Compare commits

..
Author SHA1 Message Date
Ryan Huber 729301b409 udp: fall back to SO_RCVBUF/SO_SNDBUF when FORCE is not permitted
SO_RCVBUFFORCE/SO_SNDBUFFORCE need CAP_NET_ADMIN. Without it (non-root,
or root under Docker's default caps) listen.read_buffer/write_buffer
failed with EPERM and the socket kept the system default. On EPERM retry
with the plain option, which the kernel clamps to net.core.{r,w}mem_max,
and warn when the effective size is smaller than requested.
2026-10-10 12:52:01 -04:00
Nate Brown a8c52f5c13 overlay: Handle routes Windows already has, and parse quoted route numbers (#1924) 2026-10-08 18:52:53 -05:00
Jack Doan 13af8dd6f3 Cap configured MTUs at 8937 so relayed sends never outgrow their buffers (#1899) 2026-10-08 15:41:08 -05:00
Nate Brown 5bb0622192 stats: Follow stats.interval on reload, interface gauges ride the capture loop (#1919) 2026-10-08 10:00:48 -05:00
Nate Brown f3705cbb67 graphite: Time out a send, and abandon it on stop (#1920) 2026-10-07 18:01:09 -05:00
Nate Brown 94be454f4f lighthouse: Recompute relays for me when relay.am_relay changes (#1917) 2026-10-07 14:06:27 -05:00
Nate Brown 5f26c1a7e5 Add a simple rate limiter with a helper to use it in logging (#1915) 2026-10-07 10:53:37 -05:00
Nate Brown d47c8c71f1 lighthouse: Send an update when the relays we advertise change on reload (#1916) 2026-10-07 10:52:42 -05:00
Nate Brown f57539a163 overlay/linux: Keep unsafe routes when a reload only changes tun.mtu (#1918) 2026-10-07 10:43:11 -05:00
John Maguire 8d8ae0f744 Disable DNS registration on the Windows tunnel adapter (#1914)
On a domain-joined Windows host the DNS Client registers every adapter's
addresses under the host name in AD DNS when "Register this connection's
addresses in DNS" is on, which is the default for a new adapter. When the
Nebula adapter registers, LAN clients that are not on the overlay resolve
the host name to its overlay IP and RDP, SMB and PowerShell Remoting to it
fail intermittently, with nothing pointing at the tunnel as the cause.
Windows registers on address assignment, periodically, and on every
ipconfig /registerdns.

Right after the wintun adapter is created and before Activate assigns any
address, write RegistrationEnabled=0, DisableDynamicUpdate=1 and
MaxNumberOfAddressesToRegister=0 under both the Tcpip and Tcpip6
Parameters\Interfaces\{GUID} keys. The adapter checkbox alone has proven
insufficient in practice, as has disabling dynamic updates alone, so all
three are set, matching what Tailscale and NetBird do. A failure to write
the keys is logged as a warning and does not fail adapter creation.
2026-10-07 11:42:08 -04:00
Matt Richardson f9b0853476 Attach a pending handshake's RemoteList when the handshake starts (#1909)
StartHandshake returned a hostinfo with no RemoteList; handleOutbound attached
one lazily on its first tick. Anything that touched the pending host before
then, notably create-tunnel -address, dereferenced nil and took the daemon
down. StartHandshake now attaches the lighthouse's list as it builds the
hostinfo. To keep the surrounding locking sound, a lighthouse now resolves a
punch target before taking the querier's RemoteList lock, and closeTunnel
deletes the host and evicts its RemoteList under a single hostmap write lock so
a completing handshake cannot land in between. The lock order, now documented
on the handshake manager mutex, is hostmap, then handshake manager, then
lighthouse, then RemoteList; no RemoteList lock is ever held across a hostmap
lookup.
2026-10-07 11:22:50 -04:00
Nate Brown d5062dacf0 lighthouse: Never block in QueryServer, drop queries when the channel is full (#1910) 2026-09-30 15:26:34 -05:00
57 changed files with 2500 additions and 1766 deletions

No files matched your search

+7
View File
@@ -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.
+2 -4
View File
@@ -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)
+3
View File
@@ -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{}{}
+119
View File
@@ -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"
@@ -147,6 +150,122 @@ 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)
+3 -2
View File
@@ -265,6 +265,7 @@ tun:
# 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
# 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.
@@ -410,8 +411,8 @@ 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
+18 -5
View File
@@ -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()
}
+315
View File
@@ -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")
}
+8 -11
View File
@@ -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{}
})
// 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
}
+68 -2
View File
@@ -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)
}
+5
View File
@@ -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
+28 -11
View File
@@ -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 {
@@ -582,19 +595,23 @@ 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",
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",
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
for _, relayIP := range hostinfo.relayState.CopyRelayIps() {
+24 -18
View File
@@ -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"
)
@@ -124,6 +125,10 @@ 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 +203,19 @@ 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 +248,9 @@ func NewInterface(ctx context.Context, c *InterfaceConfig) (*Interface, error) {
conntrackCacheTimeout: c.ConntrackCacheTimeout,
cpuAffinity: c.CpuAffinity,
pinThreads: c.PinThreads,
udpTxErrLimiter: udpTxErrLimiter,
udpRelayTxErrLimiter: udpRelayTxErrLimiter,
txOversizeLimiter: txOversizeLimiter,
metricHandshakes: metrics.GetOrRegisterHistogram("handshakes", nil, metrics.NewExpDecaySample(1028, 0.015)),
metricTxDropped: metrics.GetOrRegisterCounter("udp.tx.dropped", nil),
@@ -624,17 +645,15 @@ 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 +670,6 @@ 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
View File
@@ -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())
-5
View File
@@ -1,5 +0,0 @@
// Package msgx calls xnu's private batched socket calls, sendmsg_x and recvmsg_x, through libSystem.
//
// The calls are not public API, so the package never links them: it looks them up with dlsym on first use.
// The package is darwin-only, and on iOS Available is always false, so App Store builds keep to public API.
package msgx
-16
View File
@@ -1,16 +0,0 @@
package msgx
import "golang.org/x/sys/unix"
// Hdr mirrors xnu's struct msghdr_x (bsd/sys/socket_private.h), which the SDK doesn't ship.
// recvmsg_x reports each message's length in Datalen rather than in its return value.
type Hdr struct {
Name *byte
Namelen uint32
Iov *unix.Iovec
Iovlen int32
Control *byte
Controllen uint32
Flags int32
Datalen uint64
}
-77
View File
@@ -1,77 +0,0 @@
//go:build !ios
package msgx
import (
"sync"
"syscall"
"unsafe"
)
// dlsym is public libSystem API, imported the way x/sys/unix imports its libSystem stubs. The local name is
// distinct from runtime/race's dlsym import.
//
//go:cgo_import_dynamic msgx_dlsym dlsym "/usr/lib/libSystem.B.dylib"
// dlsymTrampolineAddr is the address of msgx_darwin.s's jump to dlsym.
var dlsymTrampolineAddr uintptr
// syscall6X calls the C function at fn on the system stack and returns errno when it returns a 64-bit -1.
// It is how x/sys/unix calls libSystem, and the runtime keeps it linkable for that.
