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

management/internals/modules/reverseproxy/domain/domain.go

@@ -23,6 +23,8 @@ type Domain struct {
 	// SupportsCrowdSec is populated at query time from proxy cluster capabilities.
 	// Not persisted.
 	SupportsCrowdSec *bool `gorm:"-"`
+	// SupportsPrivate is populated at query time from proxy cluster capabilities. Not persisted.
+	SupportsPrivate *bool `gorm:"-"`
 }
 
 // EventMeta returns activity event metadata for a domain

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

@@ -52,6 +52,7 @@ func domainToApi(d *domain.Domain) api.ReverseProxyDomain {
 		SupportsCustomPorts: d.SupportsCustomPorts,
 		RequireSubdomain:    d.RequireSubdomain,
 		SupportsCrowdsec:    d.SupportsCrowdSec,
+		SupportsPrivate:     d.SupportsPrivate,
 	}
 	if d.TargetCluster != "" {
 		resp.TargetCluster = &d.TargetCluster

management/internals/modules/reverseproxy/domain/manager/domain_test.go

@@ -66,8 +66,8 @@ func TestExtractClusterFromFreeDomain(t *testing.T) {
 
 func TestExtractClusterFromCustomDomains(t *testing.T) {
 	customDomains := []*domain.Domain{
-		{Domain: "example.com", TargetCluster: "eu1.proxy.netbird.io"},
-		{Domain: "proxy.corp.io", TargetCluster: "us1.proxy.netbird.io"},
+		{Domain: "example.com", TargetCluster: "eu1.proxy.netbird.io", Validated: true},
+		{Domain: "proxy.corp.io", TargetCluster: "us1.proxy.netbird.io", Validated: true},
 	}
 
 	tests := []struct {
@@ -120,19 +120,49 @@ func TestExtractClusterFromCustomDomains(t *testing.T) {
 
 	for _, tc := range tests {
 		t.Run(tc.name, func(t *testing.T) {
-			cluster, ok := extractClusterFromCustomDomains(tc.domain, customDomains)
-			assert.Equal(t, tc.wantOK, ok)
-			if ok {
-				assert.Equal(t, tc.wantVal, cluster)
+			cluster, match := extractClusterFromCustomDomains(tc.domain, customDomains)
+			if !tc.wantOK {
+				assert.Equal(t, customDomainNoMatch, match, "unrelated domain should not match any custom domain")
+				return
 			}
+			assert.Equal(t, customDomainValidated, match, "validated custom domain should resolve a cluster")
+			assert.Equal(t, tc.wantVal, cluster)
 		})
 	}
 }
 
+// An unvalidated row must never yield a cluster: the account has not shown it
+// controls the name, so no service may be bound to it.
+func TestExtractClusterFromCustomDomains_UnvalidatedDomainRefused(t *testing.T) {
+	customDomains := []*domain.Domain{
+		{Domain: "example.com", TargetCluster: "eu1.proxy.netbird.io", Validated: false},
+	}
+
+	for _, serviceDomain := range []string{"example.com", "app.example.com"} {
+		t.Run(serviceDomain, func(t *testing.T) {
+			cluster, match := extractClusterFromCustomDomains(serviceDomain, customDomains)
+			assert.Equal(t, customDomainUnvalidated, match, "unvalidated row must be reported as such")
+			assert.Empty(t, cluster, "unvalidated row must not resolve a cluster")
+		})
+	}
+}
+
+// A more specific unvalidated row must not shadow a validated parent domain.
+func TestExtractClusterFromCustomDomains_ValidatedParentWinsOverUnvalidatedChild(t *testing.T) {
+	customDomains := []*domain.Domain{
+		{Domain: "example.com", TargetCluster: "cluster-generic", Validated: true},
+		{Domain: "app.example.com", TargetCluster: "cluster-app", Validated: false},
+	}
+
+	cluster, match := extractClusterFromCustomDomains("app.example.com", customDomains)
+	assert.Equal(t, customDomainValidated, match)
+	assert.Equal(t, "cluster-generic", cluster, "validated parent domain should provide the cluster")
+}
+
 func TestExtractClusterFromCustomDomains_OverlappingDomains(t *testing.T) {
 	customDomains := []*domain.Domain{
-		{Domain: "example.com", TargetCluster: "cluster-generic"},
-		{Domain: "app.example.com", TargetCluster: "cluster-app"},
+		{Domain: "example.com", TargetCluster: "cluster-generic", Validated: true},
+		{Domain: "app.example.com", TargetCluster: "cluster-app", Validated: true},
 	}
 
 	tests := []struct {
@@ -164,8 +194,8 @@ func TestExtractClusterFromCustomDomains_OverlappingDomains(t *testing.T) {
 
 	for _, tc := range tests {
 		t.Run(tc.name, func(t *testing.T) {
-			cluster, ok := extractClusterFromCustomDomains(tc.domain, customDomains)
-			assert.True(t, ok)
+			cluster, match := extractClusterFromCustomDomains(tc.domain, customDomains)
+			assert.Equal(t, customDomainValidated, match)
 			assert.Equal(t, tc.wantVal, cluster)
 		})
 	}

