Compare commits

..

3 Commits

Author SHA1 Message Date
jnfrati
9aa1bad19e client: add optional daemon JSON socket gateway
Add positive-only service flags to enable the daemon HTTP/JSON grpc-gateway socket and persist its service configuration. Support unix and tcp listener addresses while keeping the gateway disabled by default and preserving live Unix sockets during stale cleanup.

Add gateway/socket coverage and a Docker smoke test script for validating Status over TCP and Unix JSON sockets.
2026-06-25 19:32:48 +02:00
jnfrati
5bb3ab60a8 Merge branch 'main' of github.com:netbirdio/netbird into client-json-socket 2026-06-24 14:17:07 +02:00
jnfrati
a2fd1bb0a8 add json gateway for netbird daemon 2026-05-27 19:04:55 +02:00
19 changed files with 3374 additions and 432 deletions

View File

@@ -5,6 +5,7 @@ package cmd
import (
"context"
"fmt"
"net/http"
"runtime"
"strings"
"sync"
@@ -22,15 +23,21 @@ var serviceCmd = &cobra.Command{
Short: "Manage the NetBird daemon service",
}
const defaultJSONSocket = "unix:///var/run/netbird-http.sock"
var (
serviceName string
serviceEnvVars []string
serviceName string
serviceEnvVars []string
jsonSocket string
enableJSONSocket bool
)
type program struct {
ctx context.Context
cancel context.CancelFunc
serv *grpc.Server
jsonServ *http.Server
jsonServMu sync.Mutex
serverInstance *server.Server
serverInstanceMu sync.Mutex
}
@@ -46,6 +53,8 @@ func init() {
serviceCmd.PersistentFlags().BoolVar(&updateSettingsDisabled, "disable-update-settings", false, "Disables update settings feature. If enabled, the client will not be able to change or edit any settings. To persist this setting, use: netbird service install --disable-update-settings")
serviceCmd.PersistentFlags().BoolVar(&captureEnabled, "enable-capture", false, "Enables packet capture via 'netbird debug capture'. To persist, use: netbird service install --enable-capture")
serviceCmd.PersistentFlags().BoolVar(&networksDisabled, "disable-networks", false, "Disables network selection. If enabled, the client will not allow listing, selecting, or deselecting networks. To persist, use: netbird service install --disable-networks")
serviceCmd.PersistentFlags().BoolVar(&enableJSONSocket, "enable-json-socket", false, "Enables the HTTP/JSON API socket served by grpc-gateway. To persist, use: netbird service install --enable-json-socket")
serviceCmd.PersistentFlags().StringVar(&jsonSocket, "json-socket", defaultJSONSocket, "HTTP/JSON API socket address [unix|tcp]://[path|host:port]. Requires --enable-json-socket to serve. To persist, use: netbird service install --enable-json-socket --json-socket")
rootCmd.PersistentFlags().StringVarP(&serviceName, "service", "s", defaultServiceName, "Netbird system service name")
serviceEnvDesc := `Sets extra environment variables for the service. ` +

View File

@@ -5,9 +5,6 @@ package cmd
import (
"context"
"fmt"
"net"
"os"
"strings"
"time"
"github.com/kardianos/service"
@@ -22,41 +19,56 @@ import (
"github.com/netbirdio/netbird/util"
)
func validateJSONSocketFlags() error {
if serviceCmd.PersistentFlags().Changed("json-socket") && !enableJSONSocket {
return fmt.Errorf("--json-socket requires --enable-json-socket to configure the daemon JSON gateway")
}
return nil
}
func (p *program) Start(svc service.Service) error {
// Start should not block. Do the actual work async.
log.Info("starting NetBird service") //nolint
if err := validateJSONSocketFlags(); err != nil {
return err
}
// Collect static system and platform information
system.UpdateStaticInfoAsync()
// in any case, even if configuration does not exists we run daemon to serve CLI gRPC API.
p.serv = grpc.NewServer()
split := strings.Split(daemonAddr, "://")
switch split[0] {
case "unix":
// cleanup failed close
stat, err := os.Stat(split[1])
if err == nil && !stat.IsDir() {
if err := os.Remove(split[1]); err != nil {
log.Debugf("remove socket file: %v", err)
}
}
case "tcp":
default:
return fmt.Errorf("unsupported daemon address protocol: %v", split[0])
}
listen, err := net.Listen(split[0], split[1])
daemonListener, err := listenOnAddress(daemonAddr)
if err != nil {
return fmt.Errorf("listen daemon interface: %w", err)
}
go func() {
defer listen.Close()
if split[0] == "unix" {
if err := os.Chmod(split[1], 0666); err != nil {
log.Errorf("failed setting daemon permissions: %v", split[1])
var jsonListener *socketListener
if enableJSONSocket {
jsonListener, err = listenOnAddress(jsonSocket)
if err != nil {
_ = daemonListener.Close()
return fmt.Errorf("listen daemon JSON interface: %w", err)
}
} else {
removeStaleUnixSocketForAddress(jsonSocket)
}
go func() {
defer daemonListener.Close()
if jsonListener != nil {
defer jsonListener.Close()
}
if err := daemonListener.chmodUnixSocket("daemon"); err != nil {
log.Error(err)
return
}
if jsonListener != nil {
if err := jsonListener.chmodUnixSocket("daemon JSON"); err != nil {
log.Error(err)
return
}
}
@@ -71,8 +83,16 @@ func (p *program) Start(svc service.Service) error {
p.serverInstance = serverInstance
p.serverInstanceMu.Unlock()
log.Printf("started daemon server: %v", split[1])
if err := p.serv.Serve(listen); err != nil {
if jsonListener != nil {
if err := p.startJSONGateway(jsonListener, daemonAddr); err != nil {
log.Fatalf("failed to start daemon JSON server: %v", err)
}
} else {
log.Debug("daemon JSON socket disabled")
}
log.Printf("started daemon server: %v", daemonListener.address)
if err := p.serv.Serve(daemonListener.Listener); err != nil {
log.Errorf("failed to serve daemon requests: %v", err)
}
}()
@@ -92,6 +112,20 @@ func (p *program) Stop(srv service.Service) error {
p.cancel()
p.jsonServMu.Lock()
jsonServ := p.jsonServ
p.jsonServMu.Unlock()
if jsonServ != nil {
shutdownCtx, shutdownCancel := context.WithTimeout(context.Background(), 2*time.Second)
if err := jsonServ.Shutdown(shutdownCtx); err != nil {
log.Errorf("failed to stop daemon JSON server gracefully: %v", err)
if err := jsonServ.Close(); err != nil {
log.Errorf("failed to close daemon JSON server: %v", err)
}
}
shutdownCancel()
}
if p.serv != nil {
p.serv.Stop()
}
@@ -148,6 +182,9 @@ var runCmd = &cobra.Command{
if err != nil {
return err
}
if err := validateJSONSocketFlags(); err != nil {
return err
}
return s.Run()
},
@@ -162,6 +199,9 @@ var startCmd = &cobra.Command{
if err != nil {
return err
}
if err := validateJSONSocketFlags(); err != nil {
return err
}
if err := s.Start(); err != nil {
return fmt.Errorf("start service: %w", err)
@@ -198,6 +238,9 @@ var restartCmd = &cobra.Command{
if err != nil {
return err
}
if err := validateJSONSocketFlags(); err != nil {
return err
}
if err := s.Restart(); err != nil {
return fmt.Errorf("restart service: %w", err)