//
//go:linkname syscall6X syscall.syscall6X
func syscall6X(fn, a1, a2, a3, a4, a5, a6 uintptr) (r1, r2 uintptr, err syscall.Errno)
// rtldDefault is dlfcn.h's RTLD_DEFAULT, ((void *)-2): search every image loaded in the process.
const rtldDefault = ^uintptr(1)
var (
sendmsgXName = []byte("sendmsg_x\x00")
recvmsgXName = []byte("recvmsg_x\x00")
)
type funcs struct {
sendmsgX uintptr
recvmsgX uintptr
}
var resolve = sync.OnceValue(func() funcs {
return funcs{sendmsgX: lookup(sendmsgXName), recvmsgX: lookup(recvmsgXName)}
})
// lookup returns the address of the NUL-terminated C symbol name, or 0 when no loaded image exports it.
func lookup(name []byte) uintptr {
p, _, _ := syscall6X(dlsymTrampolineAddr, rtldDefault, uintptr(unsafe.Pointer(&name[0])), 0, 0, 0, 0)
return p
}
// Available reports whether libSystem exports both sendmsg_x and recvmsg_x. When it is false, Send and Recv
// return ENOSYS.
func Available() bool {
f := resolve()
return f.sendmsgX != 0 && f.recvmsgX != 0
}
// Send passes hdrs to sendmsg_x on fd and returns how many messages the kernel took.
func Send(fd uintptr, hdrs []Hdr, flags int) (int, syscall.Errno) {
return call(resolve().sendmsgX, fd, hdrs, flags)
}
// Recv passes hdrs to recvmsg_x on fd and returns how many messages it filled.
func Recv(fd uintptr, hdrs []Hdr, flags int) (int, syscall.Errno) {
return call(resolve().recvmsgX, fd, hdrs, flags)
}
func call(fn, fd uintptr, hdrs []Hdr, flags int) (int, syscall.Errno) {
if fn == 0 {
return 0, syscall.ENOSYS
}
if len(hdrs) == 0 {
return 0, syscall.EINVAL
}
r, _, errno := syscall6X(fn, fd, uintptr(unsafe.Pointer(&hdrs[0])), uintptr(len(hdrs)), uintptr(flags), 0, 0)
if errno != 0 {
return 0, errno
}
return int(r), 0
}
-9
View File
@@ -1,9 +0,0 @@
//go:build !ios
#include "textflag.h"
// The same trampoline shape x/sys/unix uses for its libSystem stubs; it assembles for amd64 and arm64 alike.
TEXT msgx_dlsym_trampoline<>(SB),NOSPLIT,$0-0
JMP msgx_dlsym(SB)
GLOBL ·dlsymTrampolineAddr(SB), RODATA, $8
DATA ·dlsymTrampolineAddr(SB)/8, $msgx_dlsym_trampoline<>(SB)
-69
View File
@@ -1,69 +0,0 @@
//go:build !ios
package msgx
import (
"bytes"
"syscall"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"golang.org/x/sys/unix"
)
// TestAvailable pins that dlsym finds both calls in this macOS's libSystem, so the batched paths are live.
func TestAvailable(t *testing.T) {
assert.True(t, Available())
}
// TestSendRecvRoundTrip pins the libSystem calling path end to end: a batch goes out through sendmsg_x in one
// call and comes back through recvmsg_x in one call, each message whole, in order, with Datalen set.
func TestSendRecvRoundTrip(t *testing.T) {
fds, err := unix.Socketpair(unix.AF_UNIX, unix.SOCK_DGRAM, 0)
require.NoError(t, err)
t.Cleanup(func() {
_ = unix.Close(fds[0])
_ = unix.Close(fds[1])
})
// A dropped datagram fails the receive instead of hanging it.
require.NoError(t, unix.SetsockoptTimeval(fds[1], unix.SOL_SOCKET, unix.SO_RCVTIMEO, &unix.Timeval{Sec: 2}))
const n = 5
var want [n][]byte
var iovs [n]unix.Iovec
hdrs := make([]Hdr, n)
for i := range n {
want[i] = bytes.Repeat([]byte{byte(i + 1)}, 10+i)
iovs[i] = unix.Iovec{Base: &want[i][0]}
iovs[i].SetLen(len(want[i]))
hdrs[i] = Hdr{Iov: &iovs[i], Iovlen: 1}
}
sent, errno := Send(uintptr(fds[0]), hdrs, 0)
require.Zero(t, errno)
require.Equal(t, n, sent)
var bufs [n + 1][64]byte
var riovs [n + 1]unix.Iovec
rhdrs := make([]Hdr, n+1)
for i := range rhdrs {
riovs[i] = unix.Iovec{Base: &bufs[i][0]}
riovs[i].SetLen(len(bufs[i]))
rhdrs[i] = Hdr{Iov: &riovs[i], Iovlen: 1}
}
got, errno := Recv(uintptr(fds[1]), rhdrs, 0)
require.Zero(t, errno)
require.Equal(t, n, got)
for i := range n {
l := int(rhdrs[i].Datalen)
assert.Equal(t, want[i], bufs[i][:l], "message %d", i)
}
}
// TestEmptyBatch pins that an empty batch is refused before it reaches the kernel.
func TestEmptyBatch(t *testing.T) {
_, errno := Send(0, nil, 0)
assert.Equal(t, syscall.EINVAL, errno)
_, errno = Recv(0, nil, 0)
assert.Equal(t, syscall.EINVAL, errno)
}
-12
View File
@@ -1,12 +0,0 @@
package msgx
import "syscall"
// Available is always false on iOS, so App Store builds never look up private symbols.
func Available() bool { return false }
// Send always returns ENOSYS on iOS.
func Send(uintptr, []Hdr, int) (int, syscall.Errno) { return 0, syscall.ENOSYS }
// Recv always returns ENOSYS on iOS.
func Recv(uintptr, []Hdr, int) (int, syscall.Errno) { return 0, syscall.ENOSYS }
+28 -5
View File
@@ -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
+202
View File
@@ -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)
}
+31
View File
@@ -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
}
+106
View File
@@ -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
}
+5 -3
View File
@@ -300,6 +300,7 @@ func Main(c *config.C, configTest bool, buildVersion string, l *slog.Logger, dev
}
var ifce *Interface
var emitStats func()
if !configTest {
ifce, err = NewInterface(ctx, ifConfig)
if err != nil {
@@ -318,9 +319,12 @@ 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 +333,6 @@ 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)
+9 -6
View File
@@ -160,9 +160,7 @@ 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)
@@ -263,12 +261,17 @@ 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 {
// 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
func (f *Interface) sendCloseTunnel(h *HostInfo) {
-62
View File
@@ -435,65 +435,3 @@ func testPP(pkt []byte) *firewall.ParsedPacket {
}
return pp
}
// fakeTunBatchWriter records each WriteBatch call and each plain Write separately.
type fakeTunBatchWriter struct {
writes [][]byte
batches [][][]byte
}
func (w *fakeTunBatchWriter) Write(p []byte) (int, error) {
w.writes = append(w.writes, bytes.Clone(p))
return len(p), nil
}
func (w *fakeTunBatchWriter) WriteBatch(pkts [][]byte) error {
var b [][]byte
for _, p := range pkts {
b = append(b, bytes.Clone(p))
}
w.batches = append(w.batches, b)
return nil
}
// TestPassthroughUsesWriteBatch pins that a writer implementing tio.BatchWriter gets the whole
// passthrough lane in one WriteBatch per Flush, in order, and no plain Writes; an empty Flush
// makes no call.
func TestPassthroughUsesWriteBatch(t *testing.T) {
w := &fakeTunBatchWriter{}
m := newTestMultiCoalescer(t, w)
k := &keySeq{epoch: 1}
var want [][]byte
for i := range 5 {
icmp := make([]byte, 28)
icmp[0] = 0x45
icmp[3] = 28
icmp[9] = 1
icmp[20] = byte(i)
want = append(want, icmp)
if err := m.Commit(icmp, k.next(), testPP(icmp)); err != nil {
t.Fatal(err)
}
}
if err := m.Flush(); err != nil {
t.Fatal(err)
}
if err := m.Flush(); err != nil {
t.Fatal(err)
}
if len(w.writes) != 0 {
t.Fatalf("got %d plain Writes, want 0", len(w.writes))
}
if len(w.batches) != 1 {
t.Fatalf("got %d WriteBatch calls, want 1", len(w.batches))
}
for i, p := range w.batches[0] {
if !bytes.Equal(p, want[i]) {
t.Fatalf("packet %d: got % x, want % x", i, p, want[i])
}
}
if len(w.batches[0]) != len(want) {
t.Fatalf("got %d packets, want %d", len(w.batches[0]), len(want))
}
}
+2 -13
View File
@@ -2,23 +2,18 @@ package batch
import (
"io"
"github.com/slackhq/nebula/overlay/tio"
)
// Passthrough is MultiCoalescer's verbatim lane: no coalescing, packets are written at Flush in the
// order enqueued, in one WriteBatch when out implements tio.BatchWriter.
