Merge remote-tracking branch 'origin/main' into refactor/permissions-manager

management/internals/modules/reverseproxy/service/manager/api.go

@@ -37,6 +37,7 @@ func RegisterEndpoints(manager rpservice.Manager, domainManager domainmanager.Ma
 	accesslogsmanager.RegisterEndpoints(router, accessLogsManager, permissionsManager)
 
 	router.HandleFunc("/reverse-proxies/clusters", permissionsManager.WithPermission(modules.Services, operations.Read, h.getClusters)).Methods("GET", "OPTIONS")
+	router.HandleFunc("/reverse-proxies/clusters/{clusterAddress}", permissionsManager.WithPermission(modules.Services, operations.Delete, h.deleteCluster)).Methods("DELETE", "OPTIONS")
 	router.HandleFunc("/reverse-proxies/services", permissionsManager.WithPermission(modules.Services, operations.Read, h.getAllServices)).Methods("GET", "OPTIONS")
 	router.HandleFunc("/reverse-proxies/services", permissionsManager.WithPermission(modules.Services, operations.Create, h.createService)).Methods("POST", "OPTIONS")
 	router.HandleFunc("/reverse-proxies/services/{serviceId}", permissionsManager.WithPermission(modules.Services, operations.Read, h.getService)).Methods("GET", "OPTIONS")
@@ -154,7 +155,7 @@ func (h *handler) deleteService(w http.ResponseWriter, r *http.Request, userAuth
 }
 
 func (h *handler) getClusters(w http.ResponseWriter, r *http.Request, userAuth *auth.UserAuth) {
-	clusters, err := h.manager.GetActiveClusters(r.Context(), userAuth.AccountId, userAuth.UserId)
+	clusters, err := h.manager.GetClusters(r.Context(), userAuth.AccountId, userAuth.UserId)
 	if err != nil {
 		util.WriteError(r.Context(), err, w)
 		return
@@ -163,10 +164,32 @@ func (h *handler) getClusters(w http.ResponseWriter, r *http.Request, userAuth *
 	apiClusters := make([]api.ProxyCluster, 0, len(clusters))
 	for _, c := range clusters {
 		apiClusters = append(apiClusters, api.ProxyCluster{
-			Address:          c.Address,
-			ConnectedProxies: c.ConnectedProxies,
+			Id:                  c.ID,
+			Address:             c.Address,
+			Type:                api.ProxyClusterType(c.Type),
+			Online:              c.Online,
+			ConnectedProxies:    c.ConnectedProxies,
+			SupportsCustomPorts: c.SupportsCustomPorts,
+			RequireSubdomain:    c.RequireSubdomain,
+			SupportsCrowdsec:    c.SupportsCrowdSec,
+			Private:             c.Private,
 		})
 	}
 
 	util.WriteJSONObject(r.Context(), w, apiClusters)
 }
+
+func (h *handler) deleteCluster(w http.ResponseWriter, r *http.Request, userAuth *auth.UserAuth) {
+	clusterAddress := mux.Vars(r)["clusterAddress"]
+	if clusterAddress == "" {
+		util.WriteError(r.Context(), status.Errorf(status.InvalidArgument, "cluster address is required"), w)
+		return
+	}
+
+	if err := h.manager.DeleteAccountCluster(r.Context(), userAuth.AccountId, userAuth.UserId, clusterAddress); err != nil {
+		util.WriteError(r.Context(), err, w)
+		return
+	}
+
+	util.WriteJSONObject(r.Context(), w, util.EmptyObject{})
+}

management/internals/modules/reverseproxy/service/manager/domain_validation_test.go

@@ -0,0 +1,126 @@
+package manager
+
+import (
+	"context"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	"go.opentelemetry.io/otel/metric/noop"
+
+	domainmanager "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/domain/manager"
+	proxymanager "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/proxy/manager"
+	rpservice "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/service"
+	"github.com/netbirdio/netbird/management/server/activity"
+	"github.com/netbirdio/netbird/management/server/mock_server"
+	"github.com/netbirdio/netbird/management/server/store"
+	"github.com/netbirdio/netbird/shared/management/status"
+)
+
+const validationTestCluster = "eu.proxy.test"
+
+// withRealDomainManager swaps the stub cluster deriver for the real domain
+// manager backed by the same store, so service creation is gated by the actual
+// domain rows rather than by a test double that always agrees.
+func withRealDomainManager(t *testing.T, mgr *Manager, testStore store.Store) {
+	t.Helper()
+
+	ctx := context.Background()
+	proxyMgr, err := proxymanager.NewManager(testStore, noop.NewMeterProvider().Meter(""))
+	require.NoError(t, err)
+
+	_, err = proxyMgr.Connect(ctx, "proxy-1", "session-1", validationTestCluster, "127.0.0.1", nil, nil)
+	require.NoError(t, err)
+
+	accountMgr := &mock_server.MockAccountManager{
+		StoreEventFunc: func(context.Context, string, string, string, activity.ActivityDescriber, map[string]any) {},
+	}
+	mgr.clusterDeriver = domainmanager.NewManager(testStore, proxyMgr, accountMgr)
+}
+
+func newTestService(domain string) *rpservice.Service {
+	return &rpservice.Service{
+		Name:    "test-service",
+		Domain:  domain,
+		Enabled: true,
+		Mode:    rpservice.ModeHTTP,
+		Targets: []*rpservice.Target{{
+			Host:       "10.0.0.1",
+			Port:       8080,
+			Protocol:   "http",
+			TargetId:   testPeerID,
+			TargetType: "peer",
+			Enabled:    true,
+		}},
+	}
+}
+
+// A service must not bind to a domain the account has not validated, and
+// nothing may be persisted for the attempt.
+func TestCreateService_RefusesUnvalidatedDomain(t *testing.T) {
+	ctx := context.Background()
+	mgr, testStore := setupIntegrationTest(t)
+	withRealDomainManager(t, mgr, testStore)
+
+	_, err := testStore.CreateCustomDomain(ctx, testAccountID, "unproven.example.com", validationTestCluster, false)
+	require.NoError(t, err)
+
+	_, err = mgr.CreateService(ctx, testAccountID, testUserID, newTestService("unproven.example.com"))
+	require.Error(t, err, "an unvalidated domain must not bind a service")
+	assert.Contains(t, err.Error(), "not validated", "the API error should name the actual problem")
+
+	sErr, ok := status.FromError(err)
+	require.True(t, ok, "error should be a typed status error")
+	assert.Equal(t, status.PreconditionFailed, sErr.Type())
+
+	services, err := testStore.GetAccountServices(ctx, store.LockingStrengthNone, testAccountID)
+	require.NoError(t, err)
+	assert.Empty(t, services, "no service row should be written for a refused domain")
+}
+
+// The negative control: a validated domain still binds a service and derives
+// its cluster exactly as before.
+func TestCreateService_ValidatedDomainBindsService(t *testing.T) {
+	ctx := context.Background()
+	mgr, testStore := setupIntegrationTest(t)
+	withRealDomainManager(t, mgr, testStore)
+
+	_, err := testStore.CreateCustomDomain(ctx, testAccountID, "proven.example.com", validationTestCluster, true)
+	require.NoError(t, err)
+
+	created, err := mgr.CreateService(ctx, testAccountID, testUserID, newTestService("app.proven.example.com"))
+	require.NoError(t, err)
+	assert.Equal(t, validationTestCluster, created.ProxyCluster, "service should bind to the domain's target cluster")
+
+	services, err := testStore.GetAccountServices(ctx, store.LockingStrengthNone, testAccountID)
+	require.NoError(t, err)
+	require.Len(t, services, 1, "the service should be persisted")
+	assert.Equal(t, "app.proven.example.com", services[0].Domain)
+}
+
+// An update must not be a way around the creation gate: moving a live service
+// onto an unvalidated domain has to fail rather than silently keep the old
+// cluster and start serving the new hostname.
+func TestUpdateService_RefusesMoveToUnvalidatedDomain(t *testing.T) {
+	ctx := context.Background()
+	mgr, testStore := setupIntegrationTest(t)
+	withRealDomainManager(t, mgr, testStore)
+
+	_, err := testStore.CreateCustomDomain(ctx, testAccountID, "proven.example.com", validationTestCluster, true)
+	require.NoError(t, err)
+	_, err = testStore.CreateCustomDomain(ctx, testAccountID, "unproven.example.com", validationTestCluster, false)
+	require.NoError(t, err)
+
+	created, err := mgr.CreateService(ctx, testAccountID, testUserID, newTestService("app.proven.example.com"))
+	require.NoError(t, err)
+
+	moved := *created
+	moved.Domain = "app.unproven.example.com"
+	_, err = mgr.UpdateService(ctx, testAccountID, testUserID, &moved)
+	require.Error(t, err, "moving to an unvalidated domain must fail")
+	assert.Contains(t, err.Error(), "not validated")
+
+	stored, err := testStore.GetServiceByID(ctx, store.LockingStrengthNone, testAccountID, created.ID)
+	require.NoError(t, err)
+	assert.Equal(t, "app.proven.example.com", stored.Domain, "the service must keep its original domain")
+}

