Aligns new tests to signature

by moving the persistence from applySync to the map state manager

.github/workflows/golang-test-darwin.yml

@@ -45,7 +45,7 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -coverprofile=coverage.txt -tags=devcert -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined)
+        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -coverprofile=coverage.txt -tags 'devcert privileged' -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/testutil/privileged)
 
       - name: Upload coverage reports to Codecov
         uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f #v7.0.0

.github/workflows/golang-test-freebsd.yml

@@ -48,14 +48,14 @@ jobs:
             export PATH=$PATH:/usr/local/go/bin:$HOME/go/bin
             time go build -o netbird client/main.go
             # check all component except management, since we do not support management server on freebsd
-            time go test -timeout 1m -failfast ./base62/...
+            time go test -tags privileged -timeout 1m -failfast ./base62/...
             # NOTE: without -p1 `client/internal/dns` will fail because of `listen udp4 :33100: bind: address already in use`
-            time go test -timeout 8m -failfast -v -p 1 ./client/...
-            time go test -timeout 1m -failfast ./dns/...
-            time go test -timeout 1m -failfast ./encryption/...
-            time go test -timeout 1m -failfast ./formatter/...
-            time go test -timeout 1m -failfast ./client/iface/...
-            time go test -timeout 1m -failfast ./route/...
-            time go test -timeout 1m -failfast ./sharedsock/...
-            time go test -timeout 1m -failfast ./util/...
-            time go test -timeout 1m -failfast ./version/...
+            time go test -tags privileged -timeout 8m -failfast -v -p 1 ./client/...
+            time go test -tags privileged -timeout 1m -failfast ./dns/...
+            time go test -tags privileged -timeout 1m -failfast ./encryption/...
+            time go test -tags privileged -timeout 1m -failfast ./formatter/...
+            time go test -tags privileged -timeout 1m -failfast ./client/iface/...
+            time go test -tags privileged -timeout 1m -failfast ./route/...
+            time go test -tags privileged -timeout 1m -failfast ./sharedsock/...
+            time go test -tags privileged -timeout 1m -failfast ./util/...
+            time go test -tags privileged -timeout 1m -failfast ./version/...

.github/workflows/golang-test-linux.yml

@@ -158,7 +158,7 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} CI=true go test -coverprofile=coverage.txt -tags devcert -exec 'sudo' -timeout 10m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined)
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} CI=true go test -coverprofile=coverage.txt -tags devcert -timeout 10m -p 1 $(go list ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined)
 
       - name: Upload coverage reports to Codecov
         if: matrix.arch == 'amd64'
@@ -229,7 +229,7 @@ jobs:
             sh -c ' \
               apk update; apk add --no-cache \
                 ca-certificates iptables ip6tables dbus dbus-dev libpcap-dev build-base; \
-              go test -buildvcs=false -tags devcert -v -timeout 10m -p 1 $(go list -buildvcs=false ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /upload-server)
+              go test -buildvcs=false -tags "devcert privileged" -v -timeout 10m -p 1 $(go list -buildvcs=false ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /upload-server -e /client/testutil/privileged)
             '
 
   test_relay:

.github/workflows/golang-test-windows.yml

@@ -68,7 +68,7 @@ jobs:
         run: |
           $packages = go list ./... | Where-Object { $_ -notmatch '/management' } | Where-Object { $_ -notmatch '/relay' } | Where-Object { $_ -notmatch '/signal' } | Where-Object { $_ -notmatch '/proxy' } | Where-Object { $_ -notmatch '/combined' }
           $goExe = "C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe"
-          $cmd = "$goExe test -tags=devcert -timeout 10m -p 1 $($packages -join ' ') > test-out.txt 2>&1"
+          $cmd = "$goExe test -tags `"devcert privileged`" -timeout 10m -p 1 $($packages -join ' ') > test-out.txt 2>&1"
           Set-Content -Path "${{ github.workspace }}\run-tests.cmd" -Value $cmd
 
       - name: test

Makefile

@@ -1,4 +1,4 @@
-.PHONY: lint lint-all lint-install setup-hooks
+.PHONY: lint lint-all lint-install setup-hooks test-unit test-privileged
 GOLANGCI_LINT := $(shell pwd)/bin/golangci-lint
 
 # Install golangci-lint locally if needed
@@ -25,3 +25,15 @@ setup-hooks:
 	@git config core.hooksPath .githooks
 	@chmod +x .githooks/pre-push
 	@echo "✅ Git hooks configured! Pre-push will now run 'make lint'"
+
+# Host-safe unit tests: excludes the privileged-tagged tests (root / system-mutating).
+# Runs as a normal user with no sudo and leaves host networking untouched.
+test-unit:
+	@go test -tags devcert -timeout 10m ./...
+
+# Privileged suite: runs the `privileged`-tagged tests inside a --privileged
+# --cap-add=NET_ADMIN container via the ory/dockertest harness. Requires Docker.
+# Narrow the run with env vars, e.g.:
+#   PRIV_RUN=TestNftablesManager PRIV_PKGS=./client/firewall/nftables/... make test-privileged
+test-privileged:
+	@go test -tags 'devcert privileged' -timeout 30m -run TestRunPrivilegedSuiteInDocker -v ./client/testutil/privileged/...

client/cmd/service_privileged_test.go

@@ -0,0 +1,196 @@
+//go:build privileged
+
+package cmd
+
+import (
+	"context"
+	"fmt"
+	"os"
+	"runtime"
+	"testing"
+	"time"
+
+	"github.com/kardianos/service"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+const (
+	serviceStartTimeout = 10 * time.Second
+	serviceStopTimeout  = 5 * time.Second
+	statusPollInterval  = 500 * time.Millisecond
+)
+
+// waitForServiceStatus waits for service to reach expected status with timeout
+func waitForServiceStatus(expectedStatus service.Status, timeout time.Duration) (bool, error) {
+	cfg, err := newSVCConfig()
+	if err != nil {
+		return false, err
+	}
+
+	ctxSvc, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
+	if err != nil {
+		return false, err
+	}
+
+	ctx, timeoutCancel := context.WithTimeout(context.Background(), timeout)
+	defer timeoutCancel()
+
+	ticker := time.NewTicker(statusPollInterval)
+	defer ticker.Stop()
+
+	for {
+		select {
+		case <-ctx.Done():
+			return false, fmt.Errorf("timeout waiting for service status %v", expectedStatus)
+		case <-ticker.C:
+			status, err := s.Status()
+			if err != nil {
+				// Continue polling on transient errors
+				continue
+			}
+			if status == expectedStatus {
+				return true, nil
+			}
+		}
+	}
+}
+
+// TestServiceLifecycle tests the complete service lifecycle
+func TestServiceLifecycle(t *testing.T) {
+	// TODO: Add support for Windows and macOS
+	if runtime.GOOS != "linux" && runtime.GOOS != "freebsd" {
+		t.Skipf("Skipping service lifecycle test on unsupported OS: %s", runtime.GOOS)
+	}
+
+	if os.Getenv("CONTAINER") == "true" {
+		t.Skip("Skipping service lifecycle test in container environment")
+	}
+
+	originalServiceName := serviceName
+	serviceName = "netbirdtest" + fmt.Sprintf("%d", time.Now().Unix())
+	defer func() {
+		serviceName = originalServiceName
+	}()
+
+	tempDir := t.TempDir()
+	configPath = fmt.Sprintf("%s/netbird-test-config.json", tempDir)
+	logLevel = "info"
+	daemonAddr = fmt.Sprintf("unix://%s/netbird-test.sock", tempDir)
+
+	// Ensure cleanup even if a subtest fails and Stop/Uninstall subtests don't run.
+	t.Cleanup(func() {
+		cfg, err := newSVCConfig()
+		if err != nil {
+			t.Errorf("cleanup: create service config: %v", err)
+			return
+		}
+		ctxSvc, cancel := context.WithCancel(context.Background())
+		defer cancel()
+		s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
+		if err != nil {
+			t.Errorf("cleanup: create service: %v", err)
+			return
+		}
+
+		// If the subtests already cleaned up, there's nothing to do.
+		if _, err := s.Status(); err != nil {
+			return
+		}
+
+		if err := s.Stop(); err != nil {
+			t.Errorf("cleanup: stop service: %v", err)
+		}
+		if err := s.Uninstall(); err != nil {
+			t.Errorf("cleanup: uninstall service: %v", err)
+		}
+	})
+
+	ctx := context.Background()
+
+	t.Run("Install", func(t *testing.T) {
+		installCmd.SetContext(ctx)
+		err := installCmd.RunE(installCmd, []string{})
+		require.NoError(t, err)
+
+		cfg, err := newSVCConfig()
+		require.NoError(t, err)
+
+		ctxSvc, cancel := context.WithCancel(context.Background())
+		defer cancel()
+
+		s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
+		require.NoError(t, err)
+
+		status, err := s.Status()
+		assert.NoError(t, err)
+		assert.NotEqual(t, service.StatusUnknown, status)
+	})
+
+	t.Run("Start", func(t *testing.T) {
+		startCmd.SetContext(ctx)
+		err := startCmd.RunE(startCmd, []string{})
+		require.NoError(t, err)
+
+		running, err := waitForServiceStatus(service.StatusRunning, serviceStartTimeout)
+		require.NoError(t, err)
+		assert.True(t, running)
+	})
+
+	t.Run("Restart", func(t *testing.T) {
+		restartCmd.SetContext(ctx)
+		err := restartCmd.RunE(restartCmd, []string{})
+		require.NoError(t, err)
+
+		running, err := waitForServiceStatus(service.StatusRunning, serviceStartTimeout)
+		require.NoError(t, err)
+		assert.True(t, running)
+	})
+
+	t.Run("Reconfigure", func(t *testing.T) {
+		originalLogLevel := logLevel
+		logLevel = "debug"
+		defer func() {
+			logLevel = originalLogLevel
+		}()
+
+		reconfigureCmd.SetContext(ctx)
+		err := reconfigureCmd.RunE(reconfigureCmd, []string{})
+		require.NoError(t, err)
+
+		running, err := waitForServiceStatus(service.StatusRunning, serviceStartTimeout)
+		require.NoError(t, err)
+		assert.True(t, running)
+	})
+
+	t.Run("Stop", func(t *testing.T) {
+		stopCmd.SetContext(ctx)
+		err := stopCmd.RunE(stopCmd, []string{})
+		require.NoError(t, err)
+
+		stopped, err := waitForServiceStatus(service.StatusStopped, serviceStopTimeout)
+		require.NoError(t, err)
+		assert.True(t, stopped)
+	})
+
+	t.Run("Uninstall", func(t *testing.T) {
+		uninstallCmd.SetContext(ctx)
+		err := uninstallCmd.RunE(uninstallCmd, []string{})
+		require.NoError(t, err)
+
+		cfg, err := newSVCConfig()
+		require.NoError(t, err)
+
+		ctxSvc, cancel := context.WithCancel(context.Background())
+		defer cancel()
+
+		s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
+		require.NoError(t, err)
+
+		_, err = s.Status()
+		assert.Error(t, err)
+	})
+}

client/cmd/service_test.go

@@ -1,16 +1,12 @@
 package cmd
 
 import (
-	"context"
-	"fmt"
 	"os"
 	"os/signal"
 	"runtime"
 	"syscall"
 	"testing"
-	"time"
 
-	"github.com/kardianos/service"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -31,186 +27,6 @@ func TestMain(m *testing.M) {
 	os.Exit(m.Run())
 }
 