// Passthrough is MultiCoalescer's verbatim lane: no batching, packets are written at Flush in the
// order enqueued.
type Passthrough struct {
out io.Writer
bw tio.BatchWriter // nil: Flush writes one packet per Write
slots [][]byte
}
func NewPassthrough(w io.Writer) *Passthrough {
bw, _ := w.(tio.BatchWriter)
return &Passthrough{
out: w,
bw: bw,
slots: make([][]byte, 0, 128),
}
}
@@ -31,18 +26,12 @@ func (p *Passthrough) enqueue(pkt []byte) error {
func (p *Passthrough) Flush() error {
var firstErr error
if p.bw != nil {
if len(p.slots) > 0 {
firstErr = p.bw.WriteBatch(p.slots)
}
} else {
for _, s := range p.slots {
_, err := p.out.Write(s)
if err != nil && firstErr == nil {
firstErr = err
}
}
}
clear(p.slots)
p.slots = p.slots[:0]
return firstErr
+57
View File
@@ -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
View File
@@ -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 {
+116
View File
@@ -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")
}
-11
View File
@@ -145,14 +145,3 @@ func SupportsGSO(w io.Writer, want GSOProto) (GSOWriter, bool) {
return gw, false
}
}
// BatchWriter is implemented by Queues that can emit several already-segmented packets in one call.
// Backends without a batched write do not implement it and callers loop over Write.
//
// pkts are borrowed: WriteBatch must not retain them past its return.
// A returned error doesn't say which packets were delivered, so callers must not retry the batch.
// Safe for concurrent use, like Write.
type BatchWriter interface {
io.Writer
WriteBatch(pkts [][]byte) error
}
-328
View File
@@ -1,328 +0,0 @@
//go:build darwin && !ios
package tio
import (
"fmt"
"log/slog"
"os"
"sync"
"sync/atomic"
"syscall"
_ "unsafe" // for go:linkname
"github.com/slackhq/nebula/internal/msgx"
"golang.org/x/sys/unix"
)
// tunMaxPacket is the largest packet utun can return at any device MTU.
const tunMaxPacket = 65535
// tunReadBatch is the most packets one Utun.Read returns.
const tunReadBatch = 64
// tunReadArena is Utun's receive buffer. Draining stops once less than tunMaxPacket of it is left,
// so every readv has room for the largest packet utun can return.
const tunReadArena = 4 * tunMaxPacket
// tunWriteBatch is how many packets one sendmsg_x call may carry.
const tunWriteBatch = 64
// Utun is a darwin utun control socket as a Queue. Every packet crosses the socket behind a 4 byte
// address family prefix, which Read strips and the writes add.
//
// The fd lives in the runtime poller, so Close wakes a Read parked on it; darwin needs no QueueSet
// shutdown fd.
type Utun struct {
f *os.File
l *slog.Logger
// readBuf and readRet are Read's scratch, owned by the single reader.
readBuf []byte
readRet [tunReadBatch]Packet
// noSendmsgX is set when libSystem lacks sendmsg_x or the kernel refuses it; WriteBatch then writes one packet at a time.
noSendmsgX atomic.Bool
batchMu sync.Mutex
batch tunBatch
}
// tunBatch is WriteBatch's scratch, guarded by Utun.batchMu. The kernel reads every entry in place.
type tunBatch struct {
heads [tunWriteBatch][4]byte
iovs [tunWriteBatch][2]unix.Iovec
hdrs [tunWriteBatch]msgx.Hdr
pkts [tunWriteBatch][]byte
}
// NewUtun wraps a utun control socket fd and makes it non-blocking.
// On failure it does NOT close fd: the caller owns fd until NewUtun succeeds, after which Close closes it.
func NewUtun(fd int, l *slog.Logger) (*Utun, error) {
// Before os.NewFile, which only hands a non-blocking fd to the runtime poller.
if err := unix.SetNonblock(fd, true); err != nil {
return nil, fmt.Errorf("failed to set the tun fd to non-blocking mode: %w", err)
}
u := &Utun{
f: os.NewFile(uintptr(fd), "utun"),
l: l,
readBuf: make([]byte, tunReadArena),
}
u.noSendmsgX.Store(!msgx.Available())
return u, nil
}
// tunWritev and tunReadv are linkname'd to x/sys/unix's libc-routed writev/readv stubs so the
// calls go through libSystem's pinned trampoline. A raw syscall.Syscall(SYS_WRITEV/SYS_READV, ...)
// on darwin/arm64 emits an SVC #0x80 trap (see $GOROOT/src/syscall/asm_darwin_arm64.s), the path
// Apple keeps warning they will eventually disallow. We pull the low-level stubs instead of calling
// unix.Writev/unix.Readv because those take [][]byte and rebuild the []Iovec every call, which
// heap-allocates the header; linkname'ing the stubs lets us hand them our own stack-allocated
// iovecs. See golang/go#78049.
//go:linkname tunWritev golang.org/x/sys/unix.writev
//go:noescape
func tunWritev(fd int, iovecs []unix.Iovec) (n int, err error)
//go:linkname tunReadv golang.org/x/sys/unix.readv
//go:noescape
func tunReadv(fd int, iovecs []unix.Iovec) (n int, err error)
// Read waits for the utun to become readable, then reads packets until it would block, up to
// tunReadBatch, so the caller encrypts and sends them as one batch rather than one per wakeup. Each
// packet is read with its AF prefix scattered away, so the payload lands directly in readBuf.
// An error after at least one packet is dropped in favor of returning those packets; a persistent
// one recurs on the next Read.
func (u *Utun) Read() ([]Packet, error) {
rc, err := u.f.SyscallConn()
if err != nil {
return nil, err
}
var head [4]byte
n, off := 0, 0
var callErr error
err = rc.Read(func(fd uintptr) bool {
for n < tunReadBatch && len(u.readBuf)-off >= tunMaxPacket {
iovecs := [2]unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &u.readBuf[off], Len: uint64(len(u.readBuf) - off)},
}
l, e := tunReadv(int(fd), iovecs[:])
if e != nil {
if errno, ok := e.(syscall.Errno); ok && errno.Temporary() {
// Park on the poller only while there is nothing to hand back.
return n > 0
}
callErr = e
return true
}
if l < 4 {
// A datagram too short to carry the AF prefix, or end of file; stop rather than spin on it.
return true
}
end := off + l - 4
u.readRet[n] = Packet{Bytes: u.readBuf[off:end:end]}
n++
off = end
}
return true
})
if n > 0 {
return u.readRet[:n], nil
}
if err != nil {
return nil, err
}
return nil, callErr
}
// Write pushes one IP packet onto the utun device. Safe for concurrent use:
// the AF prefix and iovecs are per-call stack state, and the fd write itself
// serializes on the runtime's fd mutex (see the Queue contract in tio.go).
func (u *Utun) Write(from []byte) (int, error) {
if len(from) == 0 {
return 0, syscall.EIO
}
var head [4]byte
af, err := tunAF(from)
if err != nil {
return 0, err
}
head[3] = af
// Grab rc as a local so the compiler can devirtualize the call and keep the closure on the stack.
rc, err := u.f.SyscallConn()
if err != nil {
return 0, err
}
var n int
var callErr error
err = rc.Write(func(fd uintptr) bool {
iovecs := []unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &from[0], Len: uint64(len(from))},
}
n, callErr = tunWritev(int(fd), iovecs)
// Type-assert to syscall.Errno so the EAGAIN/EWOULDBLOCK/EINTR check doesn't box the errno
// constants into error interfaces on every call.
if errno, ok := callErr.(syscall.Errno); ok && errno.Temporary() {
return false
}
return true
})
if err != nil {
return 0, err
}
if callErr != nil {
return 0, callErr
}
return n - 4, nil
}
// tunAF returns the utun address-family prefix byte for an IP packet.
func tunAF(pkt []byte) (byte, error) {
switch pkt[0] >> 4 {
case 4:
return syscall.AF_INET, nil
case 6:
return syscall.AF_INET6, nil
default:
return 0, fmt.Errorf("unable to determine IP version from packet")
}
}
// WriteBatch writes pkts to the utun device with sendmsg_x, xnu's private batched sendmsg, up to
// tunWriteBatch packets per syscall. The kernel still hands each packet to utun on its own, so this
// saves syscalls, not per-packet kernel work. Safe for concurrent use.
//
// An empty packet or one with no IP version is skipped and reported, as Write would. Any other
// error from sendmsg_x means the kernel took an unknown prefix of that call and dropped the rest, so
// it is reported rather than retried, which could deliver packets twice.