management/internals/modules/reverseproxy/service/manager/l4_port_test.go

@@ -2,13 +2,13 @@ package manager
 
 import (
 	"context"
-	"net"
+	"net/netip"
 	"testing"
 	"time"
 
-	"github.com/golang/mock/gomock"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
+	"go.uber.org/mock/gomock"
 
 	"github.com/netbirdio/netbird/management/internals/modules/reverseproxy/proxy"
 	rpservice "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/service"
@@ -55,7 +55,8 @@ func setupL4Test(t *testing.T, customPortsSupported *bool) (*Manager, store.Stor
 				Key:       "test-key",
 				DNSLabel:  "test-peer",
 				Name:      "test-peer",
-				IP:        net.ParseIP("100.64.0.1"),
+				IP:        netip.MustParseAddr("100.64.0.1"),
+				IPv6:      netip.MustParseAddr("fd00::1"),
 				Status:    &nbpeer.PeerStatus{Connected: true, LastSeen: time.Now()},
 				Meta:      nbpeer.PeerSystemMeta{Hostname: "test-peer"},
 			},
@@ -84,7 +85,7 @@ func setupL4Test(t *testing.T, customPortsSupported *bool) (*Manager, store.Stor
 
 	accountMgr := &mock_server.MockAccountManager{
 		StoreEventFunc:         func(_ context.Context, _, _, _ string, _ activity.ActivityDescriber, _ map[string]any) {},
-		UpdateAccountPeersFunc: func(_ context.Context, _ string) {},
+		UpdateAccountPeersFunc: func(_ context.Context, _ string, _ types.UpdateReason) {},
 		GetGroupByNameFunc: func(ctx context.Context, groupName, accountID string) (*types.Group, error) {
 			return testStore.GetGroupByName(ctx, store.LockingStrengthNone, accountID, groupName)
 		},
@@ -485,6 +486,195 @@ func TestUpdate_AllowsPortChange(t *testing.T) {
 	assert.Equal(t, uint16(54321), updated.ListenPort, "explicit port change should be applied")
 }
 