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

@@ -2,22 +2,28 @@ package manager
 
 import (
 	"context"
+	"errors"
 	"fmt"
 	"net"
 	"strings"
 
 	log "github.com/sirupsen/logrus"
 
+	agentnetworkTypes "github.com/netbirdio/netbird/management/internals/modules/agentnetwork/types"
 	"github.com/netbirdio/netbird/management/internals/modules/reverseproxy/domain"
 	"github.com/netbirdio/netbird/management/server/account"
 	"github.com/netbirdio/netbird/management/server/activity"
+	nbstore "github.com/netbirdio/netbird/management/server/store"
 	"github.com/netbirdio/netbird/management/server/types"
+	"github.com/netbirdio/netbird/shared/management/status"
 )
 
 type store interface {
 	GetAccount(ctx context.Context, accountID string) (*types.Account, error)
+	GetAgentNetworkSettings(ctx context.Context, lockStrength nbstore.LockingStrength, accountID string) (*agentnetworkTypes.Settings, error)
 
 	GetCustomDomain(ctx context.Context, accountID string, domainID string) (*domain.Domain, error)
+	GetCustomDomainByName(ctx context.Context, domainName string) (*domain.Domain, error)
 	ListFreeDomains(ctx context.Context, accountID string) ([]string, error)
 	ListCustomDomains(ctx context.Context, accountID string) ([]*domain.Domain, error)
 	CreateCustomDomain(ctx context.Context, accountID string, domainName string, targetCluster string, validated bool) (*domain.Domain, error)
@@ -27,9 +33,11 @@ type store interface {
 
 type proxyManager interface {
 	GetActiveClusterAddresses(ctx context.Context) ([]string, error)
+	GetActiveClusterAddressesForAccount(ctx context.Context, accountID string) ([]string, error)
 	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 {
@@ -57,8 +65,8 @@ func (m Manager) GetDomains(ctx context.Context, accountID, userID string) ([]*d
 	var ret []*domain.Domain
 
 	// Add connected proxy clusters as free domains.
-	// The cluster address itself is the free domain base (e.g., "eu.proxy.netbird.io").
-	allowList, err := m.proxyManager.GetActiveClusterAddresses(ctx)
+	// For BYOP accounts, only their own cluster is returned; otherwise shared clusters.
+	allowList, err := m.getClusterAllowList(ctx, accountID)
 	if err != nil {
 		log.WithContext(ctx).Errorf("failed to get active proxy cluster addresses: %v", err)
 		return nil, err
@@ -78,6 +86,7 @@ func (m Manager) GetDomains(ctx context.Context, accountID, userID string) ([]*d
 		d.SupportsCustomPorts = m.proxyManager.ClusterSupportsCustomPorts(ctx, cluster)
 		d.RequireSubdomain = m.proxyManager.ClusterRequireSubdomain(ctx, cluster)
 		d.SupportsCrowdSec = m.proxyManager.ClusterSupportsCrowdSec(ctx, cluster)
+		d.SupportsPrivate = m.proxyManager.ClusterSupportsPrivate(ctx, cluster)
 		ret = append(ret, d)
 	}
 
@@ -94,6 +103,7 @@ func (m Manager) GetDomains(ctx context.Context, accountID, userID string) ([]*d
 		if d.TargetCluster != "" {
 			cd.SupportsCustomPorts = m.proxyManager.ClusterSupportsCustomPorts(ctx, d.TargetCluster)
 			cd.SupportsCrowdSec = m.proxyManager.ClusterSupportsCrowdSec(ctx, d.TargetCluster)
+			cd.SupportsPrivate = m.proxyManager.ClusterSupportsPrivate(ctx, d.TargetCluster)
 		}
 		// Custom domains never require a subdomain by default since
 		// the account owns them and should be able to use the bare domain.
@@ -104,8 +114,8 @@ func (m Manager) GetDomains(ctx context.Context, accountID, userID string) ([]*d
 }
 
 func (m Manager) CreateDomain(ctx context.Context, accountID, userID, domainName, targetCluster string) (*domain.Domain, error) {
-	// Verify the target cluster is in the available clusters
-	allowList, err := m.proxyManager.GetActiveClusterAddresses(ctx)
+	// Verify the target cluster is in the available clusters for this account
+	allowList, err := m.getClusterAllowList(ctx, accountID)
 	if err != nil {
 		return nil, fmt.Errorf("failed to get active proxy cluster addresses: %w", err)
 	}
@@ -120,6 +130,10 @@ func (m Manager) CreateDomain(ctx context.Context, accountID, userID, domainName
 		return nil, fmt.Errorf("target cluster %s is not available", targetCluster)
 	}
 
