netbird/management/internals/server/boot.go

package server

// @note this file includes all the lower level dependencies, db, http and grpc BaseServer, metrics, logger, etc.

import (
	"context"
	"crypto/tls"
	"errors"
	"net/http"
	"net/netip"
	"slices"
	"time"

	"github.com/gorilla/mux"
	grpcMiddleware "github.com/grpc-ecosystem/go-grpc-middleware/v2"
	"github.com/grpc-ecosystem/go-grpc-middleware/v2/interceptors/realip"
	"github.com/rs/cors"
	"github.com/rs/xid"
	log "github.com/sirupsen/logrus"
	"google.golang.org/grpc"
	"google.golang.org/grpc/credentials"
	"google.golang.org/grpc/keepalive"

	"github.com/netbirdio/netbird/encryption"
	"github.com/netbirdio/netbird/formatter/hook"
	"github.com/netbirdio/netbird/management/internals/modules/agentnetwork"
	"github.com/netbirdio/netbird/management/internals/modules/reverseproxy/accesslogs"
	accesslogsmanager "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/accesslogs/manager"
	proxyactivity "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/activity"
	proxyactivitymanager "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/activity/manager"
	rpservice "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/service"
	networkmapdb "github.com/netbirdio/netbird/management/internals/network_map_db"
	networkmapdbfactory "github.com/netbirdio/netbird/management/internals/network_map_db/factory"
	nbconfig "github.com/netbirdio/netbird/management/internals/server/config"
	"github.com/netbirdio/netbird/management/internals/shared/db"
	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
	"github.com/netbirdio/netbird/management/server/activity"
	activitystore "github.com/netbirdio/netbird/management/server/activity/store"
	nbcache "github.com/netbirdio/netbird/management/server/cache"
	nbContext "github.com/netbirdio/netbird/management/server/context"
	nbhttp "github.com/netbirdio/netbird/management/server/http"
	"github.com/netbirdio/netbird/management/server/idp"
	"github.com/netbirdio/netbird/management/server/store"
	"github.com/netbirdio/netbird/management/server/telemetry"
	mgmtProto "github.com/netbirdio/netbird/shared/management/proto"
	"github.com/netbirdio/netbird/shared/ratelimit"
	"github.com/netbirdio/netbird/util/crypt"
)

const apiPrefix = "/api"

var (
	kaep = keepalive.EnforcementPolicy{
		MinTime:             15 * time.Second,
		PermitWithoutStream: true,
	}

	kasp = keepalive.ServerParameters{
		MaxConnectionIdle:     15 * time.Second,
		MaxConnectionAgeGrace: 5 * time.Second,
		Time:                  5 * time.Second,
		Timeout:               2 * time.Second,
	}
)

func (s *BaseServer) Metrics() telemetry.AppMetrics {
	return Create(s, func() telemetry.AppMetrics {
		appMetrics, err := telemetry.NewDefaultAppMetrics(context.Background())
		if err != nil {
			log.Fatalf("error while creating app metrics: %s", err)
		}
		return appMetrics
	})
}

// CacheStore returns a shared cache store backed by Redis or in-memory depending on the environment.
// All consumers should reuse this store to avoid creating multiple Redis connections.
func (s *BaseServer) CacheStore() nbcache.Store {
	return Create(s, func() nbcache.Store {
		cs, err := nbcache.NewStore(context.Background(), nbcache.DefaultStoreMaxTimeout, nbcache.DefaultStoreCleanupInterval, nbcache.DefaultStoreMaxConn)
		if err != nil {
			log.Fatalf("failed to create shared cache store: %v", err)
		}
		return cs
	})
}

// DBConn opens the database connection shared by the store and the domain repositories.
func (s *BaseServer) DBConn() *db.Conn {
	return Create(s, func() *db.Conn {
		conn, err := store.OpenConn(context.Background(), s.Config.StoreConfig.Engine, s.Config.Datadir)
		if err != nil {
			log.Fatalf("failed to open database connection: %v", err)
		}
		return conn
	})
}

func (s *BaseServer) Store() store.Store {
	return Create(s, func() store.Store {
		store, err := store.NewSqlStore(context.Background(), s.DBConn(), s.Metrics(), false)
		if err != nil {
			log.Fatalf("failed to create store: %v", err)
		}

		if s.Config.DataStoreEncryptionKey != "" {
			fieldEncrypt, err := crypt.NewFieldEncrypt(s.Config.DataStoreEncryptionKey)
			if err != nil {
				log.Fatalf("failed to create field encryptor: %v", err)
			}
			store.SetFieldEncrypt(fieldEncrypt)
		}