View File

@@ -67,6 +67,10 @@ func buildServiceArguments() []string {
args = append(args, "--disable-networks")
}
if enableJSONSocket {
args = append(args, "--enable-json-socket", "--json-socket", jsonSocket)
}
return args
}
@@ -106,6 +110,10 @@ func configurePlatformSpecificSettings(svcConfig *service.Config) error {
// Create fully configured service config for install/reconfigure
func createServiceConfigForInstall() (*service.Config, error) {
if err := validateJSONSocketFlags(); err != nil {
return nil, err
}
svcConfig, err := newSVCConfig()
if err != nil {
return nil, fmt.Errorf("create service config: %w", err)

View File

@@ -0,0 +1,52 @@
//go:build !ios && !android
package cmd
import (
"context"
"errors"
"net"
"net/http"
"strings"
"time"
"github.com/grpc-ecosystem/grpc-gateway/v2/runtime"
log "github.com/sirupsen/logrus"
"google.golang.org/grpc"
"google.golang.org/grpc/credentials/insecure"
"github.com/netbirdio/netbird/client/proto"
)
func grpcGatewayEndpoint(addr string) string {
return strings.TrimPrefix(addr, "tcp://")
}
func (p *program) startJSONGateway(jsonListener *socketListener, daemonEndpoint string) error {
mux := runtime.NewServeMux()
opts := []grpc.DialOption{grpc.WithTransportCredentials(insecure.NewCredentials())}
if err := proto.RegisterDaemonServiceHandlerFromEndpoint(p.ctx, mux, grpcGatewayEndpoint(daemonEndpoint), opts); err != nil {
return err
}
jsonServer := &http.Server{
Handler: mux,
ReadHeaderTimeout: 5 * time.Second,
BaseContext: func(net.Listener) context.Context {
return p.ctx
},
}
p.jsonServMu.Lock()
p.jsonServ = jsonServer
p.jsonServMu.Unlock()
go func() {
log.Printf("started daemon JSON server: %v", jsonListener.address)
if err := jsonServer.Serve(jsonListener.Listener); err != nil && !errors.Is(err, http.ErrServerClosed) {
log.Errorf("failed to serve daemon JSON requests: %v", err)
}
}()
return nil
}

View File

@@ -0,0 +1,176 @@
//go:build !ios && !android
package cmd
import (
"net"
"os"
"path/filepath"
"runtime"
"testing"
"github.com/spf13/cobra"
"github.com/spf13/pflag"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func preserveJSONSocketTestState(t *testing.T) {
t.Helper()
origJSONSocket := jsonSocket
origEnableJSONSocket := enableJSONSocket
origChanged := map[string]bool{}
serviceCmd.PersistentFlags().VisitAll(func(flag *pflag.Flag) {
origChanged[flag.Name] = flag.Changed
})
t.Cleanup(func() {
jsonSocket = origJSONSocket
enableJSONSocket = origEnableJSONSocket
serviceCmd.PersistentFlags().VisitAll(func(flag *pflag.Flag) {
flag.Changed = origChanged[flag.Name]
})
})
}
func TestJSONSocketFlagsArePositiveEnableOnly(t *testing.T) {
assert.NotNil(t, serviceCmd.PersistentFlags().Lookup("enable-json-socket"))
assert.NotNil(t, serviceCmd.PersistentFlags().Lookup("json-socket"))
assert.Nil(t, serviceCmd.PersistentFlags().Lookup("disable-json-socket"))
assert.Equal(t, "false", serviceCmd.PersistentFlags().Lookup("enable-json-socket").DefValue)
}
func TestBuildServiceArgumentsDefaultDisablesJSONSocket(t *testing.T) {
preserveJSONSocketTestState(t)
enableJSONSocket = false
jsonSocket = "tcp://127.0.0.1:8080"
args := buildServiceArguments()
assert.NotContains(t, args, "--enable-json-socket")
assert.NotContains(t, args, "--json-socket")
}
func TestBuildServiceArgumentsIncludesJSONSocketWhenEnabled(t *testing.T) {
preserveJSONSocketTestState(t)
enableJSONSocket = true
jsonSocket = "tcp://127.0.0.1:8080"
args := buildServiceArguments()
enableIndex := indexOfArg(args, "--enable-json-socket")
jsonIndex := indexOfArg(args, "--json-socket")
require.NotEqual(t, -1, enableIndex)
require.NotEqual(t, -1, jsonIndex)
require.Less(t, enableIndex, jsonIndex)
require.Less(t, jsonIndex+1, len(args))
assert.Equal(t, "tcp://127.0.0.1:8080", args[jsonIndex+1])
}
func TestJSONSocketWithoutEnableValidation(t *testing.T) {
preserveJSONSocketTestState(t)
enableJSONSocket = false
require.NoError(t, serviceCmd.PersistentFlags().Set("json-socket", "tcp://127.0.0.1:8080"))
err := validateJSONSocketFlags()
require.Error(t, err)
assert.Contains(t, err.Error(), "--enable-json-socket")
}
func TestJSONSocketWithEnableValidation(t *testing.T) {
preserveJSONSocketTestState(t)
require.NoError(t, serviceCmd.PersistentFlags().Set("enable-json-socket", "true"))
require.NoError(t, serviceCmd.PersistentFlags().Set("json-socket", "tcp://127.0.0.1:8080"))
assert.NoError(t, validateJSONSocketFlags())
}
func TestJSONSocketServiceParamsPersistEnableAndAddress(t *testing.T) {
preserveJSONSocketTestState(t)
serviceCmd.PersistentFlags().VisitAll(func(flag *pflag.Flag) {
flag.Changed = false
})
enableJSONSocket = true
jsonSocket = "tcp://127.0.0.1:8080"
params := currentServiceParams()
require.True(t, params.EnableJSONSocket)
require.Equal(t, "tcp://127.0.0.1:8080", params.JSONSocket)
enableJSONSocket = false
jsonSocket = defaultJSONSocket
applyServiceParams(testServiceEnvCommand(), params)
assert.True(t, enableJSONSocket)
assert.Equal(t, "tcp://127.0.0.1:8080", jsonSocket)
}
func TestRemoveStaleUnixSocketDoesNotRemoveRegularFile(t *testing.T) {
path := filepath.Join(t.TempDir(), "netbird-http.sock")
require.NoError(t, os.WriteFile(path, []byte("not a socket"), 0600))
removeStaleUnixSocket(path)
data, err := os.ReadFile(path)
require.NoError(t, err)
assert.Equal(t, []byte("not a socket"), data)
}
func TestRemoveStaleUnixSocketRemovesSocket(t *testing.T) {
if runtime.GOOS == "windows" {
t.Skip("unix sockets are not available on Windows")
}
path := filepath.Join(t.TempDir(), "netbird-http.sock")
addr := &net.UnixAddr{Name: path, Net: "unix"}
listener, err := net.ListenUnix("unix", addr)
require.NoError(t, err)
listener.SetUnlinkOnClose(false)
require.NoError(t, listener.Close())
_, err = os.Lstat(path)
require.NoError(t, err, "test setup must leave a stale Unix socket path")
removeStaleUnixSocket(path)
_, err = os.Lstat(path)
assert.True(t, os.IsNotExist(err), "expected stale Unix socket to be removed, got %v", err)
}
func TestRemoveStaleUnixSocketDoesNotRemoveLiveSocket(t *testing.T) {
if runtime.GOOS == "windows" {
t.Skip("unix sockets are not available on Windows")
}
path := filepath.Join(t.TempDir(), "netbird-http.sock")
listener, err := net.Listen("unix", path)
require.NoError(t, err)
defer listener.Close()
removeStaleUnixSocket(path)
_, err = os.Lstat(path)
assert.NoError(t, err, "expected live Unix socket to be preserved")
}
func testServiceEnvCommand() *cobra.Command {
cmd := &cobra.Command{}
cmd.Flags().StringSlice("service-env", nil, "")
return cmd
}
func indexOfArg(args []string, arg string) int {
for i, candidate := range args {
if candidate == arg {
return i
}
}
return -1
}

