l.Warn("Failed to shut down prometheus stats listener", "error", err)
|
||||
}
|
||||
cancel()
|
||||
}
|
||||
cancel()
|
||||
}
|
||||
// Nothing captures once Stop returns. The last runtime is cancelled by now, whoever cleared it, so this is at most
|
||||
// the pass it was in. Nothing in a pass takes runMu
|
||||
if s.loopDone != nil {
|
||||
<-s.loopDone
|
||||
}
|
||||
}
|
||||
|
||||
// buildRuntime produces the capture functions and, for prometheus, an un-served
|
||||
// http.Server from cfg. cfg has already been validated by loadStatsConfig.
|
||||
func (s *statsServer) buildRuntime(cfg statsConfig) ([]func(), *http.Server) {
|
||||
// A graphite send in progress is abandoned when ctx is done.
|
||||
func (s *statsServer) buildRuntime(ctx context.Context, cfg statsConfig) ([]func(), *http.Server) {
|
||||
// rcrowley/go-metrics guards these registrations with a private sync.Once,
|
||||
// so subsequent reloads are no-ops.
|
||||
metrics.RegisterDebugGCStats(metrics.DefaultRegistry)
|
||||
@@ -247,6 +266,9 @@ func (s *statsServer) buildRuntime(cfg statsConfig) ([]func(), *http.Server) {
|
||||
func() { metrics.CaptureDebugGCStatsOnce(metrics.DefaultRegistry) },
|
||||
func() { metrics.CaptureRuntimeMemStatsOnce(metrics.DefaultRegistry) },
|
||||
}
|
||||
if s.emitStats != nil {
|
||||
captureFns = append(captureFns, s.emitStats)
|
||||
}
|
||||
|
||||
switch cfg.typ {
|
||||
case "graphite":
|
||||
@@ -262,7 +284,7 @@ func (s *statsServer) buildRuntime(cfg statsConfig) ([]func(), *http.Server) {
|
||||
Percentiles: []float64{0.5, 0.75, 0.95, 0.99, 0.999},
|
||||
}
|
||||
captureFns = append(captureFns, func() {
|
||||
if err := graphiteOnce(gcfg); err != nil {
|
||||
if err := graphiteOnce(ctx, gcfg, graphiteTimeout); err != nil && ctx.Err() == nil {
|
||||
s.l.Error("Graphite export failed", "error", err)
|
||||
}
|
||||
})
|
||||
@@ -309,8 +331,10 @@ func (s *statsServer) buildRuntime(cfg statsConfig) ([]func(), *http.Server) {
|
||||
return captureFns, nil
|
||||
}
|
||||
|
||||
// captureStatsLoop runs each fn on every tick of d until ctx is cancelled.
|
||||
func captureStatsLoop(ctx context.Context, d time.Duration, fns []func()) {
|
||||
// captureStatsLoop runs each fn on every tick of d until ctx is cancelled, then
|
||||
// closes done.
|
||||
func captureStatsLoop(ctx context.Context, d time.Duration, fns []func(), done chan struct{}) {
|
||||
defer close(done)
|
||||
t := time.NewTicker(d)
|
||||
defer t.Stop()
|
||||
for {
|
||||
|
||||
+294
-29
@@ -2,13 +2,18 @@ package nebula
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"io"
|
||||
"log/slog"
|
||||
"net"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"strconv"
|
||||
"sync/atomic"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/rcrowley/go-metrics"
|
||||
"github.com/slackhq/nebula/config"
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
@@ -110,15 +115,14 @@ func TestStatsServer_reload_initial_prometheus(t *testing.T) {
|
||||
}
|
||||
|
||||
func TestStatsServer_Start_graphite_blocksUntilStop(t *testing.T) {
|
||||
sink := newGraphiteSink(t)
|
||||
defer sink.Close()
|
||||
h := newGraphiteHost(t, 0)
|
||||
|
||||
s, c := newTestStatsServer(t)
|
||||
setStatsConfig(c, map[string]any{
|
||||
"type": "graphite",
|
||||
"interval": "1s",
|
||||
"protocol": "tcp",
|
||||
"host": sink.Addr(),
|
||||
"host": h.ln.Addr().String(),
|
||||
"prefix": "test",
|
||||
})
|
||||
require.NoError(t, s.reload(c, true))
|
||||
@@ -371,35 +375,19 @@ func waitForListening(t *testing.T, addr string) {
|
||||
})
|
||||
}
|
||||
|
||||
// graphiteSink is a minimal TCP accept-and-discard server so graphiteOnce
|
||||
// calls in tests don't spam error logs or wedge on connection refused.
|
||||
type graphiteSink struct {
|
||||
ln net.Listener
|
||||
}
|
||||
|
||||
func newGraphiteSink(t *testing.T) *graphiteSink {
|
||||
t.Helper()
|
||||
ln, err := net.Listen("tcp", "127.0.0.1:0")
|
||||
require.NoError(t, err)
|
||||
g := &graphiteSink{ln: ln}
|
||||
go func() {
|
||||
for {
|
||||
conn, err := ln.Accept()
|
||||
if err != nil {
|
||||
return
|
||||
}
|
||||
go func(c net.Conn) {
|
||||
_, _ = io.Copy(io.Discard, c)
|
||||
_ = c.Close()
|
||||
}(conn)
|
||||
// stopAtCleanup returns the ctx to build *s with, and stops it when the test ends. Cancelling first turns away a Start
|
||||
// a reload spawned that hasn't run yet, then Stop waits out the last pass, so nothing captures into the next test
|
||||
func stopAtCleanup(t *testing.T, s **statsServer) context.Context {
|
||||
ctx, cancel := context.WithCancel(context.Background())
|
||||
t.Cleanup(func() {
|
||||
cancel()
|
||||
if *s != nil {
|
||||
(*s).Stop()
|
||||
}
|
||||
}()
|
||||
return g
|
||||
})
|
||||
return ctx
|
||||
}
|
||||
|
||||
func (g *graphiteSink) Addr() string { return g.ln.Addr().String() }
|
||||
func (g *graphiteSink) Close() { _ = g.ln.Close() }
|
||||
|
||||
func freeTCPPort(t *testing.T) string {
|
||||
t.Helper()
|
||||
ln, err := net.Listen("tcp", "127.0.0.1:0")
|
||||
@@ -408,3 +396,280 @@ func freeTCPPort(t *testing.T) string {
|
||||
require.NoError(t, ln.Close())
|
||||
return strconv.Itoa(port)
|
||||
}
|
||||
|
||||
// Each export has to see what emitStats set in the same pass
|
||||
func TestStatsServer_emitStatsRunsBeforeExport(t *testing.T) {
|
||||
defer metrics.DefaultRegistry.Unregister("emitter.order.test")
|
||||
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
setStatsConfig(c, map[string]any{
|
||||
"type": "prometheus",
|
||||
"interval": "1s",
|
||||
