Files
netbird/client/internal/dns/server_privileged_test.go
T
Riccardo Manfrin 3e85e40be2 [client] Cache the WireGuard interface check shared by ICE agents (#8001)
* [client] Take a WireGuard detector through the interface filter

The interface filter answers whether an interface is a WireGuard device by opening
a wgctrl client and asking for it, and it does that for every interface it is given.
Nothing about that call is tied to the caller, so it can be answered by a shared
object instead of being repeated, but the filter has no way to receive one.

InterfaceFilter and the constructors that build one now take a detector, and the ICE
config carries it so that every agent can be handed the same one. Nobody supplies a
detector yet: a nil one probes on every call, which is what the filter did before, so
this changes no behaviour.

* [client] Share one WireGuard detector across every ICE agent

Creating an ICE agent builds two interface filters, one for the agent and one for
the transport net it sits on, and each is asked about every host interface. For an
interface the disallow list does not settle, answering means opening a wgctrl client,
which builds a kernel and a userspace client and resolves the netlink family, and
then a round trip that usually just reports the device does not exist. An agent is
created per peer connection attempt, so on a large network that runs constantly:
on a routing peer with ~16000 peers it measured 2.40s of a 66.59s CPU profile, 3.6%,
split evenly between opening the client and the round trip.

The engine now owns a detector and passes it to every agent through the ICE config,
so the answer for an interface is reused instead of being asked again for each agent.
It is kept for a second, short enough that a WireGuard interface appearing is picked
up before ICE settles on candidates over it.

The callers that build one filter and keep it, the relay and the UDP mux, keep
passing nil and so keep probing, which costs them nothing at their rate.

* [client] Recheck the WireGuard cache inside the singleflight group

A caller that saw an expired entry could enter the singleflight group
after another caller had already refreshed the entry and left it, and
probe the interface a second time. Read the cache again inside the group
before probing.

This also makes the concurrent probe test independent of scheduling: a
late caller finds the fresh entry instead of starting a new probe.

* [client] Drop expired WireGuard detector entries

The detector lives as long as the engine and kept an entry for every
interface name it was ever asked about. On hosts that churn interfaces,
such as container veths, the map only grew. Remove expired entries when
a new answer is stored; the map holds a few dozen names at most, so the
sweep is cheap and runs at most once per interface per TTL.

* [client] Skip the disallow-list filter test on iOS

InterfaceFilter does not apply the disallow list on iOS, so the subtest
reaches the probe there and its no-probe assertion cannot hold.
2026-10-09 11:03:34 +02:00