		return store
	})
}

// TODO dmitri: move all validation checks (e.g. config+env vars) from runtime to base server creation
// this way we don't need to spread defensive checks throughout the codebase
func (s *BaseServer) NetworkMapStore() *networkmapdb.NetworkMapDBStoreImpl {
	return Create(s, func() *networkmapdb.NetworkMapDBStoreImpl {
		store, err := networkmapdbfactory.NewNetworkMapDBStore(
			context.Background(),
			s.Config.StoreConfig.Engine,
			s.Config.Datadir,
			s.IntegratedValidator(),
			s.SettingsManager(),
		)
		// networkmap db store supports postgres and sqlite backends only
		// for other backends a fallback is used, so NotSupportedStoreEngineError
		// is not a fatal error
		if err != nil && !errors.Is(err, networkmapdbfactory.ErrNotSupportedStoreEngine) {
			log.Fatalf("failed to create network map store: %v", err)
		}
		return store
	})
}

func (s *BaseServer) EventStore() activity.Store {
	return Create(s, func() activity.Store {
		var err error
		key := s.Config.DataStoreEncryptionKey
		if key == "" {
			log.Debugf("generate new activity store encryption key")
			key, err = crypt.GenerateKey()
			if err != nil {
				log.Fatalf("failed to generate event store encryption key: %v", err)
			}
		}

		eventStore, err := activitystore.NewSqlStore(context.Background(), s.Config.Datadir, key)
		if err != nil {
			log.Fatalf("failed to initialize event store: %v", err)
		}

		return eventStore
	})
}

func (s *BaseServer) APIHandler() http.Handler {
	return Create(s, func() http.Handler {
		httpAPIHandler, err := nbhttp.NewAPIHandler(context.Background(), s.Router(), s.AccountManager(), s.NetworksManager(), s.ResourcesManager(), s.RoutesManager(), s.GroupsManager(), s.GeoLocationManager(), s.AuthManager(), s.Metrics(), s.PermissionsManager(), s.SettingsManager(), s.ZonesManager(), s.RecordsManager(), s.NetworkMapController(), s.IdpManager(), s.ServiceManager(), s.ReverseProxyDomainManager(), s.AccessLogsManager(), s.ReverseProxyGRPCServer(), s.Config.ReverseProxy.TrustedHTTPProxies, s.RateLimiter(), s.IsValidChildAccount, s.AgentNetworkManager(), nil)
		if err != nil {
			log.Fatalf("failed to create API handler: %v", err)
		}
		return httpAPIHandler
	})
}

// IDPHandler returns the HTTP handler for the embedded IdP (Dex), or nil if
// the deployment isn't using the embedded variant.
func (s *BaseServer) IDPHandler() http.Handler {
	embeddedIdP, ok := s.IdpManager().(*idp.EmbeddedIdPManager)
	if !ok || embeddedIdP == nil {
		return nil
	}
	return cors.AllowAll().Handler(embeddedIdP.Handler())
}

func (s *BaseServer) Router() *mux.Router {
	return Create(s, func() *mux.Router {
		return mux.NewRouter().PathPrefix(apiPrefix).Subrouter()
	})
}

func (s *BaseServer) RateLimiter() *ratelimit.APIRateLimiter {
	return Create(s, func() *ratelimit.APIRateLimiter {
		cfg, enabled := ratelimit.RateLimiterConfigFromEnv()
		limiter := ratelimit.NewAPIRateLimiter(cfg)
		limiter.SetEnabled(enabled)
		return limiter
	})
}

func (s *BaseServer) GRPCServer() *grpc.Server {
	return Create(s, func() *grpc.Server {
		realipOpts := realIPOptions(s.Config.ReverseProxy)
		proxyUnary, proxyStream, proxyAuthClose := nbgrpc.NewProxyAuthInterceptors(s.Store())
		s.proxyAuthClose = proxyAuthClose
		gRPCOpts := []grpc.ServerOption{
			grpc.KeepaliveEnforcementPolicy(kaep),
			grpc.KeepaliveParams(kasp),
			grpc.ChainUnaryInterceptor(realip.UnaryServerInterceptorOpts(realipOpts...), unaryInterceptor, proxyUnary),
			grpc.ChainStreamInterceptor(realip.StreamServerInterceptorOpts(realipOpts...), streamInterceptor, proxyStream),
		}

