mirror of
https://github.com/netbirdio/netbird.git
synced 2026-07-22 16:31:28 +02:00
## Describe your changes The tray session row flapped on every transient engine restart and went blank at the exact moment the session expired: - Watcher.Close zeroed the server-scoped recorder deadline on every engine restart (sleep/wake, network change, stream errors), hiding the tray row until the next successful login re-seeded it. Close now only stops the timers; the client run loop clears the recorder when it exits for real. - Status.GetSessionExpiresAt masked past deadlines as zero, so the status snapshot dropped the field right when the UI should have shown "session expired". The getter now returns the raw value and the tray renders past deadlines as "Session expired — sign in again", routing the click to the login flow. - The watcher rejected recently-expired deadlines (30s skew) delivered by login/sync, clearing the recorder instead of surfacing the expiry. Deadlines up to 30 days past are now recorded as expired; only older values are rejected as garbage. ## Issue ticket number and link ## Stack <!-- branch-stack --> ### Checklist - [x] Is it a bug fix - [ ] Is a typo/documentation fix - [ ] Is a feature enhancement - [ ] It is a refactor - [ ] Created tests that fail without the change (if possible) - [ ] This change does **not** modify the public API, gRPC protocols, functionality behavior, CLI / service flags, or introduce a new feature — **OR** I have discussed it with the NetBird team beforehand (link the issue / Slack thread in the description). See [CONTRIBUTING.md](https://github.com/netbirdio/netbird/blob/main/CONTRIBUTING.md#discuss-changes-with-the-netbird-team-first). > By submitting this pull request, you confirm that you have read and agree to the terms of the [Contributor License Agreement](https://github.com/netbirdio/netbird/blob/main/CONTRIBUTOR_LICENSE_AGREEMENT.md). ## Documentation Select exactly one: - [ ] I added/updated documentation for this change - [x] Documentation is **not needed** for this change (explain why) ### Docs PR URL (required if "docs added" is checked) Paste the PR link from https://github.com/netbirdio/docs here: https://github.com/netbirdio/docs/pull/__ <!-- codesmith:footer --> --- <a href="https://app.blacksmith.sh/netbirdio/codesmith/netbird/pr/6847"><picture><source media="(prefers-color-scheme: dark)" srcset="https://pr-comments-assets.blacksmith.sh/codesmith/view-with-codesmith-dark-v2.svg"><source media="(prefers-color-scheme: light)" srcset="https://pr-comments-assets.blacksmith.sh/codesmith/view-with-codesmith-light-v2.svg"><img alt="View with Codesmith" src="https://pr-comments-assets.blacksmith.sh/codesmith/view-with-codesmith-dark-v2.svg"></picture></a> <a href="https://backend.blacksmith.sh/track/enable-autofix?expires=1787220767&installation_model_id=427504&pr_number=6847&repository=netbirdio%2Fnetbird&return_to=https%3A%2F%2Fgithub.com%2Fnetbirdio%2Fnetbird%2Fpull%2F6847&signature=77977ad47c5eda51e1d76e59bc736203b4592462bf94076811a20ea0f7f23aee"><picture><source media="(prefers-color-scheme: dark)" srcset="https://pr-comments-assets.blacksmith.sh/codesmith/autofix-with-codesmith-dark.svg"><source media="(prefers-color-scheme: light)" srcset="https://pr-comments-assets.blacksmith.sh/codesmith/autofix-with-codesmith-light.svg"><img alt="Autofix with Codesmith" src="https://pr-comments-assets.blacksmith.sh/codesmith/autofix-with-codesmith-dark.svg"></picture></a> <sup>Need help on this PR? Tag <code>/codesmith</code> with what you need. Autofix is disabled.</sup> <!-- codesmith:autofix:disabled --> <!-- /codesmith:footer --> <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Tray “session expires” and countdown now accurately reflect expired sessions and remaining time near boundary moments (including improved rounding). * **Bug Fixes** * Recently expired deadlines are retained for correct status reporting, while very old past deadlines are rejected and cleared. * Session expiry “expires at” is preserved when the session watcher closes during reconnect scenarios. * Clicking an already-expired session in the tray now routes the user to the login screen. * **Tests** * Updated and expanded coverage for recent/ancient past handling and watcher close behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
756 lines