-const (
-	serviceStartTimeout = 10 * time.Second
-	serviceStopTimeout  = 5 * time.Second
-	statusPollInterval  = 500 * time.Millisecond
-)
-
-// waitForServiceStatus waits for service to reach expected status with timeout
-func waitForServiceStatus(expectedStatus service.Status, timeout time.Duration) (bool, error) {
-	cfg, err := newSVCConfig()
-	if err != nil {
-		return false, err
-	}
-
-	ctxSvc, cancel := context.WithCancel(context.Background())
-	defer cancel()
-
-	s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
-	if err != nil {
-		return false, err
-	}
-
-	ctx, timeoutCancel := context.WithTimeout(context.Background(), timeout)
-	defer timeoutCancel()
-
-	ticker := time.NewTicker(statusPollInterval)
-	defer ticker.Stop()
-
-	for {
-		select {
-		case <-ctx.Done():
-			return false, fmt.Errorf("timeout waiting for service status %v", expectedStatus)
-		case <-ticker.C:
-			status, err := s.Status()
-			if err != nil {
-				// Continue polling on transient errors
-				continue
-			}
-			if status == expectedStatus {
-				return true, nil
-			}
-		}
-	}
-}
-
-// TestServiceLifecycle tests the complete service lifecycle
-func TestServiceLifecycle(t *testing.T) {
-	// TODO: Add support for Windows and macOS
-	if runtime.GOOS != "linux" && runtime.GOOS != "freebsd" {
-		t.Skipf("Skipping service lifecycle test on unsupported OS: %s", runtime.GOOS)
-	}
-
-	if os.Getenv("CONTAINER") == "true" {
-		t.Skip("Skipping service lifecycle test in container environment")
-	}
-
-	originalServiceName := serviceName
-	serviceName = "netbirdtest" + fmt.Sprintf("%d", time.Now().Unix())
-	defer func() {
-		serviceName = originalServiceName
-	}()
-
-	tempDir := t.TempDir()
-	configPath = fmt.Sprintf("%s/netbird-test-config.json", tempDir)
-	logLevel = "info"
-	daemonAddr = fmt.Sprintf("unix://%s/netbird-test.sock", tempDir)
-
-	// Ensure cleanup even if a subtest fails and Stop/Uninstall subtests don't run.
-	t.Cleanup(func() {
-		cfg, err := newSVCConfig()
-		if err != nil {
-			t.Errorf("cleanup: create service config: %v", err)
-			return
-		}
-		ctxSvc, cancel := context.WithCancel(context.Background())
-		defer cancel()
-		s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
-		if err != nil {
-			t.Errorf("cleanup: create service: %v", err)
-			return
-		}
-
-		// If the subtests already cleaned up, there's nothing to do.
-		if _, err := s.Status(); err != nil {
-			return
-		}
-
-		if err := s.Stop(); err != nil {
-			t.Errorf("cleanup: stop service: %v", err)
-		}
-		if err := s.Uninstall(); err != nil {
-			t.Errorf("cleanup: uninstall service: %v", err)
-		}
-	})
-
-	ctx := context.Background()
-
-	t.Run("Install", func(t *testing.T) {
-		installCmd.SetContext(ctx)
-		err := installCmd.RunE(installCmd, []string{})
-		require.NoError(t, err)
-
-		cfg, err := newSVCConfig()
-		require.NoError(t, err)
-
-		ctxSvc, cancel := context.WithCancel(context.Background())
-		defer cancel()
-
-		s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
-		require.NoError(t, err)
-
-		status, err := s.Status()
-		assert.NoError(t, err)
-		assert.NotEqual(t, service.StatusUnknown, status)
-	})
-
-	t.Run("Start", func(t *testing.T) {
-		startCmd.SetContext(ctx)
-		err := startCmd.RunE(startCmd, []string{})
-		require.NoError(t, err)
-
-		running, err := waitForServiceStatus(service.StatusRunning, serviceStartTimeout)
-		require.NoError(t, err)
-		assert.True(t, running)
-	})
-
-	t.Run("Restart", func(t *testing.T) {
-		restartCmd.SetContext(ctx)
-		err := restartCmd.RunE(restartCmd, []string{})
-		require.NoError(t, err)
-
-		running, err := waitForServiceStatus(service.StatusRunning, serviceStartTimeout)
-		require.NoError(t, err)
-		assert.True(t, running)
-	})
-
-	t.Run("Reconfigure", func(t *testing.T) {
-		originalLogLevel := logLevel
-		logLevel = "debug"
-		defer func() {
-			logLevel = originalLogLevel
-		}()
-
-		reconfigureCmd.SetContext(ctx)
-		err := reconfigureCmd.RunE(reconfigureCmd, []string{})
-		require.NoError(t, err)
-
-		running, err := waitForServiceStatus(service.StatusRunning, serviceStartTimeout)
-		require.NoError(t, err)
-		assert.True(t, running)
-	})
-
-	t.Run("Stop", func(t *testing.T) {
-		stopCmd.SetContext(ctx)
-		err := stopCmd.RunE(stopCmd, []string{})
-		require.NoError(t, err)
-
-		stopped, err := waitForServiceStatus(service.StatusStopped, serviceStopTimeout)
-		require.NoError(t, err)
-		assert.True(t, stopped)
-	})
-
-	t.Run("Uninstall", func(t *testing.T) {
-		uninstallCmd.SetContext(ctx)
-		err := uninstallCmd.RunE(uninstallCmd, []string{})
-		require.NoError(t, err)
-
-		cfg, err := newSVCConfig()
-		require.NoError(t, err)
-
-		ctxSvc, cancel := context.WithCancel(context.Background())
-		defer cancel()
-
-		s, err := newSVC(newProgram(ctxSvc, cancel), cfg)
-		require.NoError(t, err)
-
-		_, err = s.Status()
-		assert.Error(t, err)
-	})
-}
-
 // TestServiceEnvVars tests environment variable parsing
 func TestServiceEnvVars(t *testing.T) {
 	tests := []struct {

client/firewall/iptables/manager_linux_test.go

@@ -1,3 +1,5 @@
+//go:build privileged
+
 package iptables
 
 import (

client/firewall/iptables/router_linux_test.go

@@ -1,4 +1,4 @@
-//go:build !android
+//go:build !android && privileged
 
 package iptables
 

client/firewall/nftables/manager_linux_test.go

@@ -1,3 +1,5 @@
+//go:build privileged
+
 package nftables
 
 import (

client/firewall/nftables/router_linux_test.go

@@ -1,4 +1,4 @@
-//go:build !android
+//go:build !android && privileged
 
 package nftables
 

client/iface/iface_test.go

@@ -1,3 +1,5 @@
+//go:build privileged
+
 package iface
 
 import (

client/iface/wgproxy/bind/proxy.go

@@ -136,11 +136,6 @@ func (p *ProxyBind) CloseConn() error {
 	return p.close()
 }
 
-// InjectPacket is a no-op for the userspace proxy: first-packet reinjection is kernel-only.
-func (p *ProxyBind) InjectPacket(_ []byte) error {
-	return nil
-}
-
 func (p *ProxyBind) close() error {
 	if p.remoteConn == nil {
 		return nil

client/iface/wgproxy/ebpf/wrapper.go

@@ -219,17 +219,6 @@ func (p *ProxyWrapper) RedirectAs(endpoint *net.UDPAddr) {
 	p.pausedCond.L.Unlock()
 }
 
-// InjectPacket writes b to the remote peer over the underlying transport.
-func (p *ProxyWrapper) InjectPacket(b []byte) error {
-	if p.remoteConn == nil {
-		return errors.New("proxy not started")
-	}
-	if _, err := p.remoteConn.Write(b); err != nil {
-		return err
-	}
-	return nil
-}
-
 // CloseConn close the remoteConn and automatically remove the conn instance from the map
 func (p *ProxyWrapper) CloseConn() error {
 	if p.cancel == nil {

client/iface/wgproxy/proxy.go

@@ -18,9 +18,4 @@ type Proxy interface {
 	RedirectAs(endpoint *net.UDPAddr)
 	CloseConn() error
 	SetDisconnectListener(disconnected func())
-
-	// InjectPacket writes a raw packet directly to the remote peer over the underlying transport,
-	// bypassing WireGuard. Used to replay the captured lazyconn handshake initiation. Only the
-	// kernel-mode proxies act on it; the userspace proxy is a no-op since reinjection is kernel-only.
-	InjectPacket(b []byte) error
 }

client/iface/wgproxy/proxy_linux_test.go

@@ -1,4 +1,4 @@
-//go:build linux && !android
+//go:build linux && !android && privileged
 
 package wgproxy
 

client/iface/wgproxy/proxy_seed_test.go

@@ -1,4 +1,4 @@
-//go:build !linux
+//go:build !linux || !privileged
 
 package wgproxy
 

client/iface/wgproxy/redirect_test.go

@@ -1,4 +1,4 @@
-//go:build linux && !android
+//go:build linux && !android && privileged
 
 package wgproxy
 
@@ -26,64 +26,6 @@ func compareUDPAddr(addr1, addr2 net.Addr) bool {
 	return udpAddr1.IP.Equal(udpAddr2.IP) && udpAddr1.Port == udpAddr2.Port
 }
 
-// TestRedirectAs_eBPF_IPv4 tests RedirectAs with eBPF proxy using IPv4 addresses
-func TestRedirectAs_eBPF_IPv4(t *testing.T) {
-	wgPort := 51850
-	ebpfProxy := ebpf.NewWGEBPFProxy(wgPort, 1280)
-	if err := ebpfProxy.Listen(); err != nil {
-		t.Fatalf("failed to initialize ebpf proxy: %v", err)
-	}
-	defer func() {
-		if err := ebpfProxy.Free(); err != nil {
-			t.Errorf("failed to free ebpf proxy: %v", err)
-		}
-	}()
-
-	proxy := ebpf.NewProxyWrapper(ebpfProxy)
-
-	// NetBird UDP address of the remote peer
-	nbAddr := &net.UDPAddr{
-		IP:   net.ParseIP("100.108.111.177"),
-		Port: 38746,
-	}
-
-	p2pEndpoint := &net.UDPAddr{
-		IP:   net.ParseIP("192.168.0.56"),
-		Port: 51820,
-	}
-
-	testRedirectAs(t, proxy, wgPort, nbAddr, p2pEndpoint)
-}
-
-// TestRedirectAs_eBPF_IPv6 tests RedirectAs with eBPF proxy using IPv6 addresses
-func TestRedirectAs_eBPF_IPv6(t *testing.T) {
-	wgPort := 51851
-	ebpfProxy := ebpf.NewWGEBPFProxy(wgPort, 1280)
-	if err := ebpfProxy.Listen(); err != nil {
-		t.Fatalf("failed to initialize ebpf proxy: %v", err)
-	}
-	defer func() {
-		if err := ebpfProxy.Free(); err != nil {
-			t.Errorf("failed to free ebpf proxy: %v", err)
-		}
-	}()
-
-	proxy := ebpf.NewProxyWrapper(ebpfProxy)
-
-	// NetBird UDP address of the remote peer
-	nbAddr := &net.UDPAddr{
-		IP:   net.ParseIP("100.108.111.177"),
-		Port: 38746,
-	}
-
-	p2pEndpoint := &net.UDPAddr{
-		IP:   net.ParseIP("fe80::56"),
-		Port: 51820,
-	}
-
-	testRedirectAs(t, proxy, wgPort, nbAddr, p2pEndpoint)
-}
-
 // TestRedirectAs_UDP_IPv4 tests RedirectAs with UDP proxy using IPv4 addresses
 func TestRedirectAs_UDP_IPv4(t *testing.T) {
 	wgPort := 51852
@@ -256,6 +198,64 @@ func testRedirectAs(t *testing.T, proxy Proxy, wgPort int, nbAddr, p2pEndpoint *
 	}
 }
 
+// TestRedirectAs_eBPF_IPv4 tests RedirectAs with eBPF proxy using IPv4 addresses
+func TestRedirectAs_eBPF_IPv4(t *testing.T) {
+	wgPort := 51850
+	ebpfProxy := ebpf.NewWGEBPFProxy(wgPort, 1280)
+	if err := ebpfProxy.Listen(); err != nil {
+		t.Fatalf("failed to initialize ebpf proxy: %v", err)
+	}
+	defer func() {
+		if err := ebpfProxy.Free(); err != nil {
+			t.Errorf("failed to free ebpf proxy: %v", err)
+		}
+	}()
+
+	proxy := ebpf.NewProxyWrapper(ebpfProxy)
+
+	// NetBird UDP address of the remote peer
+	nbAddr := &net.UDPAddr{
+		IP:   net.ParseIP("100.108.111.177"),
+		Port: 38746,
+	}
+
+	p2pEndpoint := &net.UDPAddr{
+		IP:   net.ParseIP("192.168.0.56"),
+		Port: 51820,
+	}
+
+	testRedirectAs(t, proxy, wgPort, nbAddr, p2pEndpoint)
+}
+
+// TestRedirectAs_eBPF_IPv6 tests RedirectAs with eBPF proxy using IPv6 addresses
+func TestRedirectAs_eBPF_IPv6(t *testing.T) {
+	wgPort := 51851
+	ebpfProxy := ebpf.NewWGEBPFProxy(wgPort, 1280)
+	if err := ebpfProxy.Listen(); err != nil {
+		t.Fatalf("failed to initialize ebpf proxy: %v", err)
+	}
+	defer func() {
+		if err := ebpfProxy.Free(); err != nil {
+			t.Errorf("failed to free ebpf proxy: %v", err)
+		}
+	}()
+
+	proxy := ebpf.NewProxyWrapper(ebpfProxy)
+
+	// NetBird UDP address of the remote peer
+	nbAddr := &net.UDPAddr{
+		IP:   net.ParseIP("100.108.111.177"),
+		Port: 38746,
+	}
+
+	p2pEndpoint := &net.UDPAddr{
+		IP:   net.ParseIP("fe80::56"),
+		Port: 51820,
+	}
+
+	testRedirectAs(t, proxy, wgPort, nbAddr, p2pEndpoint)
+}
+
 // TestRedirectAs_Multiple_Switches tests switching between multiple endpoints
 func TestRedirectAs_Multiple_Switches(t *testing.T) {
 	wgPort := 51856

client/iface/wgproxy/udp/proxy.go

@@ -147,17 +147,6 @@ func (p *WGUDPProxy) RedirectAs(endpoint *net.UDPAddr) {
 	p.sendPkg = p.srcFakerConn.SendPkg
 }
 
-// InjectPacket writes b to the remote peer over the underlying transport.
-func (p *WGUDPProxy) InjectPacket(b []byte) error {
-	if p.remoteConn == nil {
-		return errors.New("proxy not started")
-	}
-	if _, err := p.remoteConn.Write(b); err != nil {
-		return err
-	}
-	return nil
-}
-
 // CloseConn close the localConn
 func (p *WGUDPProxy) CloseConn() error {
 	if p.cancel == nil {