"listen": "127.0.0.1:0",
|
||||
"path": "/metrics",
|
||||
})
|
||||
gauge := metrics.GetOrRegisterGauge("emitter.order.test", nil)
|
||||
s, err := newStatsServerFromConfig(t.Context(), l, c, "", false, func() { gauge.Update(42) })
|
||||
require.NoError(t, err)
|
||||
|
||||
s.runMu.Lock()
|
||||
fns, srv := s.buildRuntime(t.Context(), *s.runCfg)
|
||||
s.runMu.Unlock()
|
||||
require.NotNil(t, srv)
|
||||
for _, fn := range fns {
|
||||
fn()
|
||||
}
|
||||
|
||||
rec := httptest.NewRecorder()
|
||||
srv.Handler.ServeHTTP(rec, httptest.NewRequest("GET", "/metrics", nil))
|
||||
assert.Contains(t, rec.Body.String(), "emitter_order_test 42")
|
||||
}
|
||||
|
||||
// A scrape that lands the moment the listener is up already sees what emitStats set (issue #907)
|
||||
func TestStatsServer_Start_primesBeforeServing(t *testing.T) {
|
||||
defer metrics.DefaultRegistry.Unregister("prime.before.serving")
|
||||
|
||||
port := freeTCPPort(t)
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
setStatsConfig(c, map[string]any{
|
||||
"type": "prometheus",
|
||||
"interval": "1h",
|
||||
"listen": "127.0.0.1:" + port,
|
||||
"path": "/metrics",
|
||||
})
|
||||
gauge := metrics.GetOrRegisterGauge("prime.before.serving", nil)
|
||||
// The pause would let a listener that came up before the prime serve the zero value
|
||||
var s *statsServer
|
||||
ctx := stopAtCleanup(t, &s)
|
||||
s, err := newStatsServerFromConfig(ctx, l, c, "", false, func() {
|
||||
time.Sleep(200 * time.Millisecond)
|
||||
gauge.Update(42)
|
||||
})
|
||||
require.NoError(t, err)
|
||||
|
||||
go s.Start()
|
||||
waitForListening(t, "127.0.0.1:"+port)
|
||||
|
||||
resp, err := http.Get("http://127.0.0.1:" + port + "/metrics")
|
||||
require.NoError(t, err)
|
||||
defer resp.Body.Close()
|
||||
body, err := io.ReadAll(resp.Body)
|
||||
require.NoError(t, err)
|
||||
assert.Contains(t, string(body), "prime_before_serving 42")
|
||||
}
|
||||
|
||||
// Start captures once before the first tick, the interval here is an hour
|
||||
func TestStatsServer_Start_primes(t *testing.T) {
|
||||
ln, err := net.Listen("tcp", "127.0.0.1:0")
|
||||
require.NoError(t, err)
|
||||
defer ln.Close()
|
||||
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "1h", "host": ln.Addr().String()})
|
||||
primed := make(chan struct{}, 1)
|
||||
var s *statsServer
|
||||
ctx := stopAtCleanup(t, &s)
|
||||
s, err = newStatsServerFromConfig(ctx, l, c, "", false, func() {
|
||||
select {
|
||||
case primed <- struct{}{}:
|
||||
default:
|
||||
}
|
||||
})
|
||||
require.NoError(t, err)
|
||||
|
||||
go s.Start()
|
||||
select {
|
||||
case <-primed:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("nothing was captured before the first tick")
|
||||
}
|
||||
}
|
||||
|
||||
// Graphite has to ship what emitStats set in the same pass, not the pass before
|
||||
func TestStatsServer_emitStatsRunsBeforeGraphiteExport(t *testing.T) {
|
||||
defer metrics.DefaultRegistry.Unregister("emitter.graphite.test")
|
||||
|
||||
ln, err := net.Listen("tcp", "127.0.0.1:0")
|
||||
require.NoError(t, err)
|
||||
defer ln.Close()
|
||||
received := make(chan string, 1)
|
||||
go func() {
|
||||
conn, err := ln.Accept()
|
||||
if err != nil {
|
||||
return
|
||||
}
|
||||
defer conn.Close()
|
||||
b, _ := io.ReadAll(conn)
|
||||
received <- string(b)
|
||||
}()
|
||||
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "1h", "host": ln.Addr().String()})
|
||||
gauge := metrics.GetOrRegisterGauge("emitter.graphite.test", nil)
|
||||
s, err := newStatsServerFromConfig(t.Context(), l, c, "", false, func() { gauge.Update(42) })
|
||||
require.NoError(t, err)
|
||||
|
||||
s.runMu.Lock()
|
||||
fns, _ := s.buildRuntime(t.Context(), *s.runCfg)
|
||||
s.runMu.Unlock()
|
||||
for _, fn := range fns {
|
||||
fn()
|
||||
}
|
||||
|
||||
select {
|
||||
case body := <-received:
|
||||
assert.Contains(t, body, ".emitter.graphite.test.value 42 ")
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("graphite sent nothing")
|
||||
}
|
||||
}
|
||||
|
||||
// stats.interval reloads, and the runtime a reload starts still has emitStats
|
||||
func TestStatsServer_reloadMovesTheInterval(t *testing.T) {
|
||||
var passes atomic.Int64
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
// Nothing listens here, each export fails at once
|
||||
host := "127.0.0.1:" + freeTCPPort(t)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "1h", "host": host})
|
||||
var s *statsServer
|
||||
ctx := stopAtCleanup(t, &s)
|
||||
s, err := newStatsServerFromConfig(ctx, l, c, "", false, func() { passes.Add(1) })
|
||||
require.NoError(t, err)
|
||||
go s.Start()
|
||||
waitFor(t, func() bool { return passes.Load() == 1 })
|
||||
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "20ms", "host": host})
|
||||
require.NoError(t, s.reload(c, false))
|
||||
waitFor(t, func() bool { return passes.Load() >= 6 })
|
||||
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "1h", "host": host})
|
||||
require.NoError(t, s.reload(c, false))
|
||||
// The reload's Stop waited out the 20ms loop. The new runtime's prime may land either side of this read, it is the
|
||||
// only pass left to come
|
||||
stopped := passes.Load()
|
||||
time.Sleep(300 * time.Millisecond)
|
||||
assert.LessOrEqual(t, passes.Load()-stopped, int64(1), "the 20ms loop kept running after the reload")
|
||||
}
|
||||
|
||||
// A reload's new runtime never captures while the old one is still in a pass, go-metrics keeps its GC capture state
|
||||
// in unguarded globals
|
||||