		// Append interceptors contributed by registered gRPC extensions. These
		// run after the built-in chain (ChainUnaryInterceptor is additive).
		gRPCOpts = appendExtensionInterceptors(gRPCOpts, s.grpcExtensions)

		if s.Config.HttpConfig.LetsEncryptDomain != "" {
			certManager, err := encryption.CreateCertManager(s.Config.Datadir, s.Config.HttpConfig.LetsEncryptDomain)
			if err != nil {
				log.Fatalf("failed to create certificate service: %v", err)
			}
			transportCredentials := credentials.NewTLS(certManager.TLSConfig())
			gRPCOpts = append(gRPCOpts, grpc.Creds(transportCredentials))
		} else if s.Config.HttpConfig.CertFile != "" && s.Config.HttpConfig.CertKey != "" {
			tlsConfig, err := loadTLSConfig(s.Config.HttpConfig.CertFile, s.Config.HttpConfig.CertKey)
			if err != nil {
				log.Fatalf("cannot load TLS credentials: %v", err)
			}
			transportCredentials := credentials.NewTLS(tlsConfig)
			gRPCOpts = append(gRPCOpts, grpc.Creds(transportCredentials))
		}

		gRPCAPIHandler := grpc.NewServer(gRPCOpts...)
		srv, err := nbgrpc.NewServer(s.Config, s.AccountManager(), s.SettingsManager(), s.JobManager(), s.SecretsManager(), s.Metrics(), s.AuthManager(), s.IntegratedValidator(), s.NetworkMapController(), s.OAuthConfigProvider(), s.SessionStore())
		if err != nil {
			log.Fatalf("failed to create management server: %v", err)
		}
		serviceMgr := s.ServiceManager()
		srv.SetReverseProxyManager(serviceMgr)
		if serviceMgr != nil {
			serviceMgr.StartExposeReaper(context.Background())
		}
		mgmtProto.RegisterManagementServiceServer(gRPCAPIHandler, srv)

		mgmtProto.RegisterProxyServiceServer(gRPCAPIHandler, s.ReverseProxyGRPCServer())
		log.Info("ProxyService registered on gRPC server")

		// Register services contributed by external modules via the extension seam.
		registerExtensions(gRPCAPIHandler, s.grpcExtensions)

		return gRPCAPIHandler
	})
}

func (s *BaseServer) ReverseProxyGRPCServer() *nbgrpc.ProxyServiceServer {
	return Create(s, func() *nbgrpc.ProxyServiceServer {
		proxyService := nbgrpc.NewProxyServiceServer(s.AccessLogsManager(), s.ProxyTokenStore(), s.PKCEVerifierStore(), s.proxyOIDCConfig(), s.PeersManager(), s.UsersManager(), s.IdpManager(), s.ProxyManager(), s.Store())
		s.AfterInit(func(s *BaseServer) {
			proxyService.SetServiceManager(s.ServiceManager())
			proxyService.SetActivityManager(s.ProxyActivityManager())
			proxyService.SetProxyController(s.ServiceProxyController())
			proxyService.SetAgentNetworkSynthesizer(newAgentNetworkSynthesizer(s.Store()))
			proxyService.SetAgentNetworkLimitsService(s.AgentNetworkManager())
		})
		return proxyService
	})
}

// agentNetworkSynthesizerAdapter implements nbgrpc.AgentNetworkSynthesizer by
// delegating to the agentnetwork package's store-backed synthesiser.
type agentNetworkSynthesizerAdapter struct {
	store store.Store
}

func newAgentNetworkSynthesizer(s store.Store) *agentNetworkSynthesizerAdapter {
	return &agentNetworkSynthesizerAdapter{store: s}
}

func (a *agentNetworkSynthesizerAdapter) SynthesizeServicesForCluster(ctx context.Context, clusterAddr string) ([]*rpservice.Service, error) {
	return agentnetwork.SynthesizeServicesForCluster(ctx, a.store, clusterAddr)
}

func (a *agentNetworkSynthesizerAdapter) SynthesizeServicesForAccount(ctx context.Context, accountID string) ([]*rpservice.Service, error) {
	return agentnetwork.SynthesizeServices(ctx, a.store, accountID)
}

func (a *agentNetworkSynthesizerAdapter) SynthesizeServiceForDomain(ctx context.Context, domain string) (*rpservice.Service, error) {
	return agentnetwork.SynthesizeServiceForDomain(ctx, a.store, domain)
}

func (s *BaseServer) proxyOIDCConfig() nbgrpc.ProxyOIDCConfig {
	return Create(s, func() nbgrpc.ProxyOIDCConfig {
		return nbgrpc.ProxyOIDCConfig{
			Issuer: s.Config.HttpConfig.AuthIssuer,
			// todo: double check auth clientID value
			ClientID:     s.Config.HttpConfig.AuthClientID, // Reuse dashboard client
			Scopes:       []string{"openid", "profile", "email"},
			CallbackURL:  s.Config.HttpConfig.AuthCallbackURL,
			HMACKey:      []byte(s.Config.DataStoreEncryptionKey), // Use the datastore encryption key for OIDC state HMACs, this should ensure all management instances are using the same key.
			Audience:     s.Config.HttpConfig.AuthAudience,
			KeysLocation: s.Config.HttpConfig.AuthKeysLocation,
		}
	})
}

func (s *BaseServer) ProxyTokenStore() *nbgrpc.OneTimeTokenStore {
	return Create(s, func() *nbgrpc.OneTimeTokenStore {
		tokenStore := nbgrpc.NewOneTimeTokenStore(context.Background(), s.CacheStore())
		log.Info("One-time token store initialized for proxy authentication")
		return tokenStore
	})
}

func (s *BaseServer) PKCEVerifierStore() *nbgrpc.PKCEVerifierStore {
	return Create(s, func() *nbgrpc.PKCEVerifierStore {
		return nbgrpc.NewPKCEVerifierStore(context.Background(), s.CacheStore())
	})
}