client/internal/dns/server_privileged_test.go

@@ -0,0 +1,485 @@
+//go:build privileged
+
+package dns
+
+import (
+	"context"
+	"fmt"
+	"net/netip"
+	"os"
+	"testing"
+
+	"github.com/golang/mock/gomock"
+	"github.com/miekg/dns"
+	"github.com/stretchr/testify/assert"
+	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
+
+	"github.com/netbirdio/netbird/client/iface"
+	pfmock "github.com/netbirdio/netbird/client/iface/mocks"
+	"github.com/netbirdio/netbird/client/iface/wgaddr"
+	"github.com/netbirdio/netbird/client/internal/dns/local"
+	"github.com/netbirdio/netbird/client/internal/dns/test"
+	"github.com/netbirdio/netbird/client/internal/peer"
+	"github.com/netbirdio/netbird/client/internal/stdnet"
+	nbdns "github.com/netbirdio/netbird/dns"
+)
+
+func TestUpdateDNSServer(t *testing.T) {
+
+	nameServers := []nbdns.NameServer{
+		{
+			IP:     netip.MustParseAddr("8.8.8.8"),
+			NSType: nbdns.UDPNameServerType,
+			Port:   53,
+		},
+		{
+			IP:     netip.MustParseAddr("8.8.4.4"),
+			NSType: nbdns.UDPNameServerType,
+			Port:   53,
+		},
+	}
+
+	testCases := []struct {
+		name                string
+		initUpstreamMap     []handlerWrapper
+		initLocalZones      []nbdns.CustomZone
+		initSerial          uint64
+		inputSerial         uint64
+		inputUpdate         nbdns.Config
+		shouldFail          bool
+		expectedUpstreamMap []handlerWrapper
+		expectedLocalQs     []dns.Question
+	}{
+		{
+			name:            "Initial Config Should Succeed",
+			initUpstreamMap: nil,
+			initSerial:      0,
+			inputSerial:     1,
+			inputUpdate: nbdns.Config{
+				ServiceEnable: true,
+				CustomZones: []nbdns.CustomZone{
+					{
+						Domain:  "netbird.cloud",
+						Records: zoneRecords,
+					},
+				},
+				NameServerGroups: []*nbdns.NameServerGroup{
+					{
+						Domains:     []string{"netbird.io"},
+						NameServers: nameServers,
+					},
+					{
+						NameServers: nameServers,
+						Primary:     true,
+					},
+				},
+			},
+			expectedUpstreamMap: []handlerWrapper{
+				{
+					domain:   "netbird.io",
+					priority: PriorityUpstream,
+				},
+				{
+					domain:   "netbird.cloud",
+					priority: PriorityLocal,
+				},
+				{
+					domain:   nbdns.RootZone,
+					priority: PriorityDefault,
+				},
+			},
+			expectedLocalQs: []dns.Question{{Name: "peera.netbird.cloud.", Qtype: dns.TypeA, Qclass: dns.ClassINET}},
+		},
+		{
+			name:           "New Config Should Succeed",
+			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
+			initUpstreamMap: []handlerWrapper{
+				{
+					domain:   "netbird.cloud",
+					handler:  &mockHandler{},
+					priority: PriorityUpstream,
+				},
+			},
+			initSerial:  0,
+			inputSerial: 1,
+			inputUpdate: nbdns.Config{
+				ServiceEnable: true,
+				CustomZones: []nbdns.CustomZone{
+					{
+						Domain:  "netbird.cloud",
+						Records: zoneRecords,
+					},
+				},
+				NameServerGroups: []*nbdns.NameServerGroup{
+					{
+						Domains:     []string{"netbird.io"},
+						NameServers: nameServers,
+					},
+				},
+			},
+			expectedUpstreamMap: []handlerWrapper{
+				{
+					domain:   "netbird.io",
+					priority: PriorityUpstream,
+				},
+				{
+					domain:   "netbird.cloud",
+					priority: PriorityLocal,
+				},
+			},
+			expectedLocalQs: []dns.Question{{Name: zoneRecords[0].Name, Qtype: 1, Qclass: 1}},
+		},
+		{
+			name:            "Smaller Config Serial Should Be Skipped",
+			initLocalZones:  []nbdns.CustomZone{},
+			initUpstreamMap: nil,
+			initSerial:      2,
+			inputSerial:     1,
+			shouldFail:      true,
+		},
+		{
+			name:            "Empty NS Group Domain Or Not Primary Element Should Fail",
+			initLocalZones:  []nbdns.CustomZone{},
+			initUpstreamMap: nil,
+			initSerial:      0,
+			inputSerial:     1,
+			inputUpdate: nbdns.Config{
+				ServiceEnable: true,
+				CustomZones: []nbdns.CustomZone{
+					{
+						Domain:  "netbird.cloud",
+						Records: zoneRecords,
+					},
+				},
+				NameServerGroups: []*nbdns.NameServerGroup{
+					{
+						NameServers: nameServers,
+					},
+				},
+			},
+			shouldFail: true,
+		},
+		{
+			name:            "Invalid NS Group Nameservers list Should Fail",
+			initLocalZones:  []nbdns.CustomZone{},
+			initUpstreamMap: nil,
+			initSerial:      0,
+			inputSerial:     1,
+			inputUpdate: nbdns.Config{
+				ServiceEnable: true,
+				CustomZones: []nbdns.CustomZone{
+					{
+						Domain:  "netbird.cloud",
+						Records: zoneRecords,
+					},
+				},
+				NameServerGroups: []*nbdns.NameServerGroup{
+					{
+						NameServers: nameServers,
+					},
+				},
+			},
+			shouldFail: true,
+		},
+		{
+			name:            "Invalid Custom Zone Records list Should Skip",
+			initLocalZones:  []nbdns.CustomZone{},
+			initUpstreamMap: nil,
+			initSerial:      0,
+			inputSerial:     1,
+			inputUpdate: nbdns.Config{
+				ServiceEnable: true,
+				CustomZones: []nbdns.CustomZone{
+					{
+						Domain: "netbird.cloud",
+					},
+				},
+				NameServerGroups: []*nbdns.NameServerGroup{
+					{
+						NameServers: nameServers,
+						Primary:     true,
+					},
+				},
+			},
+			expectedUpstreamMap: []handlerWrapper{{
+				domain:   ".",
+				priority: PriorityDefault,
+			}},
+		},
+		{
+			name:           "Empty Config Should Succeed and Clean Maps",
+			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
+			initUpstreamMap: []handlerWrapper{
+				{
+					domain:   zoneRecords[0].Name,
+					handler:  &mockHandler{},
+					priority: PriorityUpstream,
+				},
+			},
+			initSerial:          0,
+			inputSerial:         1,
+			inputUpdate:         nbdns.Config{ServiceEnable: true},
+			expectedUpstreamMap: nil,
+			expectedLocalQs:     []dns.Question{},
+		},
+		{
+			name:           "Disabled Service Should clean map",
+			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
+			initUpstreamMap: []handlerWrapper{
+				{
+					domain:   zoneRecords[0].Name,
+					handler:  &mockHandler{},
+					priority: PriorityUpstream,
+				},
+			},
+			initSerial:          0,
+			inputSerial:         1,
+			inputUpdate:         nbdns.Config{ServiceEnable: false},
+			expectedUpstreamMap: nil,
+			expectedLocalQs:     []dns.Question{},
+		},
+	}
+
+	for n, testCase := range testCases {
+		t.Run(testCase.name, func(t *testing.T) {
+			privKey, _ := wgtypes.GenerateKey()
+			newNet, err := stdnet.NewNet(context.Background(), nil)
+			if err != nil {
+				t.Fatal(err)
+			}
+
+			opts := iface.WGIFaceOpts{
+				IFaceName:    fmt.Sprintf("utun230%d", n),
+				Address:      wgaddr.MustParseWGAddress(fmt.Sprintf("100.66.100.%d/32", n+1)),
+				WGPort:       33100,
+				WGPrivKey:    privKey.String(),
+				MTU:          iface.DefaultMTU,
+				TransportNet: newNet,
+			}
+
+			wgIface, err := iface.NewWGIFace(opts)
+			if err != nil {
+				t.Fatal(err)
+			}
+			err = wgIface.Create()
+			if err != nil {
+				t.Fatal(err)
+			}
+			defer func() {
+				err = wgIface.Close()
+				if err != nil {
+					t.Log(err)
+				}
+			}()
+			dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
+				WgInterface:    wgIface,
+				CustomAddress:  "",
+				StatusRecorder: peer.NewRecorder("mgm"),
+				StateManager:   nil,
+				DisableSys:     false,
+			})
+			if err != nil {
+				t.Fatal(err)
+			}
+			err = dnsServer.Initialize()
+			if err != nil {
+				t.Fatal(err)
+			}
+			defer func() {
+				err = dnsServer.hostManager.restoreHostDNS()
+				if err != nil {
+					t.Log(err)
+				}
+			}()
+
+			dnsServer.dnsMuxHandlers = testCase.initUpstreamMap
+			dnsServer.localResolver.Update(testCase.initLocalZones)
+			dnsServer.updateSerial = testCase.initSerial
+
+			err = dnsServer.UpdateDNSServer(testCase.inputSerial, testCase.inputUpdate)
+			if err != nil {
+				if testCase.shouldFail {
+					return
+				}
+				t.Fatalf("update dns server should not fail, got error: %v", err)
+			}
+
+			if len(dnsServer.dnsMuxHandlers) != len(testCase.expectedUpstreamMap) {
+				t.Fatalf("update upstream failed, map size is different than expected, want %d, got %d", len(testCase.expectedUpstreamMap), len(dnsServer.dnsMuxHandlers))
+			}
+
+			for _, expected := range testCase.expectedUpstreamMap {
+				found := false
+				for _, got := range dnsServer.dnsMuxHandlers {
+					if got.domain == expected.domain && got.priority == expected.priority {
+						found = true
+						break
+					}
+				}
+				if !found {
+					t.Fatalf("update upstream failed, handler for domain=%s priority=%d not found in dnsMuxHandlers: %#v", expected.domain, expected.priority, dnsServer.dnsMuxHandlers)
+				}
+			}
+
+			var responseMSG *dns.Msg
+			responseWriter := &test.MockResponseWriter{
+				WriteMsgFunc: func(m *dns.Msg) error {
+					responseMSG = m
+					return nil
+				},
+			}
+			for _, q := range testCase.expectedLocalQs {
+				dnsServer.localResolver.ServeDNS(responseWriter, &dns.Msg{
+					Question: []dns.Question{q},
+				})
+			}
+
+			if len(testCase.expectedLocalQs) > 0 {
+				assert.NotNil(t, responseMSG, "response message should not be nil")
+				assert.Equal(t, dns.RcodeSuccess, responseMSG.Rcode, "response code should be success")
+				assert.NotEmpty(t, responseMSG.Answer, "response message should have answers")
+			}
+		})
+	}
+}
+
+func TestDNSFakeResolverHandleUpdates(t *testing.T) {
+	ov := os.Getenv("NB_WG_KERNEL_DISABLED")
+	defer t.Setenv("NB_WG_KERNEL_DISABLED", ov)
+
+	t.Setenv("NB_WG_KERNEL_DISABLED", "true")
+	newNet, err := stdnet.NewNet(context.Background(), []string{"utun2301"})
+	if err != nil {
+		t.Errorf("create stdnet: %v", err)
+		return
+	}
+
+	privKey, _ := wgtypes.GeneratePrivateKey()
+	opts := iface.WGIFaceOpts{
+		IFaceName:    "utun2301",
+		Address:      wgaddr.MustParseWGAddress("100.66.100.1/32"),
+		WGPort:       33100,
+		WGPrivKey:    privKey.String(),
+		MTU:          iface.DefaultMTU,
+		TransportNet: newNet,
+	}
+	wgIface, err := iface.NewWGIFace(opts)
+	if err != nil {
+		t.Errorf("build interface wireguard: %v", err)
+		return
+	}
+
+	err = wgIface.Create()
+	if err != nil {
+		t.Errorf("create and init wireguard interface: %v", err)
+		return
+	}
+	defer func() {
+		if err = wgIface.Close(); err != nil {
+			t.Logf("close wireguard interface: %v", err)
+		}
+	}()
+
+	ctrl := gomock.NewController(t)
+	defer ctrl.Finish()
+
+	packetfilter := pfmock.NewMockPacketFilter(ctrl)
+	packetfilter.EXPECT().FilterOutbound(gomock.Any(), gomock.Any()).AnyTimes()
+	packetfilter.EXPECT().SetUDPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
+	packetfilter.EXPECT().SetTCPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
+
+	if err := wgIface.SetFilter(packetfilter); err != nil {
+		t.Errorf("set packet filter: %v", err)
+		return
+	}
+
+	dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
+		WgInterface:    wgIface,
+		CustomAddress:  "",
+		StatusRecorder: peer.NewRecorder("mgm"),
+		StateManager:   nil,
+		DisableSys:     false,
+	})
+	if err != nil {
+		t.Errorf("create DNS server: %v", err)
+		return
+	}
+
+	err = dnsServer.Initialize()
+	if err != nil {
+		t.Errorf("run DNS server: %v", err)
+		return
+	}
+	defer func() {
+		if err = dnsServer.hostManager.restoreHostDNS(); err != nil {
+			t.Logf("restore DNS settings on the host: %v", err)
+			return
+		}
+	}()
+
+	dnsServer.dnsMuxHandlers = []handlerWrapper{
+		{
+			domain:   zoneRecords[0].Name,
+			handler:  &local.Resolver{},
+			priority: PriorityUpstream,
+		},
+	}
+	dnsServer.localResolver.Update([]nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}})
+	dnsServer.updateSerial = 0
+
+	nameServers := []nbdns.NameServer{
+		{
+			IP:     netip.MustParseAddr("8.8.8.8"),
+			NSType: nbdns.UDPNameServerType,
+			Port:   53,
+		},
+		{
+			IP:     netip.MustParseAddr("8.8.4.4"),
+			NSType: nbdns.UDPNameServerType,
+			Port:   53,
+		},
+	}
+
+	update := nbdns.Config{
+		ServiceEnable: true,
+		CustomZones: []nbdns.CustomZone{
+			{
+				Domain:  "netbird.cloud",
+				Records: zoneRecords,
+			},
+		},
+		NameServerGroups: []*nbdns.NameServerGroup{
+			{
+				Domains:     []string{"netbird.io"},
+				NameServers: nameServers,
+			},
+			{
+				NameServers: nameServers,
+				Primary:     true,
+			},
+		},
+	}
+
+	// Start the server with regular configuration
+	if err := dnsServer.UpdateDNSServer(1, update); err != nil {
+		t.Fatalf("update dns server should not fail, got error: %v", err)
+		return
+	}
+
+	update2 := update
+	update2.ServiceEnable = false
+	// Disable the server, stop the listener
+	if err := dnsServer.UpdateDNSServer(2, update2); err != nil {
+		t.Fatalf("update dns server should not fail, got error: %v", err)
+		return
+	}
+
+	update3 := update2
+	update3.NameServerGroups = update3.NameServerGroups[:1]
+	// But service still get updates and we checking that we handle
+	// internal state in the right way
+	if err := dnsServer.UpdateDNSServer(3, update3); err != nil {
+		t.Fatalf("update dns server should not fail, got error: %v", err)
+		return
+	}
+}

client/internal/dns/server_test.go

@@ -10,7 +10,6 @@ import (
 	"testing"
 	"time"
 
-	"github.com/golang/mock/gomock"
 	"github.com/miekg/dns"
 	log "github.com/sirupsen/logrus"
 	"github.com/stretchr/testify/assert"
@@ -23,7 +22,6 @@ import (
 	"github.com/netbirdio/netbird/client/iface"
 	"github.com/netbirdio/netbird/client/iface/configurer"
 	"github.com/netbirdio/netbird/client/iface/device"
-	pfmock "github.com/netbirdio/netbird/client/iface/mocks"
 	"github.com/netbirdio/netbird/client/iface/wgaddr"
 	"github.com/netbirdio/netbird/client/internal/dns/local"
 	"github.com/netbirdio/netbird/client/internal/dns/test"
@@ -104,466 +102,6 @@ func init() {
 	formatter.SetTextFormatter(log.StandardLogger())
 }
 