func TestStatsServer_reloadWaitsForTheLastPass(t *testing.T) {
|
||||
var active, most atomic.Int64
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
host := "127.0.0.1:" + freeTCPPort(t)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "10ms", "host": host})
|
||||
var s *statsServer
|
||||
ctx := stopAtCleanup(t, &s)
|
||||
s, err := newStatsServerFromConfig(ctx, l, c, "", false, func() {
|
||||
n := active.Add(1)
|
||||
for {
|
||||
m := most.Load()
|
||||
if n <= m || most.CompareAndSwap(m, n) {
|
||||
break
|
||||
}
|
||||
}
|
||||
// Long enough that a reload almost always lands mid pass
|
||||
time.Sleep(50 * time.Millisecond)
|
||||
active.Add(-1)
|
||||
})
|
||||
require.NoError(t, err)
|
||||
go s.Start()
|
||||
|
||||
for i := range 6 {
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": fmt.Sprintf("%dms", 11-i%2), "host": host})
|
||||
require.NoError(t, s.reload(c, false))
|
||||
time.Sleep(30 * time.Millisecond)
|
||||
}
|
||||
assert.Equal(t, int64(1), most.Load())
|
||||
}
|
||||
|
||||
// Nothing captures once Stop returns, a pass in progress is waited out
|
||||
func TestStatsServer_Stop_waitsForThePass(t *testing.T) {
|
||||
var inPass atomic.Bool
|
||||
entered := make(chan struct{}, 1)
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "1h", "host": "127.0.0.1:" + freeTCPPort(t)})
|
||||
var s *statsServer
|
||||
ctx := stopAtCleanup(t, &s)
|
||||
s, err := newStatsServerFromConfig(ctx, l, c, "", false, func() {
|
||||
inPass.Store(true)
|
||||
select {
|
||||
case entered <- struct{}{}:
|
||||
default:
|
||||
}
|
||||
time.Sleep(100 * time.Millisecond)
|
||||
inPass.Store(false)
|
||||
})
|
||||
require.NoError(t, err)
|
||||
|
||||
go s.Start()
|
||||
select {
|
||||
case <-entered:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("never captured")
|
||||
}
|
||||
s.Stop()
|
||||
assert.False(t, inPass.Load(), "Stop returned with a pass still running")
|
||||
}
|
||||
|
||||
// A Start that arrives while Stop waits out the last pass doesn't capture alongside it
|
||||
func TestStatsServer_Start_duringStopWaitsForThePass(t *testing.T) {
|
||||
var active, most atomic.Int64
|
||||
entered := make(chan struct{}, 1)
|
||||
l := slog.New(slog.DiscardHandler)
|
||||
c := config.NewC(l)
|
||||
setStatsConfig(c, map[string]any{"type": "graphite", "interval": "1h", "host": "127.0.0.1:" + freeTCPPort(t)})
|
||||
var s *statsServer
|
||||
ctx := stopAtCleanup(t, &s)
|
||||
s, err := newStatsServerFromConfig(ctx, l, c, "", false, func() {
|
||||
n := active.Add(1)
|
||||
for {
|
||||
m := most.Load()
|
||||
if n <= m || most.CompareAndSwap(m, n) {
|
||||
break
|
||||
}
|
||||
}
|
||||
select {
|
||||
case entered <- struct{}{}:
|
||||
default:
|
||||
}
|
||||
time.Sleep(200 * time.Millisecond)
|
||||
active.Add(-1)
|
||||
})
|
||||
require.NoError(t, err)
|
||||
|
||||
go s.Start()
|
||||
select {
|
||||
case <-entered:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("never captured")
|
||||
}
|
||||
stopped := make(chan struct{})
|
||||
go func() {
|
||||
s.Stop()
|
||||
close(stopped)
|
||||
}()
|
||||
// Stop holds runMu for its whole wait, the next Start has to arrive after it took it
|
||||
waitFor(t, func() bool {
|
||||
if s.runMu.TryLock() {
|
||||
s.runMu.Unlock()
|
||||
return false
|
||||
}
|
||||
return true
|
||||
})
|
||||
go s.Start()
|
||||
select {
|
||||
case <-stopped:
|
||||
case <-time.After(5 * time.Second):
|
||||
t.Fatal("Stop did not return")
|
||||
}
|
||||
waitFor(t, func() bool { return len(entered) == 1 })
|
||||
assert.Equal(t, int64(1), most.Load(), "a Start captured alongside the pass Stop was waiting out")
|
||||
}
|
||||
+7
-6
@@ -29,11 +29,12 @@ type Settings struct {
|
||||
type Conn interface {
|
||||
Rebind() error
|
||||
LocalAddr() (netip.AddrPort, error)
|
||||
// ListenOut invokes r for each received packet.
|
||||
// On batch-capable backends (recvmmsg), flush is called after each batch is fully delivered.
|
||||
// Callers use it to flush per-batch accumulators such as TUN write coalescers.
|
||||
// Single-packet backends call flush after each packet. flush must not be nil.
|
||||
ListenOut(r EncReader, flush func()) error
|
||||
// ListenOut invokes r for each received packet. The packets alias the receive buffers.
|
||||
// After each batch is fully delivered (after each packet on single-packet backends), handoff takes the batch's
|
||||
// buffers and returns the ones to receive the next batch into: the same ones once the caller is done with the
|
||||
// packets, after flushing per-batch accumulators such as TUN write coalescers, or replacements of the same count
|
||||
// and size to keep them. handoff must not be nil.
|
||||
ListenOut(r EncReader, handoff func(bufs [][]byte) [][]byte) error
|
||||
WriteTo(b []byte, addr netip.AddrPort) error
|
||||
// WriteBatch sends a contiguous batch of packets, each with its own
|
||||
// destination. bufs and addrs must have the same length. Linux uses
|
||||
@@ -56,7 +57,7 @@ func (NoopConn) Rebind() error {
|
||||
func (NoopConn) LocalAddr() (netip.AddrPort, error) {
|
||||
return netip.AddrPort{}, nil
|
||||
}
|
||||
func (NoopConn) ListenOut(_ EncReader, _ func()) error {
|
||||
func (NoopConn) ListenOut(_ EncReader, _ func([][]byte) [][]byte) error {
|
||||
return nil
|
||||
}
|
||||
func (NoopConn) SupportsMultipleReaders() bool {
|
||||
|
||||
+6
-127
@@ -180,121 +180,12 @@ func NewUDPStatsEmitter(udpConns []Conn) func() {
|
||||
return func() {}
|
||||
}
|
||||
|
||||
// recvBatch is how many datagrams one recvmsg_x call may return.