//
// Under mbuf exhaustion, sendmsg_x silently drops packets that writev would have delivered: the
// kernel counts each packet as sent once it is copied in, and when a later allocation fails it
// frees that prefix unsent and returns its length as a short count, which is indistinguishable
// from a real one.
func (u *Utun) WriteBatch(pkts [][]byte) error {
if u.noSendmsgX.Load() {
return u.writeEach(pkts)
}
u.batchMu.Lock()
defer u.batchMu.Unlock()
b := &u.batch
var firstErr error
for len(pkts) > 0 {
n := 0
for len(pkts) > 0 && n < tunWriteBatch {
p := pkts[0]
pkts = pkts[1:]
var af byte
err := error(syscall.EIO)
if len(p) > 0 {
af, err = tunAF(p)
}
if err != nil {
if firstErr == nil {
firstErr = err
}
continue
}
b.heads[n] = [4]byte{3: af}
b.iovs[n] = [2]unix.Iovec{
{Base: &b.heads[n][0], Len: 4},
{Base: &p[0], Len: uint64(len(p))},
}
b.hdrs[n] = msgx.Hdr{Iov: &b.iovs[n][0], Iovlen: 2}
b.pkts[n] = p
n++
}
if n == 0 {
continue
}
for off := 0; off < n; {
sent, err := u.sendmsgX(off, n)
switch {
case err == nil && sent > 0:
// The kernel stops early without an error on a packet larger than the socket's send
// buffer, or when sending a later packet fails with ENOBUFS; resend the remainder.
off += sent
case err == nil:
if werr := u.writeEach(b.pkts[off:n]); werr != nil && firstErr == nil {
firstErr = werr
}
off = n
case err == unix.EMSGSIZE:
// Only the first packet being larger than the socket's send buffer fails the whole call,
// before the kernel takes any; Write reports that one and the rest go out in the next call.
if werr := u.writeEach(b.pkts[off : off+1]); werr != nil && firstErr == nil {
firstErr = werr
}
off++
case err == unix.ENOSYS || err == unix.EPERM || err == unix.EOPNOTSUPP:
// sendmsg_x is private API: fall back to writev if a kernel or sandbox refuses it.
if u.noSendmsgX.CompareAndSwap(false, true) {
u.l.Warn("sendmsg_x unavailable on the tun device, writing one packet per syscall", "error", err)
}
if werr := u.writeEach(b.pkts[off:n]); werr != nil && firstErr == nil {
firstErr = werr
}
clear(b.iovs[:n])
clear(b.pkts[:n])
if werr := u.writeEach(pkts); werr != nil && firstErr == nil {
firstErr = werr
}
return firstErr
default:
if firstErr == nil {
firstErr = err
}
off = n
}
}
clear(b.iovs[:n])
clear(b.pkts[:n])
}
return firstErr
}
// sendmsgX hands entries off through n-1 of u.batch to sendmsg_x and returns how many the kernel took.
func (u *Utun) sendmsgX(off, n int) (int, error) {
rc, err := u.f.SyscallConn()
if err != nil {
return 0, err
}
var sent int
var errno syscall.Errno
err = rc.Write(func(fd uintptr) bool {
sent, errno = msgx.Send(fd, u.batch.hdrs[off:n], 0)
// sendmsg_x reports EAGAIN only when it took nothing; a partial batch returns its count.
return !errno.Temporary()
})
if err != nil {
return 0, err
}
if errno != 0 {
return 0, errno
}
return sent, nil
}
// writeEach writes pkts one at a time with Write.
func (u *Utun) writeEach(pkts [][]byte) error {
var firstErr error
for _, p := range pkts {
if _, err := u.Write(p); err != nil && firstErr == nil {
firstErr = err
}
}
return firstErr
}
func (u *Utun) Close() error {
return u.f.Close()
}
-198
View File
@@ -1,198 +0,0 @@
//go:build darwin && !ios
package tio
import (
"bytes"
"log/slog"
"syscall"
"testing"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"golang.org/x/sys/unix"
)
// newSocketpairUtun returns a Utun over one end of a connected AF_UNIX datagram socketpair, which
// takes sendmsg_x the way a utun control socket does, and the other end's fd for the test to use.
// sndbuf is the Utun's send buffer, which caps the largest datagram it accepts.
func newSocketpairUtun(t *testing.T, sndbuf int) (*Utun, int) {
t.Helper()
fds, err := unix.Socketpair(unix.AF_UNIX, unix.SOCK_DGRAM, 0)
require.NoError(t, err)
require.NoError(t, unix.SetsockoptInt(fds[0], unix.SOL_SOCKET, unix.SO_SNDBUF, sndbuf))
// An AF_UNIX datagram send is bounded by the receiver's buffer, so size both ends' for the tests.
require.NoError(t, unix.SetsockoptInt(fds[0], unix.SOL_SOCKET, unix.SO_RCVBUF, 1<<20))
require.NoError(t, unix.SetsockoptInt(fds[1], unix.SOL_SOCKET, unix.SO_RCVBUF, 1<<20))
// A dropped datagram fails the read-back instead of hanging it.
require.NoError(t, unix.SetsockoptTimeval(fds[1], unix.SOL_SOCKET, unix.SO_RCVTIMEO, &unix.Timeval{Sec: 2}))
u, err := NewUtun(fds[0], slog.New(slog.DiscardHandler))
if err != nil {
_ = unix.Close(fds[0])
_ = unix.Close(fds[1])
t.Fatal(err)
}
t.Cleanup(func() {
_ = u.Close()
_ = unix.Close(fds[1])
})
return u, fds[1]
}
// testTunPkt returns packet i of size bytes, alternating IPv4 and IPv6, and the datagram the utun
// device should receive for it.
func testTunPkt(i, size int) (pkt, wire []byte) {
pkt = make([]byte, size)
pkt[0] = 0x45
af := byte(syscall.AF_INET)
if i%2 == 1 {
pkt[0] = 0x60
af = syscall.AF_INET6
}
pkt[1] = byte(i)
pkt[2] = byte(i >> 8)
return pkt, append([]byte{0, 0, 0, af}, pkt...)
}
// readTunBack asserts r receives exactly want, in order.
func readTunBack(t *testing.T, r int, want [][]byte) {
t.Helper()
buf := make([]byte, 4096)
for i, w := range want {
n, _, err := unix.Recvfrom(r, buf, 0)
require.NoError(t, err, "datagram %d", i)
if !bytes.Equal(w, buf[:n]) {
t.Fatalf("datagram %d: got % x, want % x", i, buf[:n], w)
}
}
_, _, err := unix.Recvfrom(r, buf, unix.MSG_DONTWAIT)
assert.ErrorIs(t, err, unix.EAGAIN, "extra datagrams delivered")
}
// forEachTunWritePath runs fn once over sendmsg_x and once over the per-packet writev fallback.
func forEachTunWritePath(t *testing.T, fn func(t *testing.T, fallback bool)) {
for _, fallback := range []bool{false, true} {
name := "sendmsg_x"
if fallback {
name = "writev"
}
t.Run(name, func(t *testing.T) { fn(t, fallback) })
}
}
// TestUtunWriteBatch pins that WriteBatch delivers every packet whole, in order and behind the right
// utun AF prefix, across several sendmsg_x calls, and returns nil.
func TestUtunWriteBatch(t *testing.T) {
forEachTunWritePath(t, func(t *testing.T, fallback bool) {
u, r := newSocketpairUtun(t, 1<<20)
u.noSendmsgX.Store(fallback)
var pkts, want [][]byte
for i := range 3*tunWriteBatch + 5 {
p, w := testTunPkt(i, 60+i)
pkts = append(pkts, p)
want = append(want, w)
}
require.NoError(t, u.WriteBatch(pkts))
assert.Equal(t, fallback, u.noSendmsgX.Load())
readTunBack(t, r, want)
})
}
// TestUtunWriteBatchSkipsInvalid pins that an empty packet or one with no IP version is skipped
// without shifting the packets after it, and that the first such error is the one reported.
