Merge remote-tracking branch 'origin/main' into dmitri-event-aggregation

Signed-off-by: Dmitri Dolguikh <dmitri.external@netbird.io>
This commit is contained in:
Dmitri Dolguikh
2026-07-01 16:03:01 +02:00
223 changed files with 40556 additions and 1013 deletions

View File

@@ -314,6 +314,10 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
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(),
@@ -399,6 +403,7 @@ func (c *ConnectClient) run(mobileDependency MobileDependency, runningChan chan
StateManager: stateManager,
UpdateManager: c.updateManager,
ClientMetrics: c.clientMetrics,
MetricsCtx: c.ctx,
}, mobileDependency)
engine.SetSyncResponsePersistence(c.persistSyncResponse)
c.engine = engine

View File

@@ -172,6 +172,7 @@ type EngineServices struct {
StateManager *statemanager.Manager
UpdateManager *updater.Manager
ClientMetrics *metrics.ClientMetrics
MetricsCtx context.Context
}
// Engine is a mechanism responsible for reacting on Signal and Management stream events and managing connections to the remote peers.
@@ -264,6 +265,7 @@ type Engine struct {
// clientMetrics collects and pushes metrics
clientMetrics *metrics.ClientMetrics
metricsCtx context.Context
jobExecutor *jobexec.Executor
jobExecutorWG sync.WaitGroup
@@ -316,6 +318,7 @@ func NewEngine(
probeStunTurn: relay.NewStunTurnProbe(relay.DefaultCacheTTL),
jobExecutor: jobexec.NewExecutor(),
clientMetrics: services.ClientMetrics,
metricsCtx: services.MetricsCtx,
updateManager: services.UpdateManager,
syncStoreDir: config.StateDir,
}
@@ -997,6 +1000,8 @@ func (e *Engine) updateNetbirdConfig(wCfg *mgmProto.NetbirdConfig) error {
return fmt.Errorf("handle the flow configuration: %w", err)
}
e.handleMetricsUpdate(wCfg.GetMetrics())
if err := e.PopulateNetbirdConfig(wCfg, nil); err != nil {
log.Warnf("Failed to update DNS server config: %v", err)
}
@@ -1066,6 +1071,14 @@ func (e *Engine) handleFlowUpdate(config *mgmProto.FlowConfig) error {
return e.flowManager.Update(flowConfig)
}
func (e *Engine) handleMetricsUpdate(config *mgmProto.MetricsConfig) {
if config == nil {
return
}
log.Infof("received metrics configuration from management: enabled=%v", config.GetEnabled())
e.clientMetrics.UpdatePushFromMgm(e.metricsCtx, config.GetEnabled())
}
func toFlowLoggerConfig(config *mgmProto.FlowConfig) (*nftypes.FlowConfig, error) {
if config.GetInterval() == nil {
return nil, errors.New("flow interval is nil")

View File

@@ -60,6 +60,13 @@ func getMetricsInterval() time.Duration {
return interval
}
// isMetricsPushEnvSet returns true if NB_METRICS_PUSH_ENABLED is explicitly set (to any value).
// When set, the env var takes full precedence over management server configuration.
func isMetricsPushEnvSet() bool {
_, set := os.LookupEnv(EnvMetricsPushEnabled)
return set
}
func isForceSending() bool {
force, _ := strconv.ParseBool(os.Getenv(EnvMetricsForceSending))
return force

View File

@@ -7,6 +7,7 @@ import (
"fmt"
"io"
"sync"
"sync/atomic"
"time"
log "github.com/sirupsen/logrus"
@@ -75,7 +76,7 @@ type ClientMetrics struct {
agentInfo AgentInfo
mu sync.RWMutex
push *Push
push atomic.Pointer[Push]
pushMu sync.Mutex
wg sync.WaitGroup
pushCancel context.CancelFunc
@@ -167,10 +168,7 @@ func (c *ClientMetrics) UpdateAgentInfo(agentInfo AgentInfo, publicKey string) {
c.agentInfo = agentInfo
c.mu.Unlock()
c.pushMu.Lock()
push := c.push
c.pushMu.Unlock()
if push != nil {
if push := c.push.Load(); push != nil {
push.SetPeerID(agentInfo.peerID)
}
}
@@ -184,7 +182,7 @@ func (c *ClientMetrics) Export(w io.Writer) error {
return c.impl.Export(w)
}
// StartPush starts periodic pushing of metrics with the given configuration
// StartPush starts periodic pushing of metrics with the given configuration.
// Precedence: PushConfig.ServerAddress > remote config server_url
func (c *ClientMetrics) StartPush(ctx context.Context, config PushConfig) {
if c == nil {
@@ -194,11 +192,58 @@ func (c *ClientMetrics) StartPush(ctx context.Context, config PushConfig) {
c.pushMu.Lock()
defer c.pushMu.Unlock()
if c.push != nil {
if c.push.Load() != nil {
log.Warnf("metrics push already running")
return
}
c.startPushLocked(ctx, config)
}
// StopPush stops the periodic metrics push.
func (c *ClientMetrics) StopPush() {
if c == nil {
return
}
c.pushMu.Lock()
defer c.pushMu.Unlock()
c.stopPushLocked()
}
// UpdatePushFromMgm updates metrics push based on management server configuration.
// If NB_METRICS_PUSH_ENABLED is explicitly set (true or false), management config is ignored.
// When unset, management controls whether push is enabled.
func (c *ClientMetrics) UpdatePushFromMgm(ctx context.Context, enabled bool) {
if c == nil {
return
}
if isMetricsPushEnvSet() {
log.Debugf("ignoring management config, env var is explicitly set: %s", EnvMetricsPushEnabled)
return
}
c.pushMu.Lock()
defer c.pushMu.Unlock()
if enabled {
if c.push.Load() != nil {
return
}
log.Infof("enabled metrics push by management")
c.startPushLocked(ctx, PushConfigFromEnv())
} else {
if c.push.Load() == nil {
return
}
log.Infof("disabled metrics push by management")
c.stopPushLocked()
}
}
// startPushLocked starts push. Caller must hold pushMu.
func (c *ClientMetrics) startPushLocked(ctx context.Context, config PushConfig) {
c.mu.RLock()
agentVersion := c.agentInfo.Version
peerID := c.agentInfo.peerID
@@ -214,26 +259,23 @@ func (c *ClientMetrics) StartPush(ctx context.Context, config PushConfig) {
ctx, cancel := context.WithCancel(ctx)
c.pushCancel = cancel
c.push.Store(push)
c.wg.Add(1)
go func() {
defer c.wg.Done()
push.Start(ctx)
c.push.CompareAndSwap(push, nil)
}()
c.push = push
}
func (c *ClientMetrics) StopPush() {
if c == nil {
return
}
c.pushMu.Lock()
defer c.pushMu.Unlock()
if c.push == nil {
// stopPushLocked stops push. Caller must hold pushMu.
func (c *ClientMetrics) stopPushLocked() {
if c.push.Load() == nil {
return
}
c.pushCancel()
c.wg.Wait()
c.push = nil
c.push.Store(nil)
}