|
||||
const recvBatch = 64
|
||||
|
||||
// msghdrX mirrors xnu's struct msghdr_x (bsd/sys/socket_private.h). recvmsg_x reports each
|
||||
// datagram's length in Datalen rather than in its return value.
|
||||
type msghdrX struct {
|
||||
Name *byte
|
||||
Namelen uint32
|
||||
Iov *unix.Iovec
|
||||
Iovlen int32
|
||||
Control *byte
|
||||
Controllen uint32
|
||||
Flags int32
|
||||
Datalen uint64
|
||||
}
|
||||
|
||||
// ListenOut drains the socket with recvmsg_x, darwin's private batched recvmsg, so one syscall reads
|
||||
// up to recvBatch datagrams instead of one. Sends stay on sendto: xnu batches sendmsg_x through
|
||||
// the stack only on connected sockets, and on this unconnected one it loops sendit per message.
|
||||
func (u *StdConn) ListenOut(r EncReader, flush func()) error {
|
||||
rc, err := u.UDPConn.SyscallConn()
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
bufs := make([][]byte, recvBatch)
|
||||
names := make([]unix.RawSockaddrInet6, recvBatch)
|
||||
iovs := make([]unix.Iovec, recvBatch)
|
||||
hdrs := make([]msghdrX, recvBatch)
|
||||
for i := range hdrs {
|
||||
bufs[i] = make([]byte, MTU)
|
||||
iovs[i].Base = &bufs[i][0]
|
||||
iovs[i].SetLen(MTU)
|
||||
hdrs[i].Iov = &iovs[i]
|
||||
hdrs[i].Iovlen = 1
|
||||
hdrs[i].Name = (*byte)(unsafe.Pointer(&names[i]))
|
||||
}
|
||||
|
||||
for {
|
||||
var n int
|
||||
var errno syscall.Errno
|
||||
err := rc.Read(func(fd uintptr) bool {
|
||||
for i := range hdrs {
|
||||
hdrs[i].Namelen = unix.SizeofSockaddrInet6
|
||||
hdrs[i].Flags = 0
|
||||
hdrs[i].Datalen = 0
|
||||
}
|
||||
r0, _, e := unix.Syscall6(unix.SYS_RECVMSG_X, fd, uintptr(unsafe.Pointer(&hdrs[0])), recvBatch, 0, 0, 0)
|
||||
if e == unix.EAGAIN {
|
||||
return false
|
||||
}
|
||||
n, errno = int(r0), e
|
||||
return true
|
||||
})
|
||||
if err != nil {
|
||||
if errors.Is(err, net.ErrClosed) {
|
||||
return err
|
||||
}
|
||||
u.l.Error("unexpected udp socket receive error", "error", err)
|
||||
continue
|
||||
}
|
||||
if errno == unix.ENOSYS || errno == unix.EPERM || errno == unix.EOPNOTSUPP {
|
||||
// recvmsg_x is private, so a kernel or sandbox (such as an iOS extension's) may refuse it.
|
||||
u.l.Warn("recvmsg_x unavailable, reading one datagram per syscall", "error", errno)
|
||||
return u.listenOutSingle(r, flush)
|
||||
}
|
||||
if errno != 0 {
|
||||
if errno != unix.EINTR {
|
||||
u.l.Error("unexpected udp socket receive error", "error", errno)
|
||||
}
|
||||
continue
|
||||
}
|
||||
if err := checkMsghdrX(hdrs, names, n); err != nil {
|
||||
// The batch was read through a layout the kernel no longer writes, so none of it is trustworthy.
|
||||
u.l.Warn("recvmsg_x returned an unexpected header, reading one datagram per syscall", "error", err)
|
||||
return u.listenOutSingle(r, flush)
|
||||
}
|
||||
|
||||
for i := 0; i < n; i++ {
|
||||
addr, ok := sockaddrToAddrPort(&names[i])
|
||||
if !ok {
|
||||
continue
|
||||
}
|
||||
l := int(hdrs[i].Datalen)
|
||||
r(addr, bufs[i][:l:l])
|
||||
}
|
||||
flush()
|
||||
}
|
||||
}
|
||||
|
||||
// checkMsghdrX rejects a recvmsg_x result that intact msghdr_x entries can't produce, which is how a change
|
||||
// to xnu's private struct layout would show up.
|
||||
func checkMsghdrX(hdrs []msghdrX, names []unix.RawSockaddrInet6, n int) error {
|
||||
if n > len(hdrs) {
|
||||
return fmt.Errorf("returned %d datagrams for %d headers", n, len(hdrs))
|
||||
}
|
||||
for i := range n {
|
||||
if hdrs[i].Datalen > MTU {
|
||||
return fmt.Errorf("datagram %d is %d bytes, past the %d byte buffer", i, hdrs[i].Datalen, MTU)
|
||||
}
|
||||
switch {
|
||||
case names[i].Family == unix.AF_INET && hdrs[i].Namelen == unix.SizeofSockaddrInet4:
|
||||
case names[i].Family == unix.AF_INET6 && hdrs[i].Namelen == unix.SizeofSockaddrInet6:
|
||||
default:
|
||||
return fmt.Errorf("datagram %d has address family %d with length %d", i, names[i].Family, hdrs[i].Namelen)
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// listenOutSingle reads one datagram per recvfrom and flushes after each, for when recvmsg_x is refused or misbehaves.