+func TestUpdate_PreservesPortWhenCustomPortsNotSupported(t *testing.T) {
+	mgr, testStore, _ := setupL4Test(t, boolPtr(false))
+	ctx := context.Background()
+
+	existing := seedService(t, testStore, "tcp-svc", "tcp", testCluster, testCluster, 12345)
+
+	updated := &rpservice.Service{
+		ID:           existing.ID,
+		AccountID:    testAccountID,
+		Name:         "tcp-svc-renamed",
+		Mode:         "tcp",
+		Domain:       testCluster,
+		ProxyCluster: testCluster,
+		ListenPort:   0,
+		Enabled:      true,
+		Targets: []*rpservice.Target{
+			{AccountID: testAccountID, TargetId: testPeerID, TargetType: rpservice.TargetTypePeer, Protocol: "tcp", Port: 9090, Enabled: true},
+		},
+	}
+
+	_, err := mgr.persistServiceUpdate(ctx, testAccountID, updated)
+	require.NoError(t, err, "update must not be rejected by the custom-port capability check")
+	assert.Equal(t, uint16(12345), updated.ListenPort, "existing listen port should be preserved on unsupported cluster")
+}
+
+func TestUpdate_PreservesPortWhenCustomPortsUnknown(t *testing.T) {
+	mgr, testStore, _ := setupL4Test(t, nil)
+	ctx := context.Background()
+
+	existing := seedService(t, testStore, "tcp-svc", "tcp", testCluster, testCluster, 12345)
+
+	updated := &rpservice.Service{
+		ID:           existing.ID,
+		AccountID:    testAccountID,
+		Name:         "tcp-svc-renamed",
+		Mode:         "tcp",
+		Domain:       testCluster,
+		ProxyCluster: testCluster,
+		ListenPort:   0,
+		Enabled:      true,
+		Targets: []*rpservice.Target{
+			{AccountID: testAccountID, TargetId: testPeerID, TargetType: rpservice.TargetTypePeer, Protocol: "tcp", Port: 9090, Enabled: true},
+		},
+	}
+
+	_, err := mgr.persistServiceUpdate(ctx, testAccountID, updated)
+	require.NoError(t, err, "update must not be rejected when cluster capability is unknown")
+	assert.Equal(t, uint16(12345), updated.ListenPort, "existing listen port should be preserved when capability is unknown")
+}
+
+func TestUpdate_RejectsPortChangeWhenCustomPortsNotSupported(t *testing.T) {
+	mgr, testStore, _ := setupL4Test(t, boolPtr(false))
+	ctx := context.Background()
+
+	existing := seedService(t, testStore, "tcp-svc", "tcp", testCluster, testCluster, 12345)
+
+	updated := &rpservice.Service{
+		ID:           existing.ID,
+		AccountID:    testAccountID,
+		Name:         "tcp-svc",
+		Mode:         "tcp",
+		Domain:       testCluster,
+		ProxyCluster: testCluster,
+		ListenPort:   54321,
+		Enabled:      true,
+		Targets: []*rpservice.Target{
+			{AccountID: testAccountID, TargetId: testPeerID, TargetType: rpservice.TargetTypePeer, Protocol: "tcp", Port: 9090, Enabled: true},
+		},
+	}
+
+	_, err := mgr.persistServiceUpdate(ctx, testAccountID, updated)
+	require.Error(t, err, "explicit port change on update must be rejected on unsupported clusters")
+	assert.Contains(t, err.Error(), "custom ports not supported on target cluster")
+}
+
+func TestUpdate_TLSPortChangeAllowedWhenNotSupported(t *testing.T) {
+	mgr, testStore, _ := setupL4Test(t, boolPtr(false))
+	ctx := context.Background()
+
+	existing := seedService(t, testStore, "tls-svc", "tls", "app.example.com", testCluster, 443)
+
+	updated := &rpservice.Service{
+		ID:           existing.ID,
+		AccountID:    testAccountID,
+		Name:         "tls-svc",
+		Mode:         "tls",
+		Domain:       "app.example.com",
+		ProxyCluster: testCluster,
+		ListenPort:   9999,
+		Enabled:      true,
+		Targets: []*rpservice.Target{
+			{AccountID: testAccountID, TargetId: testPeerID, TargetType: rpservice.TargetTypePeer, Protocol: "tcp", Port: 8443, Enabled: true},
+		},
+	}
+
+	_, err := mgr.persistServiceUpdate(ctx, testAccountID, updated)
+	require.NoError(t, err, "TLS port change uses SNI routing and is exempt from the custom-port check")
+	assert.Equal(t, uint16(9999), updated.ListenPort, "TLS port change should be applied")
+}
+
+func TestValidateL4PortDiffOnClusterDiff(t *testing.T) {
+	tests := []struct {
+		name        string
+		mode        string
+		customPorts *bool
+		newPort     uint16
+		oldPort     uint16
+		wantErr     bool
+	}{
+		{"tcp port change unsupported", "tcp", boolPtr(false), 54321, 12345, true},
+		{"tcp port change unknown capability", "tcp", nil, 54321, 12345, true},
+		{"udp port change unsupported", "udp", boolPtr(false), 54321, 12345, true},
+		{"tcp first port assignment unsupported", "tcp", boolPtr(false), 54321, 0, true},
+		{"tcp port change supported", "tcp", boolPtr(true), 54321, 12345, false},
+		{"tcp port unchanged unsupported", "tcp", boolPtr(false), 12345, 12345, false},
+		{"tcp zero port unsupported", "tcp", boolPtr(false), 0, 12345, false},
+		{"tls port change unsupported", "tls", boolPtr(false), 9999, 443, false},
+		{"http mode ignored", "http", boolPtr(false), 54321, 12345, false},
+		{"empty mode ignored", "", boolPtr(false), 54321, 12345, false},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			newSvc := &rpservice.Service{Mode: tc.mode, ListenPort: tc.newPort, ProxyCluster: testCluster}
+			oldSvc := &rpservice.Service{Mode: tc.mode, ListenPort: tc.oldPort, ProxyCluster: testCluster}
+
+			err := validateL4PortDiffOnClusterDiff(tc.customPorts, newSvc, oldSvc)
+			if tc.wantErr {
+				assert.Error(t, err, "port diff should be rejected for %s", tc.name)
+			} else {
+				assert.NoError(t, err, "port diff should be allowed for %s", tc.name)
+			}
+		})
+	}
+}
+
+func TestUpdate_PortConflictRejected(t *testing.T) {
+	mgr, testStore, _ := setupL4Test(t, boolPtr(true))
+	ctx := context.Background()
+
+	seedService(t, testStore, "tcp-a", "tcp", "tcp-a."+testCluster, testCluster, 5432)
+	svcB := seedService(t, testStore, "tcp-b", "tcp", "tcp-b."+testCluster, testCluster, 6543)
+
+	updated := &rpservice.Service{
+		ID:           svcB.ID,
+		AccountID:    testAccountID,
+		Name:         "tcp-b",
+		Mode:         "tcp",
+		Domain:       "tcp-b." + testCluster,
+		ProxyCluster: testCluster,
+		ListenPort:   5432,
+		Enabled:      true,
+		Targets: []*rpservice.Target{
+			{AccountID: testAccountID, TargetId: testPeerID, TargetType: rpservice.TargetTypePeer, Protocol: "tcp", Port: 9090, Enabled: true},
+		},
+	}
+
+	_, err := mgr.persistServiceUpdate(ctx, testAccountID, updated)
+	require.Error(t, err, "updating to a port held by another service should be rejected")
+	assert.Contains(t, err.Error(), "already in use")
+}
+
+func TestUpdate_AutoAssignsWhenNoPort(t *testing.T) {
+	mgr, testStore, _ := setupL4Test(t, boolPtr(false))
+	ctx := context.Background()
+
+	existing := seedService(t, testStore, "tcp-svc", "tcp", testCluster, testCluster, 0)
+
+	updated := &rpservice.Service{
+		ID:           existing.ID,
+		AccountID:    testAccountID,
+		Name:         "tcp-svc",
+		Mode:         "tcp",
+		Domain:       testCluster,
+		ProxyCluster: testCluster,
+		ListenPort:   0,
+		Enabled:      true,
+		Targets: []*rpservice.Target{
+			{AccountID: testAccountID, TargetId: testPeerID, TargetType: rpservice.TargetTypePeer, Protocol: "tcp", Port: 9090, Enabled: true},
+		},
+	}
+
+	_, err := mgr.persistServiceUpdate(ctx, testAccountID, updated)
+	require.NoError(t, err)
+	assert.True(t, updated.ListenPort >= autoAssignPortMin && updated.ListenPort <= autoAssignPortMax,
+		"auto-assigned port %d should be in range [%d, %d]", updated.ListenPort, autoAssignPortMin, autoAssignPortMax)
+	assert.True(t, updated.PortAutoAssigned, "PortAutoAssigned should be set when update triggers auto-assignment")
+}
+
 func TestCreateServiceFromPeer_TCP(t *testing.T) {
 	mgr, _, _ := setupL4Test(t, boolPtr(false))
 	ctx := context.Background()

management/internals/modules/reverseproxy/service/manager/manager.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"fmt"
 	"math/rand/v2"
+	"net"
 	"net/http"
 	"os"
 	"slices"
@@ -22,6 +23,7 @@ import (
 	"github.com/netbirdio/netbird/management/server/account"
 	"github.com/netbirdio/netbird/management/server/activity"
 	"github.com/netbirdio/netbird/management/server/store"
+	"github.com/netbirdio/netbird/management/server/types"
 	"github.com/netbirdio/netbird/shared/management/status"
 )
 
@@ -76,6 +78,8 @@ type ClusterDeriver interface {
 type CapabilityProvider interface {
 	ClusterSupportsCustomPorts(ctx context.Context, clusterAddr string) *bool
 	ClusterRequireSubdomain(ctx context.Context, clusterAddr string) *bool
+	ClusterSupportsCrowdSec(ctx context.Context, clusterAddr string) *bool
+	ClusterSupportsPrivate(ctx context.Context, clusterAddr string) *bool
 }
 
 type Manager struct {
@@ -105,9 +109,31 @@ func (m *Manager) StartExposeReaper(ctx context.Context) {
 	m.exposeReaper.StartExposeReaper(ctx)
 }
 