-func TestUpdateDNSServer(t *testing.T) {
-
-	nameServers := []nbdns.NameServer{
-		{
-			IP:     netip.MustParseAddr("8.8.8.8"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-		{
-			IP:     netip.MustParseAddr("8.8.4.4"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-	}
-
-	testCases := []struct {
-		name                string
-		initUpstreamMap     []handlerWrapper
-		initLocalZones      []nbdns.CustomZone
-		initSerial          uint64
-		inputSerial         uint64
-		inputUpdate         nbdns.Config
-		shouldFail          bool
-		expectedUpstreamMap []handlerWrapper
-		expectedLocalQs     []dns.Question
-	}{
-		{
-			name:            "Initial Config Should Succeed",
-			initUpstreamMap: nil,
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						Domains:     []string{"netbird.io"},
-						NameServers: nameServers,
-					},
-					{
-						NameServers: nameServers,
-						Primary:     true,
-					},
-				},
-			},
-			expectedUpstreamMap: []handlerWrapper{
-				{
-					domain:   "netbird.io",
-					priority: PriorityUpstream,
-				},
-				{
-					domain:   "netbird.cloud",
-					priority: PriorityLocal,
-				},
-				{
-					domain:   nbdns.RootZone,
-					priority: PriorityDefault,
-				},
-			},
-			expectedLocalQs: []dns.Question{{Name: "peera.netbird.cloud.", Qtype: dns.TypeA, Qclass: dns.ClassINET}},
-		},
-		{
-			name:           "New Config Should Succeed",
-			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: 1, Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
-			initUpstreamMap: []handlerWrapper{
-				{
-					domain:   "netbird.cloud",
-					handler:  &mockHandler{},
-					priority: PriorityUpstream,
-				},
-			},
-			initSerial:  0,
-			inputSerial: 1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						Domains:     []string{"netbird.io"},
-						NameServers: nameServers,
-					},
-				},
-			},
-			expectedUpstreamMap: []handlerWrapper{
-				{
-					domain:   "netbird.io",
-					priority: PriorityUpstream,
-				},
-				{
-					domain:   "netbird.cloud",
-					priority: PriorityLocal,
-				},
-			},
-			expectedLocalQs: []dns.Question{{Name: zoneRecords[0].Name, Qtype: 1, Qclass: 1}},
-		},
-		{
-			name:            "Smaller Config Serial Should Be Skipped",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: nil,
-			initSerial:      2,
-			inputSerial:     1,
-			shouldFail:      true,
-		},
-		{
-			name:            "Empty NS Group Domain Or Not Primary Element Should Fail",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: nil,
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						NameServers: nameServers,
-					},
-				},
-			},
-			shouldFail: true,
-		},
-		{
-			name:            "Invalid NS Group Nameservers list Should Fail",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: nil,
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain:  "netbird.cloud",
-						Records: zoneRecords,
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						NameServers: nameServers,
-					},
-				},
-			},
-			shouldFail: true,
-		},
-		{
-			name:            "Invalid Custom Zone Records list Should Skip",
-			initLocalZones:  []nbdns.CustomZone{},
-			initUpstreamMap: nil,
-			initSerial:      0,
-			inputSerial:     1,
-			inputUpdate: nbdns.Config{
-				ServiceEnable: true,
-				CustomZones: []nbdns.CustomZone{
-					{
-						Domain: "netbird.cloud",
-					},
-				},
-				NameServerGroups: []*nbdns.NameServerGroup{
-					{
-						NameServers: nameServers,
-						Primary:     true,
-					},
-				},
-			},
-			expectedUpstreamMap: []handlerWrapper{{
-				domain:   ".",
-				priority: PriorityDefault,
-			}},
-		},
-		{
-			name:           "Empty Config Should Succeed and Clean Maps",
-			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
-			initUpstreamMap: []handlerWrapper{
-				{
-					domain:   zoneRecords[0].Name,
-					handler:  &mockHandler{},
-					priority: PriorityUpstream,
-				},
-			},
-			initSerial:          0,
-			inputSerial:         1,
-			inputUpdate:         nbdns.Config{ServiceEnable: true},
-			expectedUpstreamMap: nil,
-			expectedLocalQs:     []dns.Question{},
-		},
-		{
-			name:           "Disabled Service Should clean map",
-			initLocalZones: []nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}},
-			initUpstreamMap: []handlerWrapper{
-				{
-					domain:   zoneRecords[0].Name,
-					handler:  &mockHandler{},
-					priority: PriorityUpstream,
-				},
-			},
-			initSerial:          0,
-			inputSerial:         1,
-			inputUpdate:         nbdns.Config{ServiceEnable: false},
-			expectedUpstreamMap: nil,
-			expectedLocalQs:     []dns.Question{},
-		},
-	}
-
-	for n, testCase := range testCases {
-		t.Run(testCase.name, func(t *testing.T) {
-			privKey, _ := wgtypes.GenerateKey()
-			newNet, err := stdnet.NewNet(context.Background(), nil)
-			if err != nil {
-				t.Fatal(err)
-			}
-
-			opts := iface.WGIFaceOpts{
-				IFaceName:    fmt.Sprintf("utun230%d", n),
-				Address:      wgaddr.MustParseWGAddress(fmt.Sprintf("100.66.100.%d/32", n+1)),
-				WGPort:       33100,
-				WGPrivKey:    privKey.String(),
-				MTU:          iface.DefaultMTU,
-				TransportNet: newNet,
-			}
-
-			wgIface, err := iface.NewWGIFace(opts)
-			if err != nil {
-				t.Fatal(err)
-			}
-			err = wgIface.Create()
-			if err != nil {
-				t.Fatal(err)
-			}
-			defer func() {
-				err = wgIface.Close()
-				if err != nil {
-					t.Log(err)
-				}
-			}()
-			dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
-				WgInterface:    wgIface,
-				CustomAddress:  "",
-				StatusRecorder: peer.NewRecorder("mgm"),
-				StateManager:   nil,
-				DisableSys:     false,
-			})
-			if err != nil {
-				t.Fatal(err)
-			}
-			err = dnsServer.Initialize()
-			if err != nil {
-				t.Fatal(err)
-			}
-			defer func() {
-				err = dnsServer.hostManager.restoreHostDNS()
-				if err != nil {
-					t.Log(err)
-				}
-			}()
-
-			dnsServer.dnsMuxHandlers = testCase.initUpstreamMap
-			dnsServer.localResolver.Update(testCase.initLocalZones)
-			dnsServer.updateSerial = testCase.initSerial
-
-			err = dnsServer.UpdateDNSServer(testCase.inputSerial, testCase.inputUpdate)
-			if err != nil {
-				if testCase.shouldFail {
-					return
-				}
-				t.Fatalf("update dns server should not fail, got error: %v", err)
-			}
-
-			if len(dnsServer.dnsMuxHandlers) != len(testCase.expectedUpstreamMap) {
-				t.Fatalf("update upstream failed, map size is different than expected, want %d, got %d", len(testCase.expectedUpstreamMap), len(dnsServer.dnsMuxHandlers))
-			}
-
-			for _, expected := range testCase.expectedUpstreamMap {
-				found := false
-				for _, got := range dnsServer.dnsMuxHandlers {
-					if got.domain == expected.domain && got.priority == expected.priority {
-						found = true
-						break
-					}
-				}
-				if !found {
-					t.Fatalf("update upstream failed, handler for domain=%s priority=%d not found in dnsMuxHandlers: %#v", expected.domain, expected.priority, dnsServer.dnsMuxHandlers)
-				}
-			}
-
-			var responseMSG *dns.Msg
-			responseWriter := &test.MockResponseWriter{
-				WriteMsgFunc: func(m *dns.Msg) error {
-					responseMSG = m
-					return nil
-				},
-			}
-			for _, q := range testCase.expectedLocalQs {
-				dnsServer.localResolver.ServeDNS(responseWriter, &dns.Msg{
-					Question: []dns.Question{q},
-				})
-			}
-
-			if len(testCase.expectedLocalQs) > 0 {
-				assert.NotNil(t, responseMSG, "response message should not be nil")
-				assert.Equal(t, dns.RcodeSuccess, responseMSG.Rcode, "response code should be success")
-				assert.NotEmpty(t, responseMSG.Answer, "response message should have answers")
-			}
-		})
-	}
-}
-
-func TestDNSFakeResolverHandleUpdates(t *testing.T) {
-	ov := os.Getenv("NB_WG_KERNEL_DISABLED")
-	defer t.Setenv("NB_WG_KERNEL_DISABLED", ov)
-
-	t.Setenv("NB_WG_KERNEL_DISABLED", "true")
-	newNet, err := stdnet.NewNet(context.Background(), []string{"utun2301"})
-	if err != nil {
-		t.Errorf("create stdnet: %v", err)
-		return
-	}
-
-	privKey, _ := wgtypes.GeneratePrivateKey()
-	opts := iface.WGIFaceOpts{
-		IFaceName:    "utun2301",
-		Address:      wgaddr.MustParseWGAddress("100.66.100.1/32"),
-		WGPort:       33100,
-		WGPrivKey:    privKey.String(),
-		MTU:          iface.DefaultMTU,
-		TransportNet: newNet,
-	}
-	wgIface, err := iface.NewWGIFace(opts)
-	if err != nil {
-		t.Errorf("build interface wireguard: %v", err)
-		return
-	}
-
-	err = wgIface.Create()
-	if err != nil {
-		t.Errorf("create and init wireguard interface: %v", err)
-		return
-	}
-	defer func() {
-		if err = wgIface.Close(); err != nil {
-			t.Logf("close wireguard interface: %v", err)
-		}
-	}()
-
-	ctrl := gomock.NewController(t)
-	defer ctrl.Finish()
-
-	packetfilter := pfmock.NewMockPacketFilter(ctrl)
-	packetfilter.EXPECT().FilterOutbound(gomock.Any(), gomock.Any()).AnyTimes()
-	packetfilter.EXPECT().SetUDPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
-	packetfilter.EXPECT().SetTCPPacketHook(gomock.Any(), gomock.Any(), gomock.Any()).AnyTimes()
-
-	if err := wgIface.SetFilter(packetfilter); err != nil {
-		t.Errorf("set packet filter: %v", err)
-		return
-	}
-
-	dnsServer, err := NewDefaultServer(context.Background(), DefaultServerConfig{
-		WgInterface:    wgIface,
-		CustomAddress:  "",
-		StatusRecorder: peer.NewRecorder("mgm"),
-		StateManager:   nil,
-		DisableSys:     false,
-	})
-	if err != nil {
-		t.Errorf("create DNS server: %v", err)
-		return
-	}
-
-	err = dnsServer.Initialize()
-	if err != nil {
-		t.Errorf("run DNS server: %v", err)
-		return
-	}
-	defer func() {
-		if err = dnsServer.hostManager.restoreHostDNS(); err != nil {
-			t.Logf("restore DNS settings on the host: %v", err)
-			return
-		}
-	}()
-
-	dnsServer.dnsMuxHandlers = []handlerWrapper{
-		{
-			domain:   zoneRecords[0].Name,
-			handler:  &local.Resolver{},
-			priority: PriorityUpstream,
-		},
-	}
-	dnsServer.localResolver.Update([]nbdns.CustomZone{{Domain: "netbird.cloud", Records: []nbdns.SimpleRecord{{Name: "netbird.cloud", Type: int(dns.TypeA), Class: nbdns.DefaultClass, TTL: 300, RData: "10.0.0.1"}}}})
-	dnsServer.updateSerial = 0
-
-	nameServers := []nbdns.NameServer{
-		{
-			IP:     netip.MustParseAddr("8.8.8.8"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-		{
-			IP:     netip.MustParseAddr("8.8.4.4"),
-			NSType: nbdns.UDPNameServerType,
-			Port:   53,
-		},
-	}
-
-	update := nbdns.Config{
-		ServiceEnable: true,
-		CustomZones: []nbdns.CustomZone{
-			{
-				Domain:  "netbird.cloud",
-				Records: zoneRecords,
-			},
-		},
-		NameServerGroups: []*nbdns.NameServerGroup{
-			{
-				Domains:     []string{"netbird.io"},
-				NameServers: nameServers,
-			},
-			{
-				NameServers: nameServers,
-				Primary:     true,
-			},
-		},
-	}
-
-	// Start the server with regular configuration
-	if err := dnsServer.UpdateDNSServer(1, update); err != nil {
-		t.Fatalf("update dns server should not fail, got error: %v", err)
-		return
-	}
-
-	update2 := update
-	update2.ServiceEnable = false
-	// Disable the server, stop the listener
-	if err := dnsServer.UpdateDNSServer(2, update2); err != nil {
-		t.Fatalf("update dns server should not fail, got error: %v", err)
-		return
-	}
-
-	update3 := update2
-	update3.NameServerGroups = update3.NameServerGroups[:1]
-	// But service still get updates and we checking that we handle
-	// internal state in the right way
-	if err := dnsServer.UpdateDNSServer(3, update3); err != nil {
-		t.Fatalf("update dns server should not fail, got error: %v", err)
-		return
-	}
-}
-
 func TestDNSServerStartStop(t *testing.T) {
 	testCases := []struct {
 		name     string

client/internal/engine.go

@@ -210,6 +210,12 @@ type Engine struct {
 	// networkSerial is the latest CurrentSerial (state ID) of the network sent by the Management service
 	networkSerial uint64
 
+	// forwardingRules holds the ingress forward rules applied for the current target.
+	// Wholesale sections (incl. forward rules) run only on the first pass of a target;
+	// it is stashed here so the final, peer-converged pass can build the lazy-connection
+	// exclude list without recomputing them on every bounded peer pass.
+	forwardingRules []firewallManager.ForwardRule
+
 	networkMonitor *networkmonitor.NetworkMonitor
 
 	sshServer sshServer
@@ -762,7 +768,15 @@ func (e *Engine) blockLanAccess() {
 
 // modifyPeers updates peers that have been modified (e.g. IP address has been changed).
 // It closes the existing connection, removes it from the peerConns map, and creates a new one.
-func (e *Engine) modifyPeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
+// maxPeersPerSyncPass is the default per-pass cap on how many peers each of
+// removePeers/modifyPeers/addNewPeers applies, so syncMsgMux is held only for a
+// batch at a time and other subsystems can interleave between passes. It is
+// passed in (not read globally) so tests can exercise the multi-pass path.
+const maxPeersPerSyncPass = 300
+
+// modifyPeers re-applies up to maxBatch changed peers per call. It returns true
+// when more changed peers remained than the cap, so the caller re-runs.
+func (e *Engine) modifyPeers(peersUpdate []*mgmProto.RemotePeerConfig, maxBatch int) (bool, error) {
 