func TestUtunWriteBatchSkipsInvalid(t *testing.T) {
forEachTunWritePath(t, func(t *testing.T, fallback bool) {
u, r := newSocketpairUtun(t, 1<<20)
u.noSendmsgX.Store(fallback)
var pkts, want [][]byte
for i := range 20 {
p, w := testTunPkt(i, 60)
pkts = append(pkts, p)
want = append(want, w)
}
pkts = append(pkts[:10], append([][]byte{{0x10, 1, 2}, {}}, pkts[10:]...)...)
err := u.WriteBatch(pkts)
assert.ErrorContains(t, err, "IP version")
readTunBack(t, r, want)
})
}
// TestUtunWriteBatchOversize pins that a packet larger than the socket's send buffer drops only
// itself and is reported, whether sendmsg_x reaches it mid-call (a short count) or first (EMSGSIZE
// for the whole call). The send buffer is 1024 bytes and the oversize packet 2000.
func TestUtunWriteBatchOversize(t *testing.T) {
for _, tc := range []struct {
name string
at int
}{
{"mid-call", tunWriteBatch + 6},
{"first-in-call", tunWriteBatch},
} {
t.Run(tc.name, func(t *testing.T) {
forEachTunWritePath(t, func(t *testing.T, fallback bool) {
u, r := newSocketpairUtun(t, 1024)
u.noSendmsgX.Store(fallback)
var pkts, want [][]byte
for i := range 2*tunWriteBatch + 10 {
size := 100
if i == tc.at {
size = 2000
}
p, w := testTunPkt(i, size)
pkts = append(pkts, p)
if i != tc.at {
want = append(want, w)
}
}
err := u.WriteBatch(pkts)
assert.ErrorIs(t, err, unix.EMSGSIZE)
assert.Equal(t, fallback, u.noSendmsgX.Load())
readTunBack(t, r, want)
})
})
}
}
// TestUtunReadDrains pins that Read hands back every queued packet, AF prefix stripped and in
// order, capped at tunReadBatch per call, and doesn't wait for more once the queue is empty.
func TestUtunReadDrains(t *testing.T) {
u, w := newSocketpairUtun(t, 1<<20)
total := tunReadBatch + 3
var want [][]byte
for i := range total {
pkt, wire := testTunPkt(i, 60+i)
_, err := unix.Write(w, wire)
require.NoError(t, err)
want = append(want, pkt)
}
var got [][]byte
for _, size := range []int{tunReadBatch, 3} {
pkts, err := u.Read()
require.NoError(t, err)
require.Len(t, pkts, size)
for _, p := range pkts {
got = append(got, p.Clone().Bytes)
}
}
assert.Equal(t, want, got)
}
// TestUtunReadClipsPackets pins that each returned packet's capacity ends at its own bytes, so
// appending to one can't overwrite the next.
func TestUtunReadClipsPackets(t *testing.T) {
u, w := newSocketpairUtun(t, 1<<20)
for i := range 2 {
_, wire := testTunPkt(i, 100)
_, err := unix.Write(w, wire)
require.NoError(t, err)
}
pkts, err := u.Read()
require.NoError(t, err)
require.Len(t, pkts, 2)
assert.Equal(t, len(pkts[0].Bytes), cap(pkts[0].Bytes))
}
+23 -1
View File
@@ -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
@@ -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
}
+1 -1
View File
@@ -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
}
+118 -15
View File
@@ -7,7 +7,9 @@ import (
"fmt"
"log/slog"
"net/netip"
"os"
"sync/atomic"
"syscall"
"unsafe"
"github.com/gaissmai/bart"
@@ -20,7 +22,7 @@ import (
)
type tun struct {
queue *tio.Utun
f *os.File
Device string
vpnNetworks []netip.Prefix
DefaultMTU int
@@ -115,16 +117,16 @@ func newTun(c *config.C, l *slog.Logger, vpnNetworks []netip.Prefix, _ bool) (*t
return nil, fmt.Errorf("failed to retrieve tun name: %w", err)
}
queue, err := tio.NewUtun(fd, l)
err = unix.SetNonblock(fd, true)
if err != nil {
return nil, err
return nil, fmt.Errorf("SetNonblock: %v", err)
}
t := &tun{
queue: queue,
f: os.NewFile(uintptr(fd), ""),
Device: name,
vpnNetworks: vpnNetworks,
DefaultMTU: c.GetInt("tun.mtu", DefaultMTU),
DefaultMTU: getMTU(l, c),
l: l,
}
@@ -154,24 +156,24 @@ func (t *tun) deviceBytes() (o [16]byte) {
// network extension is handed its device. Everything about moving packets is shared with newTun,
// only the setup differs: the host owns addressing and routing here.
func newTunFromFd(c *config.C, l *slog.Logger, deviceFd int, vpnNetworks []netip.Prefix) (*tun, error) {
queue, err := tio.NewUtun(deviceFd, l)
if err != nil {
if err := unix.SetNonblock(deviceFd, true); err != nil {
// We own the fd from the moment it is handed to us
_ = unix.Close(deviceFd)
return nil, err
return nil, fmt.Errorf("failed to set the tun fd to non-blocking mode: %w", err)
}
file := os.NewFile(uintptr(deviceFd), "/dev/tun")
t := &tun{
queue: queue,
f: file,
Device: utunNameFromFd(deviceFd),
vpnNetworks: vpnNetworks,
DefaultMTU: c.GetInt("tun.mtu", DefaultMTU),
DefaultMTU: getMTU(l, c),
hostOwned: true,
l: l,
}
if err := t.reload(c, true); err != nil {
_ = queue.Close()
_ = file.Close()
return nil, err
}
@@ -195,8 +197,8 @@ func utunNameFromFd(fd int) string {
}
func (t *tun) Close() error {
if t.queue != nil {
return t.queue.Close()
if t.f != nil {
return t.f.Close()
}
return nil
}
@@ -348,7 +350,7 @@ func (t *tun) activate6(network netip.Prefix) 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
}
@@ -554,6 +556,107 @@ func delRoute(prefix netip.Prefix, gateway netroute.Addr) error {
return nil
}
// tunWritev and tunReadv are linkname'd to x/sys/unix's libc-routed writev/readv stubs so the
// calls go through libSystem's pinned trampoline. A raw syscall.Syscall(SYS_WRITEV/SYS_READV, ...)
// on darwin/arm64 emits an SVC #0x80 trap (see $GOROOT/src/syscall/asm_darwin_arm64.s), the path
// Apple keeps warning they will eventually disallow. We pull the low-level stubs instead of calling
// unix.Writev/unix.Readv because those take [][]byte and rebuild the []Iovec every call, which
// heap-allocates the header; linkname'ing the stubs lets us hand them our own stack-allocated
// iovecs. See golang/go#78049.
//go:linkname tunWritev golang.org/x/sys/unix.writev
//go:noescape
func tunWritev(fd int, iovecs []unix.Iovec) (n int, err error)
//go:linkname tunReadv golang.org/x/sys/unix.readv
//go:noescape
func tunReadv(fd int, iovecs []unix.Iovec) (n int, err error)
// Read pulls one IP packet off the utun device, scattering the 4 byte protocol header away from
// the packet so the payload lands directly in to.
func (t *tun) Read(to []byte) (int, error) {
var head [4]byte
rc, err := t.f.SyscallConn()
if err != nil {
return 0, err
}
var n int
var callErr error
err = rc.Read(func(fd uintptr) bool {
iovecs := []unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &to[0], Len: uint64(len(to))},
}
n, callErr = tunReadv(int(fd), iovecs)
if errno, ok := callErr.(syscall.Errno); ok && errno.Temporary() {
return false
}
return true
})
if err != nil {
return 0, err
}
if callErr != nil {
return 0, callErr
}
if n < 4 {
return 0, nil
}
return n - 4, nil
}
// Write pushes one IP packet onto the utun device. Safe for concurrent use:
// the AF prefix and iovecs are per-call stack state, and the fd write itself
// serializes on the runtime's fd mutex (see the Queue contract in tio.go).
func (t *tun) Write(from []byte) (int, error) {
if len(from) == 0 {
return 0, syscall.EIO
}
ipVer := from[0] >> 4
var head [4]byte
switch ipVer {
case 4:
head[3] = syscall.AF_INET
case 6:
head[3] = syscall.AF_INET6
default:
return 0, fmt.Errorf("unable to determine IP version from packet")
}
// Grab rc as a local so the compiler can devirtualize the call and keep the closure on the stack.