-// GetActiveClusters returns all active proxy clusters with their connected proxy count.
-func (m *Manager) GetActiveClusters(ctx context.Context, accountID, userID string) ([]proxy.Cluster, error) {
-	return m.store.GetActiveProxyClusters(ctx)
+// GetClusters returns every proxy cluster visible to the account
+// (shared + its own BYOP), regardless of whether any proxy in the
+// cluster is currently heartbeating. Each cluster is enriched with the
+// capability flags reported by its active proxies so the dashboard can
+// render feature support without a second round-trip.
+func (m *Manager) GetClusters(ctx context.Context, accountID, userID string) ([]proxy.Cluster, error) {
+	clusters, err := m.store.GetProxyClusters(ctx, accountID)
+	if err != nil {
+		return nil, err
+	}
+
+	for i := range clusters {
+		clusters[i].SupportsCustomPorts = m.capabilities.ClusterSupportsCustomPorts(ctx, clusters[i].Address)
+		clusters[i].RequireSubdomain = m.capabilities.ClusterRequireSubdomain(ctx, clusters[i].Address)
+		clusters[i].SupportsCrowdSec = m.capabilities.ClusterSupportsCrowdSec(ctx, clusters[i].Address)
+		clusters[i].Private = m.capabilities.ClusterSupportsPrivate(ctx, clusters[i].Address)
+	}
+
+	return clusters, nil
+}
+
+// DeleteAccountCluster removes all proxy registrations for the given cluster address
+// owned by the account.
+func (m *Manager) DeleteAccountCluster(ctx context.Context, accountID, userID, clusterAddress string) error {
+	return m.store.DeleteAccountCluster(ctx, clusterAddress, accountID)
 }
 
 func (m *Manager) GetAllServices(ctx context.Context, accountID, userID string) ([]*service.Service, error) {
@@ -155,6 +181,9 @@ func (m *Manager) replaceHostByLookup(ctx context.Context, accountID string, s *
 			target.Host = resource.Domain
 		case service.TargetTypeSubnet:
 			// For subnets we do not do any lookups on the resource
+		case service.TargetTypeCluster:
+			// Cluster targets carry the upstream address on target_id; the
+			// proxy resolves the destination at request time.
 		default:
 			return fmt.Errorf("unknown target type: %s", target.TargetType)
 		}
@@ -194,7 +223,7 @@ func (m *Manager) CreateService(ctx context.Context, accountID, userID string, s
 
 	m.proxyController.SendServiceUpdateToCluster(ctx, accountID, s.ToProtoMapping(service.Create, "", m.proxyController.GetOIDCValidationConfig()), s.ProxyCluster)
 
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationCreate})
 
 	return s, nil
 }
@@ -253,6 +282,10 @@ func (m *Manager) validateSubdomainRequirement(ctx context.Context, domain, clus
 func (m *Manager) persistNewService(ctx context.Context, accountID string, svc *service.Service) error {
 	customPorts := m.clusterCustomPorts(ctx, svc)
 
+	if err := validateTargetReferences(ctx, m.store, accountID, svc.Targets); err != nil {
+		return err
+	}
+
 	return m.store.ExecuteInTransaction(ctx, func(transaction store.Store) error {
 		if svc.Domain != "" {
 			if err := m.checkDomainAvailable(ctx, transaction, svc.Domain, ""); err != nil {
@@ -260,7 +293,7 @@ func (m *Manager) persistNewService(ctx context.Context, accountID string, svc *
 			}
 		}
 
-		if err := m.ensureL4Port(ctx, transaction, svc, customPorts); err != nil {
+		if err := m.ensureL4Port(ctx, transaction, svc, customPorts, false); err != nil {
 			return err
 		}
 
@@ -268,10 +301,6 @@ func (m *Manager) persistNewService(ctx context.Context, accountID string, svc *
 			return err
 		}
 
-		if err := validateTargetReferences(ctx, transaction, accountID, svc.Targets); err != nil {
-			return err
-		}
-
 		if err := transaction.CreateService(ctx, svc); err != nil {
 			return fmt.Errorf("create service: %w", err)
 		}
@@ -293,11 +322,11 @@ func (m *Manager) clusterCustomPorts(ctx context.Context, svc *service.Service)
 
 // ensureL4Port auto-assigns a listen port when needed and validates cluster support.
 // customPorts must be pre-computed via clusterCustomPorts before entering a transaction.
-func (m *Manager) ensureL4Port(ctx context.Context, tx store.Store, svc *service.Service, customPorts *bool) error {
+func (m *Manager) ensureL4Port(ctx context.Context, tx store.Store, svc *service.Service, customPorts *bool, serviceUpdate bool) error {
 	if !service.IsL4Protocol(svc.Mode) {
 		return nil
 	}
-	if service.IsPortBasedProtocol(svc.Mode) && svc.ListenPort > 0 && (customPorts == nil || !*customPorts) {
+	if service.IsPortBasedProtocol(svc.Mode) && svc.ListenPort > 0 && !serviceUpdate && (customPorts == nil || !*customPorts) {
 		if svc.Source != service.SourceEphemeral {
 			return status.Errorf(status.InvalidArgument, "custom ports not supported on cluster %s", svc.ProxyCluster)
 		}
@@ -382,12 +411,16 @@ func (m *Manager) assignPort(ctx context.Context, tx store.Store, cluster string
 func (m *Manager) persistNewEphemeralService(ctx context.Context, accountID, peerID string, svc *service.Service) error {
 	customPorts := m.clusterCustomPorts(ctx, svc)
 
+	if err := validateTargetReferences(ctx, m.store, accountID, svc.Targets); err != nil {
+		return err
+	}
+
 	return m.store.ExecuteInTransaction(ctx, func(transaction store.Store) error {
 		if err := m.validateEphemeralPreconditions(ctx, transaction, accountID, peerID, svc); err != nil {
 			return err
 		}
 
-		if err := m.ensureL4Port(ctx, transaction, svc, customPorts); err != nil {
+		if err := m.ensureL4Port(ctx, transaction, svc, customPorts, false); err != nil {
 			return err
 		}
 
@@ -395,10 +428,6 @@ func (m *Manager) persistNewEphemeralService(ctx context.Context, accountID, pee
 			return err
 		}
 
-		if err := validateTargetReferences(ctx, transaction, accountID, svc.Targets); err != nil {
-			return err
-		}
-
 		if err := transaction.CreateService(ctx, svc); err != nil {
 			return fmt.Errorf("create service: %w", err)
 		}
@@ -470,7 +499,7 @@ func (m *Manager) UpdateService(ctx context.Context, accountID, userID string, s
 	}
 
 	m.sendServiceUpdateNotifications(ctx, accountID, service, updateInfo)
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationUpdate})
 
 	return service, nil
 }
@@ -491,10 +520,22 @@ func (m *Manager) persistServiceUpdate(ctx context.Context, accountID string, se
 	svcForCaps.ProxyCluster = effectiveCluster
 	customPorts := m.clusterCustomPorts(ctx, &svcForCaps)
 
+	if err := validateTargetReferences(ctx, m.store, accountID, service.Targets); err != nil {
+		return nil, err
+	}
+
+	// Validate subdomain requirement *before* the transaction: the underlying
+	// capability lookup talks to the main DB pool, and SQLite's single-connection
+	// pool would self-deadlock if this ran while the tx already held the only
+	// connection.
+	if err := m.validateSubdomainRequirement(ctx, service.Domain, effectiveCluster); err != nil {
+		return nil, err
+	}
+
 	var updateInfo serviceUpdateInfo
 
 	err = m.store.ExecuteInTransaction(ctx, func(transaction store.Store) error {
-		return m.executeServiceUpdate(ctx, transaction, accountID, service, &updateInfo, customPorts)
+		return m.executeServiceUpdate(ctx, transaction, accountID, service, &updateInfo, customPorts, effectiveCluster)
 	})
 
 	return &updateInfo, err
@@ -512,19 +553,22 @@ func (m *Manager) resolveEffectiveCluster(ctx context.Context, accountID string,
 		return existing.ProxyCluster, nil
 	}
 