483 lines
13 KiB
Go

//go:build privileged
package dns
import (
"context"
"fmt"
"net/netip"
"os"
"testing"
"github.com/miekg/dns"
"github.com/stretchr/testify/assert"
"go.uber.org/mock/gomock"
"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
"github.com/netbirdio/netbird/client/iface"
pfmock "github.com/netbirdio/netbird/client/iface/mocks"
"github.com/netbirdio/netbird/client/iface/wgaddr"
"github.com/netbirdio/netbird/client/internal/dns/local"
"github.com/netbirdio/netbird/client/internal/dns/test"
"github.com/netbirdio/netbird/client/internal/peer"
"github.com/netbirdio/netbird/client/internal/stdnet"
nbdns "github.com/netbirdio/netbird/dns"
)
// testIFaceBlackList mirrors the overlay prefixes profilemanager.DefaultInterfaceBlacklist
// carries. Declared here rather than imported because profilemanager imports this package.
var testIFaceBlackList = []string{"wt", "utun", "tun0"}
func TestUpdateDNSServer(t *testing.T) {
nameServers := []nbdns.NameServer{
{
IP: netip.MustParseAddr("8.8.8.8"),
NSType: nbdns.UDPNameServerType,
Port: 53,
},
{
IP: netip.MustParseAddr("8.8.4.4"),
NSType: nbdns.UDPNameServerType,
Port: 53,
},
}
testCases := []struct {
name string
initUpstreamMap []handlerWrapper
initLocalZones []nbdns.CustomZone
initSerial uint64
inputSerial uint64
inputUpdate nbdns.Config
shouldFail bool
expectedUpstreamMap []handlerWrapper
expectedLocalQs []dns.Question
}{
{
name: "Initial Config Should Succeed",
initUpstreamMap: nil,
initSerial: 0,
inputSerial: 1,
inputUpdate: nbdns.Config{
ServiceEnable: true,
CustomZones: []nbdns.CustomZone{
{
Domain: "netbird.cloud",
Records: zoneRecords,
},
},
NameServerGroups: []*nbdns.NameServerGroup{
{
Domains: []string{"netbird.io"},
NameServers: nameServers,
},
{
NameServers: nameServers,
Primary: true,
},
},
},
expectedUpstreamMap: []handlerWrapper{
{
domain: "netbird.io",
priority: PriorityUpstream,
},
{
domain: "netbird.cloud",
priority: PriorityLocal,
},
{
domain: nbdns.RootZone,
priority: PriorityDefault,
},
},
expectedLocalQs: []dns.Question{{Name: "peera.netbird.cloud.", Qtype: dns.TypeA, Qclass: dns.ClassINET}},
},
{
name: "New Config Should Succeed",
initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
initUpstreamMap: []handlerWrapper{
{
domain: "netbird.cloud",
handler: &mockHandler{},
priority: PriorityUpstream,
},
},
initSerial: 0,
inputSerial: 1,
inputUpdate: nbdns.Config{
ServiceEnable: true,
CustomZones: []nbdns.CustomZone{
{
Domain: "netbird.cloud",
Records: zoneRecords,
},
},
NameServerGroups: []*nbdns.NameServerGroup{
{
Domains: []string{"netbird.io"},
NameServers: nameServers,
},
},
},
expectedUpstreamMap: []handlerWrapper{
{
domain: "netbird.io",
priority: PriorityUpstream,
},
{
domain: "netbird.cloud",
priority: PriorityLocal,
},
},
expectedLocalQs: []dns.Question{{Name: zoneRecords[0].Name, Qtype: 1, Qclass: 1}},
},
{
name: "Smaller Config Serial Should Be Skipped",
initLocalZones: []nbdns.CustomZone{},
initUpstreamMap: nil,
initSerial: 2,
inputSerial: 1,
shouldFail: true,
},
{
name: "Empty NS Group Domain Or Not Primary Element Should Fail",
initLocalZones: []nbdns.CustomZone{},
initUpstreamMap: nil,
initSerial: 0,
inputSerial: 1,
inputUpdate: nbdns.Config{
ServiceEnable: true,
CustomZones: []nbdns.CustomZone{
{
Domain: "netbird.cloud",
Records: zoneRecords,
},
},
NameServerGroups: []*nbdns.NameServerGroup{
{
NameServers: nameServers,
},
},
},
shouldFail: true,
},
{
name: "Invalid NS Group Nameservers list Should Fail",
initLocalZones: []nbdns.CustomZone{},
initUpstreamMap: nil,
initSerial: 0,
inputSerial: 1,
inputUpdate: nbdns.Config{