 	// first, check if peers have been modified
 	var modified []*mgmProto.RemotePeerConfig
@@ -792,26 +806,32 @@ func (e *Engine) modifyPeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
 		}
 	}
 
+	more := false
+	if len(modified) > maxBatch {
+		modified = modified[:maxBatch]
+		more = true
+	}
+
 	// second, close all modified connections and remove them from the state map
 	for _, p := range modified {
-		err := e.removePeer(p.GetWgPubKey())
-		if err != nil {
-			return err
+		if err := e.removePeer(p.GetWgPubKey()); err != nil {
+			return false, err
 		}
 	}
 	// third, add the peer connections again
 	for _, p := range modified {
-		err := e.addNewPeer(p)
-		if err != nil {
-			return err
+		if err := e.addNewPeer(p); err != nil {
+			return false, err
 		}
 	}
-	return nil
+	return more, nil
 }
 
 // removePeers finds and removes peers that do not exist anymore in the network map received from the Management Service.
 // It also removes peers that have been modified (e.g. change of IP address). They will be added again in addPeers method.
-func (e *Engine) removePeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
+// removePeers removes up to maxBatch peers per call. It returns true when more
+// peers remained to remove than the cap, so the caller re-runs.
+func (e *Engine) removePeers(peersUpdate []*mgmProto.RemotePeerConfig, maxBatch int) (bool, error) {
 	newPeers := make([]string, 0, len(peersUpdate))
 	for _, p := range peersUpdate {
 		newPeers = append(newPeers, p.GetWgPubKey())
@@ -819,14 +839,19 @@ func (e *Engine) removePeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
 
 	toRemove := util.SliceDiff(e.peerStore.PeersPubKey(), newPeers)
 
+	more := false
+	if len(toRemove) > maxBatch {
+		toRemove = toRemove[:maxBatch]
+		more = true
+	}
+
 	for _, p := range toRemove {
-		err := e.removePeer(p)
-		if err != nil {
-			return err
+		if err := e.removePeer(p); err != nil {
+			return false, err
 		}
 		log.Infof("removed peer %s", p)
 	}
-	return nil
+	return more, nil
 }
 
 func (e *Engine) removeAllPeers() error {
@@ -895,19 +920,17 @@ func (e *Engine) handleAutoUpdateVersion(autoUpdateSettings *mgmProto.AutoUpdate
 	e.updateManager.SetVersion(autoUpdateSettings.Version, autoUpdateSettings.AlwaysUpdate)
 }
 
-func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
-	started := time.Now()
-	defer func() {
-		duration := time.Since(started)
-		log.Infof("sync finished in %s", duration)
-		e.clientMetrics.RecordSyncDuration(e.ctx, duration)
-	}()
+// applySyncPass applies one bounded pass of the sync update under syncMsgMux and
+// returns true if more peers remained than the per-pass cap. It is driven by the
+// mapStateManager, which re-invokes it (releasing the lock between passes) until
+// the update is fully applied.
+func (e *Engine) applySyncPass(update *mgmProto.SyncResponse, firstPass bool) (bool, error) {
 	e.syncMsgMux.Lock()
 	defer e.syncMsgMux.Unlock()
 
 	// Check context INSIDE lock to ensure atomicity with shutdown
 	if e.ctx.Err() != nil {
-		return e.ctx.Err()
+		return false, e.ctx.Err()
 	}
 
 	if update.NetworkMap != nil && update.NetworkMap.PeerConfig != nil {
@@ -915,7 +938,7 @@ func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
 	}
 
 	if err := e.updateNetbirdConfig(update.GetNetbirdConfig()); err != nil {
-		return err
+		return false, err
 	}
 
 	// Posture checks are bound to the network map presence:
@@ -925,23 +948,22 @@ func (e *Engine) handleSync(update *mgmProto.SyncResponse) error {
 	//                                        leave the previously applied checks untouched
 	nm := update.GetNetworkMap()
 	if nm == nil {
-		return nil
+		return false, nil
 	}
 
 	if err := e.updateChecksIfNew(update.Checks); err != nil {
-		return err
+		return false, err
 	}
 
-	e.persistSyncResponse(update)
-
 	// only apply new changes and ignore old ones
-	if err := e.updateNetworkMap(nm); err != nil {
-		return err
+	more, err := e.updateNetworkMap(nm, maxPeersPerSyncPass, firstPass)
+	if err != nil {
+		return false, err
 	}
 
 	e.statusRecorder.PublishEvent(cProto.SystemEvent_INFO, cProto.SystemEvent_SYSTEM, "Network map updated", "", nil)
 
-	return nil
+	return more, nil
 }
 
 // updateNetbirdConfig applies the management-provided NetBird configuration:
@@ -987,6 +1009,13 @@ func (e *Engine) updateNetbirdConfig(wCfg *mgmProto.NetbirdConfig) error {
 // (not syncMsgMux) is held for the whole Set so the store cannot be cleared (disabled /
 // engine close) mid-call and have this write resurrect a file that was just removed.
 func (e *Engine) persistSyncResponse(update *mgmProto.SyncResponse) {
+	// Only persist updates that carry a network map. Config-only updates (e.g. relay
+	// token rotation, STUN/TURN) have a nil NetworkMap; persisting them would overwrite
+	// the last full map on disk and break restore-on-restart.
+	if update.GetNetworkMap() == nil {
+		return
+	}
+
 	e.syncRespMux.RLock()
 	defer e.syncRespMux.RUnlock()
 
@@ -1278,7 +1307,19 @@ func (e *Engine) receiveManagementEvents() {
 			e.config.DisableSSHAuth,
 		)
 
-		err = e.mgmClient.Sync(e.ctx, info, e.handleSync)
+		// The map-state manager converges the latest update in the background in
+		// bounded passes; the stream callback only hands it the newest target.
+		manager := newMapStateManager(e.applySyncPass, e.persistSyncResponse, func(d time.Duration) {
+			log.Infof("sync finished in %s", d)
+			e.clientMetrics.RecordSyncDuration(e.ctx, d)
+		})
+		e.shutdownWg.Add(1)
+		go func() {
+			defer e.shutdownWg.Done()
+			manager.run(e.ctx)
+		}()
+
+		err = e.mgmClient.Sync(e.ctx, info, manager.SetTarget)
 		if err != nil {
 			// happens if management is unavailable for a long time.
 			// We want to cancel the operation of the whole client
@@ -1329,21 +1370,104 @@ func (e *Engine) updateTURNs(turns []*mgmProto.ProtectedHostConfig) error {
 	return nil
 }
 
-func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
+// updateNetworkMap applies the wholesale parts (config, routes, ACL, DNS) in full
+// and up to maxBatch peers per phase. It returns true when more peers remained
+// than the cap, so the caller re-runs until convergence.
+func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap, maxBatch int, firstPass bool) (bool, error) {
 	// intentionally leave it before checking serial because for now it can happen that peer IP changed but serial didn't
 	if networkMap.GetPeerConfig() != nil {
 		err := e.updateConfig(networkMap.GetPeerConfig())
 		if err != nil {
-			return err
+			return false, err
 		}
 	}
 
 	serial := networkMap.GetSerial()
 	if e.networkSerial > serial {
 		log.Debugf("received outdated NetworkMap with serial %d, ignoring", serial)
-		return nil
+		return false, nil
 	}
 
+	// Wholesale sections (firewall/ACL, DNS, routes, forward rules) are applied
+	// up-front and only once per target: they are cheap, local, idempotent and must
+	// be in place before peers come up (fail-closed). On the bounded re-runs that only
+	// drain the remaining peer batches they are skipped — the applied forward rules are
+	// reused from e.forwardingRules for the lazy-exclude finalize.
+	if firstPass {
+		e.applyWholesale(networkMap, serial)
+	}
+
+	log.Debugf("got peers update from Management Service, total peers to connect to = %d", len(networkMap.GetRemotePeers()))
+
+	e.updateOfflinePeers(networkMap.GetOfflinePeers())
+
+	// Filter out own peer from the remote peers list
+	localPubKey := e.config.WgPrivateKey.PublicKey().String()
+	remotePeers := make([]*mgmProto.RemotePeerConfig, 0, len(networkMap.GetRemotePeers()))
+	for _, p := range networkMap.GetRemotePeers() {
+		if p.GetWgPubKey() != localPubKey {
+			remotePeers = append(remotePeers, p)
+		}
+	}
+
+	// needMore signals the caller to re-run when a peer phase hit its per-pass cap.
+	needMore := false
+
+	// cleanup request, most likely our peer has been deleted
+	if networkMap.GetRemotePeersIsEmpty() {
+		err := e.removeAllPeers()
+		e.statusRecorder.FinishPeerListModifications()
+		if err != nil {
+			return false, err
+		}
+	} else {
+		removeMore, err := e.removePeers(remotePeers, maxBatch)
+		if err != nil {
+			return false, err
+		}
+
+		modifyMore, err := e.modifyPeers(remotePeers, maxBatch)
+		if err != nil {
+			return false, err
+		}
+
+		addMore, err := e.addNewPeers(remotePeers, maxBatch)
+		if err != nil {
+			return false, err
+		}
+
+		needMore = removeMore || modifyMore || addMore
+
+		e.statusRecorder.FinishPeerListModifications()
+
+		e.updatePeerSSHHostKeys(remotePeers)
+
+		if err := e.updateSSHClientConfig(remotePeers); err != nil {
+			log.Warnf("failed to update SSH client config: %v", err)
+		}
+
+		e.updateSSHServerAuth(networkMap.GetSshAuth())
+	}
+
+	// Set the exclude list only once peers have fully converged (this pass added
+	// the last batch). It needs all target peers present in the store, and
+	// ExcludePeer has replace-semantics — a partial set mid-convergence would be wrong.
+	if !needMore {
+		excludedLazyPeers := e.toExcludedLazyPeers(e.forwardingRules, remotePeers)
+		e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
+	}
+
+	e.networkSerial = serial
+
+	return needMore, nil
+}
+
+// applyWholesale applies the cheap, local, idempotent map sections — lazy feature
+// flag, firewall/legacy management, DNS, routes, ACL filtering, DNS forwarder and
+// ingress forward rules — that must be in place before peers come up. It runs once
+// per target (first pass only); the resulting forward rules are stashed in
+// e.forwardingRules for the lazy-exclude finalize on the peer-converged pass.
+func (e *Engine) applyWholesale(networkMap *mgmProto.NetworkMap, serial uint64) {
 	if err := e.connMgr.UpdatedRemoteFeatureFlag(e.ctx, networkMap.GetPeerConfig().GetLazyConnectionEnabled()); err != nil {
 		log.Errorf("failed to update lazy connection feature flag: %v", err)
 	}
@@ -1404,61 +1528,7 @@ func (e *Engine) updateNetworkMap(networkMap *mgmProto.NetworkMap) error {
 	if err != nil {
 		log.Errorf("failed to update forward rules, err: %v", err)
 	}
-
-	log.Debugf("got peers update from Management Service, total peers to connect to = %d", len(networkMap.GetRemotePeers()))
-
-	e.updateOfflinePeers(networkMap.GetOfflinePeers())
-
-	// Filter out own peer from the remote peers list
-	localPubKey := e.config.WgPrivateKey.PublicKey().String()
-	remotePeers := make([]*mgmProto.RemotePeerConfig, 0, len(networkMap.GetRemotePeers()))
-	for _, p := range networkMap.GetRemotePeers() {
-		if p.GetWgPubKey() != localPubKey {
-			remotePeers = append(remotePeers, p)
-		}
-	}
-
-	// cleanup request, most likely our peer has been deleted
-	if networkMap.GetRemotePeersIsEmpty() {
-		err := e.removeAllPeers()
-		e.statusRecorder.FinishPeerListModifications()
-		if err != nil {
-			return err
-		}
-	} else {
-		err := e.removePeers(remotePeers)
-		if err != nil {
-			return err
-		}
-
-		err = e.modifyPeers(remotePeers)
-		if err != nil {
-			return err
-		}
-
-		err = e.addNewPeers(remotePeers)
-		if err != nil {
-			return err
-		}
-
-		e.statusRecorder.FinishPeerListModifications()
-
-		e.updatePeerSSHHostKeys(remotePeers)
-
-		if err := e.updateSSHClientConfig(remotePeers); err != nil {
-			log.Warnf("failed to update SSH client config: %v", err)
-		}
-
-		e.updateSSHServerAuth(networkMap.GetSshAuth())
-	}
-
-	// must set the exclude list after the peers are added. Without it the manager can not figure out the peers parameters from the store
-	excludedLazyPeers := e.toExcludedLazyPeers(forwardingRules, remotePeers)
-	e.connMgr.SetExcludeList(e.ctx, excludedLazyPeers)
-
-	e.networkSerial = serial
-
-	return nil
+	e.forwardingRules = forwardingRules
 }
 
 func toDNSFeatureFlag(networkMap *mgmProto.NetworkMap) bool {
@@ -1638,14 +1708,23 @@ func addrToString(addr netip.Addr) string {
 }
 
 // addNewPeers adds peers that were not know before but arrived from the Management service with the update
-func (e *Engine) addNewPeers(peersUpdate []*mgmProto.RemotePeerConfig) error {
+// addNewPeers adds up to maxBatch not-yet-present peers per call. It returns true
+// when more new peers remained than the cap, so the caller re-runs.
+func (e *Engine) addNewPeers(peersUpdate []*mgmProto.RemotePeerConfig, maxBatch int) (bool, error) {
+	added := 0
 	for _, p := range peersUpdate {
-		err := e.addNewPeer(p)
-		if err != nil {
-			return err
+		if _, ok := e.peerStore.PeerConn(p.GetWgPubKey()); ok {
+			continue // already present (cheap skip), does not count toward the cap
 		}
+		if added >= maxBatch {
+			return true, nil // at least one more new peer remains
+		}
+		if err := e.addNewPeer(p); err != nil {
+			return false, err
+		}
+		added++
 	}
-	return nil
+	return false, nil
 }
 