-	if m.clusterDeriver != nil {
-		derived, err := m.clusterDeriver.DeriveClusterFromDomain(ctx, accountID, svc.Domain)
-		if err != nil {
-			log.WithError(err).Warnf("could not derive cluster from domain %s", svc.Domain)
-		} else {
-			return derived, nil
-		}
+	if m.clusterDeriver == nil {
+		return existing.ProxyCluster, nil
 	}
 
-	return existing.ProxyCluster, nil
+	// Falling back to the old cluster here would let an update move a service
+	// onto a domain the account has not validated, bypassing the check that
+	// creation makes.
+	derived, err := m.clusterDeriver.DeriveClusterFromDomain(ctx, accountID, svc.Domain)
+	if err != nil {
+		return "", status.Errorf(status.PreconditionFailed, "could not derive cluster from domain %s: %v", svc.Domain, err)
+	}
+
+	return derived, nil
 }
 
-func (m *Manager) executeServiceUpdate(ctx context.Context, transaction store.Store, accountID string, service *service.Service, updateInfo *serviceUpdateInfo, customPorts *bool) error {
+func (m *Manager) executeServiceUpdate(ctx context.Context, transaction store.Store, accountID string, service *service.Service, updateInfo *serviceUpdateInfo, customPorts *bool, effectiveCluster string) error {
 	existingService, err := transaction.GetServiceByID(ctx, store.LockingStrengthUpdate, accountID, service.ID)
 	if err != nil {
 		return err
@@ -542,17 +586,13 @@ func (m *Manager) executeServiceUpdate(ctx context.Context, transaction store.St
 	updateInfo.domainChanged = existingService.Domain != service.Domain
 
 	if updateInfo.domainChanged {
-		if err := m.handleDomainChange(ctx, transaction, accountID, service); err != nil {
+		if err := m.handleDomainChange(ctx, transaction, service, effectiveCluster); err != nil {
 			return err
 		}
 	} else {
 		service.ProxyCluster = existingService.ProxyCluster
 	}
 
-	if err := m.validateSubdomainRequirement(ctx, service.Domain, service.ProxyCluster); err != nil {
-		return err
-	}
-
 	m.preserveExistingAuthSecrets(service, existingService)
 	if err := validateHeaderAuthValues(service.Auth.HeaderAuths); err != nil {
 		return err
@@ -561,15 +601,22 @@ func (m *Manager) executeServiceUpdate(ctx context.Context, transaction store.St
 	m.preserveListenPort(service, existingService)
 	updateInfo.serviceEnabledChanged = existingService.Enabled != service.Enabled
 
-	if err := m.ensureL4Port(ctx, transaction, service, customPorts); err != nil {
+	// if the service is being updated, and we decide in the future to allow mode update,
+	// we should reconsider the currently assigned port if not 0 for clusters that don't support custom ports
+	if err := validateL4PortDiffOnClusterDiff(customPorts, service, existingService); err != nil {
 		return err
 	}
+
+	if err := m.ensureL4Port(ctx, transaction, service, customPorts, true); err != nil {
+		return err
+	}
+
+	// we can try carrying the previous service port into a new cluster, if this becomes a problem for multiple users,
+	// we should reconsider adding another check
 	if err := m.checkPortConflict(ctx, transaction, service); err != nil {
 		return err
 	}
-	if err := validateTargetReferences(ctx, transaction, accountID, service.Targets); err != nil {
-		return err
-	}
+
 	if err := transaction.UpdateService(ctx, service); err != nil {
 		return fmt.Errorf("update service: %w", err)
 	}
@@ -577,20 +624,33 @@ func (m *Manager) executeServiceUpdate(ctx context.Context, transaction store.St
 	return nil
 }
 
-func (m *Manager) handleDomainChange(ctx context.Context, transaction store.Store, accountID string, svc *service.Service) error {
+// validateL4PortDiffOnClusterDiff checks if custom L4 ports are configured and validates port changes across clusters.
+// It ensures no port changes if custom ports are unsupported for a given cluster and protocol mode.
+// Returns an error if validation fails, otherwise returns nil.
+func validateL4PortDiffOnClusterDiff(customPorts *bool, newSVC, oldSVC *service.Service) error {
+	if !service.IsPortBasedProtocol(newSVC.Mode) || (customPorts != nil && *customPorts) {
+		return nil
+	}
+
+	if newSVC.ListenPort != 0 && newSVC.ListenPort != oldSVC.ListenPort {
+		return status.Errorf(status.InvalidArgument, "custom ports not supported on target cluster %s", newSVC.ProxyCluster)
+	}
+
+	return nil
+}
+
+// handleDomainChange validates the new domain is free inside the transaction
+// and applies the pre-resolved cluster (computed outside the tx by
+// resolveEffectiveCluster). It must NOT call clusterDeriver here: that talks
+// to the main DB pool and would self-deadlock under SQLite (max_open_conns=1)
+// because the transaction already holds the only connection.
+func (m *Manager) handleDomainChange(ctx context.Context, transaction store.Store, svc *service.Service, effectiveCluster string) error {
 	if err := m.checkDomainAvailable(ctx, transaction, svc.Domain, svc.ID); err != nil {
 		return err
 	}
-
-	if m.clusterDeriver != nil {
-		newCluster, err := m.clusterDeriver.DeriveClusterFromDomain(ctx, accountID, svc.Domain)
-		if err != nil {
-			log.WithError(err).Warnf("could not derive cluster from domain %s", svc.Domain)
-		} else {
-			svc.ProxyCluster = newCluster
-		}
+	if effectiveCluster != "" {
+		svc.ProxyCluster = effectiveCluster
 	}
-
 	return nil
 }
 
@@ -699,6 +759,10 @@ func validateTargetReferences(ctx context.Context, transaction store.Store, acco
 			if err := validateResourceTarget(ctx, transaction, accountID, target); err != nil {
 				return err
 			}
+		case service.TargetTypeCluster:
+			if err := validateClusterTarget(target); err != nil {
+				return err
+			}
 		default:
 			return status.Errorf(status.InvalidArgument, "unknown target type %q for target %q", target.TargetType, target.TargetId)
 		}
@@ -706,6 +770,13 @@ func validateTargetReferences(ctx context.Context, transaction store.Store, acco
 	return nil
 }
 
+func validateClusterTarget(target *service.Target) error {
+	if !target.Options.DirectUpstream {
+		return status.Errorf(status.InvalidArgument, "cluster target %s has direct upstream disabled", target.Host)
+	}
+	return nil
+}
+
 func validatePeerTarget(ctx context.Context, transaction store.Store, accountID string, target *service.Target) error {
 	if _, err := transaction.GetPeerByID(ctx, store.LockingStrengthShare, accountID, target.TargetId); err != nil {
 		if sErr, ok := status.FromError(err); ok && sErr.Type() == status.NotFound {
@@ -766,7 +837,7 @@ func (m *Manager) DeleteService(ctx context.Context, accountID, userID, serviceI
 
 	m.proxyController.SendServiceUpdateToCluster(ctx, accountID, s.ToProtoMapping(service.Delete, "", m.proxyController.GetOIDCValidationConfig()), s.ProxyCluster)
 