+	if err := m.checkDomainAvailable(ctx, domainName); err != nil {
+		return nil, err
+	}
+
 	// Attempt an initial validation against the specified cluster only
 	var validated bool
 	if m.validator.IsValid(ctx, domainName, []string{targetCluster}) {
@@ -136,6 +150,23 @@ func (m Manager) CreateDomain(ctx context.Context, accountID, userID, domainName
 	return d, nil
 }
 
+// checkDomainAvailable reports whether the domain is free to claim. The unique
+// index on the column is the real guard; this turns the violation into a
+// conflict the caller can act on instead of a database error, and says nothing
+// about which account holds the domain.
+func (m Manager) checkDomainAvailable(ctx context.Context, domainName string) error {
+	_, err := m.store.GetCustomDomainByName(ctx, domainName)
+	if err == nil {
+		return status.Errorf(status.AlreadyExists, "domain %s is already registered", domainName)
+	}
+
+	if sErr, ok := status.FromError(err); ok && sErr.Type() == status.NotFound {
+		return nil
+	}
+
+	return fmt.Errorf("look up domain: %w", err)
+}
+
 func (m Manager) DeleteDomain(ctx context.Context, accountID, userID, domainID string) error {
 	d, err := m.store.GetCustomDomain(ctx, accountID, domainID)
 	if err != nil {
@@ -228,7 +259,7 @@ func (m Manager) GetClusterDomains() []string {
 // For free domains (those ending with a known cluster suffix), the cluster is extracted from the domain.
 // For custom domains, the cluster is determined by checking the registered custom domain's target cluster.
 func (m Manager) DeriveClusterFromDomain(ctx context.Context, accountID, domain string) (string, error) {
-	allowList, err := m.proxyManager.GetActiveClusterAddresses(ctx)
+	allowList, err := m.getClusterAllowList(ctx, accountID)
 	if err != nil {
 		return "", fmt.Errorf("failed to get active proxy cluster addresses: %w", err)
 	}
@@ -245,27 +276,114 @@ func (m Manager) DeriveClusterFromDomain(ctx context.Context, accountID, domain
 		return "", fmt.Errorf("list custom domains: %w", err)
 	}
 
-	targetCluster, valid := extractClusterFromCustomDomains(domain, customDomains)
-	if valid {
+	targetCluster, match := extractClusterFromCustomDomains(domain, customDomains)
+	switch match {
+	case customDomainValidated:
 		return targetCluster, nil
+	case customDomainUnvalidated:
+		return "", status.Errorf(status.PreconditionFailed, "domain %s is not validated", domain)
 	}
 
 	return "", fmt.Errorf("domain %s does not match any available proxy cluster", domain)
 }
 
-func extractClusterFromCustomDomains(serviceDomain string, customDomains []*domain.Domain) (string, bool) {
+func (m Manager) getClusterAllowList(ctx context.Context, accountID string) ([]string, error) {
+	byopAddresses, err := m.proxyManager.GetActiveClusterAddressesForAccount(ctx, accountID)
+	if err != nil {
+		return nil, fmt.Errorf("get BYOP cluster addresses: %w", err)
+	}
+	publicAddresses, err := m.proxyManager.GetActiveClusterAddresses(ctx)
+	if err != nil {
+		return nil, fmt.Errorf("get public cluster addresses: %w", err)
+	}
+	reserved, err := m.reservedGatewayAddress(ctx, accountID)
+	if err != nil {
+		return nil, err
+	}
+	seen := make(map[string]struct{}, len(byopAddresses)+len(publicAddresses))
+	merged := make([]string, 0, len(byopAddresses)+len(publicAddresses))
+	for _, addr := range byopAddresses {
+		if _, ok := seen[addr]; ok || addr == reserved {
+			continue
+		}
+		seen[addr] = struct{}{}
+		merged = append(merged, addr)
+	}
+	for _, addr := range publicAddresses {
+		if _, ok := seen[addr]; ok || addr == reserved {
+			continue
+		}
+		seen[addr] = struct{}{}
+		merged = append(merged, addr)
+	}
+	return merged, nil
+}
+
+// reservedGatewayAddress returns the account's agent-network gateway address
+// when its settings pin is self-addressed — a proxy dedicated to serving
+// exactly the gateway. Dropping that address from the cluster allow list keeps
+// it from being offered as a cluster for ordinary services, and because the
+// free-domain suffix match is depth-independent, dropping the address rejects
+// every name beneath it as well as the bare one. Only the account's own
+// gateway address can ever appear in its allow list (another tenant's gateway
+// proxy is account-scoped to them), so this single-address exclusion is
+// sufficient. Returns "" when the account has no settings row or a labeled
+// (shared-cluster) pin.
+func (m Manager) reservedGatewayAddress(ctx context.Context, accountID string) (string, error) {
+	settings, err := m.store.GetAgentNetworkSettings(ctx, nbstore.LockingStrengthNone, accountID)
+	if err != nil {
+		var sErr *status.Error
+		if errors.As(err, &sErr) && sErr.Type() == status.NotFound {
+			return "", nil
+		}
+		return "", fmt.Errorf("get agent network settings: %w", err)
+	}
+	if settings == nil || !settings.Dedicated() {
+		return "", nil
+	}
+	return settings.ProxyAddress, nil
+}
+
+// customDomainMatch describes how a service domain relates to the account's
+// custom domain rows.
+type customDomainMatch int
+
+const (
+	customDomainNoMatch customDomainMatch = iota
+	customDomainUnvalidated
+	customDomainValidated
+)
+
+// extractClusterFromCustomDomains finds the longest custom domain covering the
+// service domain and reports its target cluster. Only a validated row yields a
+// cluster: until the CNAME check has passed the account has not shown it
+// controls the name, so no traffic may be routed for it.
+func extractClusterFromCustomDomains(serviceDomain string, customDomains []*domain.Domain) (string, customDomainMatch) {
 	bestCluster := ""
 	bestLen := -1
+	matched := false
 	for _, cd := range customDomains {
 		if serviceDomain != cd.Domain && !strings.HasSuffix(serviceDomain, "."+cd.Domain) {
 			continue
 		}
+		matched = true
+		if !cd.Validated {
+			continue
+		}
 		if l := len(cd.Domain); l > bestLen {
 			bestLen = l
 			bestCluster = cd.TargetCluster
 		}
 	}
-	return bestCluster, bestLen >= 0
+
+	switch {
+	case bestLen >= 0:
+		return bestCluster, customDomainValidated
+	case matched:
+		return "", customDomainUnvalidated
+	default:
+		return "", customDomainNoMatch
+	}
 }
 