View File

@@ -23,6 +23,7 @@ const serviceParamsFile = "service.json"
type serviceParams struct {
LogLevel string `json:"log_level"`
DaemonAddr string `json:"daemon_addr"`
JSONSocket string `json:"json_socket"`
ManagementURL string `json:"management_url,omitempty"`
ConfigPath string `json:"config_path,omitempty"`
LogFiles []string `json:"log_files,omitempty"`
@@ -30,6 +31,7 @@ type serviceParams struct {
DisableUpdateSettings bool `json:"disable_update_settings,omitempty"`
EnableCapture bool `json:"enable_capture,omitempty"`
DisableNetworks bool `json:"disable_networks,omitempty"`
EnableJSONSocket bool `json:"enable_json_socket,omitempty"`
ServiceEnvVars map[string]string `json:"service_env_vars,omitempty"`
}
@@ -75,6 +77,7 @@ func currentServiceParams() *serviceParams {
params := &serviceParams{
LogLevel: logLevel,
DaemonAddr: daemonAddr,
JSONSocket: jsonSocket,
ManagementURL: managementURL,
ConfigPath: configPath,
LogFiles: logFiles,
@@ -82,6 +85,7 @@ func currentServiceParams() *serviceParams {
DisableUpdateSettings: updateSettingsDisabled,
EnableCapture: captureEnabled,
DisableNetworks: networksDisabled,
EnableJSONSocket: enableJSONSocket,
}
if len(serviceEnvVars) > 0 {
@@ -113,9 +117,8 @@ func applyServiceParams(cmd *cobra.Command, params *serviceParams) {
return
}
// For fields with non-empty defaults (log-level, daemon-addr), keep the
// != "" guard so that an older service.json missing the field doesn't
// clobber the default with an empty string.
// For fields with non-empty defaults, keep the != "" guard so that an older
// service.json missing the field doesn't clobber the default with an empty string.
if !rootCmd.PersistentFlags().Changed("log-level") && params.LogLevel != "" {
logLevel = params.LogLevel
}
@@ -124,6 +127,14 @@ func applyServiceParams(cmd *cobra.Command, params *serviceParams) {
daemonAddr = params.DaemonAddr
}
if !serviceCmd.PersistentFlags().Changed("json-socket") && params.JSONSocket != "" {
jsonSocket = params.JSONSocket
}
if !serviceCmd.PersistentFlags().Changed("enable-json-socket") {
enableJSONSocket = params.EnableJSONSocket
}
// For optional fields where empty means "use default", always apply so
// that an explicit clear (--management-url "") persists across reinstalls.
if !rootCmd.PersistentFlags().Changed("management-url") {