|
||||
func (u *StdConn) listenOutSingle(r EncReader, flush func()) error {
|
||||
buffer := make([]byte, MTU)
|
||||
func (u *StdConn) ListenOut(r EncReader, handoff func(bufs [][]byte) [][]byte) error {
|
||||
bufs := [][]byte{make([]byte, MTU)}
|
||||
buffer := bufs[0]
|
||||
|
||||
for {
|
||||
// Just read one packet at a time
|
||||
n, rua, err := u.ReadFromUDPAddrPort(buffer)
|
||||
if err != nil {
|
||||
if errors.Is(err, net.ErrClosed) {
|
||||
@@ -305,23 +196,11 @@ func (u *StdConn) listenOutSingle(r EncReader, flush func()) error {
|
||||
}
|
||||
|
||||
r(netip.AddrPortFrom(rua.Addr().Unmap(), rua.Port()), buffer[:n:n])
|
||||
flush()
|
||||
bufs = handoff(bufs)
|
||||
buffer = bufs[0]
|
||||
}
|
||||
}
|
||||
|
||||
func sockaddrToAddrPort(sa *unix.RawSockaddrInet6) (netip.AddrPort, bool) {
|
||||
switch sa.Family {
|
||||
case unix.AF_INET:
|
||||
sa4 := (*unix.RawSockaddrInet4)(unsafe.Pointer(sa))
|
||||
port := binary.BigEndian.Uint16((*[2]byte)(unsafe.Pointer(&sa4.Port))[:])
|
||||
return netip.AddrPortFrom(netip.AddrFrom4(sa4.Addr), port), true
|
||||
case unix.AF_INET6:
|
||||
port := binary.BigEndian.Uint16((*[2]byte)(unsafe.Pointer(&sa.Port))[:])
|
||||
return netip.AddrPortFrom(netip.AddrFrom16(sa.Addr).Unmap(), port), true
|
||||
}
|
||||
return netip.AddrPort{}, false
|
||||
}
|
||||
|
||||
func (u *StdConn) SupportsMultipleReaders() bool {
|
||||
return false
|
||||
}
|
||||
|
||||
+5
-3
@@ -86,8 +86,9 @@ type rawMessage struct {
|
||||
Len uint32
|
||||
}
|
||||
|
||||
func (u *GenericConn) ListenOut(r EncReader, flush func()) error {
|
||||
buffer := make([]byte, MTU)
|
||||
func (u *GenericConn) ListenOut(r EncReader, handoff func(bufs [][]byte) [][]byte) error {
|
||||
bufs := [][]byte{make([]byte, MTU)}
|
||||
buffer := bufs[0]
|
||||
|
||||
var lastRecvErr time.Time
|
||||
|
||||
@@ -107,7 +108,8 @@ func (u *GenericConn) ListenOut(r EncReader, flush func()) error {
|
||||
}
|
||||
|
||||
r(netip.AddrPortFrom(rua.Addr().Unmap(), rua.Port()), buffer[:n:n])
|
||||
flush()
|
||||
bufs = handoff(bufs)
|
||||
buffer = bufs[0]
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
+24
-4
@@ -245,11 +245,18 @@ func getFrom(names [][]byte, i int, isV4 bool) netip.AddrPort {
|
||||
return netip.AddrPortFrom(ip.Unmap(), binary.BigEndian.Uint16(names[i][2:4]))
|
||||
}
|
||||
|
||||
func (u *StdConn) ListenOut(r EncReader, flush func()) error {
|
||||
bufSize := MTU
|
||||
// HandoffShape is the count and size of the buffers ListenOut hands over after each batch.
|
||||
func (u *StdConn) HandoffShape() (n, size int) {
|
||||
if u.groSupported {
|
||||
return u.batch, udpGROBufferSize
|
||||
}
|
||||
return u.batch, MTU
|
||||
}
|
||||
|
||||
func (u *StdConn) ListenOut(r EncReader, handoff func(bufs [][]byte) [][]byte) error {
|
||||
_, bufSize := u.HandoffShape()
|
||||
cmsgSpace := 0
|
||||
if u.groSupported {
|
||||
bufSize = udpGROBufferSize
|
||||
cmsgSpace = unix.CmsgSpace(udpGROCmsgPayload)
|
||||
}
|
||||
msgs, buffers, names, _ := prepareRawMessages(u.batch, bufSize, cmsgSpace)
|
||||
@@ -283,7 +290,20 @@ func (u *StdConn) ListenOut(r EncReader, flush func()) error {
|
||||
deliverSegments(r, from, payload, segSize)
|
||||
}
|
||||
|
||||
flush()
|
||||
next := handoff(buffers)
|
||||
if len(next) != len(buffers) {
|
||||
panic(fmt.Sprintf("udp: handoff returned %d buffers, need %d", len(next), len(buffers)))
|
||||
}
|
||||
if &next[0] != &buffers[0] {
|
||||
// Replacements, point the recvmmsg iovecs at them
|
||||
for i := range next {
|
||||
if len(next[i]) < bufSize {
|
||||
panic(fmt.Sprintf("udp: handoff buffer is %d bytes, need %d", len(next[i]), bufSize))
|
||||
}
|
||||
msgs[i].Hdr.Iov.Base = &next[i][0]
|
||||
}
|
||||
buffers = next
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -226,7 +226,7 @@ func TestOffloadsDisabledRxDelivers(t *testing.T) {
|
||||
_ = sc.ListenOut(func(from netip.AddrPort, payload []byte) {
|
||||
// payload aliases the shared recv buffer row; copy before handing off.
|
||||
rxCh <- rxPkt{from, slices.Clone(payload)}
|
||||
}, func() {})
|
||||
}, func(bufs [][]byte) [][]byte { return bufs })
|
||||
}()
|
||||
|
||||
tx, err := net.DialUDP("udp4", nil, net.UDPAddrFromAddrPort(addr))
|
||||
|
||||
@@ -142,7 +142,7 @@ func runTeardownCase(t *testing.T, batch int, name string, traffic func(send net
|
||||
go func() {
|
||||
loopDone <- sc.ListenOut(func(netip.AddrPort, []byte) {
|
||||
received.Add(1)
|
||||
}, func() {})
|
||||
}, func(bufs [][]byte) [][]byte { return bufs })
|
||||
}()
|
||||
|
||||
send, err := net.Dial("udp", addr.String())
|
||||
|
||||
@@ -1,85 +0,0 @@
|
||||
//go:build !e2e_testing
|
||||
|
||||
package udp
|
||||
|
||||
import (
|
||||
"log/slog"
|
||||
"net"
|
||||
"net/netip"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"testing"
|
||||
)
|
||||
|
||||
// BenchmarkListenOut measures the receive path on loopback: senders flood datagrams sized like a full tunnel
|
||||
// packet, and each op is one datagram delivered to the reader. There are enough senders that the reader, not the
|
||||
// senders, sets the pace. flushes/pkt is 1 when every receive is its own syscall and falls as receives batch. The
|
||||
// senders share the machine, so ns/op is only comparable between runs on the same host.
|
||||
func BenchmarkListenOut(b *testing.B) {
|
||||
for _, tc := range []struct{ name, listen string }{
|
||||
{"v4", "127.0.0.1"},
|
||||
{"v6", "::1"},
|
||||
} {
|
||||
b.Run(tc.name, func(b *testing.B) {
|
||||
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.AddrPortFrom(netip.MustParseAddr(tc.listen), 0)})
|
||||
if err != nil {
|
||||
b.Fatal(err)
|
||||
}
|
||||
la, err := c.LocalAddr()
|
||||
if err != nil {
|
||||
b.Fatal(err)
|
||||
}
|
||||
const size, senders = 1300, 4
|
||||
txs := make([]net.Conn, senders)
|
||||
for i := range txs {
|
||||
if txs[i], err = net.Dial("udp", la.String()); err != nil {
|
||||
b.Fatal(err)
|
||||
}
|
||||
defer txs[i].Close()
|
||||
}
|
||||
|
||||
target := int64(b.N)
|
||||
var got, flushes atomic.Int64
|
||||
reached := make(chan struct{})
|
||||
done := make(chan struct{})
|
||||
go func() {
|
||||
_ = c.ListenOut(func(netip.AddrPort, []byte) {
|
||||
if got.Add(1) == target {
|
||||
close(reached)
|
||||
}
|
||||
}, func() { flushes.Add(1) })
|
||||
close(done)
|
||||
}()
|
||||
|
||||
// Writes fail with ENOBUFS while the send queue is full; the flood only has to keep the receiver busy.