rc, err := t.f.SyscallConn()
if err != nil {
return 0, err
}
var n int
var callErr error
err = rc.Write(func(fd uintptr) bool {
iovecs := []unix.Iovec{
{Base: &head[0], Len: 4},
{Base: &from[0], Len: uint64(len(from))},
}
n, callErr = tunWritev(int(fd), iovecs)
// Type-assert to syscall.Errno so the EAGAIN/EWOULDBLOCK/EINTR check doesn't box the errno
// constants into error interfaces on every call.
if errno, ok := callErr.(syscall.Errno); ok && errno.Temporary() {
return false
}
return true
})
if err != nil {
return 0, err
}
if callErr != nil {
return 0, callErr
}
return n - 4, nil
}
func (t *tun) Networks() []netip.Prefix {
return t.vpnNetworks
}
@@ -563,5 +666,5 @@ func (t *tun) Name() string {
}
func (t *tun) Queues(int) ([]tio.Queue, error) {
return []tio.Queue{t.queue}, nil
return []tio.Queue{tio.NewSingleQueue(t, defaultBatchBufSize)}, nil
}
+2 -2
View File
@@ -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
}
+1 -1
View File
@@ -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
}
+12 -3
View File
@@ -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
}
}
+2 -2
View File
@@ -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
}
+2 -2
View File
@@ -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
}
+56
View File
@@ -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)
}
+1 -1
View File
@@ -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
}
+79 -5
View File
@@ -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 {
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()
+168
View File
@@ -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") })
}
+91
View File
@@ -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
}
}
}
+264
View File
@@ -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())
}
}
+6 -2
View File
@@ -684,10 +684,14 @@ func sshCreateTunnel(ifce *Interface, fs any, a []string, w sshd.StringWriter) e
}
}
hostInfo = ifce.handshakeManager.StartHandshake(vpnAddr, nil)
// 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() {
hostInfo.SetRemote(addr)
hh.hostinfo.SetRemote(addr)
}
})
return w.WriteLine("Created")
}
+38 -14
View File
@@ -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 {
// 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 {
@@ -218,13 +231,12 @@ 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
}
if rt != nil {
rt.cancel()
if rt.listener != nil {
shutdownCtx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
@@ -234,10 +246,17 @@ func (s *statsServer) Stop() {
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
View File
@@ -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))
@@ -371,34 +375,18 @@ 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
// 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()
}
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
})
return ctx
}
go func(c net.Conn) {
_, _ = io.Copy(io.Discard, c)
_ = c.Close()
}(conn)
}
}()
return g
}
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()
@@ -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")
}
+4 -261
View File
@@ -10,12 +10,10 @@ import (
"log/slog"
"net"
"net/netip"
"sync/atomic"
"syscall"
"unsafe"
"github.com/slackhq/nebula/config"
"github.com/slackhq/nebula/internal/msgx"
"golang.org/x/sys/unix"
)
@@ -24,9 +22,6 @@ type StdConn struct {
isV4 bool
sysFd uintptr
l *slog.Logger
noSendmsgX atomic.Bool
sb *sendScratch
}
var _ Conn = &StdConn{}
@@ -40,7 +35,6 @@ func NewListener(l *slog.Logger, s Settings) (Conn, error) {
if uc, ok := pc.(*net.UDPConn); ok {
c := &StdConn{UDPConn: uc, l: l}
c.noSendmsgX.Store(!msgx.Available())
rc, err := uc.SyscallConn()
if err != nil {
@@ -145,138 +139,11 @@ func (u *StdConn) WriteTo(b []byte, ap netip.AddrPort) error {
}
}
// sendBatch is how many datagrams one sendmsg_x call may carry.
const sendBatch = 64
// sendScratch is WriteBatch's scratch. The kernel reads every entry in place.
type sendScratch struct {
names [sendBatch]unix.RawSockaddrInet6
iovs [sendBatch]unix.Iovec
hdrs [sendBatch]msgx.Hdr
// idx maps each entry back to its index in WriteBatch's bufs.
idx [sendBatch]int
}
// WriteBatch sends bufs with sendmsg_x, darwin's private batched sendmsg, so one syscall carries up to
// sendBatch datagrams. On this unconnected socket xnu still runs each message through sendit, so the
// saving is the syscall crossings, not the per-send policy checks.
//
// xnu's per-message loop stops at the first message that fails. For the errors in sentNothing it reports
// how many went out; for any other it returns only the errno, with the messages before it already sent.
// Each call therefore carries one run of datagrams to the same destination: a destination the kernel
// refuses fails on its first datagram, so a failed call costs that run and nothing else, and no datagram
// is ever resent after the kernel may have taken it.
//
func (u *StdConn) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, error) {
// An un-sendable destination costs its own packet, never the ones behind it in the batch.
// TODO: WriteTo maps EWOULDBLOCK to an error, so a full send buffer
// silently drops the rest of a burst (linux blocks instead). Poll for
// writability on EAGAIN before giving up on the remainder.
func (u *StdConn) WriteBatch(bufs [][]byte, addrs []netip.AddrPort) (int, error) {
if u.noSendmsgX.Load() {
return u.writeEach(bufs, addrs), nil
}
if u.sb == nil {
u.sb = new(sendScratch)
}
b := u.sb
written := 0
for i := 0; i < len(bufs); {
n := 0
for ; i < len(bufs) && n < sendBatch; i++ {
namelen, err := u.putSockaddr(&b.names[n], addrs[i])
if err != nil {
u.l.Debug("failed to write packet in batch", "udpAddr", addrs[i], "error", err)
continue
}
b.iovs[n] = unix.Iovec{}
if len(bufs[i]) > 0 {
b.iovs[n].Base = &bufs[i][0]
b.iovs[n].SetLen(len(bufs[i]))
}
b.hdrs[n] = msgx.Hdr{Name: (*byte)(unsafe.Pointer(&b.names[n])), Namelen: namelen, Iov: &b.iovs[n], Iovlen: 1}
b.idx[n] = i
n++
}
for off := 0; off < n; {
dst := addrs[b.idx[off]]
end := off + 1
for end < n && addrs[b.idx[end]] == dst {
end++
}
sent, errno := u.sendmsgX(off, end)
switch {
case errno == 0 && sent > 0:
written += sent
off += sent
continue
case sentNothing(errno):
// The datagram at off failed on its own account. sendto delivers or reports it, as without
// batching, and the rest of the run goes back to sendmsg_x.
j := b.idx[off]
written += u.writeEach(bufs[j:j+1], addrs[j:j+1])
off++
continue
case errno == 0:
case errno == unix.ENOSYS || errno == unix.EOPNOTSUPP:
u.fallBack(errno)
case errno == unix.EPERM || errno == unix.EINVAL:
// Either can refuse one destination, but a datagram that sendto delivers was refused by
// sendmsg_x itself, as a sandbox or an older kernel would.
j := b.idx[off]
if u.writeEach(bufs[j:j+1], addrs[j:j+1]) == 0 {
off = end
continue
}
written++
off++
u.fallBack(errno)
default:
u.l.Debug("failed to write packets in batch", "udpAddr", dst, "packets", end-off, "error", errno)
off = end
continue
}
// The kernel took nothing from this call, so the rest can go out one datagram at a time.
for _, j := range b.idx[off:n] {
written += u.writeEach(bufs[j:j+1], addrs[j:j+1])
}
off = n
if u.noSendmsgX.Load() {
clear(b.iovs[:n])
return written + u.writeEach(bufs[i:], addrs[i:]), nil
}
}
clear(b.iovs[:n])
}
return written, nil
}
// sendmsgX hands entries off through end-1 of u.sb to sendmsg_x and returns how many the kernel took.