ServiceEnable: true,
CustomZones: []nbdns.CustomZone{
{
Domain: "netbird.cloud",
Records: zoneRecords,
},
},
NameServerGroups: []*nbdns.NameServerGroup{
{
NameServers: nameServers,
},
},
},
shouldFail: true,
},
{
name: "Invalid Custom Zone Records list Should Skip",
initLocalZones: []nbdns.CustomZone{},
initUpstreamMap: nil,
initSerial: 0,
inputSerial: 1,
inputUpdate: nbdns.Config{
ServiceEnable: true,
CustomZones: []nbdns.CustomZone{
{
Domain: "netbird.cloud",
},
},
NameServerGroups: []*nbdns.NameServerGroup{
{
NameServers: nameServers,
Primary: true,
},
},
},
expectedUpstreamMap: []handlerWrapper{{
domain: ".",
priority: PriorityDefault,
}},
},
{
name: "Empty Config Should Succeed and Clean Maps",
initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
initUpstreamMap: []handlerWrapper{
{
domain: zoneRecords[0].Name,
handler: &mockHandler{},
priority: PriorityUpstream,
},
},
initSerial: 0,
inputSerial: 1,
inputUpdate: nbdns.Config{ServiceEnable: true},
expectedUpstreamMap: nil,
expectedLocalQs: []dns.Question{},
},
{
name: "Disabled Service Should clean map",
initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
initUpstreamMap: []handlerWrapper{
{
domain: zoneRecords[0].Name,
handler: &mockHandler{},
priority: PriorityUpstream,
},
},
initSerial: 0,
inputSerial: 1,
inputUpdate: nbdns.Config{ServiceEnable: false},
expectedUpstreamMap: nil,
expectedLocalQs: []dns.Question{},
},
}
for n, testCase := range testCases {
t.Run(testCase.name, func(t *testing.T) {
privKey, _ := wgtypes.GenerateKey()
newNet := stdnet.NewNet(context.Background(), testIFaceBlackList, nil)
opts := iface.WGIFaceOpts{
IFaceName: fmt.Sprintf("utun230%d", n),
Address: wgaddr.MustParseWGAddress(fmt.Sprintf("100.66.100.%d/32", n+1)),
WGPort: 33100,
WGPrivKey: privKey.String(),
MTU: iface.DefaultMTU,
TransportNet: newNet,
}
wgIface, err := iface.NewWGIFace(opts)
if err != nil {
t.Fatal(err)
}
err = wgIface.Create()
if err != nil {
t.Fatal(err)
}
defer func() {
err = wgIface.Close()
if err != nil {
t.Log(err)
}
}()
dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
WgInterface: wgIface,
CustomAddress: "",
StatusRecorder: peer.NewRecorder("mgm"),
StateManager: nil,
DisableSys: false,
})
if err != nil {
t.Fatal(err)
}
err = dnsServer.Initialize()
if err != nil {
t.Fatal(err)
}
defer func() {
err = dnsServer.hostManager.restoreHostDNS()
if err != nil {
t.Log(err)
}
}()
dnsServer.dnsMuxHandlers = testCase.initUpstreamMap
dnsServer.localResolver.Update(testCase.initLocalZones)
dnsServer.updateSerial = testCase.initSerial
err = dnsServer.UpdateDNSServer(testCase.inputSerial, testCase.inputUpdate)
if err != nil {
if testCase.shouldFail {
return
}
t.Fatalf("update dns server should not fail, got error: %v", err)
}
if len(dnsServer.dnsMuxHandlers) != len(testCase.expectedUpstreamMap) {
t.Fatalf("update upstream failed, map size is different than expected, want %d, got %d", len(testCase.expectedUpstreamMap), len(dnsServer.dnsMuxHandlers))
}
for _, expected := range testCase.expectedUpstreamMap {
found := false
for _, got := range dnsServer.dnsMuxHandlers {
if got.domain == expected.domain && got.priority == expected.priority {
found = true
break
}
}
if !found {
t.Fatalf("update upstream failed, handler for domain=%s priority=%d not found in dnsMuxHandlers: %#v", expected.domain, expected.priority, dnsServer.dnsMuxHandlers)
}
}
var responseMSG *dns.Msg
responseWriter := &test.MockResponseWriter{
WriteMsgFunc: func(m *dns.Msg) error {
responseMSG = m
return nil
},
}
for _, q := range testCase.expectedLocalQs {