23 KiB
Go
756 lines
23 KiB
Go
package internal
|
|
|
|
import (
|
|
"context"
|
|
"errors"
|
|
"fmt"
|
|
"net"
|
|
"net/netip"
|
|
"path/filepath"
|
|
"runtime"
|
|
"runtime/debug"
|
|
"strings"
|
|
"sync"
|
|
"sync/atomic"
|
|
"time"
|
|
|
|
"github.com/cenkalti/backoff/v4"
|
|
log "github.com/sirupsen/logrus"
|
|
|
|
"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
|
|
"google.golang.org/grpc/codes"
|
|
gstatus "google.golang.org/grpc/status"
|
|
|
|
"github.com/netbirdio/netbird/client/iface/wgaddr"
|
|
|
|
"github.com/netbirdio/netbird/client/iface"
|
|
"github.com/netbirdio/netbird/client/iface/device"
|
|
"github.com/netbirdio/netbird/client/iface/netstack"
|
|
"github.com/netbirdio/netbird/client/internal/dns"
|
|
"github.com/netbirdio/netbird/client/internal/lazyconn"
|
|
"github.com/netbirdio/netbird/client/internal/listener"
|
|
"github.com/netbirdio/netbird/client/internal/metrics"
|
|
"github.com/netbirdio/netbird/client/internal/peer"
|
|
"github.com/netbirdio/netbird/client/internal/profilemanager"
|
|
"github.com/netbirdio/netbird/client/internal/statemanager"
|
|
"github.com/netbirdio/netbird/client/internal/stdnet"
|
|
"github.com/netbirdio/netbird/client/internal/updater"
|
|
"github.com/netbirdio/netbird/client/internal/updater/installer"
|
|
nbnet "github.com/netbirdio/netbird/client/net"
|
|
cProto "github.com/netbirdio/netbird/client/proto"
|
|
"github.com/netbirdio/netbird/client/ssh"
|
|
sshconfig "github.com/netbirdio/netbird/client/ssh/config"
|
|
"github.com/netbirdio/netbird/client/system"
|
|
mgm "github.com/netbirdio/netbird/shared/management/client"
|
|
mgmProto "github.com/netbirdio/netbird/shared/management/proto"
|
|
"github.com/netbirdio/netbird/shared/relay/auth/hmac"
|
|
relayClient "github.com/netbirdio/netbird/shared/relay/client"
|
|
signal "github.com/netbirdio/netbird/shared/signal/client"
|
|
"github.com/netbirdio/netbird/util"
|
|
"github.com/netbirdio/netbird/version"
|
|
)
|
|
|
|
// androidRunOverride is set on Android to inject mobile dependencies
|
|
// when using embed.Client (which calls Run() with empty MobileDependency).
|
|
var androidRunOverride func(c *ConnectClient, runningChan chan struct{}, logPath string) error
|
|
|
|
type ConnectClient struct {
|
|
ctx context.Context
|
|
runCancel context.CancelFunc
|
|
runExited chan struct{}
|
|
runOnce sync.Once
|
|
runStarted atomic.Bool
|
|
config *profilemanager.Config
|
|
statusRecorder *peer.Status
|
|
|
|
engine *Engine
|
|
engineMutex sync.Mutex
|
|
clientMetrics *metrics.ClientMetrics
|
|
updateManager *updater.Manager
|
|
|
|
persistSyncResponse bool
|
|
}
|
|
|
|
func NewConnectClient(
|
|
ctx context.Context,
|
|
config *profilemanager.Config,
|
|
statusRecorder *peer.Status,
|
|
) *ConnectClient {
|
|
// Derive the run context here so Stop owns the cancel that unblocks the run
|
|
// loop. runCancel is set once at construction, so Stop can call it without
|
|
// racing the run loop's startup. Callers therefore need not cancel before Stop.
|
|
runCtx, runCancel := context.WithCancel(ctx)
|
|
return &ConnectClient{
|
|
ctx: runCtx,
|
|
runCancel: runCancel,
|
|
runExited: make(chan struct{}),
|
|
config: config,
|
|
statusRecorder: statusRecorder,
|
|
engineMutex: sync.Mutex{},
|
|
}
|
|
}
|
|
|
|
func (c *ConnectClient) SetUpdateManager(um *updater.Manager) {
|
|
c.updateManager = um
|
|
}
|
|
|
|
// Run with main logic.