 // ExtractClusterFromFreeDomain extracts the cluster address from a free domain.

management/internals/modules/reverseproxy/domain/manager/manager_realstore_test.go

@@ -0,0 +1,298 @@
+package manager
+
+import (
+	"context"
+	"fmt"
+	"sync"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	"go.opentelemetry.io/otel/metric/noop"
+
+	"github.com/netbirdio/netbird/management/internals/modules/reverseproxy/domain"
+	proxymanager "github.com/netbirdio/netbird/management/internals/modules/reverseproxy/proxy/manager"
+	"github.com/netbirdio/netbird/management/server/activity"
+	"github.com/netbirdio/netbird/management/server/mock_server"
+	nbstore "github.com/netbirdio/netbird/management/server/store"
+	"github.com/netbirdio/netbird/management/server/types"
+	"github.com/netbirdio/netbird/shared/management/status"
+)
+
+const (
+	testCluster    = "eu.proxy.test"
+	accountA       = "account-a"
+	accountAUser   = "account-a-admin"
+	accountB       = "account-b"
+	accountBUser   = "account-b-admin"
+	accountAMember = "account-a-member"
+)
+
+// stubResolver answers CNAME lookups from a table the test controls, so a
+// domain can point at the cluster or nowhere without touching a real resolver.
+type stubResolver struct {
+	mu     sync.Mutex
+	cnames map[string]string
+}
+
+func (r *stubResolver) LookupCNAME(_ context.Context, host string) (string, error) {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+
+	cname, ok := r.cnames[host]
+	if !ok {
+		return "", fmt.Errorf("lookup %s: no such host", host)
+	}
+	return cname + ".", nil
+}
+
+func (r *stubResolver) set(host, cname string) {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+	r.cnames[host] = cname
+}
+
+type domainTestEnv struct {
+	manager  Manager
+	store    nbstore.Store
+	resolver *stubResolver
+}
+
+// setupDomainTest builds the domain manager on a real SQLite store with two
+// accounts and one active public proxy cluster.
+func setupDomainTest(t *testing.T) *domainTestEnv {
+	t.Helper()
+
+	ctx := context.Background()
+	testStore, cleanup, err := nbstore.NewTestStoreFromSQL(ctx, "", t.TempDir())
+	require.NoError(t, err)
+	t.Cleanup(cleanup)
+
+	for accountID, userID := range map[string]string{accountA: accountAUser, accountB: accountBUser} {
+		users := map[string]*types.User{
+			userID: {
+				Id:        userID,
+				AccountID: accountID,
+				Role:      types.UserRoleAdmin,
+			},
+		}
+		if accountID == accountA {
+			// A real member of the account whose role denies Services:Create, so
+			// permission denial is exercised as ok=false rather than as a lookup
+			// error for a user who is not in the account at all.
+			users[accountAMember] = &types.User{
+				Id:        accountAMember,
+				AccountID: accountID,
+				Role:      types.UserRoleUser,
+			}
+		}
+
+		require.NoError(t, testStore.SaveAccount(ctx, &types.Account{
+			Id:        accountID,
+			CreatedBy: userID,
+			Settings:  &types.Settings{},
+			Users:     users,
+		}))
+	}
+
+	proxyMgr, err := proxymanager.NewManager(testStore, noop.NewMeterProvider().Meter(""))
+	require.NoError(t, err)
+
+	_, err = proxyMgr.Connect(ctx, "proxy-1", "session-1", testCluster, "127.0.0.1", nil, nil)
+	require.NoError(t, err)
+
+	resolver := &stubResolver{cnames: make(map[string]string)}
+
+	mgr := Manager{
+		store:        testStore,
+		proxyManager: proxyMgr,
+		validator:    domain.Validator{Resolver: resolver},
+		accountManager: &mock_server.MockAccountManager{
+			StoreEventFunc: func(context.Context, string, string, string, activity.ActivityDescriber, map[string]any) {},
+		},
+	}
+
+	return &domainTestEnv{manager: mgr, store: testStore, resolver: resolver}
+}
+
+// storedDomain reads a domain row back through the store so assertions are made
+// on what was persisted rather than on the value the manager returned.
+func storedDomain(t *testing.T, s nbstore.Store, accountID, domainName string) *domain.Domain {