View File

@@ -41,6 +41,8 @@ func TestSaveAndLoadServiceParams(t *testing.T) {
params := &serviceParams{
LogLevel: "debug",
DaemonAddr: "unix:///var/run/netbird.sock",
JSONSocket: "tcp://127.0.0.1:8080",
EnableJSONSocket: true,
ManagementURL: "https://my.server.com",
ConfigPath: "/etc/netbird/config.json",
LogFiles: []string{"/var/log/netbird/client.log", "console"},
@@ -63,6 +65,8 @@ func TestSaveAndLoadServiceParams(t *testing.T) {
assert.Equal(t, params.LogLevel, loaded.LogLevel)
assert.Equal(t, params.DaemonAddr, loaded.DaemonAddr)
assert.Equal(t, params.JSONSocket, loaded.JSONSocket)
assert.Equal(t, params.EnableJSONSocket, loaded.EnableJSONSocket)
assert.Equal(t, params.ManagementURL, loaded.ManagementURL)
assert.Equal(t, params.ConfigPath, loaded.ConfigPath)
assert.Equal(t, params.LogFiles, loaded.LogFiles)
@@ -101,6 +105,8 @@ func TestLoadServiceParams_InvalidJSON(t *testing.T) {
func TestCurrentServiceParams(t *testing.T) {
origLogLevel := logLevel
origDaemonAddr := daemonAddr
origJSONSocket := jsonSocket
origEnableJSONSocket := enableJSONSocket
origManagementURL := managementURL
origConfigPath := configPath
origLogFiles := logFiles
@@ -110,6 +116,8 @@ func TestCurrentServiceParams(t *testing.T) {
t.Cleanup(func() {
logLevel = origLogLevel
daemonAddr = origDaemonAddr
jsonSocket = origJSONSocket
enableJSONSocket = origEnableJSONSocket
managementURL = origManagementURL
configPath = origConfigPath
logFiles = origLogFiles
@@ -120,6 +128,8 @@ func TestCurrentServiceParams(t *testing.T) {
logLevel = "trace"
daemonAddr = "tcp://127.0.0.1:9999"
jsonSocket = "tcp://127.0.0.1:8080"
enableJSONSocket = true
managementURL = "https://mgmt.example.com"
configPath = "/tmp/test-config.json"
logFiles = []string{"/tmp/test.log"}
@@ -131,6 +141,8 @@ func TestCurrentServiceParams(t *testing.T) {
assert.Equal(t, "trace", params.LogLevel)
assert.Equal(t, "tcp://127.0.0.1:9999", params.DaemonAddr)
assert.Equal(t, "tcp://127.0.0.1:8080", params.JSONSocket)
assert.True(t, params.EnableJSONSocket)
assert.Equal(t, "https://mgmt.example.com", params.ManagementURL)
assert.Equal(t, "/tmp/test-config.json", params.ConfigPath)
assert.Equal(t, []string{"/tmp/test.log"}, params.LogFiles)
@@ -142,6 +154,8 @@ func TestCurrentServiceParams(t *testing.T) {
func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
origLogLevel := logLevel
origDaemonAddr := daemonAddr
origJSONSocket := jsonSocket
origEnableJSONSocket := enableJSONSocket
origManagementURL := managementURL
origConfigPath := configPath
origLogFiles := logFiles
@@ -151,6 +165,8 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
t.Cleanup(func() {
logLevel = origLogLevel
daemonAddr = origDaemonAddr
jsonSocket = origJSONSocket
enableJSONSocket = origEnableJSONSocket
managementURL = origManagementURL
configPath = origConfigPath
logFiles = origLogFiles
@@ -162,6 +178,8 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
// Reset all flags to defaults.
logLevel = "info"
daemonAddr = "unix:///var/run/netbird.sock"
jsonSocket = defaultJSONSocket
enableJSONSocket = false
managementURL = ""
configPath = "/etc/netbird/config.json"
logFiles = []string{"/var/log/netbird/client.log"}
@@ -184,6 +202,8 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
saved := &serviceParams{
LogLevel: "debug",
DaemonAddr: "tcp://127.0.0.1:5555",
JSONSocket: "tcp://127.0.0.1:8080",
EnableJSONSocket: true,
ManagementURL: "https://saved.example.com",
ConfigPath: "/saved/config.json",
LogFiles: []string{"/saved/client.log"},
@@ -201,6 +221,8 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
// All other fields were not Changed, so they should use saved values.
assert.Equal(t, "tcp://127.0.0.1:5555", daemonAddr)
assert.Equal(t, "tcp://127.0.0.1:8080", jsonSocket)
assert.True(t, enableJSONSocket)
assert.Equal(t, "https://saved.example.com", managementURL)
assert.Equal(t, "/saved/config.json", configPath)
assert.Equal(t, []string{"/saved/client.log"}, logFiles)
@@ -212,14 +234,17 @@ func TestApplyServiceParams_OnlyUnchangedFlags(t *testing.T) {
func TestApplyServiceParams_BooleanRevertToFalse(t *testing.T) {
origProfilesDisabled := profilesDisabled
origUpdateSettingsDisabled := updateSettingsDisabled
origEnableJSONSocket := enableJSONSocket
t.Cleanup(func() {
profilesDisabled = origProfilesDisabled
updateSettingsDisabled = origUpdateSettingsDisabled
enableJSONSocket = origEnableJSONSocket
})
// Simulate current state where booleans are true (e.g. set by previous install).
profilesDisabled = true
updateSettingsDisabled = true
enableJSONSocket = true
// Reset Changed state so flags appear unset.
serviceCmd.PersistentFlags().VisitAll(func(f *pflag.Flag) {
@@ -238,6 +263,7 @@ func TestApplyServiceParams_BooleanRevertToFalse(t *testing.T) {
assert.False(t, profilesDisabled, "saved false should override current true")
assert.False(t, updateSettingsDisabled, "saved false should override current true")
assert.False(t, enableJSONSocket, "saved false should override current true")
}
func TestApplyServiceParams_ClearManagementURL(t *testing.T) {
@@ -530,6 +556,7 @@ func fieldToGlobalVar(field string) string {
m := map[string]string{
"LogLevel": "logLevel",
"DaemonAddr": "daemonAddr",
"JSONSocket": "jsonSocket",
"ManagementURL": "managementURL",
"ConfigPath": "configPath",
"LogFiles": "logFiles",
@@ -537,6 +564,7 @@ func fieldToGlobalVar(field string) string {
"DisableUpdateSettings": "updateSettingsDisabled",
"EnableCapture": "captureEnabled",
"DisableNetworks": "networksDisabled",
"EnableJSONSocket": "enableJSONSocket",
"ServiceEnvVars": "serviceEnvVars",
}
if v, ok := m[field]; ok {

View File

@@ -0,0 +1,111 @@
//go:build !ios && !android
package cmd
import (
"errors"
"fmt"
"net"
"os"
"strings"
"syscall"
"time"
log "github.com/sirupsen/logrus"
)
type socketListener struct {
net.Listener
network string
address string
}
func listenOnAddress(addr string) (*socketListener, error) {
network, address, err := parseListenAddress(addr)
if err != nil {
return nil, err
}
if network == "unix" {
removeStaleUnixSocket(address)
}
listener, err := net.Listen(network, address)
if err != nil {
return nil, err
}
return &socketListener{Listener: listener, network: network, address: address}, nil
}
func parseListenAddress(addr string) (string, string, error) {
network, address, ok := strings.Cut(addr, "://")
if !ok || network == "" || address == "" {
return "", "", fmt.Errorf("address must be in [unix|tcp]://[path|host:port] format: %q", addr)
}
switch network {
case "unix", "tcp":
return network, address, nil
default:
return "", "", fmt.Errorf("unsupported daemon address protocol: %v", network)
}
}
func removeStaleUnixSocket(path string) {
stat, err := os.Lstat(path)
if err != nil {
if !os.IsNotExist(err) {
log.Debugf("stat socket file: %v", err)
}
return
}
if stat.Mode()&os.ModeSocket == 0 {
return
}
if !isStaleUnixSocket(path) {
return
}
if err := os.Remove(path); err != nil {
log.Debugf("remove socket file: %v", err)
}
}
func isStaleUnixSocket(path string) bool {
conn, err := net.DialTimeout("unix", path, 100*time.Millisecond)
if err == nil {
if closeErr := conn.Close(); closeErr != nil {
log.Debugf("close unix socket probe: %v", closeErr)
}
return false
}
if os.IsNotExist(err) || os.IsPermission(err) || os.IsTimeout(err) {
log.Debugf("not removing unix socket %s after probe error: %v", path, err)
return false
}
return errors.Is(err, syscall.ECONNREFUSED)
}
func removeStaleUnixSocketForAddress(addr string) {
network, address, err := parseListenAddress(addr)
if err != nil || network != "unix" {
return
}
removeStaleUnixSocket(address)
}
func (l *socketListener) chmodUnixSocket(description string) error {
if l == nil || l.network != "unix" {
return nil
}
if err := os.Chmod(l.address, 0666); err != nil {
return fmt.Errorf("failed setting %s permissions for %s: %w", description, l.address, err)
}
return nil
}