|
|
func (c *ConnectClient) Run(runningChan chan struct{}, logPath string) error {
|
|
if androidRunOverride != nil {
|
|
return androidRunOverride(c, runningChan, logPath)
|
|
}
|
|
return c.run(MobileDependency{}, runningChan, logPath)
|
|
}
|
|
|
|
// RunOnAndroid with main logic on mobile system
|
|
func (c *ConnectClient) RunOnAndroid(
|
|
tunAdapter device.TunAdapter,
|
|
iFaceDiscover stdnet.ExternalIFaceDiscover,
|
|
networkChangeListener listener.NetworkChangeListener,
|
|
dnsAddresses []netip.AddrPort,
|
|
dnsReadyListener dns.ReadyListener,
|
|
stateFilePath string,
|
|
cacheDir string,
|
|
) error {
|
|
// in case of non Android os these variables will be nil
|
|
mobileDependency := MobileDependency{
|
|
TunAdapter: tunAdapter,
|
|
IFaceDiscover: iFaceDiscover,
|
|
NetworkChangeListener: networkChangeListener,
|
|
HostDNSAddresses: dnsAddresses,
|
|
DnsReadyListener: dnsReadyListener,
|
|
StateFilePath: stateFilePath,
|
|
TempDir: cacheDir,
|
|
}
|
|
return c.run(mobileDependency, nil, "")
|
|
}
|
|
|
|
func (c *ConnectClient) RunOniOS(
|
|
fileDescriptor int32,
|
|
networkChangeListener listener.NetworkChangeListener,
|
|
dnsManager dns.IosDnsManager,
|
|
stateFilePath string,
|
|
cacheDir string,
|
|
logFilePath string,
|
|
) error {
|
|
// Set GC percent to 5% to reduce memory usage as iOS only allows 50MB of memory for the extension.
|
|
debug.SetGCPercent(5)
|
|
|
|
mobileDependency := MobileDependency{
|
|
FileDescriptor: fileDescriptor,
|
|
NetworkChangeListener: networkChangeListener,
|
|
DnsManager: dnsManager,
|
|
StateFilePath: stateFilePath,
|
|
TempDir: cacheDir,
|
|
}
|
|
return c.run(mobileDependency, nil, logFilePath)
|
|
}
|
|
|
|
func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan struct{}, logPath string) error {
|
|
// Mark the loop as started and signal exit on return so Stop can wait for
|
|
// the loop to finish (and skip the wait if the loop never ran).
|
|
c.runStarted.Store(true)
|
|
defer c.runOnce.Do(func() { close(c.runExited) })
|
|
|
|
defer func() {
|
|
if r := recover(); r != nil {
|
|
rec := c.statusRecorder
|
|
if rec != nil {
|
|
rec.PublishEvent(
|
|
cProto.SystemEvent_CRITICAL, cProto.SystemEvent_SYSTEM,
|
|
"panic occurred",
|
|
"The Netbird service panicked. Please restart the service and submit a bug report with the client logs.",
|
|
nil,
|
|
)
|
|
}
|
|
|
|
log.Panicf("Panic occurred: %v, stack trace: %s", r, string(debug.Stack()))
|
|
}
|
|
}()
|
|
|
|
// Stop metrics push on exit
|
|
defer func() {
|
|
if c.clientMetrics != nil {
|
|
c.clientMetrics.StopPush()
|
|
}
|
|
}()
|
|
|
|
log.Infof("starting NetBird client version %s on %s/%s", version.NetbirdVersion(), runtime.GOOS, runtime.GOARCH)
|
|
|
|
nbnet.Init()
|
|
|
|
// Initialize metrics once at startup (always active for debug bundles)
|
|
if c.clientMetrics == nil {
|
|
agentInfo := metrics.AgentInfo{
|
|
DeploymentType: metrics.DeploymentTypeUnknown,
|
|
Version: version.NetbirdVersion(),
|
|
OS: runtime.GOOS,
|
|
Arch: runtime.GOARCH,
|
|
}
|
|
c.clientMetrics = metrics.NewClientMetrics(agentInfo)
|
|
log.Debugf("initialized client metrics")
|
|
|
|
// Start metrics push if enabled (uses daemon context, persists across engine restarts)
|
|
if metrics.IsMetricsPushEnabled() {
|
|
c.clientMetrics.StartPush(c.ctx, metrics.PushConfigFromEnv())
|
|
}
|
|
}
|
|
|
|
backOff := &backoff.ExponentialBackOff{
|
|
InitialInterval: time.Second,
|
|
RandomizationFactor: 1,
|
|
Multiplier: 1.7,
|
|
MaxInterval: 15 * time.Second,
|
|
MaxElapsedTime: 3 * 30 * 24 * time.Hour, // 3 months
|
|
Stop: backoff.Stop,
|
|
Clock: backoff.SystemClock,
|
|
}
|
|
|
|
state := CtxGetState(c.ctx)
|
|
defer func() {
|
|
s, err := state.Status()
|
|
if err != nil || s != StatusNeedsLogin {
|
|
state.Set(StatusIdle)
|
|
}
|
|
}()
|
|
|
|
wrapErr := state.Wrap
|
|
myPrivateKey, err := wgtypes.ParseKey(c.config.PrivateKey)