|
||||
stop := make(chan struct{})
|
||||
var sent sync.WaitGroup
|
||||
for _, tx := range txs {
|
||||
sent.Go(func() {
|
||||
payload := make([]byte, size)
|
||||
for {
|
||||
select {
|
||||
case <-stop:
|
||||
return
|
||||
default:
|
||||
_, _ = tx.Write(payload)
|
||||
}
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
b.SetBytes(size)
|
||||
b.ResetTimer()
|
||||
<-reached
|
||||
b.StopTimer()
|
||||
|
||||
close(stop)
|
||||
sent.Wait()
|
||||
if err := c.Close(); err != nil {
|
||||
b.Fatal(err)
|
||||
}
|
||||
<-done
|
||||
b.ReportMetric(float64(flushes.Load())/float64(got.Load()), "flushes/pkt")
|
||||
})
|
||||
}
|
||||
}
|
||||
@@ -1,302 +0,0 @@
|
||||
//go:build !e2e_testing
|
||||
|
||||
package udp
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"fmt"
|
||||
"log/slog"
|
||||
"net"
|
||||
"net/netip"
|
||||
"sync"
|
||||
"syscall"
|
||||
"testing"
|
||||
"time"
|
||||
"unsafe"
|
||||
|
||||
"github.com/stretchr/testify/assert"
|
||||
"github.com/stretchr/testify/require"
|
||||
"golang.org/x/sys/unix"
|
||||
)
|
||||
|
||||
// A burst larger than recvBatch arrives whole, with each datagram's own bytes and source address, and
|
||||
// flush runs once per recvmsg_x batch rather than once per datagram.
|
||||
func TestListenOutBatchesBurst(t *testing.T) {
|
||||
for _, tc := range []struct{ name, listen string }{
|
||||
{"v4", "127.0.0.1"},
|
||||
{"v6", "::1"},
|
||||
} {
|
||||
t.Run(tc.name, func(t *testing.T) {
|
||||
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.AddrPortFrom(netip.MustParseAddr(tc.listen), 0)})
|
||||
require.NoError(t, err)
|
||||
la, err := c.LocalAddr()
|
||||
require.NoError(t, err)
|
||||
|
||||
tx, err := net.Dial("udp", la.String())
|
||||
require.NoError(t, err)
|
||||
defer tx.Close()
|
||||
txAddr := netip.MustParseAddrPort(tx.LocalAddr().String())
|
||||
|
||||
const total = recvBatch*3 + 7
|
||||
var mu sync.Mutex
|
||||
var got []string
|
||||
var froms []netip.AddrPort
|
||||
flushes := 0
|
||||
// Queued before the reader starts, so recvmsg_x sees full batches.
|
||||
for i := 0; i < total; i++ {
|
||||
_, err := tx.Write([]byte(fmt.Sprintf("pkt-%03d", i)))
|
||||
require.NoError(t, err)
|
||||
}
|
||||
|
||||
done := make(chan struct{})
|
||||
go func() {
|
||||
_ = c.ListenOut(func(addr netip.AddrPort, payload []byte) {
|
||||
mu.Lock()
|
||||
got = append(got, string(payload))
|
||||
froms = append(froms, addr)
|
||||
mu.Unlock()
|
||||
}, func() {
|
||||
mu.Lock()
|
||||
flushes++
|
||||
mu.Unlock()
|
||||
})
|
||||
close(done)
|
||||
}()
|
||||
|
||||
require.Eventually(t, func() bool {
|
||||
mu.Lock()
|
||||
defer mu.Unlock()
|
||||
return len(got) == total
|
||||
}, 5*time.Second, 10*time.Millisecond)
|
||||
|
||||
require.NoError(t, c.Close())
|
||||
<-done
|
||||
|
||||
for i, p := range got {
|
||||
assert.Equal(t, fmt.Sprintf("pkt-%03d", i), p)
|
||||
assert.Equal(t, txAddr, froms[i])
|
||||
}
|
||||
assert.Equal(t, (total+recvBatch-1)/recvBatch, flushes)
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
// msghdrX round-trips through xnu's recvmsg_x on a loopback socket, v4 and v6. The kernel reads Name, Namelen,
|
||||
// Iov and Iovlen from each entry and writes Namelen, Flags and Datalen back, so a field at the wrong offset or a
|
||||
// wrong stride between entries shows up as a wrong length, sender, or payload.
|
||||
// TestListenOutSingleFallback pins the path ListenOut takes when the kernel refuses recvmsg_x.