 // addNewPeer add peer if connection doesn't exist

client/internal/engine_privileged_test.go

@@ -0,0 +1,565 @@
+//go:build privileged
+
+package internal
+
+import (
+	"context"
+	"fmt"
+	"net"
+	"runtime"
+	"strings"
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/golang/mock/gomock"
+	"github.com/google/uuid"
+	log "github.com/sirupsen/logrus"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	"go.opentelemetry.io/otel"
+	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
+	"google.golang.org/grpc"
+	"google.golang.org/grpc/keepalive"
+
+	"github.com/netbirdio/netbird/client/iface"
+	"github.com/netbirdio/netbird/client/iface/device"
+	"github.com/netbirdio/netbird/client/iface/wgaddr"
+	"github.com/netbirdio/netbird/client/internal/dns"
+	"github.com/netbirdio/netbird/client/internal/peer"
+	nbssh "github.com/netbirdio/netbird/client/ssh"
+	"github.com/netbirdio/netbird/client/system"
+	nbdns "github.com/netbirdio/netbird/dns"
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/controller"
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/update_channel"
+	"github.com/netbirdio/netbird/management/internals/modules/peers"
+	"github.com/netbirdio/netbird/management/internals/modules/peers/ephemeral/manager"
+	"github.com/netbirdio/netbird/management/internals/server/config"
+	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
+	"github.com/netbirdio/netbird/management/server"
+	"github.com/netbirdio/netbird/management/server/activity"
+	nbcache "github.com/netbirdio/netbird/management/server/cache"
+	"github.com/netbirdio/netbird/management/server/groups"
+	"github.com/netbirdio/netbird/management/server/integrations/integrated_validator/validator"
+	"github.com/netbirdio/netbird/management/server/integrations/port_forwarding"
+	"github.com/netbirdio/netbird/management/server/job"
+	"github.com/netbirdio/netbird/management/server/permissions"
+	"github.com/netbirdio/netbird/management/server/settings"
+	"github.com/netbirdio/netbird/management/server/store"
+	"github.com/netbirdio/netbird/management/server/telemetry"
+	"github.com/netbirdio/netbird/management/server/types"
+	mgmt "github.com/netbirdio/netbird/shared/management/client"
+	mgmtProto "github.com/netbirdio/netbird/shared/management/proto"
+	relayClient "github.com/netbirdio/netbird/shared/relay/client"
+	signal "github.com/netbirdio/netbird/shared/signal/client"
+	"github.com/netbirdio/netbird/shared/signal/proto"
+	signalServer "github.com/netbirdio/netbird/signal/server"
+	"github.com/netbirdio/netbird/util"
+)
+
+func TestEngine_SSH(t *testing.T) {
+	key, err := wgtypes.GeneratePrivateKey()
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	sshKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
+	defer cancel()
+
+	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
+	engine := NewEngine(
+		ctx, cancel,
+		&EngineConfig{
+			WgIfaceName:      "utun101",
+			WgAddr:           wgaddr.MustParseWGAddress("100.64.0.1/24"),
+			WgPrivateKey:     key,
+			WgPort:           33100,
+			ServerSSHAllowed: true,
+			MTU:              iface.DefaultMTU,
+			SSHKey:           sshKey,
+		},
+		EngineServices{
+			SignalClient:   &signal.MockClient{},
+			MgmClient:      &mgmt.MockClient{},
+			RelayManager:   relayMgr,
+			StatusRecorder: peer.NewRecorder("https://mgm"),
+		},
+		MobileDependency{},
+	)
+
+	engine.dnsServer = &dns.MockServer{
+		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
+	}
+
+	err = engine.Start(nil, nil)
+	require.NoError(t, err)
+
+	defer func() {
+		err := engine.Stop()
+		if err != nil {
+			return
+		}
+	}()
+
+	peerWithSSH := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "MNHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
+		AllowedIps: []string{"100.64.0.21/24"},
+		SshConfig: &mgmtProto.SSHConfig{
+			SshPubKey: []byte("ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIFATYCqaQw/9id1Qkq3n16JYhDhXraI6Pc1fgB8ynEfQ"),
+		},
+	}
+
+	// SSH server is not enabled so SSH config of a remote peer should be ignored
+	networkMap := &mgmtProto.NetworkMap{
+		Serial:             6,
+		PeerConfig:         nil,
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	assert.Nil(t, engine.sshServer)
+
+	// SSH server is enabled, therefore SSH config should be applied
+	networkMap = &mgmtProto.NetworkMap{
+		Serial: 7,
+		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
+			SshConfig: &mgmtProto.SSHConfig{
+				SshEnabled: true,
+				JwtConfig: &mgmtProto.JWTConfig{
+					Issuer:       "test-issuer",
+					Audience:     "test-audience",
+					KeysLocation: "test-keys",
+					MaxTokenAge:  3600,
+				},
+			}},
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	time.Sleep(250 * time.Millisecond)
+	assert.NotNil(t, engine.sshServer)
+
+	// now remove peer
+	networkMap = &mgmtProto.NetworkMap{
+		Serial:             8,
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	// time.Sleep(250 * time.Millisecond)
+	assert.NotNil(t, engine.sshServer)
+
+	// now disable SSH server
+	networkMap = &mgmtProto.NetworkMap{
+		Serial: 9,
+		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
+			SshConfig: &mgmtProto.SSHConfig{SshEnabled: false}},
+		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
+		RemotePeersIsEmpty: false,
+	}
+
+	_, err = engine.updateNetworkMap(networkMap, maxPeersPerSyncPass, true)
+	require.NoError(t, err)
+
+	assert.Nil(t, engine.sshServer)
+}
+
+func TestEngine_Sync(t *testing.T) {
+	key, err := wgtypes.GeneratePrivateKey()
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
+	defer cancel()
+
+	// feed updates to Engine via mocked Management client
+	updates := make(chan *mgmtProto.SyncResponse)
+	defer close(updates)
+	syncFunc := func(ctx context.Context, info *system.Info, msgHandler func(msg *mgmtProto.SyncResponse) error) error {
+		for msg := range updates {
+			err := msgHandler(msg)
+			if err != nil {
+				t.Fatal(err)
+			}
+		}
+		return nil
+	}
+	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
+	engine := NewEngine(ctx, cancel, &EngineConfig{
+		WgIfaceName:  "utun103",
+		WgAddr:       wgaddr.MustParseWGAddress("100.64.0.1/24"),
+		WgPrivateKey: key,
+		WgPort:       33100,
+		MTU:          iface.DefaultMTU,
+	}, EngineServices{
+		SignalClient:   &signal.MockClient{},
+		MgmClient:      &mgmt.MockClient{SyncFunc: syncFunc},
+		RelayManager:   relayMgr,
+		StatusRecorder: peer.NewRecorder("https://mgm"),
+	}, MobileDependency{})
+	engine.ctx = ctx
+
+	engine.dnsServer = &dns.MockServer{
+		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
+	}
+
+	defer func() {
+		err := engine.Stop()
+		if err != nil {
+			return
+		}
+	}()
+
+	err = engine.Start(nil, nil)
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+
+	peer1 := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "RRHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
+		AllowedIps: []string{"100.64.0.10/24"},
+	}
+	peer2 := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "LLHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
+		AllowedIps: []string{"100.64.0.11/24"},
+	}
+	peer3 := &mgmtProto.RemotePeerConfig{
+		WgPubKey:   "GGHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
+		AllowedIps: []string{"100.64.0.12/24"},
+	}
+	// 1st update with just 1 peer and serial larger than the current serial of the engine => apply update
+	updates <- &mgmtProto.SyncResponse{
+		NetworkMap: &mgmtProto.NetworkMap{
+			Serial:             10,
+			PeerConfig:         nil,
+			RemotePeers:        []*mgmtProto.RemotePeerConfig{peer1, peer2, peer3},
+			RemotePeersIsEmpty: false,
+		},
+	}
+
+	timeout := time.After(time.Second * 2)
+	for {
+		select {
+		case <-timeout:
+			t.Fatalf("timeout while waiting for test to finish")
+			return
+		default:
+		}
+
+		if getPeers(engine) == 3 && engine.networkSerial == 10 {
+			break
+		}
+	}
+}
+
+func TestEngine_MultiplePeers(t *testing.T) {
+	// log.SetLevel(log.DebugLevel)
+
+	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
+	defer cancel()
+
+	sigServer, signalAddr, err := startSignal(t)
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+	defer sigServer.Stop()
+	mgmtServer, mgmtAddr, err := startManagement(t, t.TempDir(), "../testdata/store.sql")
+	if err != nil {
+		t.Fatal(err)
+		return
+	}
+	defer mgmtServer.GracefulStop()
+
+	setupKey := "A2C8E62B-38F5-4553-B31E-DD66C696CEBB"
+
+	mu := sync.Mutex{}
+	engines := []*Engine{}
+	numPeers := 10
+	wg := sync.WaitGroup{}
+	wg.Add(numPeers)
+	// create and start peers
+	for i := 0; i < numPeers; i++ {
+		j := i
+		go func() {
+			engine, err := createEngine(ctx, cancel, setupKey, j, mgmtAddr, signalAddr)
+			if err != nil {
+				wg.Done()
+				t.Errorf("unable to create the engine for peer %d with error %v", j, err)
+				return
+			}
+			engine.dnsServer = &dns.MockServer{}
+			mu.Lock()
+			defer mu.Unlock()
+			guid := fmt.Sprintf("{%s}", uuid.New().String())
+			device.CustomWindowsGUIDString = strings.ToLower(guid)
+			err = engine.Start(nil, nil)
+			if err != nil {
+				t.Errorf("unable to start engine for peer %d with error %v", j, err)
+				wg.Done()
+				return
+			}
+			engines = append(engines, engine)
+			wg.Done()
+		}()
+	}
+
+	// wait until all have been created and started
+	wg.Wait()
+	if len(engines) != numPeers {
+		t.Fatal("not all peers were started")
+	}
+	// check whether all the peer have expected peers connected
+
+	expectedConnected := numPeers * (numPeers - 1)
+
+	// adjust according to timeouts
+	timeout := 50 * time.Second
+	timeoutChan := time.After(timeout)
+	ticker := time.NewTicker(time.Second)
+	defer ticker.Stop()
+loop:
+	for {
+		select {
+		case <-timeoutChan:
+			t.Fatalf("waiting for expected connections timeout after %s", timeout.String())
+			break loop
+		case <-ticker.C:
+			totalConnected := 0
+			for _, engine := range engines {
+				totalConnected += getConnectedPeers(engine)
+			}
+			if totalConnected == expectedConnected {
+				log.Infof("total connected=%d", totalConnected)
+				break loop
+			}
+			log.Infof("total connected=%d", totalConnected)
+		}
+	}
+	// cleanup test
+	for n, peerEngine := range engines {
+		t.Logf("stopping peer with interface %s from multipeer test, loopIndex %d", peerEngine.wgInterface.Name(), n)
+		errStop := peerEngine.mgmClient.Close()
+		if errStop != nil {
+			log.Infoln("got error trying to close management clients from engine: ", errStop)
+		}
+		errStop = peerEngine.Stop()
+		if errStop != nil {
+			log.Infoln("got error trying to close testing peers engine: ", errStop)
+		}
+	}
+}
+
+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 createEngine(ctx context.Context, cancel context.CancelFunc, setupKey string, i int, mgmtAddr string, signalAddr string) (*Engine, error) {
+	key, err := wgtypes.GeneratePrivateKey()
+	if err != nil {
+		return nil, err
+	}
+	mgmtClient, err := mgmt.NewClient(ctx, mgmtAddr, key, false)
+	if err != nil {
+		return nil, err
+	}
+	signalClient, err := signal.NewClient(ctx, signalAddr, key, false)
+	if err != nil {
+		return nil, err
+	}
+
+	info := system.GetInfo(ctx)
+	resp, err := mgmtClient.Register(setupKey, "", info, nil, nil)
+	if err != nil {
+		return nil, err
+	}
+
+	var ifaceName string
+	if runtime.GOOS == "darwin" {
+		ifaceName = fmt.Sprintf("utun1%d", i)
+	} else {
+		ifaceName = fmt.Sprintf("wt%d", i)
+	}
+
+	wgPort := 33100 + i
+	conf := &EngineConfig{
+		WgIfaceName:  ifaceName,
+		WgAddr:       wgaddr.MustParseWGAddress(resp.PeerConfig.Address),
+		WgPrivateKey: key,
+		WgPort:       wgPort,
+		MTU:          iface.DefaultMTU,
+	}
+
+	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
+	e, err := NewEngine(ctx, cancel, conf, EngineServices{
+		SignalClient:   signalClient,
+		MgmClient:      mgmtClient,
+		RelayManager:   relayMgr,
+		StatusRecorder: peer.NewRecorder("https://mgm"),
+	}, MobileDependency{}), nil
+	e.ctx = ctx
+	return e, err
+}
+
+func startSignal(t *testing.T) (*grpc.Server, string, error) {
+	t.Helper()
+
+	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	if err != nil {
+		log.Fatalf("failed to listen: %v", err)
+	}
+
+	srv, err := signalServer.NewServer(context.Background(), otel.Meter(""))
+	require.NoError(t, err)
+	proto.RegisterSignalExchangeServer(s, srv)
+
+	go func() {
+		if err = s.Serve(lis); err != nil {
+			log.Fatalf("failed to serve: %v", err)
+		}
+	}()
+
+	return s, lis.Addr().String(), nil
+}
+
+func startManagement(t *testing.T, dataDir, testFile string) (*grpc.Server, string, error) {
+	t.Helper()
+
+	config := &config.Config{
+		Stuns:      []*config.Host{},
+		TURNConfig: &config.TURNConfig{},
+		Relay: &config.Relay{
+			Addresses:      []string{"127.0.0.1:1234"},
+			CredentialsTTL: util.Duration{Duration: time.Hour},
+			Secret:         "222222222222222222",
+		},
+		Signal: &config.Host{
+			Proto: "http",
+			URI:   "localhost:10000",
+		},
+		Datadir:    dataDir,
+		HttpConfig: nil,
+	}
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	if err != nil {
+		return nil, "", err
+	}
+	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
+
+	store, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), testFile, config.Datadir)
+	if err != nil {
+		return nil, "", err
+	}
+	t.Cleanup(cleanUp)
+
+	eventStore := &activity.InMemoryEventStore{}
+	if err != nil {
+		return nil, "", err
+	}
+
+	permissionsManager := permissions.NewManager(store)
+	peersManager := peers.NewManager(store, permissionsManager)
+	jobManager := job.NewJobManager(nil, store, peersManager)
+
+	cacheStore, err := nbcache.NewStore(context.Background(), 100*time.Millisecond, 300*time.Millisecond, 100)
+	if err != nil {
+		return nil, "", err
+	}
+
+	ia, _ := validator.NewIntegratedValidator(context.Background(), peersManager, nil, eventStore, cacheStore)
+
+	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
+	require.NoError(t, err)
+
+	ctrl := gomock.NewController(t)
+	t.Cleanup(ctrl.Finish)
+	settingsMockManager := settings.NewMockManager(ctrl)
+	settingsMockManager.EXPECT().
+		GetSettings(gomock.Any(), gomock.Any(), gomock.Any()).
+		Return(&types.Settings{}, nil).
+		AnyTimes()
+	settingsMockManager.EXPECT().
+		GetExtraSettings(gomock.Any(), gomock.Any()).
+		Return(&types.ExtraSettings{}, nil).
+		AnyTimes()
+
+	groupsManager := groups.NewManagerMock()
+
+	updateManager := update_channel.NewPeersUpdateManager(metrics)
+	requestBuffer := server.NewAccountRequestBuffer(context.Background(), store)
+	networkMapController := controller.NewController(context.Background(), store, metrics, updateManager, requestBuffer, server.MockIntegratedValidator{}, settingsMockManager, "netbird.selfhosted", port_forwarding.NewControllerMock(), manager.NewEphemeralManager(store, peersManager), config)
+	accountManager, err := server.BuildManager(context.Background(), config, store, networkMapController, jobManager, nil, "", eventStore, nil, false, ia, metrics, port_forwarding.NewControllerMock(), settingsMockManager, permissionsManager, false, cacheStore)
+	if err != nil {
+		return nil, "", err
+	}
+
+	secretsManager, err := nbgrpc.NewTimeBasedAuthSecretsManager(updateManager, config.TURNConfig, config.Relay, settingsMockManager, groupsManager)
+	if err != nil {
+		return nil, "", err
+	}
+	mgmtServer, err := nbgrpc.NewServer(config, accountManager, settingsMockManager, jobManager, secretsManager, nil, nil, &server.MockIntegratedValidator{}, networkMapController, nil, nil)
+	if err != nil {
+		return nil, "", err
+	}
+	mgmtProto.RegisterManagementServiceServer(s, mgmtServer)
+	go func() {
+		if err = s.Serve(lis); err != nil {
+			log.Fatalf("failed to serve: %v", err)
+		}
+	}()
+
+	return s, lis.Addr().String(), nil
+}
+
+// getConnectedPeers returns a connection Status or nil if peer connection wasn't found
+func getConnectedPeers(e *Engine) int {
+	e.syncMsgMux.Lock()
+	defer e.syncMsgMux.Unlock()
+	i := 0
+	for _, id := range e.peerStore.PeersPubKey() {
+		conn, _ := e.peerStore.PeerConn(id)
+		if conn.IsConnected() {
+			i++
+		}
+	}
+	return i
+}
+
+func getPeers(e *Engine) int {
+	e.syncMsgMux.Lock()
+	defer e.syncMsgMux.Unlock()
+
+	return len(e.peerStore.PeersPubKey())
+}