|
|
if err != nil {
|
|
log.Errorf("failed parsing Wireguard key %s: [%s]", c.config.PrivateKey, err.Error())
|
|
return wrapErr(err)
|
|
}
|
|
|
|
var mgmTlsEnabled bool
|
|
if c.config.ManagementURL.Scheme == "https" {
|
|
mgmTlsEnabled = true
|
|
}
|
|
|
|
publicSSHKey, err := ssh.GeneratePublicKey([]byte(c.config.SSHKey))
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
var path string
|
|
if runtime.GOOS == "ios" || runtime.GOOS == "android" {
|
|
// On mobile, use the provided state file path directly
|
|
if !fileExists(mobileDependency.StateFilePath) {
|
|
if err := createFile(mobileDependency.StateFilePath); err != nil {
|
|
log.Errorf("failed to create state file: %v", err)
|
|
// we are not exiting as we can run without the state manager
|
|
}
|
|
}
|
|
path = mobileDependency.StateFilePath
|
|
} else {
|
|
sm := profilemanager.NewServiceManager("")
|
|
path = sm.GetStatePath()
|
|
}
|
|
stateManager := statemanager.New(path)
|
|
stateManager.RegisterState(&sshconfig.ShutdownState{})
|
|
|
|
if c.updateManager != nil {
|
|
c.updateManager.CheckUpdateSuccess(c.ctx)
|
|
}
|
|
|
|
inst := installer.New()
|
|
if err := inst.CleanUpInstallerFiles(); err != nil {
|
|
log.Errorf("failed to clean up temporary installer file: %v", err)
|
|
}
|
|
|
|
defer func() {
|
|
c.statusRecorder.SetSessionExpiresAt(time.Time{})
|
|
c.statusRecorder.ClientStop()
|
|
}()
|
|
operation := func() error {
|
|
// if context cancelled we not start new backoff cycle
|
|
if c.ctx.Err() != nil {
|
|
return nil
|
|
}
|
|
|
|
state.Set(StatusConnecting)
|
|
|
|
engineCtx, cancel := context.WithCancel(c.ctx)
|
|
defer func() {
|
|
_, err := state.Status()
|
|
c.statusRecorder.MarkManagementDisconnected(err)
|
|
c.statusRecorder.CleanLocalPeerState()
|
|
cancel()
|
|
}()
|
|
|
|
log.Debugf("connecting to the Management service %s", c.config.ManagementURL.Host)
|
|
mgmClient, err := mgm.NewClient(engineCtx, c.config.ManagementURL.Host, myPrivateKey, mgmTlsEnabled)
|
|
if err != nil {
|
|
// On daemon shutdown / Down() the parent context is cancelled
|
|
// and the dial fails with "context canceled". Wrapping that
|
|
// into state would leave the snapshot stuck at Connecting+err
|
|
// until the backoff loop wakes up — instead let the operation
|
|
// return cleanly so the deferred state.Set(StatusIdle) takes
|
|
// effect on the next iteration.
|
|
if c.ctx.Err() != nil {
|
|
return nil
|
|
}
|
|
return wrapErr(gstatus.Errorf(codes.FailedPrecondition, "failed connecting to Management Service : %s", err))
|
|
}
|
|
mgmNotifier := statusRecorderToMgmConnStateNotifier(c.statusRecorder)
|
|
mgmClient.SetConnStateListener(mgmNotifier)
|
|
|
|
// Update metrics with actual deployment type after connection
|
|
deploymentType := metrics.DetermineDeploymentType(mgmClient.GetServerURL())
|
|
agentInfo := metrics.AgentInfo{
|
|
DeploymentType: deploymentType,
|
|
Version: version.NetbirdVersion(),
|
|
OS: runtime.GOOS,
|
|
Arch: runtime.GOARCH,
|
|
}
|
|
c.clientMetrics.UpdateAgentInfo(agentInfo, myPrivateKey.PublicKey().String())
|
|
|
|
log.Debugf("connected to the Management service %s", c.config.ManagementURL.Host)
|
|
defer func() {
|
|
if err = mgmClient.Close(); err != nil {
|
|
log.Warnf("failed to close the Management service client %v", err)
|
|
}
|
|
}()
|
|
|
|
// connect (just a connection, no stream yet) and login to Management Service to get an initial global Netbird config
|
|
loginStarted := time.Now()
|
|
loginResp, err := loginToManagement(engineCtx, mgmClient, publicSSHKey, c.config)
|
|
if err != nil {
|
|
c.clientMetrics.RecordLoginDuration(engineCtx, time.Since(loginStarted), false)
|
|
log.Debug(err)
|
|
if s, ok := gstatus.FromError(err); ok && (s.Code() == codes.PermissionDenied) {
|
|
state.Set(StatusNeedsLogin)
|
|
c.runCancel()
|
|
return backoff.Permanent(wrapErr(err)) // unrecoverable error