// ProxyActivityManager records reverse proxy usage for activity accounting.
func (s *BaseServer) ProxyActivityManager() proxyactivity.Manager {
	return Create(s, func() proxyactivity.Manager {
		return proxyactivitymanager.NewManager(s.Store())
	})
}

func (s *BaseServer) AccessLogsManager() accesslogs.Manager {
	return Create(s, func() accesslogs.Manager {
		accessLogManager := accesslogsmanager.NewManager(accesslogsmanager.NewRepository(s.DBConn()), s.Store(), s.PermissionsManager(), s.GeoLocationManager())
		accessLogManager.StartPeriodicCleanup(
			context.Background(),
			s.Config.ReverseProxy.AccessLogRetentionDays,
			s.Config.ReverseProxy.AccessLogCleanupIntervalHours,
		)
		return accessLogManager
	})
}

func loadTLSConfig(certFile, certKey string) (*tls.Config, error) {
	// Load server's certificate and private key
	serverCert, err := tls.LoadX509KeyPair(certFile, certKey)
	if err != nil {
		return nil, err
	}

	// NewDefaultAppMetrics the credentials and return it
	config := &tls.Config{
		Certificates: []tls.Certificate{serverCert},
		ClientAuth:   tls.NoClientCert,
		NextProtos: []string{
			"h2", "http/1.1", // enable HTTP/2
		},
	}

	return config, nil
}

func unaryInterceptor(
	ctx context.Context,
	req interface{},
	info *grpc.UnaryServerInfo,
	handler grpc.UnaryHandler,
) (interface{}, error) {
	reqID := xid.New().String()
	//nolint
	ctx = context.WithValue(ctx, hook.ExecutionContextKey, hook.GRPCSource)
	//nolint
	ctx = context.WithValue(ctx, nbContext.RequestIDKey, reqID)
	return handler(ctx, req)
}

func streamInterceptor(
	srv interface{},
	ss grpc.ServerStream,
	info *grpc.StreamServerInfo,
	handler grpc.StreamHandler,
) error {
	reqID := xid.New().String()
	wrapped := grpcMiddleware.WrapServerStream(ss)
	//nolint
	ctx := context.WithValue(ss.Context(), hook.ExecutionContextKey, hook.GRPCSource)
	//nolint
	wrapped.WrappedContext = context.WithValue(ctx, nbContext.RequestIDKey, reqID)
	return handler(srv, wrapped)
}

// realIPOptions builds the real-IP middleware options.
//
// Empty TrustedPeers trusts all IPv4 and IPv6 sources. Configure TrustedPeers
// with the reverse proxy address or network.
//
// X-Forwarded-For takes precedence over X-Real-IP.
func realIPOptions(cfg nbconfig.ReverseProxy) []realip.Option {
	trustedPeers := cfg.TrustedPeers
	if len(trustedPeers) == 0 {
		trustedPeers = []netip.Prefix{
			netip.MustParsePrefix("0.0.0.0/0"),
			netip.MustParsePrefix("::/0"),
		}
	}
	if idx := slices.IndexFunc(trustedPeers, func(p netip.Prefix) bool { return p.Bits() == 0 }); idx >= 0 {
		log.WithContext(context.Background()).Warnf("TrustedPeers contains the default route %s, which trusts "+
			"X-Forwarded-For from every client and allows connection IP spoofing. Set TrustedPeers to the address "+
			"of your reverse proxy.", trustedPeers[idx])
	}
	if cfg.TrustedHTTPProxiesCount > 0 {
		log.WithContext(context.Background()).Warn(
			"TrustedHTTPProxiesCount skips X-Forwarded-For entries by position before TrustedHTTPProxies filters by address. " +
				"An incorrect count may skip the real client IP and produce an incorrect source address.",
		)
	}

	return []realip.Option{
		realip.WithTrustedPeers(trustedPeers),
		realip.WithTrustedProxies(cfg.TrustedHTTPProxies),
		realip.WithTrustedProxiesCount(cfg.TrustedHTTPProxiesCount),
		realip.WithHeaders([]string{realip.XForwardedFor, realip.XRealIp}),
	}
}