View File

@@ -895,16 +895,6 @@ func (e *Engine) handleAutoUpdateVersion(autoUpdateSettings *mgmProto.AutoUpdate
e.updateManager.SetVersion(autoUpdateSettings.Version, autoUpdateSettings.AlwaysUpdate)
}
// phase times a sync sub-phase: it returns a function that records the elapsed
// duration when called. Starting the timer at the call site keeps inter-phase
// glue code out of the measurement.
func (e *Engine) phase(name string) func() {
start := time.Now()
return func() {
e.clientMetrics.RecordSyncPhase(e.ctx, name, time.Since(start))
}
}
func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
started := time.Now()
defer func() {
@@ -924,10 +914,7 @@ func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
e.handleAutoUpdateVersion(update.NetworkMap.PeerConfig.AutoUpdate)
}
done := e.phase("netbird_config")
err := e.updateNetbirdConfig(update.GetNetbirdConfig())
done()
if err != nil {
if err := e.updateNetbirdConfig(update.GetNetbirdConfig()); err != nil {
return err
}
@@ -941,16 +928,11 @@ func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
return nil
}
done = e.phase("checks")
err = e.updateChecksIfNew(update.Checks)
done()
if err != nil {
if err := e.updateChecksIfNew(update.Checks); err != nil {
return err
}
done = e.phase("persist")
e.persistSyncResponse(update)
done()
// only apply new changes and ignore old ones
if err := e.updateNetworkMap(nm); err != nil {
@@ -1375,16 +1357,13 @@ func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
dnsConfig := toDNSConfig(protoDNSConfig, e.wgInterface.Address())
done := e.phase("dns_server")
if err := e.dnsServer.UpdateDNSServer(serial, dnsConfig); err != nil {
log.Errorf("failed to update dns server, err: %v", err)
}
done()
e.routeManager.SetDNSForwarderPort(dnsConfig.ForwarderPort)
// apply routes first, route related actions might depend on routing being enabled
done = e.phase("routes_classify")
routes := toRoutes(networkMap.GetRoutes())
serverRoutes, clientRoutes := e.routeManager.ClassifyRoutes(routes)
@@ -1393,60 +1372,29 @@ func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
e.connMgr.UpdateRouteHAMap(clientRoutes)
log.Debugf("updated lazy connection manager with %d HA groups", len(clientRoutes))
}
done()
done = e.phase("routes_apply")
dnsRouteFeatureFlag := toDNSFeatureFlag(networkMap)
if err := e.routeManager.UpdateRoutes(serial, serverRoutes, clientRoutes, dnsRouteFeatureFlag); err != nil {
log.Errorf("failed to update routes: %v", err)
}
done()
done = e.phase("filtering")
if e.acl != nil {
e.acl.ApplyFiltering(networkMap, dnsRouteFeatureFlag)
}
done()
done = e.phase("dns_forwarder")
fwdEntries := toRouteDomains(e.config.WgPrivateKey.PublicKey().String(), routes)
e.updateDNSForwarder(dnsRouteFeatureFlag, fwdEntries)
done()
// Ingress forward rules
done = e.phase("forward_rules")
forwardingRules, err := e.updateForwardRules(networkMap.GetForwardingRules())
if err != nil {
log.Errorf("failed to update forward rules, err: %v", err)
}
done()
log.Debugf("got peers update from Management Service, total peers to connect to = %d", len(networkMap.GetRemotePeers()))
done = e.phase("offline_peers")
e.updateOfflinePeers(networkMap.GetOfflinePeers())
done()
remotePeers, err := e.reconcilePeers(networkMap)
if err != nil {
return err
}
// must set the exclude list after the peers are added. Without it the manager can not figure out the peers parameters from the store
done = e.phase("lazy_exclude")
excludedLazyPeers := e.toExcludedLazyPeers(forwardingRules, remotePeers)
e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
done()
e.networkSerial = serial
return nil
}
// reconcilePeers applies the remote peer list from the network map (removing,
// modifying and adding peers, then updating SSH config) and returns the remote
// peers with our own peer filtered out, for use by later sync steps.
func (e *Engine) reconcilePeers(networkMap *mgmProto.NetworkMap) ([]*mgmProto.RemotePeerConfig, error) {
// Filter out own peer from the remote peers list
localPubKey := e.config.WgPrivateKey.PublicKey().String()
remotePeers := make([]*mgmProto.RemotePeerConfig, 0, len(networkMap.GetRemotePeers()))
@@ -1461,43 +1409,42 @@ func (e *Engine) reconcilePeers(networkMap *mgmProto.NetworkMap) ([]*mgmProto.Re
err := e.removeAllPeers()
e.statusRecorder.FinishPeerListModifications()
if err != nil {
return nil, err
return err
}
return remotePeers, nil
} else {
err := e.removePeers(remotePeers)
if err != nil {
return err
}
err = e.modifyPeers(remotePeers)
if err != nil {
return err
}
err = e.addNewPeers(remotePeers)
if err != nil {
return err
}
e.statusRecorder.FinishPeerListModifications()
e.updatePeerSSHHostKeys(remotePeers)
if err := e.updateSSHClientConfig(remotePeers); err != nil {
log.Warnf("failed to update SSH client config: %v", err)
}
e.updateSSHServerAuth(networkMap.GetSshAuth())
}
done := e.phase("removed_peers")
err := e.removePeers(remotePeers)
done()
if err != nil {
return nil, err
}
// must set the exclude list after the peers are added. Without it the manager can not figure out the peers parameters from the store
excludedLazyPeers := e.toExcludedLazyPeers(forwardingRules, remotePeers)
e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
done = e.phase("modified_peers")
err = e.modifyPeers(remotePeers)
done()
if err != nil {
return nil, err
}
e.networkSerial = serial
done = e.phase("added_peers")
err = e.addNewPeers(remotePeers)
done()
if err != nil {
return nil, err
}
e.statusRecorder.FinishPeerListModifications()
e.updatePeerSSHHostKeys(remotePeers)
if err := e.updateSSHClientConfig(remotePeers); err != nil {
log.Warnf("failed to update SSH client config: %v", err)
}
e.updateSSHServerAuth(networkMap.GetSshAuth())
return remotePeers, nil
return nil
}
func toDNSFeatureFlag(networkMap *mgmProto.NetworkMap) bool {

View File

@@ -120,30 +120,6 @@ func (m *influxDBMetrics) RecordSyncDuration(_ context.Context, agentInfo AgentI
m.trimLocked()
}
func (m *influxDBMetrics) RecordSyncPhase(_ context.Context, agentInfo AgentInfo, phase string, duration time.Duration) {
tags := fmt.Sprintf("deployment_type=%s,version=%s,os=%s,arch=%s,peer_id=%s,phase=%s",
agentInfo.DeploymentType.String(),
agentInfo.Version,
agentInfo.OS,
agentInfo.Arch,
agentInfo.peerID,
phase,
)
m.mu.Lock()
defer m.mu.Unlock()
m.samples = append(m.samples, influxSample{
measurement: "netbird_sync_phase",
tags: tags,
fields: map[string]float64{
"duration_seconds": duration.Seconds(),
},
timestamp: time.Now(),
})
m.trimLocked()
}
func (m *influxDBMetrics) RecordLoginDuration(_ context.Context, agentInfo AgentInfo, duration time.Duration, success bool) {
result := "success"
if !success {