|
|
}
|
|
return wrapErr(err)
|
|
}
|
|
c.clientMetrics.RecordLoginDuration(engineCtx, time.Since(loginStarted), true)
|
|
c.statusRecorder.MarkManagementConnected()
|
|
|
|
if metricsConfig := loginResp.GetNetbirdConfig().GetMetrics(); metricsConfig != nil {
|
|
c.clientMetrics.UpdatePushFromMgm(c.ctx, metricsConfig.GetEnabled())
|
|
}
|
|
|
|
localPeerState := peer.LocalPeerState{
|
|
IP: loginResp.GetPeerConfig().GetAddress(),
|
|
PubKey: myPrivateKey.PublicKey().String(),
|
|
KernelInterface: device.WireGuardModuleIsLoaded() && !netstack.IsEnabled(),
|
|
FQDN: loginResp.GetPeerConfig().GetFqdn(),
|
|
}
|
|
c.statusRecorder.UpdateLocalPeerState(localPeerState)
|
|
|
|
signalURL := fmt.Sprintf("%s://%s",
|
|
strings.ToLower(loginResp.GetNetbirdConfig().GetSignal().GetProtocol().String()),
|
|
loginResp.GetNetbirdConfig().GetSignal().GetUri(),
|
|
)
|
|
|
|
c.statusRecorder.UpdateSignalAddress(signalURL)
|
|
|
|
c.statusRecorder.MarkSignalDisconnected(nil)
|
|
defer func() {
|
|
_, err := state.Status()
|
|
c.statusRecorder.MarkSignalDisconnected(err)
|
|
}()
|
|
|
|
// with the global Netbird config in hand connect (just a connection, no stream yet) Signal
|
|
signalClient, err := connectToSignal(engineCtx, loginResp.GetNetbirdConfig(), myPrivateKey)
|
|
if err != nil {
|
|
log.Error(err)
|
|
return wrapErr(err)
|
|
}
|
|
defer func() {
|
|
err = signalClient.Close()
|
|
if err != nil {
|
|
log.Warnf("failed closing Signal service client %v", err)
|
|
}
|
|
}()
|
|
|
|
signalNotifier := statusRecorderToSignalConnStateNotifier(c.statusRecorder)
|
|
signalClient.SetConnStateListener(signalNotifier)
|
|
|
|
c.statusRecorder.MarkSignalConnected()
|
|
|
|
relayURLs, token := parseRelayInfo(loginResp)
|
|
if override, ok := peer.OverrideRelayURLs(); ok {
|
|
log.Infof("overriding relay URLs from %s: %v", peer.EnvKeyNBHomeRelayServers, override)
|
|
relayURLs = override
|
|
}
|
|
peerConfig := loginResp.GetPeerConfig()
|
|
|
|
engineConfig, err := createEngineConfig(myPrivateKey, c.config, peerConfig, logPath)
|
|
if err != nil {
|
|
log.Error(err)
|
|
return wrapErr(err)
|
|
}
|
|
engineConfig.TempDir = mobileDependency.TempDir
|
|
// Leave StateDir empty when there is no state path so a disk-backed
|
|
// syncstore falls back to os.TempDir() instead of filepath.Dir("") == ".".
|
|
if path != "" {
|
|
engineConfig.StateDir = filepath.Dir(path)
|
|
}
|
|
|
|
relayManager := relayClient.NewManager(engineCtx, relayURLs, myPrivateKey.PublicKey().String(), engineConfig.MTU)
|
|
c.statusRecorder.SetRelayMgr(relayManager)
|
|
if len(relayURLs) > 0 {
|
|
if token != nil {
|
|
if err := relayManager.UpdateToken(token); err != nil {
|
|
log.Errorf("failed to update token: %s", err)
|
|
return wrapErr(err)
|
|
}
|
|
}
|
|
log.Infof("connecting to the Relay service(s): %s", strings.Join(relayURLs, ", "))
|
|
if err = relayManager.Serve(); err != nil {
|
|
log.Error(err)
|
|
}
|
|
}
|
|
|
|
checks := loginResp.GetChecks()
|
|
|
|
c.engineMutex.Lock()
|
|
engine := NewEngine(engineCtx, cancel, engineConfig, EngineServices{
|
|
SignalClient: signalClient,
|
|
MgmClient: mgmClient,
|
|
RelayManager: relayManager,
|
|
StatusRecorder: c.statusRecorder,
|
|
Checks: checks,
|
|
StateManager: stateManager,
|
|
UpdateManager: c.updateManager,
|
|
ClientMetrics: c.clientMetrics,
|
|
MetricsCtx: c.ctx,
|
|
}, mobileDependency)
|
|
engine.SetSyncResponsePersistence(c.persistSyncResponse)
|
|
c.engine = engine
|
|
c.engineMutex.Unlock()
|
|
|
|
if err := engine.Start(loginResp.GetNetbirdConfig(), c.config.ManagementURL); err != nil {
|
|
log.Errorf("error while starting Netbird Connection Engine: %s", err)
|
|
return wrapErr(err)
|
|
}
|
|
|
|
// Seed the session-expiry deadline from the LoginResponse. Subsequent
|
|
// changes flow in through SyncResponse and are applied in handleSync.