dnsServer.localResolver.ServeDNS(responseWriter, &dns.Msg{
Question: []dns.Question{q},
})
}
if len(testCase.expectedLocalQs) > 0 {
assert.NotNil(t, responseMSG, "response message should not be nil")
assert.Equal(t, dns.RcodeSuccess, responseMSG.Rcode, "response code should be success")
assert.NotEmpty(t, responseMSG.Answer, "response message should have answers")
}
})
}
}
func TestDNSFakeResolverHandleUpdates(t *testing.T) {
ov := os.Getenv("NB_WG_KERNEL_DISABLED")
defer t.Setenv("NB_WG_KERNEL_DISABLED", ov)
t.Setenv("NB_WG_KERNEL_DISABLED", "true")
newNet := stdnet.NewNet(context.Background(), []string{"utun2301"}, nil)
privKey, _ := wgtypes.GeneratePrivateKey()
opts := iface.WGIFaceOpts{
IFaceName: "utun2301",
Address: wgaddr.MustParseWGAddress("100.66.100.1/32"),
WGPort: 33100,
WGPrivKey: privKey.String(),
MTU: iface.DefaultMTU,
TransportNet: newNet,
}
wgIface, err := iface.NewWGIFace(opts)
if err != nil {
t.Errorf("build interface wireguard: %v", err)
return
}
err = wgIface.Create()
if err != nil {
t.Errorf("create and init wireguard interface: %v", err)
return
}
defer func() {
if err = wgIface.Close(); err != nil {
t.Logf("close wireguard interface: %v", err)
}
}()
ctrl := gomock.NewController(t)
defer ctrl.Finish()
packetfilter := pfmock.NewMockPacketFilter(ctrl)
packetfilter.EXPECT().FilterOutbound(gomock.Any(), gomock.Any()).AnyTimes()
packetfilter.EXPECT().SetUDPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
packetfilter.EXPECT().SetTCPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
if err := wgIface.SetFilter(packetfilter); err != nil {
t.Errorf("set packet filter: %v", err)
return
}
dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
WgInterface: wgIface,
CustomAddress: "",
StatusRecorder: peer.NewRecorder("mgm"),
StateManager: nil,
DisableSys: false,
})
if err != nil {
t.Errorf("create DNS server: %v", err)
return
}
err = dnsServer.Initialize()
if err != nil {
t.Errorf("run DNS server: %v", err)
return
}
defer func() {
if err = dnsServer.hostManager.restoreHostDNS(); err != nil {
t.Logf("restore DNS settings on the host: %v", err)
return
}
}()
dnsServer.dnsMuxHandlers = []handlerWrapper{
{
domain: zoneRecords[0].Name,
handler: &local.Resolver{},
priority: PriorityUpstream,
},
}
dnsServer.localResolver.Update([]nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}})
dnsServer.updateSerial = 0
nameServers := []nbdns.NameServer{
{
IP: netip.MustParseAddr("8.8.8.8"),
NSType: nbdns.UDPNameServerType,
Port: 53,
},
{
IP: netip.MustParseAddr("8.8.4.4"),
NSType: nbdns.UDPNameServerType,
Port: 53,
},
}
update := nbdns.Config{
ServiceEnable: true,
CustomZones: []nbdns.CustomZone{
{
Domain: "netbird.cloud",
Records: zoneRecords,
},
},
NameServerGroups: []*nbdns.NameServerGroup{
{
Domains: []string{"netbird.io"},
NameServers: nameServers,
},
{
NameServers: nameServers,
Primary: true,
},
},
}
// Start the server with regular configuration
if err := dnsServer.UpdateDNSServer(1, update); err != nil {
t.Fatalf("update dns server should not fail, got error: %v", err)
return
}
update2 := update
update2.ServiceEnable = false
// Disable the server, stop the listener
if err := dnsServer.UpdateDNSServer(2, update2); err != nil {
t.Fatalf("update dns server should not fail, got error: %v", err)
return
}
update3 := update2
update3.NameServerGroups = update3.NameServerGroups[:1]
// But service still get updates and we checking that we handle
// internal state in the right way
if err := dnsServer.UpdateDNSServer(3, update3); err != nil {
t.Fatalf("update dns server should not fail, got error: %v", err)
return
}
}