+	t.Helper()
+
+	domains, err := s.ListCustomDomains(context.Background(), accountID)
+	require.NoError(t, err)
+	for _, d := range domains {
+		if d.Domain == domainName {
+			return d
+		}
+	}
+	return nil
+}
+
+// A domain whose CNAME check fails is stored unvalidated and must not resolve a
+// cluster, which is what service creation gates on.
+func TestCreateDomain_FailedLookupIsNotServable(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+
+	created, err := env.manager.CreateDomain(ctx, accountA, accountAUser, "apps.example.com", testCluster)
+	require.NoError(t, err)
+	assert.False(t, created.Validated, "a domain whose CNAME lookup fails must not be created validated")
+
+	stored := storedDomain(t, env.store, accountA, "apps.example.com")
+	require.NotNil(t, stored, "domain row should exist")
+	assert.False(t, stored.Validated, "persisted row must be unvalidated")
+
+	cluster, err := env.manager.DeriveClusterFromDomain(ctx, accountA, "apps.example.com")
+	require.Error(t, err, "an unvalidated domain must not resolve a cluster")
+	assert.Empty(t, cluster)
+	assert.Contains(t, err.Error(), "not validated", "error should tell the caller what to fix")
+
+	sErr, ok := status.FromError(err)
+	require.True(t, ok, "error should be a typed status error")
+	assert.Equal(t, status.PreconditionFailed, sErr.Type())
+
+	_, err = env.manager.DeriveClusterFromDomain(ctx, accountA, "sub.apps.example.com")
+	assert.Error(t, err, "subdomains of an unvalidated custom domain are not servable either")
+}
+
+// A second account claiming a registered domain gets a clean conflict, not a
+// database error surfaced as a 500.
+func TestCreateDomain_DuplicateIsAConflict(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+
+	_, err := env.manager.CreateDomain(ctx, accountA, accountAUser, "shared.example.com", testCluster)
+	require.NoError(t, err)
+
+	_, err = env.manager.CreateDomain(ctx, accountB, accountBUser, "shared.example.com", testCluster)
+	require.Error(t, err)
+
+	sErr, ok := status.FromError(err)
+	require.True(t, ok, "conflict must be a typed status error, not a raw database error")
+	assert.Equal(t, status.AlreadyExists, sErr.Type(), "conflict should map to 409, not 500")
+	assert.NotContains(t, sErr.Message, accountA, "the response must not reveal the holding account")
+
+	assert.Nil(t, storedDomain(t, env.store, accountB, "shared.example.com"), "no row should be written on conflict")
+}
+
+// The same account re-adding one of its own domains is a conflict too.
+func TestCreateDomain_SameAccountDuplicateIsAConflict(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+
+	_, err := env.manager.CreateDomain(ctx, accountA, accountAUser, "dup.example.com", testCluster)
+	require.NoError(t, err)
+
+	_, err = env.manager.CreateDomain(ctx, accountA, accountAUser, "dup.example.com", testCluster)
+	require.Error(t, err)
+
+	sErr, ok := status.FromError(err)
+	require.True(t, ok)
+	assert.Equal(t, status.AlreadyExists, sErr.Type())
+}
+
+// The negative control: a validated domain still derives its cluster, for the
+// bare name and for subdomains, exactly as before.
+func TestCreateDomain_ValidatedDomainDerivesCluster(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+	env.resolver.set("validation.valid.example.com", testCluster)
+
+	created, err := env.manager.CreateDomain(ctx, accountA, accountAUser, "valid.example.com", testCluster)
+	require.NoError(t, err)
+	require.True(t, created.Validated, "a matching CNAME should validate on create")
+
+	cluster, err := env.manager.DeriveClusterFromDomain(ctx, accountA, "valid.example.com")
+	require.NoError(t, err)
+	assert.Equal(t, testCluster, cluster)
+
+	cluster, err = env.manager.DeriveClusterFromDomain(ctx, accountA, "app.valid.example.com")
+	require.NoError(t, err)
+	assert.Equal(t, testCluster, cluster, "subdomains of a validated custom domain resolve too")