|
|
engine.ApplySessionDeadline(loginResp.GetSessionExpiresAt())
|
|
|
|
log.Infof("Netbird engine started, the IP is: %s", peerConfig.GetAddress())
|
|
state.Set(StatusConnected)
|
|
|
|
if runningChan != nil {
|
|
select {
|
|
case <-runningChan:
|
|
default:
|
|
close(runningChan)
|
|
}
|
|
}
|
|
|
|
<-engineCtx.Done()
|
|
|
|
c.engineMutex.Lock()
|
|
c.engine = nil
|
|
c.engineMutex.Unlock()
|
|
|
|
log.Infof("ensuring wg interface is removed, Netbird engine context cancelled")
|
|
|
|
if err := engine.Stop(); err != nil {
|
|
log.Errorf("Failed to stop engine: %v", err)
|
|
}
|
|
c.statusRecorder.ClientTeardown()
|
|
|
|
backOff.Reset()
|
|
|
|
log.Info("stopped NetBird client")
|
|
|
|
if _, err := state.Status(); errors.Is(err, ErrResetConnection) {
|
|
return err
|
|
}
|
|
|
|
return nil
|
|
}
|
|
|
|
c.statusRecorder.ClientStart()
|
|
// Wrap the backoff with c.ctx so Down()/actCancel propagates into the
|
|
// inter-attempt sleep — otherwise a 15s MaxInterval can keep the retry
|
|
// loop alive long after the caller asked to give up, leaving the
|
|
// status stream stuck at Connecting.
|
|
err = backoff.Retry(operation, backoff.WithContext(backOff, c.ctx))
|
|
if err != nil {
|
|
log.Debugf("exiting client retry loop due to unrecoverable error: %s", err)
|
|
if s, ok := gstatus.FromError(err); ok && (s.Code() == codes.PermissionDenied) {
|
|
state.Set(StatusNeedsLogin)
|
|
c.runCancel()
|
|
}
|
|
return err
|
|
}
|
|
return nil
|
|
}
|
|
|
|
func parseRelayInfo(loginResp *mgmProto.LoginResponse) ([]string, *hmac.Token) {
|
|
relayCfg := loginResp.GetNetbirdConfig().GetRelay()
|
|
if relayCfg == nil {
|
|
return nil, nil
|
|
}
|
|
|
|
token := &hmac.Token{
|
|
Payload: relayCfg.GetTokenPayload(),
|
|
Signature: relayCfg.GetTokenSignature(),
|
|
}
|
|
|
|
return relayCfg.GetUrls(), token
|
|
}
|
|
|
|
func (c *ConnectClient) Engine() *Engine {
|
|
if c == nil {
|
|
return nil
|
|
}
|
|
var e *Engine
|
|
c.engineMutex.Lock()
|
|
e = c.engine
|
|
c.engineMutex.Unlock()
|
|
return e
|
|
}
|
|
|
|
// GetLatestSyncResponse returns the latest sync response from the engine.
|
|
func (c *ConnectClient) GetLatestSyncResponse() (*mgmProto.SyncResponse, error) {
|
|
engine := c.Engine()
|
|
if engine == nil {
|
|
return nil, errors.New("engine is not initialized")
|
|
}
|
|
|
|
syncResponse, err := engine.GetLatestSyncResponse()
|
|
if err != nil {
|
|
return nil, fmt.Errorf("get latest sync response: %w", err)
|
|
}
|
|
|
|
if syncResponse == nil {
|
|
return nil, errors.New("sync response is not available")
|
|
}
|
|
|
|
return syncResponse, nil
|
|
}
|
|
|
|
// SetLogLevel sets the log level for the firewall manager if the engine is running.
|
|
func (c *ConnectClient) SetLogLevel(level log.Level) {
|
|
engine := c.Engine()
|
|
if engine == nil {
|
|
return
|
|
}
|
|
|
|
fwManager := engine.GetFirewallManager()
|
|
if fwManager != nil {
|
|
fwManager.SetLogLevel(level)
|
|
}
|
|
}
|
|
|
|
// Status returns the current client status
|
|
func (c *ConnectClient) Status() StatusType {
|
|
if c == nil {
|
|
return StatusIdle
|
|
}
|
|
status, err := CtxGetState(c.ctx).Status()
|
|
if err != nil {
|
|
return StatusIdle
|
|
}
|
|
|
|
return status
|
|
}
|
|
|
|
func (c *ConnectClient) Stop() error {
|
|
c.runCancel()
|
|
if c.runStarted.Load() {
|
|
<-c.runExited
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// SetSyncResponsePersistence enables or disables sync response persistence.