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationDelete})
 
 	return nil
 }
@@ -781,6 +852,10 @@ func (m *Manager) DeleteAllServices(ctx context.Context, accountID, userID strin
 		}
 
 		for _, svc := range services {
+			if err = transaction.DeleteServiceTargets(ctx, accountID, svc.ID); err != nil {
+				return fmt.Errorf("failed to delete service targets: %w", err)
+			}
+
 			if err = transaction.DeleteService(ctx, accountID, svc.ID); err != nil {
 				return fmt.Errorf("failed to delete service: %w", err)
 			}
@@ -799,7 +874,7 @@ func (m *Manager) DeleteAllServices(ctx context.Context, accountID, userID strin
 		m.proxyController.SendServiceUpdateToCluster(ctx, accountID, svc.ToProtoMapping(service.Delete, "", oidcCfg), svc.ProxyCluster)
 	}
 
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationDelete})
 
 	return nil
 }
@@ -855,7 +930,7 @@ func (m *Manager) ReloadService(ctx context.Context, accountID, serviceID string
 
 	m.proxyController.SendServiceUpdateToCluster(ctx, accountID, s.ToProtoMapping(service.Update, "", m.proxyController.GetOIDCValidationConfig()), s.ProxyCluster)
 
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationUpdate})
 
 	return nil
 }
@@ -866,12 +941,14 @@ func (m *Manager) ReloadAllServicesForAccount(ctx context.Context, accountID str
 		return fmt.Errorf("failed to get services: %w", err)
 	}
 
+	oidcCfg := m.proxyController.GetOIDCValidationConfig()
+
 	for _, s := range services {
 		err = m.replaceHostByLookup(ctx, accountID, s)
 		if err != nil {
 			return fmt.Errorf("failed to replace host by lookup for service %s: %w", s.ID, err)
 		}
-		m.proxyController.SendServiceUpdateToCluster(ctx, accountID, s.ToProtoMapping(service.Update, "", m.proxyController.GetOIDCValidationConfig()), s.ProxyCluster)
+		m.proxyController.SendServiceUpdateToCluster(ctx, accountID, s.ToProtoMapping(service.Update, "", oidcCfg), s.ProxyCluster)
 	}
 
 	return nil
@@ -923,6 +1000,10 @@ func (m *Manager) GetAccountServices(ctx context.Context, accountID string) ([]*
 	return services, nil
 }
 
+func (m *Manager) GetServiceByDomain(ctx context.Context, domain string) (*service.Service, error) {
+	return m.store.GetServiceByDomain(ctx, domain)
+}
+
 func (m *Manager) GetServiceIDByTargetID(ctx context.Context, accountID string, resourceID string) (string, error) {
 	target, err := m.store.GetServiceTargetByTargetID(ctx, store.LockingStrengthNone, accountID, resourceID)
 	if err != nil {
@@ -1037,11 +1118,11 @@ func (m *Manager) CreateServiceFromPeer(ctx context.Context, accountID, peerID s
 	}
 
 	m.proxyController.SendServiceUpdateToCluster(ctx, accountID, svc.ToProtoMapping(service.Create, "", m.proxyController.GetOIDCValidationConfig()), svc.ProxyCluster)
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationCreate})
 
 	serviceURL := "https://" + svc.Domain
 	if service.IsL4Protocol(svc.Mode) {
-		serviceURL = fmt.Sprintf("%s://%s:%d", svc.Mode, svc.Domain, svc.ListenPort)
+		serviceURL = fmt.Sprintf("%s://%s", svc.Mode, net.JoinHostPort(svc.Domain, strconv.Itoa(int(svc.ListenPort))))
 	}
 
 	return &service.ExposeServiceResponse{
@@ -1127,6 +1208,10 @@ func (m *Manager) deletePeerService(ctx context.Context, accountID, peerID, serv
 			return status.Errorf(status.PermissionDenied, "cannot delete service exposed by another peer")
 		}
 
+		if err = transaction.DeleteServiceTargets(ctx, accountID, serviceID); err != nil {
+			return fmt.Errorf("delete service targets: %w", err)
+		}
+
 		if err = transaction.DeleteService(ctx, accountID, serviceID); err != nil {
 			return fmt.Errorf("delete service: %w", err)
 		}
@@ -1149,7 +1234,7 @@ func (m *Manager) deletePeerService(ctx context.Context, accountID, peerID, serv
 
 	m.proxyController.SendServiceUpdateToCluster(ctx, accountID, svc.ToProtoMapping(service.Delete, "", m.proxyController.GetOIDCValidationConfig()), svc.ProxyCluster)
 
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationDelete})
 
 	return nil
 }
@@ -1176,6 +1261,10 @@ func (m *Manager) deleteExpiredPeerService(ctx context.Context, accountID, peerI
 			return nil
 		}
 
+		if err = transaction.DeleteServiceTargets(ctx, accountID, serviceID); err != nil {
+			return fmt.Errorf("delete service targets: %w", err)
+		}
+
 		if err = transaction.DeleteService(ctx, accountID, serviceID); err != nil {
 			return fmt.Errorf("delete service: %w", err)
 		}
@@ -1200,7 +1289,7 @@ func (m *Manager) deleteExpiredPeerService(ctx context.Context, accountID, peerI
 	meta := addPeerInfoToEventMeta(svc.EventMeta(), peer)
 	m.accountManager.StoreEvent(ctx, peerID, serviceID, accountID, activity.PeerServiceExposeExpired, meta)
 	m.proxyController.SendServiceUpdateToCluster(ctx, accountID, svc.ToProtoMapping(service.Delete, "", m.proxyController.GetOIDCValidationConfig()), svc.ProxyCluster)
-	m.accountManager.UpdateAccountPeers(ctx, accountID)
+	m.accountManager.UpdateAccountPeers(ctx, accountID, types.UpdateReason{Resource: types.UpdateResourceService, Operation: types.UpdateOperationDelete})
 
 	return nil
 }
@@ -1210,7 +1299,7 @@ func addPeerInfoToEventMeta(meta map[string]any, peer *nbpeer.Peer) map[string]a
 		return meta
 	}
 	meta["peer_name"] = peer.Name
-	if peer.IP != nil {
+	if peer.IP.IsValid() {
 		meta["peer_ip"] = peer.IP.String()
 	}
 	return meta

management/internals/modules/reverseproxy/service/manager/manager_test.go

@@ -3,15 +3,14 @@ package manager
 import (
 	"context"
 	"errors"
-	"net"
+	"net/netip"
 	"testing"
 	"time"
 
-	cachestore "github.com/eko/gocache/lib/v4/store"
-	"github.com/golang/mock/gomock"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 	"go.opentelemetry.io/otel/metric/noop"
+	"go.uber.org/mock/gomock"
 
 	"github.com/netbirdio/netbird/management/internals/modules/reverseproxy/proxy"
 	proxymanager "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/proxy/manager"
@@ -28,7 +27,7 @@ import (
 	"github.com/netbirdio/netbird/shared/management/status"
 )
 