// The socket is non-blocking, so a full send buffer returns EAGAIN rather than waiting.
func (u *StdConn) sendmsgX(off, end int) (int, syscall.Errno) {
for {
sent, errno := msgx.Send(u.sysFd, u.sb.hdrs[off:end], 0)
if errno == unix.EINTR {
continue
}
return sent, errno
}
}
// sentNothing reports whether sendmsg_x failing with errno means no datagram went out: once any has, xnu
// turns these errors into a short count instead.
func sentNothing(errno syscall.Errno) bool {
return errno == unix.EWOULDBLOCK || errno == unix.ENOBUFS || errno == unix.EMSGSIZE
}
// fallBack switches WriteBatch to one sendto per datagram for the life of the socket.
func (u *StdConn) fallBack(errno syscall.Errno) {
if u.noSendmsgX.CompareAndSwap(false, true) {
u.l.Warn("sendmsg_x unavailable, sending one datagram per syscall", "error", errno)
}
}
// writeEach sends bufs one at a time with WriteTo and returns how many were written.
func (u *StdConn) writeEach(bufs [][]byte, addrs []netip.AddrPort) int {
written := 0
for i, b := range bufs {
if err := u.WriteTo(b, addrs[i]); err == nil {
@@ -285,23 +152,7 @@ func (u *StdConn) writeEach(bufs [][]byte, addrs []netip.AddrPort) int {
u.l.Debug("failed to write packet in batch", "udpAddr", addrs[i], "error", err)
}
}
return written
}
// putSockaddr writes ap into sa in the socket's address family, as WriteTo would, and returns its length.
func (u *StdConn) putSockaddr(sa *unix.RawSockaddrInet6, ap netip.AddrPort) (uint32, error) {
if u.isV4 {
if ap.Addr().Is6() {
return 0, ErrInvalidIPv6RemoteForSocket
}
sa4 := (*unix.RawSockaddrInet4)(unsafe.Pointer(sa))
*sa4 = unix.RawSockaddrInet4{Family: unix.AF_INET, Addr: ap.Addr().As4()}
binary.BigEndian.PutUint16((*[2]byte)(unsafe.Pointer(&sa4.Port))[:], ap.Port())
return unix.SizeofSockaddrInet4, nil
}
*sa = unix.RawSockaddrInet6{Family: unix.AF_INET6, Addr: ap.Addr().As16()}
binary.BigEndian.PutUint16((*[2]byte)(unsafe.Pointer(&sa.Port))[:], ap.Port())
return unix.SizeofSockaddrInet6, nil
return written, nil
}
func (u *StdConn) LocalAddr() (netip.AddrPort, error) {
@@ -329,106 +180,11 @@ func NewUDPStatsEmitter(udpConns []Conn) func() {
return func() {}
}
// recvBatch is how many datagrams one recvmsg_x call may return.
const recvBatch = 64
// ListenOut drains the socket with recvmsg_x, darwin's private batched recvmsg, so one syscall reads
// up to recvBatch datagrams instead of one.
func (u *StdConn) ListenOut(r EncReader, flush func()) error {
if !msgx.Available() {
return u.listenOutSingle(r, flush)
}
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([]msgx.Hdr, 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
}
n, errno = msgx.Recv(fd, hdrs, 0)
return errno != unix.EAGAIN
})
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 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 []msgx.Hdr, 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)
for {
// Just read one packet at a time
n, rua, err := u.ReadFromUDPAddrPort(buffer)
if err != nil {
if errors.Is(err, net.ErrClosed) {
@@ -443,19 +199,6 @@ func (u *StdConn) listenOutSingle(r EncReader, flush func()) error {
}
}
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
}
+24 -4
View File
@@ -135,12 +135,23 @@ func (u *StdConn) Rebind() error {
return nil
}
// SetRecvBuffer tries SO_RCVBUFFORCE, which needs CAP_NET_ADMIN. Without it we
// fall back to SO_RCVBUF, which the kernel clamps to net.core.rmem_max.
func (u *StdConn) SetRecvBuffer(n int) error {
return unix.SetsockoptInt(u.sysFd, unix.SOL_SOCKET, unix.SO_RCVBUFFORCE, n)
err := unix.SetsockoptInt(u.sysFd, unix.SOL_SOCKET, unix.SO_RCVBUFFORCE, n)
if errors.Is(err, unix.EPERM) {
err = unix.SetsockoptInt(u.sysFd, unix.SOL_SOCKET, unix.SO_RCVBUF, n)
}
return err
}
// SetSendBuffer is SetRecvBuffer for SO_SNDBUFFORCE/SO_SNDBUF and net.core.wmem_max.
func (u *StdConn) SetSendBuffer(n int) error {
return unix.SetsockoptInt(u.sysFd, unix.SOL_SOCKET, unix.SO_SNDBUFFORCE, n)
err := unix.SetsockoptInt(u.sysFd, unix.SOL_SOCKET, unix.SO_SNDBUFFORCE, n)
if errors.Is(err, unix.EPERM) {
err = unix.SetsockoptInt(u.sysFd, unix.SOL_SOCKET, unix.SO_SNDBUF, n)
}
return err
}
func (u *StdConn) SetSoMark(mark int) error {
@@ -396,8 +407,13 @@ func (u *StdConn) ReloadConfig(c *config.C) {
b := c.GetInt("listen.read_buffer", 0)
if b > 0 {
if err := u.SetRecvBuffer(b); err == nil {
if s, err := u.GetRecvBuffer(); err == nil {
// The kernel reports double the requested size
if s, err := u.GetRecvBuffer(); err == nil && s >= 2*b {
u.l.Info("listen.read_buffer was set", "size", s)
} else if err == nil {
u.l.Warn("listen.read_buffer was limited by the system",
"requested", b, "size", s,
"hint", "raise net.core.rmem_max or grant CAP_NET_ADMIN")
} else {
u.l.Warn("Failed to get listen.read_buffer", "error", err)
}
@@ -409,8 +425,12 @@ func (u *StdConn) ReloadConfig(c *config.C) {
b = c.GetInt("listen.write_buffer", 0)
if b > 0 {
if err := u.SetSendBuffer(b); err == nil {
if s, err := u.GetSendBuffer(); err == nil {
if s, err := u.GetSendBuffer(); err == nil && s >= 2*b {
u.l.Info("listen.write_buffer was set", "size", s)
} else if err == nil {
u.l.Warn("listen.write_buffer was limited by the system",
"requested", b, "size", s,
"hint", "raise net.core.wmem_max or grant CAP_NET_ADMIN")
} else {
u.l.Warn("Failed to get listen.write_buffer", "error", err)
}
+124
View File
@@ -0,0 +1,124 @@
//go:build linux && !android && !e2e_testing
package udp
import (
"bytes"
"fmt"
"log/slog"
"net/netip"
"os"
"strconv"
"strings"
"testing"
"github.com/slackhq/nebula/config"
)
func readSysctlInt(t *testing.T, path string) int {
t.Helper()
b, err := os.ReadFile(path)
if err != nil {
t.Skipf("can't read %s: %v", path, err)
}
n, err := strconv.Atoi(strings.TrimSpace(string(b)))
if err != nil {
t.Skipf("can't parse %s: %v", path, err)
}
return n
}
// TestSetBufferAboveSysctlMax asks for more than net.core.{r,w}mem_max. With
// CAP_NET_ADMIN the FORCE variant takes it as is; without it we must fall back
// to the plain option and get the clamped size instead of EPERM.
func TestSetBufferAboveSysctlMax(t *testing.T) {
cases := []struct {
name string
sysctl string
set func(*StdConn, int) error
get func(*StdConn) (int, error)
}{
{"read", "/proc/sys/net/core/rmem_max", (*StdConn).SetRecvBuffer, (*StdConn).GetRecvBuffer},
{"write", "/proc/sys/net/core/wmem_max", (*StdConn).SetSendBuffer, (*StdConn).GetSendBuffer},
}
for _, tc := range cases {
t.Run(tc.name, func(t *testing.T) {
sysMax := readSysctlInt(t, tc.sysctl)
c, err := NewListener(testLogger(), Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0"), Batch: 1})
if err != nil {
t.Skipf("NewListener: %v", err)
}
sc := c.(*StdConn)
defer sc.Close()
want := sysMax + 1<<20
if err := tc.set(sc, want); err != nil {
t.Fatalf("set %d: %v", want, err)
}
got, err := tc.get(sc)
if err != nil {
t.Fatalf("get: %v", err)
}
// The kernel doubles the value; either path gives at least the clamped size.
if got < 2*sysMax {
t.Fatalf("effective size %d, want >= %d", got, 2*sysMax)
}
t.Logf("requested %d, %s=%d, effective %d (forced=%v)", want, tc.sysctl, sysMax, got, got >= 2*want)
})
}
}
// TestReloadConfigBufferLogLevel checks that a clamped buffer is a Warn, and a
// buffer that fits is still the Info it always was.