+}
+
+// Validating a domain flips the gate: the same lookup that failed before now
+// resolves a cluster.
+func TestValidateDomain_UnlocksClusterDerivation(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+
+	created, err := env.manager.CreateDomain(ctx, accountA, accountAUser, "later.example.com", testCluster)
+	require.NoError(t, err)
+	require.False(t, created.Validated)
+
+	_, err = env.manager.DeriveClusterFromDomain(ctx, accountA, "later.example.com")
+	require.Error(t, err)
+
+	env.resolver.set("validation.later.example.com", testCluster)
+	env.manager.ValidateDomain(ctx, accountA, accountAUser, created.ID)
+
+	require.True(t, storedDomain(t, env.store, accountA, "later.example.com").Validated)
+
+	cluster, err := env.manager.DeriveClusterFromDomain(ctx, accountA, "later.example.com")
+	require.NoError(t, err)
+	assert.Equal(t, testCluster, cluster)
+}
+
+// Free cluster domains are unaffected by the custom domain gate.
+func TestDeriveClusterFromDomain_FreeDomainUnaffected(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+
+	cluster, err := env.manager.DeriveClusterFromDomain(ctx, accountA, "myapp.abc123."+testCluster)
+	require.NoError(t, err)
+	assert.Equal(t, testCluster, cluster)
+}
+
+// The manager pre-check exists to turn a conflict into a 409, but the unique
+// index on the column is what actually guarantees the domain is claimed once.
+//
+// Two requests can clear the pre-check concurrently and race to the insert.
+// Inserting twice through the store reaches the same code path the loser of
+// that race takes, without the nondeterminism of driving it from goroutines,
+// and the loser must still see a conflict rather than an internal error.
+func TestStore_DuplicateDomainRejectedByIndexAsConflict(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+
+	_, err := env.store.CreateCustomDomain(ctx, accountA, "indexed.example.com", testCluster, false)
+	require.NoError(t, err)
+
+	_, err = env.store.CreateCustomDomain(ctx, accountB, "indexed.example.com", testCluster, false)
+	require.Error(t, err, "the unique index must reject the same domain in a second account")
+
+	sErr, ok := status.FromError(err)
+	require.True(t, ok, "the losing insert must return a typed status error")
+	assert.Equal(t, status.AlreadyExists, sErr.Type(), "a lost race is a 409, not a 500")
+}
+
+// Validation runs asynchronously, so it can finish after the domain was
+// deleted and then write a stale row back. gorm's Save falls back to an insert
+// when an update affects no rows, which would resurrect the domain as
+// validated; UpdateCustomDomain avoids that by selecting explicit columns.
+// This pins that behaviour, since dropping the Select would reintroduce it.
+func TestUpdateCustomDomain_DoesNotResurrectDeletedDomain(t *testing.T) {
+	ctx := context.Background()
+	env := setupDomainTest(t)
+
+	created, err := env.manager.CreateDomain(ctx, accountA, accountAUser, "racy.example.com", testCluster)
+	require.NoError(t, err)
+
+	stale := storedDomain(t, env.store, accountA, "racy.example.com")
+	require.NotNil(t, stale)
+
+	require.NoError(t, env.manager.DeleteDomain(ctx, accountA, accountAUser, created.ID))
+	require.Nil(t, storedDomain(t, env.store, accountA, "racy.example.com"), "the domain should be gone")
+
+	// What an in-flight validation would write once its CNAME check succeeded.
+	// The write has to succeed for the assertion below to mean anything: a
+	// rejected write would leave the domain absent for the wrong reason.
+	stale.Validated = true
+	_, err = env.store.UpdateCustomDomain(ctx, accountA, stale)
+	require.NoError(t, err, "the update itself must succeed, so absence is not just a failed write")
+
+	assert.Nil(t, storedDomain(t, env.store, accountA, "racy.example.com"),
+		"a late validation write must not recreate a deleted domain")
+}