client/internal/engine_test.go

@@ -6,37 +6,18 @@ import (
 	"net"
 	"net/netip"
 	"os"
-	"runtime"
 	"strings"
 	"sync"
 	"testing"
 	"time"
 
-	"github.com/golang/mock/gomock"
-	"github.com/google/uuid"
-	log "github.com/sirupsen/logrus"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
-	"go.opentelemetry.io/otel"
 	wgdevice "golang.zx2c4.com/wireguard/device"
 	"golang.zx2c4.com/wireguard/tun/netstack"
 	"golang.zx2c4.com/wireguard/wgctrl/wgtypes"
-	"google.golang.org/grpc"
-	"google.golang.org/grpc/keepalive"
 
 	"github.com/netbirdio/netbird/client/internal/stdnet"
-	"github.com/netbirdio/netbird/management/server/job"
-
-	"github.com/netbirdio/netbird/management/server/integrations/integrated_validator/validator"
-
-	"github.com/netbirdio/netbird/management/internals/controllers/network_map/controller"
-	"github.com/netbirdio/netbird/management/internals/controllers/network_map/update_channel"
-	"github.com/netbirdio/netbird/management/internals/modules/peers"
-	"github.com/netbirdio/netbird/management/internals/modules/peers/ephemeral/manager"
-	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
-
-	"github.com/netbirdio/netbird/management/internals/server/config"
-	"github.com/netbirdio/netbird/management/server/groups"
 
 	"github.com/netbirdio/netbird/client/iface"
 	"github.com/netbirdio/netbird/client/iface/configurer"
@@ -50,18 +31,7 @@ import (
 	icemaker "github.com/netbirdio/netbird/client/internal/peer/ice"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
 	"github.com/netbirdio/netbird/client/internal/routemanager"
-	nbssh "github.com/netbirdio/netbird/client/ssh"
-	"github.com/netbirdio/netbird/client/system"
 	nbdns "github.com/netbirdio/netbird/dns"
-	"github.com/netbirdio/netbird/management/server"
-	"github.com/netbirdio/netbird/management/server/activity"
-	nbcache "github.com/netbirdio/netbird/management/server/cache"
-	"github.com/netbirdio/netbird/management/server/integrations/port_forwarding"
-	"github.com/netbirdio/netbird/management/server/permissions"
-	"github.com/netbirdio/netbird/management/server/settings"
-	"github.com/netbirdio/netbird/management/server/store"
-	"github.com/netbirdio/netbird/management/server/telemetry"
-	"github.com/netbirdio/netbird/management/server/types"
 	"github.com/netbirdio/netbird/monotime"
 	"github.com/netbirdio/netbird/route"
 	mgmt "github.com/netbirdio/netbird/shared/management/client"
@@ -69,25 +39,9 @@ import (
 	"github.com/netbirdio/netbird/shared/netiputil"
 	relayClient "github.com/netbirdio/netbird/shared/relay/client"
 	signal "github.com/netbirdio/netbird/shared/signal/client"
-	"github.com/netbirdio/netbird/shared/signal/proto"
-	signalServer "github.com/netbirdio/netbird/signal/server"
 	"github.com/netbirdio/netbird/util"
 )
 
-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,
-	}
-)
-
 type MockWGIface struct {
 	CreateFunc                 func() error
 	CreateOnAndroidFunc        func(routeRange []string, ip string, domains []string) error
@@ -224,10 +178,6 @@ func (m *MockWGIface) LastActivities() map[string]monotime.Time {
 	return nil
 }
 
-func (m *MockWGIface) MTU() uint16 {
-	return 1280
-}
-
 func (m *MockWGIface) SetPresharedKey(peerKey string, psk wgtypes.Key, updateOnly bool) error {
 	return nil
 }
@@ -238,129 +188,6 @@ func TestMain(m *testing.M) {
 	os.Exit(code)
 }
 
-func TestEngine_SSH(t *testing.T) {
-	key, err := wgtypes.GeneratePrivateKey()
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-
-	sshKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-
-	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
-	defer cancel()
-
-	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
-	engine := NewEngine(
-		ctx, cancel,
-		&EngineConfig{
-			WgIfaceName:      "utun101",
-			WgAddr:           wgaddr.MustParseWGAddress("100.64.0.1/24"),
-			WgPrivateKey:     key,
-			WgPort:           33100,
-			ServerSSHAllowed: true,
-			MTU:              iface.DefaultMTU,
-			SSHKey:           sshKey,
-		},
-		EngineServices{
-			SignalClient:   &signal.MockClient{},
-			MgmClient:      &mgmt.MockClient{},
-			RelayManager:   relayMgr,
-			StatusRecorder: peer.NewRecorder("https://mgm"),
-		},
-		MobileDependency{},
-	)
-
-	engine.dnsServer = &dns.MockServer{
-		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
-	}
-
-	err = engine.Start(nil, nil)
-	require.NoError(t, err)
-
-	defer func() {
-		err := engine.Stop()
-		if err != nil {
-			return
-		}
-	}()
-
-	peerWithSSH := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "MNHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
-		AllowedIps: []string{"100.64.0.21/24"},
-		SshConfig: &mgmtProto.SSHConfig{
-			SshPubKey: []byte("ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIFATYCqaQw/9id1Qkq3n16JYhDhXraI6Pc1fgB8ynEfQ"),
-		},
-	}
-
-	// SSH server is not enabled so SSH config of a remote peer should be ignored
-	networkMap := &mgmtProto.NetworkMap{
-		Serial:             6,
-		PeerConfig:         nil,
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	assert.Nil(t, engine.sshServer)
-
-	// SSH server is enabled, therefore SSH config should be applied
-	networkMap = &mgmtProto.NetworkMap{
-		Serial: 7,
-		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
-			SshConfig: &mgmtProto.SSHConfig{
-				SshEnabled: true,
-				JwtConfig: &mgmtProto.JWTConfig{
-					Issuer:       "test-issuer",
-					Audience:     "test-audience",
-					KeysLocation: "test-keys",
-					MaxTokenAge:  3600,
-				},
-			}},
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	time.Sleep(250 * time.Millisecond)
-	assert.NotNil(t, engine.sshServer)
-
-	// now remove peer
-	networkMap = &mgmtProto.NetworkMap{
-		Serial:             8,
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	// time.Sleep(250 * time.Millisecond)
-	assert.NotNil(t, engine.sshServer)
-
-	// now disable SSH server
-	networkMap = &mgmtProto.NetworkMap{
-		Serial: 9,
-		PeerConfig: &mgmtProto.PeerConfig{Address: "100.64.0.1/24",
-			SshConfig: &mgmtProto.SSHConfig{SshEnabled: false}},
-		RemotePeers:        []*mgmtProto.RemotePeerConfig{peerWithSSH},
-		RemotePeersIsEmpty: false,
-	}
-
-	err = engine.updateNetworkMap(networkMap)
-	require.NoError(t, err)
-
-	assert.Nil(t, engine.sshServer)
-}
-
 func TestEngine_SSHUpdateLogic(t *testing.T) {
 	// Test that SSH server start/stop logic works based on config
 	engine := &Engine{
@@ -606,7 +433,7 @@ func TestEngine_UpdateNetworkMap(t *testing.T) {
 
 	for _, c := range []testCase{case1, case2, case3, case4, case5, case6} {
 		t.Run(c.name, func(t *testing.T) {
-			err = engine.updateNetworkMap(c.networkMap)
+			_, err = engine.updateNetworkMap(c.networkMap, maxPeersPerSyncPass, true)
 			if err != nil {
 				t.Fatal(err)
 				return
@@ -633,97 +460,47 @@ func TestEngine_UpdateNetworkMap(t *testing.T) {
 			}
 		})
 	}
-}
 
-func TestEngine_Sync(t *testing.T) {
-	key, err := wgtypes.GeneratePrivateKey()
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-
-	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
-	defer cancel()
-
-	// feed updates to Engine via mocked Management client
-	updates := make(chan *mgmtProto.SyncResponse)
-	defer close(updates)
-	syncFunc := func(ctx context.Context, info *system.Info, msgHandler func(msg *mgmtProto.SyncResponse) error) error {
-		for msg := range updates {
-			err := msgHandler(msg)
-			if err != nil {
-				t.Fatal(err)
-			}
-		}
-		return nil
-	}
-	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
-	engine := NewEngine(ctx, cancel, &EngineConfig{
-		WgIfaceName:  "utun103",
-		WgAddr:       wgaddr.MustParseWGAddress("100.64.0.1/24"),
-		WgPrivateKey: key,
-		WgPort:       33100,
-		MTU:          iface.DefaultMTU,
-	}, EngineServices{
-		SignalClient:   &signal.MockClient{},
-		MgmClient:      &mgmt.MockClient{SyncFunc: syncFunc},
-		RelayManager:   relayMgr,
-		StatusRecorder: peer.NewRecorder("https://mgm"),
-	}, MobileDependency{})
-	engine.ctx = ctx
-
-	engine.dnsServer = &dns.MockServer{
-		UpdateDNSServerFunc: func(serial uint64, update nbdns.Config) error { return nil },
-	}
-
-	defer func() {
-		err := engine.Stop()
-		if err != nil {
-			return
-		}
-	}()
-
-	err = engine.Start(nil, nil)
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-
-	peer1 := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "RRHf3Ma6z6mdLbriAJbqhX7+nM/B71lgw2+91q3LfhU=",
-		AllowedIps: []string{"100.64.0.10/24"},
-	}
-	peer2 := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "LLHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
-		AllowedIps: []string{"100.64.0.11/24"},
-	}
-	peer3 := &mgmtProto.RemotePeerConfig{
-		WgPubKey:   "GGHf3Ma6z6mdLbriAJbqhX9+nM/B71lgw2+91q3LlhU=",
-		AllowedIps: []string{"100.64.0.12/24"},
-	}
-	// 1st update with just 1 peer and serial larger than the current serial of the engine => apply update
-	updates <- &mgmtProto.SyncResponse{
-		NetworkMap: &mgmtProto.NetworkMap{
-			Serial:             10,
-			PeerConfig:         nil,
-			RemotePeers:        []*mgmtProto.RemotePeerConfig{peer1, peer2, peer3},
-			RemotePeersIsEmpty: false,
-		},
-	}
-
-	timeout := time.After(time.Second * 2)
-	for {
-		select {
-		case <-timeout:
-			t.Fatalf("timeout while waiting for test to finish")
-			return
-		default:
+	// chunked apply: with a per-pass cap smaller than the number of peers, a
+	// single updateNetworkMap applies one batch and reports more==true; the
+	// caller re-runs until convergence. (engine currently holds 0 peers.)
+	t.Run("chunked add converges over multiple passes", func(t *testing.T) {
+		nm := &mgmtProto.NetworkMap{
+			Serial:      6,
+			RemotePeers: []*mgmtProto.RemotePeerConfig{peer1, peer2, peer3},
 		}
 