func TestReloadConfigBufferLogLevel(t *testing.T) {
sysMax := readSysctlInt(t, "/proc/sys/net/core/rmem_max")
if wmax := readSysctlInt(t, "/proc/sys/net/core/wmem_max"); wmax < sysMax {
sysMax = wmax
}
cases := []struct {
name string
size int
want []string
}{
{"fits", 64 * 1024, []string{"level=INFO msg=\"listen.read_buffer was set\"", "level=INFO msg=\"listen.write_buffer was set\""}},
{"too big", sysMax + 1<<20, []string{"level=WARN msg=\"listen.read_buffer was limited by the system\"", "level=WARN msg=\"listen.write_buffer was limited by the system\""}},
}
for _, tc := range cases {
t.Run(tc.name, func(t *testing.T) {
var buf bytes.Buffer
l := slog.New(slog.NewTextHandler(&buf, nil))
c, err := NewListener(l, Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0"), Batch: 1})
if err != nil {
t.Skipf("NewListener: %v", err)
}
sc := c.(*StdConn)
defer sc.Close()
if tc.size > sysMax {
// With CAP_NET_ADMIN the FORCE variant takes any size, nothing gets limited.
if err := sc.SetRecvBuffer(tc.size); err == nil {
if got, _ := sc.GetRecvBuffer(); got >= 2*tc.size {
t.Skip("FORCE path took the full size, can't observe the limit")
}
}
}
cfg := config.NewC(l)
if err := cfg.LoadString(fmt.Sprintf("listen:\n read_buffer: %d\n write_buffer: %d\n", tc.size, tc.size)); err != nil {
t.Fatal(err)
}
buf.Reset()
sc.ReloadConfig(cfg)
out := buf.String()
for _, w := range tc.want {
if !strings.Contains(out, w) {
t.Errorf("missing %s in:\n%s", w, out)
}
}
if strings.Contains(out, "level=ERROR") {
t.Errorf("unexpected error log:\n%s", out)
}
t.Log(out)
})
}
}
-85
View File
@@ -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")
})
}
}
-302
View File
@@ -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/slackhq/nebula/internal/msgx"
"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)
})
}
}
// msgx.Hdr 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() ([]msgx.Hdr, []unix.RawSockaddrInet6) {
hdrs := make([]msgx.Hdr, 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([]msgx.Hdr, []unix.RawSockaddrInet6)
}{
{"count past headers", 3, func([]msgx.Hdr, []unix.RawSockaddrInet6) {}},
{"datalen past buffer", 2, func(h []msgx.Hdr, _ []unix.RawSockaddrInet6) { h[1].Datalen = MTU + 1 }},
{"unknown family", 2, func(_ []msgx.Hdr, a []unix.RawSockaddrInet6) { a[0].Family = unix.AF_UNIX }},
{"family and length disagree", 2, func(h []msgx.Hdr, _ []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([]msgx.Hdr, 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] = msgx.Hdr{
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 {
n, errno = msgx.Recv(fd, hdrs[:nh], 0)
return errno != 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
}
+1 -1
View File
@@ -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
)
-150
View File
@@ -1,150 +0,0 @@
//go:build !e2e_testing
package udp
import (
"fmt"
"log/slog"
"net"
"net/netip"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// recvAll reads n datagrams from pc, or fails the test after a second without one.
func recvAll(t *testing.T, pc net.PacketConn, n int) []string {
t.Helper()
var got []string
buf := make([]byte, 64)
for len(got) < n {
require.NoError(t, pc.SetReadDeadline(time.Now().Add(time.Second)))
l, _, err := pc.ReadFrom(buf)
require.NoError(t, err, "after %d of %d datagrams", len(got), n)
got = append(got, string(buf[:l]))
}
return got
}
// A batch longer than sendBatch reaches each peer whole and in order through sendmsg_x, without falling
// back to sendto: a run to one peer that spans two calls, then datagrams alternating between peers.
func TestWriteBatchSendsEachToItsPeer(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)
defer c.Close()
var peers [2]net.PacketConn
var dsts [2]netip.AddrPort
for i := range peers {
peers[i], err = net.ListenPacket("udp", net.JoinHostPort(tc.listen, "0"))
require.NoError(t, err)
defer peers[i].Close()
dsts[i] = netip.MustParseAddrPort(peers[i].LocalAddr().String())
}
const total = sendBatch*2 + 5
bufs := make([][]byte, total)
addrs := make([]netip.AddrPort, total)
var want [2][]string
for i := range bufs {
bufs[i] = fmt.Appendf(nil, "pkt-%03d", i)
p := i % 2
if i < sendBatch+6 {
p = 0
}
addrs[i] = dsts[p]
want[p] = append(want[p], string(bufs[i]))
}
n, err := c.WriteBatch(bufs, addrs)
require.NoError(t, err)
assert.Equal(t, total, n)
assert.False(t, c.(*StdConn).noSendmsgX.Load(), "fell back to sendto")
for i := range peers {
assert.Equal(t, want[i], recvAll(t, peers[i], len(want[i])))
}
})
}
}
// Destinations that can't be sent to cost only their own packets: an IPv6 peer on an IPv4 socket is
// skipped before the call, and a broadcast address without SO_BROADCAST is refused by the kernel
// mid-batch. Neither stops the packets around it, delivers any twice, or turns sendmsg_x off.
func TestWriteBatchSkipsUnsendableDestinations(t *testing.T) {
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0")})
require.NoError(t, err)
defer c.Close()
peer, err := net.ListenPacket("udp", "127.0.0.1:0")
require.NoError(t, err)
defer peer.Close()
good := netip.MustParseAddrPort(peer.LocalAddr().String())
bcast := netip.AddrPortFrom(netip.MustParseAddr("255.255.255.255"), good.Port())
v6 := netip.MustParseAddrPort("[::1]:4242")
addrs := []netip.AddrPort{good, bcast, bcast, good, v6, good, bcast}
bufs := make([][]byte, len(addrs))
for i := range bufs {
bufs[i] = fmt.Appendf(nil, "pkt-%d", i)
}
n, err := c.WriteBatch(bufs, addrs)
require.NoError(t, err)
assert.Equal(t, 3, n)
assert.False(t, c.(*StdConn).noSendmsgX.Load(), "fell back to sendto")
assert.Equal(t, []string{"pkt-0", "pkt-3", "pkt-5"}, recvAll(t, peer, 3))
require.NoError(t, peer.SetReadDeadline(time.Now().Add(100*time.Millisecond)))
_, _, err = peer.ReadFrom(make([]byte, 64))
assert.Error(t, err, "a datagram arrived twice")
}
// A datagram the kernel refuses as its own fault (EMSGSIZE here, as ENOBUFS would be under a full
// interface queue) costs only itself, whether it fails after others in the same call (a short count) or
// first (the errno alone): the rest of its run still arrives, in order, and sendmsg_x stays on.
func TestWriteBatchLosesOnlyTheFailedDatagram(t *testing.T) {
for _, tc := range []struct {
name string
big int
}{
{"mid-call", 2},
{"first", 0},
} {
t.Run(tc.name, func(t *testing.T) {
c, err := NewListener(slog.New(slog.DiscardHandler), Settings{Listen: netip.MustParseAddrPort("127.0.0.1:0")})
require.NoError(t, err)
defer c.Close()
peer, err := net.ListenPacket("udp", "127.0.0.1:0")
require.NoError(t, err)
defer peer.Close()
dst := netip.MustParseAddrPort(peer.LocalAddr().String())
bufs := make([][]byte, 5)
addrs := make([]netip.AddrPort, len(bufs))
var want []string
for i := range bufs {
addrs[i] = dst
if i == tc.big {
// Past the 65535-byte IP datagram limit, so the kernel can only refuse it.
bufs[i] = make([]byte, 70000)
continue
}
bufs[i] = fmt.Appendf(nil, "pkt-%d", i)
want = append(want, string(bufs[i]))
}
n, err := c.WriteBatch(bufs, addrs)
require.NoError(t, err)
assert.Equal(t, len(want), n)
assert.False(t, c.(*StdConn).noSendmsgX.Load(), "fell back to sendto")
assert.Equal(t, want, recvAll(t, peer, len(want)))
})
}
}