|
||||
func TestListenOutSingleFallback(t *testing.T) {
|
||||
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0")})
|
||||
require.NoError(t, err)
|
||||
la, err := c.LocalAddr()
|
||||
require.NoError(t, err)
|
||||
tx, err := net.Dial("udp", la.String())
|
||||
require.NoError(t, err)
|
||||
defer tx.Close()
|
||||
txAddr := netip.MustParseAddrPort(tx.LocalAddr().String())
|
||||
|
||||
const total = 5
|
||||
for i := 0; i < total; i++ {
|
||||
_, err := tx.Write([]byte(fmt.Sprintf("pkt-%03d", i)))
|
||||
require.NoError(t, err)
|
||||
}
|
||||
|
||||
var mu sync.Mutex
|
||||
var got []string
|
||||
flushes := 0
|
||||
done := make(chan struct{})
|
||||
go func() {
|
||||
_ = c.(*StdConn).listenOutSingle(func(addr netip.AddrPort, payload []byte) {
|
||||
assert.Equal(t, txAddr, addr)
|
||||
mu.Lock()
|
||||
got = append(got, string(payload))
|
||||
mu.Unlock()
|
||||
}, func() {
|
||||
mu.Lock()
|
||||
flushes++
|
||||
mu.Unlock()
|
||||
})
|
||||
close(done)
|
||||
}()
|
||||
|
||||
require.Eventually(t, func() bool {
|
||||
mu.Lock()
|
||||
defer mu.Unlock()
|
||||
return len(got) == total
|
||||
}, 5*time.Second, 10*time.Millisecond)
|
||||
require.NoError(t, c.Close())
|
||||
<-done
|
||||
|
||||
for i, p := range got {
|
||||
assert.Equal(t, fmt.Sprintf("pkt-%03d", i), p)
|
||||
}
|
||||
assert.Equal(t, total, flushes)
|
||||
}
|
||||
|
||||
func TestCheckMsghdrX(t *testing.T) {
|
||||
valid := func() ([]msghdrX, []unix.RawSockaddrInet6) {
|
||||
hdrs := make([]msghdrX, 2)
|
||||
names := make([]unix.RawSockaddrInet6, 2)
|
||||
hdrs[0].Namelen, hdrs[0].Datalen = unix.SizeofSockaddrInet4, 1
|
||||
names[0].Family = unix.AF_INET
|
||||
hdrs[1].Namelen, hdrs[1].Datalen = unix.SizeofSockaddrInet6, MTU
|
||||
names[1].Family = unix.AF_INET6
|
||||
return hdrs, names
|
||||
}
|
||||
|
||||
hdrs, names := valid()
|
||||
require.NoError(t, checkMsghdrX(hdrs, names, 2))
|
||||
require.NoError(t, checkMsghdrX(hdrs, names, 0))
|
||||
|
||||
for _, tc := range []struct {
|
||||
name string
|
||||
n int
|
||||
mutate func([]msghdrX, []unix.RawSockaddrInet6)
|
||||
}{
|
||||
{"count past headers", 3, func([]msghdrX, []unix.RawSockaddrInet6) {}},
|
||||
{"datalen past buffer", 2, func(h []msghdrX, _ []unix.RawSockaddrInet6) { h[1].Datalen = MTU + 1 }},
|
||||
{"unknown family", 2, func(_ []msghdrX, a []unix.RawSockaddrInet6) { a[0].Family = unix.AF_UNIX }},
|
||||
{"family and length disagree", 2, func(h []msghdrX, _ []unix.RawSockaddrInet6) { h[0].Namelen = unix.SizeofSockaddrInet6 }},
|
||||
} {
|
||||
t.Run(tc.name, func(t *testing.T) {
|
||||
hdrs, names := valid()
|
||||
tc.mutate(hdrs, names)
|
||||
assert.Error(t, checkMsghdrX(hdrs, names, tc.n))
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
func TestMsghdrXKernelRoundTrip(t *testing.T) {
|
||||
for _, v6 := range []bool{false, true} {
|
||||
t.Run(fmt.Sprintf("v6=%v", v6), func(t *testing.T) {
|
||||
// Loopback delivery goes through the dlil input thread, so the settle gives all three time to
|
||||
// queue and come back from one call, which is what exercises the stride.
|
||||
got := recvmsgXRoundTrip(t, v6, []int{1, 300, 1400}, 4, 50*time.Millisecond)
|
||||
assert.Equal(t, 3, got, "largest batch from one recvmsg_x call")
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
// Any burst, of any datagram lengths up to the kernel's maxdgram, into any number of headers, comes back in order
|
||||
// with every entry's fields intact, and headers past the returned count untouched.
|
||||
func FuzzMsghdrXKernelRoundTrip(f *testing.F) {
|
||||
f.Add(false, uint8(3), []byte{0, 1, 1, 44, 5, 120})
|
||||
f.Add(true, uint8(0), []byte{0, 0, 0x23, 0x29}) // one empty datagram, then one at MTU
|
||||
f.Add(false, uint8(1), []byte{0x23, 0x2a, 0x24, 0x00, 0}) // past MTU: cut to MTU, no MSG_TRUNC
|
||||
f.Add(true, uint8(recvBatch-1), bytes.Repeat([]byte{0x05, 0xdc}, recvBatch))
|
||||
|
||||
f.Fuzz(func(t *testing.T, v6 bool, spare uint8, sizes []byte) {
|
||||
// Two bytes per datagram, big-endian, folded into [0, maxdgram].
|
||||
var lens []int
|
||||
for i := 0; i+1 < len(sizes) && len(lens) < recvBatch; i += 2 {
|
||||
lens = append(lens, (int(sizes[i])<<8|int(sizes[i+1]))%(maxDgram+1))
|
||||
}
|
||||
if len(lens) == 0 {
|
||||
return
|
||||
}
|
||||
recvmsgXRoundTrip(t, v6, lens, 1+int(spare)%recvBatch, time.Millisecond)
|
||||
})
|
||||
}
|
||||
|
||||
// maxDgram is darwin's default net.inet.udp.maxdgram, the largest datagram a loopback sender may write.
|
||||
const maxDgram = 9216
|
||||
|
||||
// recvmsgXRoundTrip sends one datagram per entry of lens to a fresh loopback socket, drains them with recvmsg_x
|
||||
// into nh headers, and checks every field the kernel writes. Neighboring datagrams differ in sender, length and
|
||||
// fill byte, so no field can pass by reading a neighbor's value. It returns the largest batch one call returned.
|
||||
func recvmsgXRoundTrip(t *testing.T, v6 bool, lens []int, nh int, settle time.Duration) int {
|
||||
t.Helper()
|
||||
network, lo := "udp4", netip.MustParseAddr("127.0.0.1")
|
||||
if v6 {
|
||||
network, lo = "udp6", netip.IPv6Loopback()
|
||||
}
|
||||
rx, err := net.ListenUDP(network, net.UDPAddrFromAddrPort(netip.AddrPortFrom(lo, 0)))
|
||||
require.NoError(t, err)
|
||||
defer rx.Close()
|
||||
|
||||
txs := make([]*net.UDPConn, 3)
|
||||
for i := range txs {
|
||||
txs[i], err = net.DialUDP(network, nil, rx.LocalAddr().(*net.UDPAddr))
|
||||
require.NoError(t, err)
|
||||
defer txs[i].Close()
|
||||
}
|
||||
senders := make([]netip.AddrPort, len(lens))
|
||||
for i, l := range lens {
|
||||
tx := txs[i%len(txs)]
|
||||
senders[i] = tx.LocalAddr().(*net.UDPAddr).AddrPort()
|
||||
_, err := tx.Write(bytes.Repeat([]byte{byte(i + 1)}, l))
|
||||
require.NoError(t, err)
|
||||
}
|
||||
time.Sleep(settle)
|
||||
|
||||
// Namelen goes in as the v6 size on both families, so the v4 size coming back proves the kernel wrote it.
|
||||
const sentinel = 0xa5a5a5a5a5a5a5a5
|
||||
wantNamelen := uint32(unix.SizeofSockaddrInet4)
|
||||
if v6 {
|
||||
wantNamelen = unix.SizeofSockaddrInet6
|
||||
}
|
||||
bufs := make([][]byte, nh)
|
||||
names := make([]unix.RawSockaddrInet6, nh)
|
||||
iovs := make([]unix.Iovec, nh)
|
||||
hdrs := make([]msghdrX, nh)
|
||||
for i := range hdrs {
|
||||
bufs[i] = make([]byte, MTU)
|
||||
iovs[i].Base = &bufs[i][0]
|
||||
iovs[i].SetLen(MTU)
|
||||
}
|
||||
|
||||
// A datagram dropped on loopback would otherwise block rc.Read forever.