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

@@ -0,0 +1,297 @@
+package manager
+
+import (
+	"context"
+	"errors"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	agentnetworkTypes "github.com/netbirdio/netbird/management/internals/modules/agentnetwork/types"
+	"github.com/netbirdio/netbird/management/internals/modules/reverseproxy/domain"
+	nbstore "github.com/netbirdio/netbird/management/server/store"
+	"github.com/netbirdio/netbird/management/server/types"
+	"github.com/netbirdio/netbird/shared/management/status"
+)
+
+type mockProxyManager struct {
+	getActiveClusterAddressesFunc           func(ctx context.Context) ([]string, error)
+	getActiveClusterAddressesForAccountFunc func(ctx context.Context, accountID string) ([]string, error)
+}
+
+func (m *mockProxyManager) GetActiveClusterAddresses(ctx context.Context) ([]string, error) {
+	if m.getActiveClusterAddressesFunc != nil {
+		return m.getActiveClusterAddressesFunc(ctx)
+	}
+	return nil, nil
+}
+
+func (m *mockProxyManager) GetActiveClusterAddressesForAccount(ctx context.Context, accountID string) ([]string, error) {
+	if m.getActiveClusterAddressesForAccountFunc != nil {
+		return m.getActiveClusterAddressesForAccountFunc(ctx, accountID)
+	}
+	return nil, nil
+}
+
+func (m *mockProxyManager) ClusterSupportsCustomPorts(_ context.Context, _ string) *bool {
+	return nil
+}
+
+func (m *mockProxyManager) ClusterRequireSubdomain(_ context.Context, _ string) *bool {
+	return nil
+}
+
+func (m *mockProxyManager) ClusterSupportsCrowdSec(_ context.Context, _ string) *bool {
+	return nil
+}
+
+func (m *mockProxyManager) ClusterSupportsPrivate(_ context.Context, _ string) *bool {
+	return nil
+}
+
+func TestGetClusterAllowList_BYOPMergedWithPublic(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, accID string) ([]string, error) {
+			assert.Equal(t, "acc-123", accID)
+			return []string{"byop.example.com"}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return []string{"eu.proxy.netbird.io"}, nil
+		},
+	}
+
+	mgr := Manager{store: &stubStore{}, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.NoError(t, err)
+	assert.Equal(t, []string{"byop.example.com", "eu.proxy.netbird.io"}, result)
+}
+
+func TestGetClusterAllowList_DeduplicatesBYOPAndPublic(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return []string{"shared.example.com", "byop.example.com"}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return []string{"shared.example.com", "eu.proxy.netbird.io"}, nil
+		},
+	}
+
+	mgr := Manager{store: &stubStore{}, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.NoError(t, err)
+	assert.Equal(t, []string{"shared.example.com", "byop.example.com", "eu.proxy.netbird.io"}, result)
+}
+
+func TestGetClusterAllowList_NoBYOP_FallbackToShared(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return nil, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return []string{"eu.proxy.netbird.io", "us.proxy.netbird.io"}, nil
+		},
+	}
+
+	mgr := Manager{store: &stubStore{}, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.NoError(t, err)
+	assert.Equal(t, []string{"eu.proxy.netbird.io", "us.proxy.netbird.io"}, result)
+}
+
+func TestGetClusterAllowList_BYOPError_ReturnsError(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return nil, errors.New("db error")
+		},
+	}
+
+	mgr := Manager{store: &stubStore{}, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.Error(t, err)
+	assert.Nil(t, result)
+	assert.Contains(t, err.Error(), "BYOP cluster addresses")
+}
+
+func TestGetClusterAllowList_PublicError_ReturnsError(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return []string{"byop.example.com"}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return nil, errors.New("db error")
+		},
+	}
+
+	mgr := Manager{store: &stubStore{}, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.Error(t, err)
+	assert.Nil(t, result)
+	assert.Contains(t, err.Error(), "public cluster addresses")
+}
+
+func TestGetClusterAllowList_BYOPEmptySlice_FallbackToShared(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return []string{}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return []string{"eu.proxy.netbird.io"}, nil
+		},
+	}
+
+	mgr := Manager{store: &stubStore{}, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.NoError(t, err)
+	assert.Equal(t, []string{"eu.proxy.netbird.io"}, result)
+}
+
+func TestGetClusterAllowList_PublicEmpty_BYOPOnly(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return []string{"byop.example.com"}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return nil, nil
+		},
+	}
+
+	mgr := Manager{store: &stubStore{}, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.NoError(t, err)
+	assert.Equal(t, []string{"byop.example.com"}, result)
+}
+
+// stubStore satisfies the manager's narrow store interface for allow-list
+// tests. Only the agent-network settings lookup participates; the default (a
+// nil func) reads as "no settings row", the state most accounts are in.
+type stubStore struct {