View File

@@ -1,259 +0,0 @@
{
"annotations": {
"list": []
},
"editable": true,
"fiscalYearStartMonth": 0,
"graphTooltip": 1,
"links": [],
"refresh": "",
"schemaVersion": 39,
"tags": [
"netbird",
"sync"
],
"templating": {
"list": [
{
"current": {
"text": "All",
"value": "$__all"
},
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"definition": "import \"influxdata/influxdb/schema\"\nschema.tagValues(bucket: \"metrics\", tag: \"version\")",
"includeAll": true,
"label": "version",
"multi": true,
"name": "version",
"query": "import \"influxdata/influxdb/schema\"\nschema.tagValues(bucket: \"metrics\", tag: \"version\")",
"refresh": 2,
"type": "query",
"allValue": ".*"
}
]
},
"time": {
"from": "now-2d",
"to": "now"
},
"timepicker": {},
"timezone": "",
"title": "NetBird Sync Phases (where time goes)",
"uid": "netbird-sync-phases",
"version": 1,
"panels": [
{
"id": 1,
"title": "Time per phase over time (stacked, ms)",
"type": "timeseries",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"gridPos": {
"h": 10,
"w": 24,
"x": 0,
"y": 0
},
"fieldConfig": {
"defaults": {
"unit": "ms",
"custom": {
"drawStyle": "bars",
"stacking": {
"mode": "normal",
"group": "A"
},
"fillOpacity": 80,
"lineWidth": 0
}
},
"overrides": []
},
"options": {
"legend": {
"displayMode": "table",
"placement": "right",
"calcs": [
"max",
"mean"
]
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"refId": "A",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"query": "from(bucket: \"metrics\")\n |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n |> filter(fn: (r) => r._measurement == \"netbird_sync_phase\" and r._field == \"duration_seconds\")\n |> filter(fn: (r) => r.version =~ /${version:regex}/)\n |> map(fn: (r) => ({ r with _value: r._value * 1000.0 }))\n |> keep(columns: [\"_time\", \"_value\", \"phase\"])\n |> group(columns: [\"phase\"])"
}
]
},
{
"id": 2,
"title": "p95 per phase (ms)",
"type": "bargauge",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"gridPos": {
"h": 11,
"w": 12,
"x": 0,
"y": 10
},
"fieldConfig": {
"defaults": {
"unit": "ms",
"color": {
"mode": "continuous-GrYlRd"
}
},
"overrides": []
},
"options": {
"displayMode": "gradient",
"orientation": "horizontal",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"showUnfilled": true
},
"targets": [
{
"refId": "A",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"query": "from(bucket: \"metrics\")\n |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n |> filter(fn: (r) => r._measurement == \"netbird_sync_phase\" and r._field == \"duration_seconds\")\n |> filter(fn: (r) => r.version =~ /${version:regex}/)\n |> map(fn: (r) => ({ r with _value: r._value * 1000.0 }))\n |> group(columns: [\"phase\"])\n |> quantile(q: 0.95)\n |> group()\n |> sort(columns: [\"_value\"], desc: true)"
}
]
},
{
"id": 3,
"title": "Per-phase stats (ms): mean / p95 / max",
"type": "table",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"gridPos": {
"h": 11,
"w": 12,
"x": 12,
"y": 10
},
"fieldConfig": {
"defaults": {
"unit": "ms"
},
"overrides": []
},
"options": {
"showHeader": true,
"sortBy": [
{
"displayName": "max",
"desc": true
}
]
},
"transformations": [
{
"id": "merge",
"options": {}
}
],
"targets": [
{
"refId": "mean",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"query": "from(bucket: \"metrics\")\n |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n |> filter(fn: (r) => r._measurement == \"netbird_sync_phase\" and r._field == \"duration_seconds\")\n |> filter(fn: (r) => r.version =~ /${version:regex}/)\n |> map(fn: (r) => ({ r with _value: r._value * 1000.0 }))\n |> group(columns: [\"phase\"])\n |> mean()\n |> group()\n |> keep(columns: [\"phase\", \"_value\"])\n |> rename(columns: {_value: \"mean\"})"
},
{
"refId": "p95",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"query": "from(bucket: \"metrics\")\n |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n |> filter(fn: (r) => r._measurement == \"netbird_sync_phase\" and r._field == \"duration_seconds\")\n |> filter(fn: (r) => r.version =~ /${version:regex}/)\n |> map(fn: (r) => ({ r with _value: r._value * 1000.0 }))\n |> group(columns: [\"phase\"])\n |> quantile(q: 0.95)\n |> group()\n |> keep(columns: [\"phase\", \"_value\"])\n |> rename(columns: {_value: \"p95\"})"
},
{
"refId": "max",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"query": "from(bucket: \"metrics\")\n |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n |> filter(fn: (r) => r._measurement == \"netbird_sync_phase\" and r._field == \"duration_seconds\")\n |> filter(fn: (r) => r.version =~ /${version:regex}/)\n |> map(fn: (r) => ({ r with _value: r._value * 1000.0 }))\n |> group(columns: [\"phase\"])\n |> max()\n |> group()\n |> keep(columns: [\"phase\", \"_value\"])\n |> rename(columns: {_value: \"max\"})"
}
]
},
{
"id": 4,
"title": "Total sync duration (netbird_sync, ms) \u2014 reference",
"type": "timeseries",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"gridPos": {
"h": 8,
"w": 24,
"x": 0,
"y": 21
},
"fieldConfig": {
"defaults": {
"unit": "ms",
"custom": {
"drawStyle": "points",
"pointSize": 5
}
},
"overrides": []
},
"options": {
"legend": {
"displayMode": "table",
"placement": "right",
"calcs": [
"max",
"mean"
]
},
"tooltip": {
"mode": "single"
}
},
"targets": [
{
"refId": "A",
"datasource": {
"type": "influxdb",
"uid": "influxdb"
},
"query": "from(bucket: \"metrics\")\n |> range(start: v.timeRangeStart, stop: v.timeRangeStop)\n |> filter(fn: (r) => r._measurement == \"netbird_sync\" and r._field == \"duration_seconds\")\n |> filter(fn: (r) => r.version =~ /${version:regex}/)\n |> map(fn: (r) => ({ r with _value: r._value * 1000.0 }))\n |> keep(columns: [\"_time\", \"_value\", \"version\"])\n |> group(columns: [\"version\"])"
}
]
}
]
}

View File

@@ -59,19 +59,6 @@ var allowedMeasurements = map[string]measurementSpec{
"peer_id": true,
},
},
"netbird_sync_phase": {
allowedFields: map[string]bool{
"duration_seconds": true,
},
allowedTags: map[string]bool{
"deployment_type": true,
"version": true,
"os": true,
"arch": true,
"peer_id": true,
"phase": true,
},
},
"netbird_login": {
allowedFields: map[string]bool{
"duration_seconds": true,