|
||||
require.NoError(t, rx.SetReadDeadline(time.Now().Add(5*time.Second)))
|
||||
rc, err := rx.SyscallConn()
|
||||
require.NoError(t, err)
|
||||
largest := 0
|
||||
for next := 0; next < len(lens); {
|
||||
for i := range hdrs {
|
||||
clear(bufs[i])
|
||||
hdrs[i] = msghdrX{
|
||||
Name: (*byte)(unsafe.Pointer(&names[i])),
|
||||
Namelen: unix.SizeofSockaddrInet6,
|
||||
Iov: &iovs[i],
|
||||
Iovlen: 1,
|
||||
Flags: -1,
|
||||
Datalen: sentinel,
|
||||
}
|
||||
}
|
||||
var n int
|
||||
var errno syscall.Errno
|
||||
require.NoError(t, rc.Read(func(fd uintptr) bool {
|
||||
r0, _, e := unix.Syscall6(unix.SYS_RECVMSG_X, fd, uintptr(unsafe.Pointer(&hdrs[0])), uintptr(nh), 0, 0, 0)
|
||||
n, errno = int(r0), e
|
||||
return e != unix.EAGAIN
|
||||
}), "received %d of %d datagrams", next, len(lens))
|
||||
require.Zero(t, errno)
|
||||
require.Positive(t, n)
|
||||
require.LessOrEqual(t, n, min(nh, len(lens)-next))
|
||||
largest = max(largest, n)
|
||||
require.NoError(t, checkMsghdrX(hdrs[:nh], names[:nh], n), "ListenOut would fall back on this batch")
|
||||
|
||||
for i := range n {
|
||||
d := next + i
|
||||
h := &hdrs[i]
|
||||
want := min(lens[d], MTU)
|
||||
require.EqualValues(t, want, h.Datalen, "datagram %d Datalen", d)
|
||||
require.Equal(t, wantNamelen, h.Namelen, "datagram %d Namelen", d)
|
||||
from, ok := sockaddrToAddrPort(&names[i])
|
||||
require.True(t, ok, "datagram %d family %d", d, names[i].Family)
|
||||
require.Equal(t, senders[d], from, "datagram %d sender", d)
|
||||
// Flags goes in as -1 and must come back 0. xnu's list receive (soreceive_m_list) never sets
|
||||
// MSG_TRUNC, so a datagram past MTU arrives silently cut to MTU, as recvfrom through net.UDPConn does.
|
||||
require.Zero(t, h.Flags, "datagram %d Flags", d)
|
||||
require.Equal(t, bytes.Repeat([]byte{byte(d + 1)}, want), bufs[i][:want], "datagram %d payload", d)
|
||||
if want < MTU {
|
||||
require.Zero(t, bufs[i][want], "datagram %d wrote past Datalen", d)
|
||||
}
|
||||
}
|
||||
for i := n; i < nh; i++ {
|
||||
require.EqualValues(t, uint64(sentinel), hdrs[i].Datalen, "spare header %d Datalen", i)
|
||||
require.EqualValues(t, unix.SizeofSockaddrInet6, hdrs[i].Namelen, "spare header %d Namelen", i)
|
||||
}
|
||||
next += n
|
||||
}
|
||||
return largest
|
||||
}
|
||||
@@ -30,7 +30,7 @@ func procyield(cycles uint32)
|
||||
|
||||
const (
|
||||
packetsPerRing = 1024
|
||||
// Caps tun.mtu at MTU-32 direct, MTU-64 relayed, unenforced anywhere else. 17.6MB page locked per socket.
|
||||
// Fits a relayed packet at overlay.MaxMTU, which tun.mtu and route mtus are capped to. 17.6MB page locked per socket.
|
||||
bytesPerPacket = MTU
|
||||
receiveSpins = 15
|
||||
)
|
||||
@@ -142,8 +142,9 @@ func (u *RIOConn) bind(l *slog.Logger, sa windows.Sockaddr) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
func (u *RIOConn) ListenOut(r EncReader, flush func()) error {
|
||||
buffer := make([]byte, MTU)
|
||||
func (u *RIOConn) ListenOut(r EncReader, handoff func(bufs [][]byte) [][]byte) error {
|
||||
bufs := [][]byte{make([]byte, MTU)}
|
||||
buffer := bufs[0]
|
||||
|
||||
var lastRecvErr time.Time
|
||||
|
||||
@@ -164,7 +165,8 @@ func (u *RIOConn) ListenOut(r EncReader, flush func()) error {
|
||||
}
|
||||
|
||||
r(netip.AddrPortFrom(netip.AddrFrom16(rua.Addr).Unmap(), (rua.Port>>8)|((rua.Port&0xff)<<8)), buffer[:n:n])
|
||||
flush()
|
||||
bufs = handoff(bufs)
|
||||
buffer = bufs[0]
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
+12
-4
@@ -182,16 +182,24 @@ func (u *TesterConn) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, err
|
||||
return written, nil
|
||||
}
|
||||
|
||||
func (u *TesterConn) ListenOut(r EncReader, flush func()) error {
|
||||
// HandoffShape is one buffer per handoff; the pool resizes the replacements it gets back as it needs.
|
||||
func (u *TesterConn) HandoffShape() (n, size int) { return 1, MTU }
|
||||
|
||||
// ListenOut hands each packet's buffer over as a batch of one, the buffer it gets back goes to the pool. Like StdConn
|
||||
// it carries on with the slice handoff returns, since a reader that keeps the packets keeps the one it was given.
|
||||
// The batcher borrows plaintext decrypted in place inside p.Data until it flushes, inside handoff, so the packet
|
||||
// must stay alive until handoff returns.
|
||||
func (u *TesterConn) ListenOut(r EncReader, handoff func(bufs [][]byte) [][]byte) error {
|
||||
bufs := make([][]byte, 1)
|
||||
for {
|
||||
select {
|
||||
case <-u.done:
|
||||
return os.ErrClosed
|
||||
case p := <-u.RxPackets:
|
||||
r(p.From, p.Data[:len(p.Data):len(p.Data)])
|
||||
// The batcher borrows plaintext decrypted in place inside p.Data
|
||||
// until Flush, so the packet must stay alive across flush()
|
||||
flush()
|
||||
bufs[0] = p.Data
|
||||
bufs = handoff(bufs)
|
||||
p.Data = bufs[0]
|
||||
p.Release()
|
||||
}
|
||||
}
|
||||
|
||||
Reference in new issue
Block a user