|
|
// When enabled, the last received sync response will be stored and can be retrieved
|
|
// through the Engine's GetLatestSyncResponse method. When disabled, any stored
|
|
// sync response will be cleared.
|
|
func (c *ConnectClient) SetSyncResponsePersistence(enabled bool) {
|
|
c.engineMutex.Lock()
|
|
c.persistSyncResponse = enabled
|
|
c.engineMutex.Unlock()
|
|
|
|
engine := c.Engine()
|
|
if engine != nil {
|
|
engine.SetSyncResponsePersistence(enabled)
|
|
}
|
|
}
|
|
|
|
// createEngineConfig converts configuration received from Management Service to EngineConfig
|
|
func createEngineConfig(key wgtypes.Key, config *profilemanager.Config, peerConfig *mgmProto.PeerConfig, logPath string) (*EngineConfig, error) {
|
|
nm := false
|
|
if config.NetworkMonitor != nil {
|
|
nm = *config.NetworkMonitor
|
|
}
|
|
wgAddr, err := wgaddr.ParseWGAddress(peerConfig.Address)
|
|
if err != nil {
|
|
return nil, fmt.Errorf("parse overlay address %q: %w", peerConfig.Address, err)
|
|
}
|
|
|
|
if !config.DisableIPv6 {
|
|
if err := wgAddr.SetIPv6FromCompact(peerConfig.GetAddressV6()); err != nil {
|
|
log.Warn(err)
|
|
}
|
|
}
|
|
|
|
engineConf := &EngineConfig{
|
|
WgIfaceName: config.WgIface,
|
|
WgAddr: wgAddr,
|
|
IFaceBlackList: config.IFaceBlackList,
|
|
DisableIPv6Discovery: config.DisableIPv6Discovery,
|
|
WgPrivateKey: key,
|
|
WgPort: config.WgPort,
|
|
NetworkMonitor: nm,
|
|
SSHKey: []byte(config.SSHKey),
|
|
NATExternalIPs: config.NATExternalIPs,
|
|
CustomDNSAddress: config.CustomDNSAddress,
|
|
RosenpassEnabled: config.RosenpassEnabled,
|
|
RosenpassPermissive: config.RosenpassPermissive,
|
|
ServerSSHAllowed: util.ReturnBoolWithDefaultTrue(config.ServerSSHAllowed),
|
|
EnableSSHRoot: config.EnableSSHRoot,
|
|
EnableSSHSFTP: config.EnableSSHSFTP,
|
|
EnableSSHLocalPortForwarding: config.EnableSSHLocalPortForwarding,
|
|
EnableSSHRemotePortForwarding: config.EnableSSHRemotePortForwarding,
|
|
DisableSSHAuth: config.DisableSSHAuth,
|
|
DNSRouteInterval: config.DNSRouteInterval,
|
|
|
|
DisableClientRoutes: config.DisableClientRoutes,
|
|
DisableServerRoutes: config.DisableServerRoutes || config.BlockInbound,
|
|
DisableDNS: config.DisableDNS,
|
|
DisableFirewall: config.DisableFirewall,
|
|
BlockLANAccess: config.BlockLANAccess,
|
|
BlockInbound: config.BlockInbound,
|
|
DisableIPv6: config.DisableIPv6,
|
|
|
|
LazyConnection: lazyconn.ParseState(config.LazyConnection),
|
|
|
|
MTU: selectMTU(config.MTU, peerConfig.Mtu),
|
|
LogPath: logPath,
|
|
|
|
ProfileConfig: config,
|
|
}
|
|
|
|
if config.PreSharedKey != "" {
|
|
preSharedKey, err := wgtypes.ParseKey(config.PreSharedKey)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
engineConf.PreSharedKey = &preSharedKey
|
|
}
|
|
|
|
port, err := freePort(config.WgPort)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if port != config.WgPort {
|
|
log.Infof("using %d as wireguard port: %d is in use", port, config.WgPort)
|
|
}
|
|
engineConf.WgPort = port
|
|
|
|
return engineConf, nil
|
|
}
|
|
|
|
func selectMTU(localMTU uint16, peerMTU int32) uint16 {
|
|
var finalMTU uint16 = iface.DefaultMTU
|
|
if localMTU > 0 {
|
|
finalMTU = localMTU
|
|
} else if peerMTU > 0 {
|
|
finalMTU = uint16(peerMTU)
|
|
}
|
|
|
|
// Set global DNS MTU
|
|
dns.SetCurrentMTU(finalMTU)
|
|
|
|
return finalMTU
|
|
}
|
|
|
|
// connectToSignal creates Signal Service client and established a connection