+	getAgentNetworkSettingsFunc func(ctx context.Context, accountID string) (*agentnetworkTypes.Settings, error)
+}
+
+func (s *stubStore) GetAccount(context.Context, string) (*types.Account, error) {
+	panic("not used in allow-list tests")
+}
+
+func (s *stubStore) GetAgentNetworkSettings(ctx context.Context, _ nbstore.LockingStrength, accountID string) (*agentnetworkTypes.Settings, error) {
+	if s.getAgentNetworkSettingsFunc != nil {
+		return s.getAgentNetworkSettingsFunc(ctx, accountID)
+	}
+	return nil, status.Errorf(status.NotFound, "agent network settings for account %s not found", accountID)
+}
+
+func (s *stubStore) GetCustomDomain(context.Context, string, string) (*domain.Domain, error) {
+	panic("not used in allow-list tests")
+}
+
+func (s *stubStore) GetCustomDomainByName(context.Context, string) (*domain.Domain, error) {
+	panic("not used in allow-list tests")
+}
+
+func (s *stubStore) ListFreeDomains(context.Context, string) ([]string, error) {
+	panic("not used in allow-list tests")
+}
+
+func (s *stubStore) ListCustomDomains(context.Context, string) ([]*domain.Domain, error) {
+	panic("not used in allow-list tests")
+}
+
+func (s *stubStore) CreateCustomDomain(context.Context, string, string, string, bool) (*domain.Domain, error) {
+	panic("not used in allow-list tests")
+}
+
+func (s *stubStore) UpdateCustomDomain(context.Context, string, *domain.Domain) (*domain.Domain, error) {
+	panic("not used in allow-list tests")
+}
+
+func (s *stubStore) DeleteCustomDomain(context.Context, string, string) error {
+	panic("not used in allow-list tests")
+}
+
+// TestGetClusterAllowList_DedicatedGatewayAddressExcluded pins invariant (B)'s
+// chokepoint: a self-addressed settings pin reserves the account's gateway
+// address, so it is dropped from the allow list — which, because the
+// free-domain suffix match is depth-independent, rejects every name beneath
+// it as well as the bare one. Other addresses are unaffected.
+func TestGetClusterAllowList_DedicatedGatewayAddressExcluded(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return []string{"brave-otter.gateway.example.com", "byop.example.com"}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return []string{"eu.proxy.netbird.io"}, nil
+		},
+	}
+	st := &stubStore{
+		getAgentNetworkSettingsFunc: func(_ context.Context, accountID string) (*agentnetworkTypes.Settings, error) {
+			assert.Equal(t, "acc-123", accountID,
+				"the exclusion must look up the requesting account's own settings")
+			return &agentnetworkTypes.Settings{
+				AccountID:    accountID,
+				Domain:       "brave-otter.gateway.example.com",
+				ProxyAddress: "brave-otter.gateway.example.com",
+			}, nil
+		},
+	}
+
+	mgr := Manager{store: st, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.NoError(t, err)
+	assert.Equal(t, []string{"byop.example.com", "eu.proxy.netbird.io"}, result,
+		"the dedicated gateway address must be reserved from cluster selection")
+}
+
+// TestGetClusterAllowList_LabeledPinDoesNotExclude pins the counterpart: a
+// labeled pin means the gateway rides on a shared cluster serving ordinary
+// services too, so nothing is reserved.
+func TestGetClusterAllowList_LabeledPinDoesNotExclude(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return []string{"byop.example.com"}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return []string{"eu.proxy.netbird.io"}, nil
+		},
+	}
+	st := &stubStore{
+		getAgentNetworkSettingsFunc: func(_ context.Context, accountID string) (*agentnetworkTypes.Settings, error) {
+			return &agentnetworkTypes.Settings{
+				AccountID:    accountID,
+				Domain:       "violet.eu.proxy.netbird.io",
+				ProxyAddress: "eu.proxy.netbird.io",
+			}, nil
+		},
+	}
+
+	mgr := Manager{store: st, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.NoError(t, err)
+	assert.Equal(t, []string{"byop.example.com", "eu.proxy.netbird.io"}, result,
+		"a labeled pin reserves nothing")
+}
+
+// TestGetClusterAllowList_SettingsLookupError_ReturnsError pins that a store
+// outage is surfaced rather than silently treated as "nothing reserved" —
+// failing open here would offer a reserved gateway address for ordinary
+// services.
+func TestGetClusterAllowList_SettingsLookupError_ReturnsError(t *testing.T) {
+	pm := &mockProxyManager{
+		getActiveClusterAddressesForAccountFunc: func(_ context.Context, _ string) ([]string, error) {
+			return []string{"byop.example.com"}, nil
+		},
+		getActiveClusterAddressesFunc: func(_ context.Context) ([]string, error) {
+			return []string{"eu.proxy.netbird.io"}, nil
+		},
+	}
+	st := &stubStore{
+		getAgentNetworkSettingsFunc: func(_ context.Context, _ string) (*agentnetworkTypes.Settings, error) {
+			return nil, status.Errorf(status.Internal, "store outage")
+		},
+	}
+
+	mgr := Manager{store: st, proxyManager: pm}
+	result, err := mgr.getClusterAllowList(context.Background(), "acc-123")
+	require.Error(t, err)
+	assert.Nil(t, result)
+	assert.Contains(t, err.Error(), "agent network settings")
+}