-func testCacheStore(t *testing.T) cachestore.StoreInterface {
+func testCacheStore(t *testing.T) nbcache.Store {
 	t.Helper()
 	s, err := nbcache.NewStore(context.Background(), 30*time.Minute, 10*time.Minute, 100)
 	require.NoError(t, err)
@@ -292,6 +291,7 @@ func TestPersistNewService(t *testing.T) {
 		assert.Equal(t, status.AlreadyExists, sErr.Type())
 	})
 }
+
 func TestPreserveExistingAuthSecrets(t *testing.T) {
 	mgr := &Manager{}
 
@@ -402,7 +402,8 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 	testPeer := &nbpeer.Peer{
 		ID:   ownerPeerID,
 		Name: "test-peer",
-		IP:   net.ParseIP("100.64.0.1"),
+		IP:   netip.MustParseAddr("100.64.0.1"),
+		IPv6: netip.MustParseAddr("fd00::1"),
 	}
 
 	newEphemeralService := func() *rpservice.Service {
@@ -430,7 +431,7 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 		t.Helper()
 		tokenStore := nbgrpc.NewOneTimeTokenStore(context.Background(), testCacheStore(t))
 		pkceStore := nbgrpc.NewPKCEVerifierStore(context.Background(), testCacheStore(t))
-		srv := nbgrpc.NewProxyServiceServer(nil, tokenStore, pkceStore, nbgrpc.ProxyOIDCConfig{}, nil, nil, nil)
+		srv := nbgrpc.NewProxyServiceServer(nil, tokenStore, pkceStore, nbgrpc.ProxyOIDCConfig{}, nil, nil, nil, nil, nil)
 		return srv
 	}
 
@@ -444,7 +445,7 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 			StoreEventFunc: func(_ context.Context, _, _, _ string, activityID activity.ActivityDescriber, _ map[string]any) {
 				storedActivity = activityID.(activity.Activity)
 			},
-			UpdateAccountPeersFunc: func(_ context.Context, _ string) {},
+			UpdateAccountPeersFunc: func(_ context.Context, _ string, _ types.UpdateReason) {},
 		}
 
 		mockStore.EXPECT().
@@ -454,6 +455,9 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 				txMock.EXPECT().
 					GetServiceByID(ctx, store.LockingStrengthUpdate, accountID, serviceID).
 					Return(newEphemeralService(), nil)
+				txMock.EXPECT().
+					DeleteServiceTargets(ctx, accountID, serviceID).
+					Return(nil)
 				txMock.EXPECT().
 					DeleteService(ctx, accountID, serviceID).
 					Return(nil)
@@ -546,7 +550,7 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 			StoreEventFunc: func(_ context.Context, _, _, _ string, activityID activity.ActivityDescriber, _ map[string]any) {
 				storedActivity = activityID.(activity.Activity)
 			},
-			UpdateAccountPeersFunc: func(_ context.Context, _ string) {},
+			UpdateAccountPeersFunc: func(_ context.Context, _ string, _ types.UpdateReason) {},
 		}
 
 		mockStore.EXPECT().
@@ -556,6 +560,9 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 				txMock.EXPECT().
 					GetServiceByID(ctx, store.LockingStrengthUpdate, accountID, serviceID).
 					Return(newEphemeralService(), nil)
+				txMock.EXPECT().
+					DeleteServiceTargets(ctx, accountID, serviceID).
+					Return(nil)
 				txMock.EXPECT().
 					DeleteService(ctx, accountID, serviceID).
 					Return(nil)
@@ -590,7 +597,7 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 			StoreEventFunc: func(_ context.Context, _, _, _ string, _ activity.ActivityDescriber, meta map[string]any) {
 				storedMeta = meta
 			},
-			UpdateAccountPeersFunc: func(_ context.Context, _ string) {},
+			UpdateAccountPeersFunc: func(_ context.Context, _ string, _ types.UpdateReason) {},
 		}
 
 		mockStore.EXPECT().
@@ -600,6 +607,9 @@ func TestDeletePeerService_SourcePeerValidation(t *testing.T) {
 				txMock.EXPECT().
 					GetServiceByID(ctx, store.LockingStrengthUpdate, accountID, serviceID).
 					Return(newEphemeralService(), nil)
+				txMock.EXPECT().
+					DeleteServiceTargets(ctx, accountID, serviceID).
+					Return(nil)
 				txMock.EXPECT().
 					DeleteService(ctx, accountID, serviceID).
 					Return(nil)
@@ -679,7 +689,8 @@ func setupIntegrationTest(t *testing.T) (*Manager, store.Store) {
 				Key:       "test-key",
 				DNSLabel:  "test-peer",
 				Name:      "test-peer",
-				IP:        net.ParseIP("100.64.0.1"),
+				IP:        netip.MustParseAddr("100.64.0.1"),
+				IPv6:      netip.MustParseAddr("fd00::1"),
 				Status:    &nbpeer.PeerStatus{Connected: true, LastSeen: time.Now()},
 				Meta:      nbpeer.PeerSystemMeta{Hostname: "test-peer"},
 			},
@@ -699,7 +710,7 @@ func setupIntegrationTest(t *testing.T) (*Manager, store.Store) {
 
 	accountMgr := &mock_server.MockAccountManager{
 		StoreEventFunc:         func(_ context.Context, _, _, _ string, _ activity.ActivityDescriber, _ map[string]any) {},
-		UpdateAccountPeersFunc: func(_ context.Context, _ string) {},
+		UpdateAccountPeersFunc: func(_ context.Context, _ string, _ types.UpdateReason) {},
 		GetGroupByNameFunc: func(ctx context.Context, groupName, accountID string) (*types.Group, error) {
 			return testStore.GetGroupByName(ctx, store.LockingStrengthNone, accountID, groupName)
 		},
@@ -707,7 +718,7 @@ func setupIntegrationTest(t *testing.T) (*Manager, store.Store) {
 
 	tokenStore := nbgrpc.NewOneTimeTokenStore(ctx, testCacheStore(t))
 	pkceStore := nbgrpc.NewPKCEVerifierStore(ctx, testCacheStore(t))
-	proxySrv := nbgrpc.NewProxyServiceServer(nil, tokenStore, pkceStore, nbgrpc.ProxyOIDCConfig{}, nil, nil, nil)
+	proxySrv := nbgrpc.NewProxyServiceServer(nil, tokenStore, pkceStore, nbgrpc.ProxyOIDCConfig{}, nil, nil, nil, nil, nil)
 
 	proxyController, err := proxymanager.NewGRPCController(proxySrv, noop.NewMeterProvider().Meter(""))
 	require.NoError(t, err)
@@ -745,7 +756,8 @@ func Test_validateExposePermission(t *testing.T) {
 			Key:       "other-key",
 			DNSLabel:  "other-peer",
 			Name:      "other-peer",
-			IP:        net.ParseIP("100.64.0.2"),
+			IP:        netip.MustParseAddr("100.64.0.2"),
+			IPv6:      netip.MustParseAddr("fd00::2"),
 			Status:    &nbpeer.PeerStatus{LastSeen: time.Now()},
 			Meta:      nbpeer.PeerSystemMeta{Hostname: "other-peer"},
 		})
@@ -1128,7 +1140,7 @@ func TestDeleteService_DeletesTargets(t *testing.T) {
 