|
|
func connectToSignal(ctx context.Context, wtConfig *mgmProto.NetbirdConfig, ourPrivateKey wgtypes.Key) (*signal.GrpcClient, error) {
|
|
var sigTLSEnabled bool
|
|
if wtConfig.Signal.Protocol == mgmProto.HostConfig_HTTPS {
|
|
sigTLSEnabled = true
|
|
} else {
|
|
sigTLSEnabled = false
|
|
}
|
|
|
|
signalClient, err := signal.NewClient(ctx, wtConfig.Signal.Uri, ourPrivateKey, sigTLSEnabled)
|
|
if err != nil {
|
|
log.Errorf("error while connecting to the Signal Exchange Service %s: %s", wtConfig.Signal.Uri, err)
|
|
return nil, gstatus.Errorf(codes.FailedPrecondition, "failed connecting to Signal Service : %s", err)
|
|
}
|
|
|
|
return signalClient, nil
|
|
}
|
|
|
|
// loginToManagement creates Management ServiceDependencies client, establishes a connection, logs-in and gets a global Netbird config (signal, turn, stun hosts, etc)
|
|
func loginToManagement(ctx context.Context, client mgm.Client, pubSSHKey []byte, config *profilemanager.Config) (*mgmProto.LoginResponse, error) {
|
|
sysInfo := system.GetInfo(ctx)
|
|
sysInfo.SetFlags(
|
|
config.RosenpassEnabled,
|
|
config.RosenpassPermissive,
|
|
config.ServerSSHAllowed,
|
|
config.DisableClientRoutes,
|
|
config.DisableServerRoutes,
|
|
config.DisableDNS,
|
|
config.DisableFirewall,
|
|
config.BlockLANAccess,
|
|
config.BlockInbound,
|
|
config.DisableIPv6,
|
|
config.EnableSSHRoot,
|
|
config.EnableSSHSFTP,
|
|
config.EnableSSHLocalPortForwarding,
|
|
config.EnableSSHRemotePortForwarding,
|
|
config.DisableSSHAuth,
|
|
)
|
|
return client.Login(sysInfo, pubSSHKey, config.DNSLabels)
|
|
}
|
|
|
|
func statusRecorderToMgmConnStateNotifier(statusRecorder *peer.Status) mgm.ConnStateNotifier {
|
|
var sri interface{} = statusRecorder
|
|
mgmNotifier, _ := sri.(mgm.ConnStateNotifier)
|
|
return mgmNotifier
|
|
}
|
|
|
|
func statusRecorderToSignalConnStateNotifier(statusRecorder *peer.Status) signal.ConnStateNotifier {
|
|
var sri interface{} = statusRecorder
|
|
notifier, _ := sri.(signal.ConnStateNotifier)
|
|
return notifier
|
|
}
|
|
|
|
// freePort attempts to determine if the provided port is available, if not it will ask the system for a free port.
|
|
func freePort(initPort int) (int, error) {
|
|
addr := net.UDPAddr{Port: initPort}
|
|
|
|
conn, err := net.ListenUDP("udp", &addr)
|
|
if err == nil {
|
|
returnPort := conn.LocalAddr().(*net.UDPAddr).Port
|
|
closeConnWithLog(conn)
|
|
return returnPort, nil
|
|
}
|
|
|
|
// if the port is already in use, ask the system for a free port
|
|
addr.Port = 0
|
|
conn, err = net.ListenUDP("udp", &addr)
|
|
if err != nil {
|
|
return 0, fmt.Errorf("unable to get a free port: %v", err)
|
|
}
|
|
|
|
udpAddr, ok := conn.LocalAddr().(*net.UDPAddr)
|
|
if !ok {
|
|
return 0, errors.New("wrong address type when getting a free port")
|
|
}
|
|
closeConnWithLog(conn)
|
|
return udpAddr.Port, nil
|
|
}
|
|
|
|
func closeConnWithLog(conn *net.UDPConn) {
|
|
startClosing := time.Now()
|
|
err := conn.Close()
|
|
if err != nil {
|
|
log.Warnf("closing probe port %d failed: %v. NetBird will still attempt to use this port for connection.", conn.LocalAddr().(*net.UDPAddr).Port, err)
|
|
}
|
|
if time.Since(startClosing) > time.Second {
|
|
log.Warnf("closing the testing port %d took %s. Usually it is safe to ignore, but continuous warnings may indicate a problem.", conn.LocalAddr().(*net.UDPAddr).Port, time.Since(startClosing))
|
|
}
|
|
}
|