View File

@@ -56,9 +56,6 @@ type metricsImplementation interface {
// RecordSyncDuration records how long it took to process a sync message
RecordSyncDuration(ctx context.Context, agentInfo AgentInfo, duration time.Duration)
// RecordSyncPhase records how long a single sub-phase of sync processing took
RecordSyncPhase(ctx context.Context, agentInfo AgentInfo, phase string, duration time.Duration)
// RecordLoginDuration records how long the login to management took
RecordLoginDuration(ctx context.Context, agentInfo AgentInfo, duration time.Duration, success bool)
@@ -130,18 +127,6 @@ func (c *ClientMetrics) RecordSyncDuration(ctx context.Context, duration time.Du
c.impl.RecordSyncDuration(ctx, agentInfo, duration)
}
// RecordSyncPhase records the duration of a single sub-phase of sync processing
func (c *ClientMetrics) RecordSyncPhase(ctx context.Context, phase string, duration time.Duration) {
if c == nil {
return
}
c.mu.RLock()
agentInfo := c.agentInfo
c.mu.RUnlock()
c.impl.RecordSyncPhase(ctx, agentInfo, phase, duration)
}
// RecordLoginDuration records how long the login to management server took
func (c *ClientMetrics) RecordLoginDuration(ctx context.Context, duration time.Duration, success bool) {
if c == nil {

View File

@@ -70,9 +70,6 @@ func (m *mockMetrics) RecordConnectionStages(_ context.Context, _ AgentInfo, _ s
func (m *mockMetrics) RecordSyncDuration(_ context.Context, _ AgentInfo, _ time.Duration) {
}
func (m *mockMetrics) RecordSyncPhase(_ context.Context, _ AgentInfo, _ string, _ time.Duration) {
}
func (m *mockMetrics) RecordLoginDuration(_ context.Context, _ AgentInfo, _ time.Duration, _ bool) {
}

2560
client/proto/daemon.pb.gw.go Normal file

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,80 @@
package proto
import (
"context"
"net"
"net/http"
"net/http/httptest"
"strings"
"testing"
gatewayruntime "github.com/grpc-ecosystem/grpc-gateway/v2/runtime"
"google.golang.org/grpc"
"google.golang.org/grpc/credentials/insecure"
"google.golang.org/grpc/test/bufconn"
)
func TestGatewayServerRoutesCoverDaemonRPCs(t *testing.T) {
mux := gatewayruntime.NewServeMux()
if err := RegisterDaemonServiceHandlerServer(context.Background(), mux, UnimplementedDaemonServiceServer{}); err != nil {
t.Fatalf("register daemon gateway server handlers: %v", err)
}
assertAllDaemonGatewayRoutesRegistered(t, mux)
}
func TestGatewayClientRoutesCoverDaemonRPCs(t *testing.T) {
listener := bufconn.Listen(1024 * 1024)
server := grpc.NewServer()
RegisterDaemonServiceServer(server, UnimplementedDaemonServiceServer{})
go func() {
if err := server.Serve(listener); err != nil && err != grpc.ErrServerStopped {
t.Errorf("serve bufconn gRPC server: %v", err)
}
}()
t.Cleanup(func() {
server.Stop()
_ = listener.Close()
})
ctx, cancel := context.WithCancel(context.Background())
defer cancel()
mux := gatewayruntime.NewServeMux()
opts := []grpc.DialOption{
grpc.WithContextDialer(func(context.Context, string) (net.Conn, error) {
return listener.Dial()
}),
grpc.WithTransportCredentials(insecure.NewCredentials()),
}
if err := RegisterDaemonServiceHandlerFromEndpoint(ctx, mux, "passthrough:///bufnet", opts); err != nil {
t.Fatalf("register daemon gateway client handlers: %v", err)
}
assertAllDaemonGatewayRoutesRegistered(t, mux)
}
func assertAllDaemonGatewayRoutesRegistered(t *testing.T, mux http.Handler) {
t.Helper()
for _, method := range DaemonService_ServiceDesc.Methods {
assertGatewayRouteRegistered(t, mux, method.MethodName)
}
for _, stream := range DaemonService_ServiceDesc.Streams {
assertGatewayRouteRegistered(t, mux, stream.StreamName)
}
}
func assertGatewayRouteRegistered(t *testing.T, mux http.Handler, methodName string) {
t.Helper()
path := "/daemon.DaemonService/" + methodName
req := httptest.NewRequest(http.MethodPost, path, strings.NewReader("{}"))
req.Header.Set("Content-Type", "application/json")
res := httptest.NewRecorder()
mux.ServeHTTP(res, req)
if res.Code == http.StatusNotFound {
t.Fatalf("gateway route for %s is not registered", methodName)
}
}

View File

@@ -12,5 +12,11 @@ script_path=$(dirname "$(realpath "$0")")
cd "$script_path"
go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.36.6
go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@v1.6.1
protoc -I ./ ./daemon.proto --go_out=../ --go-grpc_out=../ --experimental_allow_proto3_optional
go install github.com/grpc-ecosystem/grpc-gateway/v2/protoc-gen-grpc-gateway@v2.26.3
protoc -I ./ ./daemon.proto \
--go_out=../ \
--go-grpc_out=../ \
--grpc-gateway_out=../ \
--grpc-gateway_opt=generate_unbound_methods=true \
--experimental_allow_proto3_optional
cd "$old_pwd"