 	tokenStore := nbgrpc.NewOneTimeTokenStore(ctx, testCacheStore(t))
 	pkceStore := nbgrpc.NewPKCEVerifierStore(ctx, testCacheStore(t))
-	proxySrv := nbgrpc.NewProxyServiceServer(nil, tokenStore, pkceStore, nbgrpc.ProxyOIDCConfig{}, nil, nil, nil)
+	proxySrv := nbgrpc.NewProxyServiceServer(nil, tokenStore, pkceStore, nbgrpc.ProxyOIDCConfig{}, nil, nil, nil, nil, nil)
 
 	proxyController, err := proxymanager.NewGRPCController(proxySrv, noop.NewMeterProvider().Meter(""))
 	require.NoError(t, err)
@@ -1162,7 +1174,7 @@ func TestDeleteService_DeletesTargets(t *testing.T) {
 	mockAcct.EXPECT().
 		StoreEvent(ctx, userID, service.ID, accountID, activity.ServiceDeleted, gomock.Any())
 	mockAcct.EXPECT().
-		UpdateAccountPeers(ctx, accountID)
+		UpdateAccountPeers(ctx, accountID, gomock.Any())
 
 	err = mgr.DeleteService(ctx, accountID, userID, service.ID)
 	require.NoError(t, err)
@@ -1178,6 +1190,67 @@ func TestDeleteService_DeletesTargets(t *testing.T) {
 	assert.Len(t, targets, 0, "All targets should be deleted when service is deleted")
 }
 
+func TestDeleteExpiredPeerService_DeletesTargets(t *testing.T) {
+	ctx := context.Background()
+	mgr, testStore := setupIntegrationTest(t)
+
+	resp, err := mgr.CreateServiceFromPeer(ctx, testAccountID, testPeerID, &rpservice.ExposeServiceRequest{
+		Port: 8080,
+		Mode: "http",
+	})
+	require.NoError(t, err)
+
+	svcID := resolveServiceIDByDomain(t, testStore, resp.Domain)
+
+	targets, err := testStore.GetTargetsByServiceID(ctx, store.LockingStrengthNone, testAccountID, svcID)
+	require.NoError(t, err)
+	require.Len(t, targets, 1, "ephemeral peer-exposed service should have exactly one persisted target before reaping")
+
+	expireEphemeralService(t, testStore, testAccountID, resp.Domain)
+	err = mgr.deleteExpiredPeerService(ctx, testAccountID, testPeerID, svcID)
+	require.NoError(t, err)
+
+	_, err = testStore.GetServiceByDomain(ctx, resp.Domain)
+	require.Error(t, err, "expired peer-exposed service should be deleted")
+	s, ok := status.FromError(err)
+	require.True(t, ok)
+	assert.Equal(t, status.NotFound, s.Type())
+
+	targets, err = testStore.GetTargetsByServiceID(ctx, store.LockingStrengthNone, testAccountID, svcID)
+	require.NoError(t, err)
+	assert.Len(t, targets, 0, "orphaned target rows must be deleted when an expired peer-exposed service is reaped")
+}
+
+func TestDeleteServiceFromPeer_DeletesTargets(t *testing.T) {
+	ctx := context.Background()
+	mgr, testStore := setupIntegrationTest(t)
+
+	resp, err := mgr.CreateServiceFromPeer(ctx, testAccountID, testPeerID, &rpservice.ExposeServiceRequest{
+		Port: 8080,
+		Mode: "http",
+	})
+	require.NoError(t, err)
+
+	svcID := resolveServiceIDByDomain(t, testStore, resp.Domain)
+
+	targets, err := testStore.GetTargetsByServiceID(ctx, store.LockingStrengthNone, testAccountID, svcID)
+	require.NoError(t, err)
+	require.Len(t, targets, 1, "ephemeral peer-exposed service should have exactly one persisted target before stopping")
+
+	err = mgr.StopServiceFromPeer(ctx, testAccountID, testPeerID, svcID)
+	require.NoError(t, err)
+
+	_, err = testStore.GetServiceByDomain(ctx, resp.Domain)
+	require.Error(t, err, "stopped peer-exposed service should be deleted")
+	s, ok := status.FromError(err)
+	require.True(t, ok)
+	assert.Equal(t, status.NotFound, s.Type())
+
+	targets, err = testStore.GetTargetsByServiceID(ctx, store.LockingStrengthNone, testAccountID, svcID)
+	require.NoError(t, err)
+	assert.Len(t, targets, 0, "orphaned target rows must be deleted when a peer stops its exposed service")
+}
+
 func TestValidateProtocolChange(t *testing.T) {
 	tests := []struct {
 		name    string
@@ -1330,3 +1403,66 @@ func TestValidateSubdomainRequirement(t *testing.T) {
 		})
 	}
 }
+
+func TestValidateTargetReferences_ClusterTargetSkipsLookup(t *testing.T) {
+	ctx := context.Background()
+	ctrl := gomock.NewController(t)
+	mockStore := store.NewMockStore(ctrl)
+	accountID := "test-account"
+
+	// No peer or resource lookups must be issued for cluster targets.
+	targets := []*rpservice.Target{
+		{
+			TargetId:   "eu.proxy.netbird.io",
+			TargetType: rpservice.TargetTypeCluster,
+			Options:    rpservice.TargetOptions{DirectUpstream: true},
+		},
+	}
+	require.NoError(t, validateTargetReferences(ctx, mockStore, accountID, targets), "cluster target must validate without store lookups")
+}
+
+// TestValidateTargetReferences_ClusterTargetRequiresDirectUpstream pins the
+// store-side check that cluster targets must opt into the host-stack dial
+// path. Without DirectUpstream the proxy would route this target through
+// the embedded NetBird client and fail on every request.
+func TestValidateTargetReferences_ClusterTargetRequiresDirectUpstream(t *testing.T) {
+	ctx := context.Background()
+	ctrl := gomock.NewController(t)
+	mockStore := store.NewMockStore(ctrl)
+	accountID := "test-account"
+
+	targets := []*rpservice.Target{
+		{
+			TargetId:   "eu.proxy.netbird.io",
+			TargetType: rpservice.TargetTypeCluster,
+			Host:       "backend.lan",
+		},
+	}
+	err := validateTargetReferences(ctx, mockStore, accountID, targets)
+	require.Error(t, err, "cluster target without direct_upstream must be rejected")
+	assert.ErrorContains(t, err, "direct upstream disabled")
+}
+
+func TestReplaceHostByLookup_SkipsClusterTarget(t *testing.T) {
+	ctx := context.Background()
+	ctrl := gomock.NewController(t)
+	mockStore := store.NewMockStore(ctrl)
+	accountID := "test-account"
+
+	mgr := &Manager{store: mockStore}
+
+	svc := &rpservice.Service{
+		ID:        "svc-1",
+		AccountID: accountID,
+		Targets: []*rpservice.Target{
+			{
+				TargetId:   "eu.proxy.netbird.io",
+				TargetType: rpservice.TargetTypeCluster,
+				Host:       "127.0.0.1",
+			},
+		},
+	}
+
+	require.NoError(t, mgr.replaceHostByLookup(ctx, accountID, svc), "cluster target must not trigger peer/resource lookup")
+	assert.Equal(t, "127.0.0.1", svc.Targets[0].Host, "operator-supplied host must be preserved for cluster target")
+}