View File

@@ -386,7 +386,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.NetworkMonitor != nil && input.NetworkMonitor != config.NetworkMonitor {
if input.NetworkMonitor != nil && (config.NetworkMonitor == nil || *input.NetworkMonitor != *config.NetworkMonitor) {
log.Infof("switching Network Monitor to %t", *input.NetworkMonitor)
config.NetworkMonitor = input.NetworkMonitor
updated = true
@@ -454,7 +454,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.EnableSSHRoot != nil && input.EnableSSHRoot != config.EnableSSHRoot {
if input.EnableSSHRoot != nil && (config.EnableSSHRoot == nil || *input.EnableSSHRoot != *config.EnableSSHRoot) {
if *input.EnableSSHRoot {
log.Infof("enabling SSH root login")
} else {
@@ -464,7 +464,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.EnableSSHSFTP != nil && input.EnableSSHSFTP != config.EnableSSHSFTP {
if input.EnableSSHSFTP != nil && (config.EnableSSHSFTP == nil || *input.EnableSSHSFTP != *config.EnableSSHSFTP) {
if *input.EnableSSHSFTP {
log.Infof("enabling SSH SFTP subsystem")
} else {
@@ -474,7 +474,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.EnableSSHLocalPortForwarding != nil && input.EnableSSHLocalPortForwarding != config.EnableSSHLocalPortForwarding {
if input.EnableSSHLocalPortForwarding != nil && (config.EnableSSHLocalPortForwarding == nil || *input.EnableSSHLocalPortForwarding != *config.EnableSSHLocalPortForwarding) {
if *input.EnableSSHLocalPortForwarding {
log.Infof("enabling SSH local port forwarding")
} else {
@@ -484,7 +484,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.EnableSSHRemotePortForwarding != nil && input.EnableSSHRemotePortForwarding != config.EnableSSHRemotePortForwarding {
if input.EnableSSHRemotePortForwarding != nil && (config.EnableSSHRemotePortForwarding == nil || *input.EnableSSHRemotePortForwarding != *config.EnableSSHRemotePortForwarding) {
if *input.EnableSSHRemotePortForwarding {
log.Infof("enabling SSH remote port forwarding")
} else {
@@ -494,7 +494,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.DisableSSHAuth != nil && input.DisableSSHAuth != config.DisableSSHAuth {
if input.DisableSSHAuth != nil && (config.DisableSSHAuth == nil || *input.DisableSSHAuth != *config.DisableSSHAuth) {
if *input.DisableSSHAuth {
log.Infof("disabling SSH authentication")
} else {
@@ -504,7 +504,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.SSHJWTCacheTTL != nil && input.SSHJWTCacheTTL != config.SSHJWTCacheTTL {
if input.SSHJWTCacheTTL != nil && (config.SSHJWTCacheTTL == nil || *input.SSHJWTCacheTTL != *config.SSHJWTCacheTTL) {
log.Infof("updating SSH JWT cache TTL to %d seconds", *input.SSHJWTCacheTTL)
config.SSHJWTCacheTTL = input.SSHJWTCacheTTL
updated = true
@@ -587,7 +587,7 @@ func (config *Config) apply(input ConfigInput) (updated bool, err error) {
updated = true
}
if input.DisableNotifications != nil && input.DisableNotifications != config.DisableNotifications {
if input.DisableNotifications != nil && (config.DisableNotifications == nil || *input.DisableNotifications != *config.DisableNotifications) {
if *input.DisableNotifications {
log.Infof("disabling notifications")
} else {