223
client/test/json-socket-docker.sh Executable file
View File

@@ -0,0 +1,223 @@
#!/usr/bin/env bash
set -eEuo pipefail
usage() {
cat <<'EOF'
Usage: client/test/json-socket-docker.sh [tcp|unix|both]
Builds the NetBird client Docker image from the local source tree, starts
`netbird service run` in a container with --enable-json-socket, and verifies
that the HTTP/JSON daemon gateway responds to Status requests.
Modes:
tcp Validate tcp://0.0.0.0:8080 via a published localhost port (default)
unix Validate unix:///sock/netbird-http.sock via a bind-mounted socket dir
both Run both validations
Environment:
CONTAINER_RUNTIME docker or podman. Auto-detected if unset.
IMAGE Image tag to build. Default: netbird-json-socket-test:local
TARGETARCH Go/Docker target arch. Default: `go env GOARCH`
PLATFORM Docker platform. Default: linux/$TARGETARCH
WAIT_TIMEOUT Seconds to wait for the JSON socket. Default: 30
EOF
}
if [[ "${1:-}" == "-h" || "${1:-}" == "--help" ]]; then
usage
exit 0
fi
MODE="${1:-tcp}"
case "${MODE}" in
tcp|unix|both) ;;
*)
usage >&2
echo "invalid mode: ${MODE}" >&2
exit 2
;;
esac
RUNTIME="${CONTAINER_RUNTIME:-}"
if [[ -z "${RUNTIME}" ]]; then
if command -v docker >/dev/null 2>&1; then
RUNTIME=docker
elif command -v podman >/dev/null 2>&1; then
RUNTIME=podman
else
echo "docker or podman is required" >&2
exit 127
fi
fi
if ! command -v "${RUNTIME}" >/dev/null 2>&1; then
echo "container runtime not found: ${RUNTIME}" >&2
exit 127
fi
if ! command -v curl >/dev/null 2>&1; then
echo "curl is required" >&2
exit 127
fi
ROOT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)"
IMAGE="${IMAGE:-netbird-json-socket-test:local}"
TARGETARCH="${TARGETARCH:-$(go env GOARCH)}"
PLATFORM="${PLATFORM:-linux/${TARGETARCH}}"
WAIT_TIMEOUT="${WAIT_TIMEOUT:-30}"
TMP_DIR="$(mktemp -d)"
CONTAINERS=()
cleanup() {
local status=$?
for container in "${CONTAINERS[@]:-}"; do
"${RUNTIME}" rm -f "${container}" >/dev/null 2>&1 || true
done
rm -rf "${TMP_DIR}"
exit "${status}"
}
trap cleanup EXIT
build_image() {
echo "==> Building Linux ${TARGETARCH} netbird binary"
mkdir -p "${TMP_DIR}/context/client"
cp "${ROOT_DIR}/client/Dockerfile" "${TMP_DIR}/context/Dockerfile"
cp "${ROOT_DIR}/client/netbird-entrypoint.sh" "${TMP_DIR}/context/client/netbird-entrypoint.sh"
(cd "${ROOT_DIR}" && CGO_ENABLED=0 GOOS=linux GOARCH="${TARGETARCH}" go build -o "${TMP_DIR}/context/netbird" ./client)
echo "==> Building ${IMAGE} for ${PLATFORM}"
"${RUNTIME}" build \
--platform "${PLATFORM}" \
--build-arg NETBIRD_BINARY=netbird \
-t "${IMAGE}" \
-f "${TMP_DIR}/context/Dockerfile" \
"${TMP_DIR}/context"
}
pick_port() {
python3 - <<'PY'
import socket
sock = socket.socket()
sock.bind(("127.0.0.1", 0))
print(sock.getsockname()[1])
sock.close()
PY
}
assert_status_json() {
local response_file="$1"
if command -v python3 >/dev/null 2>&1; then
python3 - "${response_file}" <<'PY'
import json
import sys
with open(sys.argv[1], encoding="utf-8") as fh:
data = json.load(fh)
if not data.get("status"):
raise SystemExit("missing non-empty status field")
if "daemonVersion" not in data:
raise SystemExit("missing daemonVersion field")
print(f"status={data['status']} daemonVersion={data['daemonVersion']}")
PY
else
grep -q '"status"' "${response_file}"
grep -q '"daemonVersion"' "${response_file}"
cat "${response_file}"
fi
}
container_logs() {
local container="$1"
echo "---- ${container} logs ----" >&2
"${RUNTIME}" logs "${container}" >&2 || true
echo "--------------------------" >&2
}
wait_for_http_status() {
local container="$1"
local response="${TMP_DIR}/${container}.json"
local curl_err="${TMP_DIR}/${container}.curl.err"
shift
local deadline=$((SECONDS + WAIT_TIMEOUT))
while (( SECONDS < deadline )); do
if curl -fsS "$@" \
-X POST \
-H 'Content-Type: application/json' \
-d '{}' \
-o "${response}" \
2>"${curl_err}"; then
assert_status_json "${response}"
return 0
fi
if ! "${RUNTIME}" ps --format '{{.Names}}' | grep -Fxq "${container}"; then
echo "container exited before JSON socket became ready" >&2
container_logs "${container}"
return 1
fi
sleep 1
done
echo "timed out waiting for JSON socket after ${WAIT_TIMEOUT}s" >&2
cat "${curl_err}" >&2 || true
container_logs "${container}"
return 1
}
run_netbird_container() {
local container="$1"
local json_socket="$2"
shift 2
CONTAINERS+=("${container}")
"${RUNTIME}" run --rm -d \
--name "${container}" \
-e NB_STATE_DIR=/tmp/netbird-state \
--entrypoint /usr/local/bin/netbird \
"$@" \
"${IMAGE}" \
--log-file console \
--daemon-addr unix:///tmp/netbird.sock \
service run \
--enable-json-socket \
--json-socket "${json_socket}" >/dev/null
}
run_tcp_test() {
local port container
port="$(pick_port)"
container="nb-json-socket-tcp-$RANDOM-$RANDOM"
echo "==> Validating TCP JSON socket on 127.0.0.1:${port}"
run_netbird_container "${container}" "tcp://0.0.0.0:8080" -p "127.0.0.1:${port}:8080"
wait_for_http_status "${container}" "http://127.0.0.1:${port}/daemon.DaemonService/Status"
}
run_unix_test() {
local sock_dir sock_path container
sock_dir="${TMP_DIR}/sock"
sock_path="${sock_dir}/netbird-http.sock"
container="nb-json-socket-unix-$RANDOM-$RANDOM"
mkdir -p "${sock_dir}"
echo "==> Validating Unix JSON socket at ${sock_path}"
run_netbird_container "${container}" "unix:///sock/netbird-http.sock" -v "${sock_dir}:/sock"
wait_for_http_status "${container}" --unix-socket "${sock_path}" "http://unix/daemon.DaemonService/Status"
}
build_image
case "${MODE}" in
tcp)
run_tcp_test
;;
unix)
run_unix_test
;;
both)
run_tcp_test
run_unix_test
;;
esac
echo "==> Docker JSON socket validation passed (${MODE})"

2
go.mod
View File

@@ -66,6 +66,7 @@ require (
github.com/google/nftables v0.3.0
github.com/gopacket/gopacket v1.4.0
github.com/grpc-ecosystem/go-grpc-middleware/v2 v2.0.2-0.20240212192251-757544f21357
github.com/grpc-ecosystem/grpc-gateway/v2 v2.26.3
github.com/hashicorp/go-multierror v1.1.1
github.com/hashicorp/go-secure-stdlib/base62 v0.1.2
github.com/hashicorp/go-version v1.7.0
@@ -330,6 +331,7 @@ require (
golang.org/x/text v0.36.0 // indirect
golang.org/x/tools v0.43.0 // indirect
golang.zx2c4.com/wintun v0.0.0-20230126152724-0fa3db229ce2 // indirect
google.golang.org/genproto/googleapis/api v0.0.0-20260319201613-d00831a3d3e7 // indirect
google.golang.org/genproto/googleapis/rpc v0.0.0-20260401024825-9d38bb4040a9 // indirect
gopkg.in/square/go-jose.v2 v2.6.0 // indirect
gopkg.in/tomb.v1 v1.0.0-20141024135613-dd632973f1e7 // indirect