-		if getPeers(engine) == 3 && engine.networkSerial == 10 {
-			break
+		more, err := engine.updateNetworkMap(nm, 1, true)
+		require.NoError(t, err)
+		require.True(t, more, "pass 1 should signal more")
+		require.Len(t, engine.peerStore.PeersPubKey(), 1)
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.True(t, more, "pass 2 should signal more")
+		require.Len(t, engine.peerStore.PeersPubKey(), 2)
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.False(t, more, "pass 3 should converge")
+		require.Len(t, engine.peerStore.PeersPubKey(), 3)
+	})
+
+	t.Run("chunked remove converges over multiple passes", func(t *testing.T) {
+		nm := &mgmtProto.NetworkMap{
+			Serial:      7,
+			RemotePeers: []*mgmtProto.RemotePeerConfig{peer1}, // remove peer2, peer3
 		}
-	}
+
+		more, err := engine.updateNetworkMap(nm, 1, true)
+		require.NoError(t, err)
+		require.True(t, more, "pass 1 should signal more (2 to remove, cap 1)")
+
+		more, err = engine.updateNetworkMap(nm, 1, false)
+		require.NoError(t, err)
+		require.False(t, more, "pass 2 should converge")
+		require.Len(t, engine.peerStore.PeersPubKey(), 1)
+	})
 }
 
 func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
@@ -894,7 +671,7 @@ func TestEngine_UpdateNetworkMapWithRoutes(t *testing.T) {
 				}
 			}()
 
-			err = engine.updateNetworkMap(testCase.networkMap)
+			_, err = engine.updateNetworkMap(testCase.networkMap, maxPeersPerSyncPass, true)
 			assert.NoError(t, err, "shouldn't return error")
 			assert.Equal(t, testCase.expectedSerial, input.inputSerial, "serial should match")
 			assert.Len(t, input.clientRoutes, testCase.expectedLen, "clientRoutes len should match")
@@ -1098,7 +875,7 @@ func TestEngine_UpdateNetworkMapWithDNSUpdate(t *testing.T) {
 				}
 			}()
 
-			err = engine.updateNetworkMap(testCase.networkMap)
+			_, err = engine.updateNetworkMap(testCase.networkMap, maxPeersPerSyncPass, true)
 			assert.NoError(t, err, "shouldn't return error")
 			assert.Equal(t, testCase.expectedSerial, input.inputSerial, "serial should match")
 			assert.Len(t, input.inputNSGroups, testCase.expectedZonesLen, "zones len should match")
@@ -1109,104 +886,6 @@ func TestEngine_UpdateNetworkMapWithDNSUpdate(t *testing.T) {
 	}
 }
 
-func TestEngine_MultiplePeers(t *testing.T) {
-	// log.SetLevel(log.DebugLevel)
-
-	ctx, cancel := context.WithCancel(CtxInitState(context.Background()))
-	defer cancel()
-
-	sigServer, signalAddr, err := startSignal(t)
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-	defer sigServer.Stop()
-	mgmtServer, mgmtAddr, err := startManagement(t, t.TempDir(), "../testdata/store.sql")
-	if err != nil {
-		t.Fatal(err)
-		return
-	}
-	defer mgmtServer.GracefulStop()
-
-	setupKey := "A2C8E62B-38F5-4553-B31E-DD66C696CEBB"
-
-	mu := sync.Mutex{}
-	engines := []*Engine{}
-	numPeers := 10
-	wg := sync.WaitGroup{}
-	wg.Add(numPeers)
-	// create and start peers
-	for i := 0; i < numPeers; i++ {
-		j := i
-		go func() {
-			engine, err := createEngine(ctx, cancel, setupKey, j, mgmtAddr, signalAddr)
-			if err != nil {
-				wg.Done()
-				t.Errorf("unable to create the engine for peer %d with error %v", j, err)
-				return
-			}
-			engine.dnsServer = &dns.MockServer{}
-			mu.Lock()
-			defer mu.Unlock()
-			guid := fmt.Sprintf("{%s}", uuid.New().String())
-			device.CustomWindowsGUIDString = strings.ToLower(guid)
-			err = engine.Start(nil, nil)
-			if err != nil {
-				t.Errorf("unable to start engine for peer %d with error %v", j, err)
-				wg.Done()
-				return
-			}
-			engines = append(engines, engine)
-			wg.Done()
-		}()
-	}
-
-	// wait until all have been created and started
-	wg.Wait()
-	if len(engines) != numPeers {
-		t.Fatal("not all peers was started")
-	}
-	// check whether all the peer have expected peers connected
-
-	expectedConnected := numPeers * (numPeers - 1)
-
-	// adjust according to timeouts
-	timeout := 50 * time.Second
-	timeoutChan := time.After(timeout)
-	ticker := time.NewTicker(time.Second)
-	defer ticker.Stop()
-loop:
-	for {
-		select {
-		case <-timeoutChan:
-			t.Fatalf("waiting for expected connections timeout after %s", timeout.String())
-			break loop
-		case <-ticker.C:
-			totalConnected := 0
-			for _, engine := range engines {
-				totalConnected += getConnectedPeers(engine)
-			}
-			if totalConnected == expectedConnected {
-				log.Infof("total connected=%d", totalConnected)
-				break loop
-			}
-			log.Infof("total connected=%d", totalConnected)
-		}
-	}
-	// cleanup test
-	for n, peerEngine := range engines {
-		t.Logf("stopping peer with interface %s from multipeer test, loopIndex %d", peerEngine.wgInterface.Name(), n)
-		errStop := peerEngine.mgmClient.Close()
-		if errStop != nil {
-			log.Infoln("got error trying to close management clients from engine: ", errStop)
-		}
-		errStop = peerEngine.Stop()
-		if errStop != nil {
-			log.Infoln("got error trying to close testing peers engine: ", errStop)
-		}
-	}
-}
-
 func Test_ParseNATExternalIPMappings(t *testing.T) {
 	ifaceList, err := net.Interfaces()
 	if err != nil {
@@ -1530,187 +1209,6 @@ func TestCompareNetIPLists(t *testing.T) {
 	}
 }
 
-func createEngine(ctx context.Context, cancel context.CancelFunc, setupKey string, i int, mgmtAddr string, signalAddr string) (*Engine, error) {
-	key, err := wgtypes.GeneratePrivateKey()
-	if err != nil {
-		return nil, err
-	}
-	mgmtClient, err := mgmt.NewClient(ctx, mgmtAddr, key, false)
-	if err != nil {
-		return nil, err
-	}
-	signalClient, err := signal.NewClient(ctx, signalAddr, key, false)
-	if err != nil {
-		return nil, err
-	}
-
-	info := system.GetInfo(ctx)
-	resp, err := mgmtClient.Register(setupKey, "", info, nil, nil)
-	if err != nil {
-		return nil, err
-	}
-
-	var ifaceName string
-	if runtime.GOOS == "darwin" {
-		ifaceName = fmt.Sprintf("utun1%d", i)
-	} else {
-		ifaceName = fmt.Sprintf("wt%d", i)
-	}
-
-	wgPort := 33100 + i
-	conf := &EngineConfig{
-		WgIfaceName:  ifaceName,
-		WgAddr:       wgaddr.MustParseWGAddress(resp.PeerConfig.Address),
-		WgPrivateKey: key,
-		WgPort:       wgPort,
-		MTU:          iface.DefaultMTU,
-	}
-
-	relayMgr := relayClient.NewManager(ctx, nil, key.PublicKey().String(), iface.DefaultMTU)
-	e, err := NewEngine(ctx, cancel, conf, EngineServices{
-		SignalClient:   signalClient,
-		MgmClient:      mgmtClient,
-		RelayManager:   relayMgr,
-		StatusRecorder: peer.NewRecorder("https://mgm"),
-	}, MobileDependency{}), nil
-	e.ctx = ctx
-	return e, err
-}
-
-func startSignal(t *testing.T) (*grpc.Server, string, error) {
-	t.Helper()
-
-	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
-
-	lis, err := net.Listen("tcp", "localhost:0")
-	if err != nil {
-		log.Fatalf("failed to listen: %v", err)
-	}
-
-	srv, err := signalServer.NewServer(context.Background(), otel.Meter(""))
-	require.NoError(t, err)
-	proto.RegisterSignalExchangeServer(s, srv)
-
-	go func() {
-		if err = s.Serve(lis); err != nil {
-			log.Fatalf("failed to serve: %v", err)
-		}
-	}()
-
-	return s, lis.Addr().String(), nil
-}
-
-func startManagement(t *testing.T, dataDir, testFile string) (*grpc.Server, string, error) {
-	t.Helper()
-
-	config := &config.Config{
-		Stuns:      []*config.Host{},
-		TURNConfig: &config.TURNConfig{},
-		Relay: &config.Relay{
-			Addresses:      []string{"127.0.0.1:1234"},
-			CredentialsTTL: util.Duration{Duration: time.Hour},
-			Secret:         "222222222222222222",
-		},
-		Signal: &config.Host{
-			Proto: "http",
-			URI:   "localhost:10000",
-		},
-		Datadir:    dataDir,
-		HttpConfig: nil,
-	}
-
-	lis, err := net.Listen("tcp", "localhost:0")
-	if err != nil {
-		return nil, "", err
-	}
-	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
-
-	store, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), testFile, config.Datadir)
-	if err != nil {
-		return nil, "", err
-	}
-	t.Cleanup(cleanUp)
-
-	eventStore := &activity.InMemoryEventStore{}
-	if err != nil {
-		return nil, "", err
-	}
-
-	permissionsManager := permissions.NewManager(store)
-	peersManager := peers.NewManager(store, permissionsManager)
-	jobManager := job.NewJobManager(nil, store, peersManager)
-
-	cacheStore, err := nbcache.NewStore(context.Background(), 100*time.Millisecond, 300*time.Millisecond, 100)
-	if err != nil {
-		return nil, "", err
-	}
-
-	ia, _ := validator.NewIntegratedValidator(context.Background(), peersManager, nil, eventStore, cacheStore)
-
-	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
-	require.NoError(t, err)
-
-	ctrl := gomock.NewController(t)
-	t.Cleanup(ctrl.Finish)
-	settingsMockManager := settings.NewMockManager(ctrl)
-	settingsMockManager.EXPECT().
-		GetSettings(gomock.Any(), gomock.Any(), gomock.Any()).
-		Return(&types.Settings{}, nil).
-		AnyTimes()
-	settingsMockManager.EXPECT().
-		GetExtraSettings(gomock.Any(), gomock.Any()).
-		Return(&types.ExtraSettings{}, nil).
-		AnyTimes()
-
-	groupsManager := groups.NewManagerMock()
-
-	updateManager := update_channel.NewPeersUpdateManager(metrics)
-	requestBuffer := server.NewAccountRequestBuffer(context.Background(), store)
-	networkMapController := controller.NewController(context.Background(), store, metrics, updateManager, requestBuffer, server.MockIntegratedValidator{}, settingsMockManager, "netbird.selfhosted", port_forwarding.NewControllerMock(), manager.NewEphemeralManager(store, peersManager), config)
-	accountManager, err := server.BuildManager(context.Background(), config, store, networkMapController, jobManager, nil, "", eventStore, nil, false, ia, metrics, port_forwarding.NewControllerMock(), settingsMockManager, permissionsManager, false, cacheStore)
-	if err != nil {
-		return nil, "", err
-	}
-
-	secretsManager, err := nbgrpc.NewTimeBasedAuthSecretsManager(updateManager, config.TURNConfig, config.Relay, settingsMockManager, groupsManager)
-	if err != nil {
-		return nil, "", err
-	}
-	mgmtServer, err := nbgrpc.NewServer(config, accountManager, settingsMockManager, jobManager, secretsManager, nil, nil, &server.MockIntegratedValidator{}, networkMapController, nil, nil)
-	if err != nil {
-		return nil, "", err
-	}
-	mgmtProto.RegisterManagementServiceServer(s, mgmtServer)
-	go func() {
-		if err = s.Serve(lis); err != nil {
-			log.Fatalf("failed to serve: %v", err)
-		}
-	}()
-
-	return s, lis.Addr().String(), nil
-}
-
-// getConnectedPeers returns a connection Status or nil if peer connection wasn't found
-func getConnectedPeers(e *Engine) int {
-	e.syncMsgMux.Lock()
-	defer e.syncMsgMux.Unlock()
-	i := 0
-	for _, id := range e.peerStore.PeersPubKey() {
-		conn, _ := e.peerStore.PeerConn(id)
-		if conn.IsConnected() {
-			i++
-		}
-	}
-	return i
-}
-
-func getPeers(e *Engine) int {
-	e.syncMsgMux.Lock()
-	defer e.syncMsgMux.Unlock()
-
-	return len(e.peerStore.PeersPubKey())
-}
-
 func mustEncodePrefix(t *testing.T, p netip.Prefix) []byte {
 	t.Helper()
 	b, err := netiputil.EncodePrefix(p)

client/internal/iface_common.go

@@ -44,5 +44,4 @@ type wgIfaceBase interface {
 	FullStats() (*configurer.Stats, error)
 	LastActivities() map[string]monotime.Time
 	SetPresharedKey(peerKey string, psk wgtypes.Key, updateOnly bool) error
-	MTU() uint16
 }

client/internal/lazyconn/activity/listener_bind.go

@@ -124,11 +124,6 @@ func (d *BindListener) ReadPackets() {
 	d.done.Done()
 }
 
-// CapturedPacket is unused in userspace bind mode: first-packet reinjection is kernel-only.
-func (d *BindListener) CapturedPacket() []byte {
-	return nil
-}
-
 // Close stops the listener and cleans up resources.
 func (d *BindListener) Close() {
 	d.peerCfg.Log.Infof("closing activity listener (LazyConn)")

client/internal/lazyconn/activity/listener_bind_test.go

@@ -68,10 +68,6 @@ func (m *MockWGIfaceBind) GetBind() device.EndpointManager {
 	return m.endpointMgr
 }
 
-func (m *MockWGIfaceBind) MTU() uint16 {
-	return 1280
-}
-
 func TestBindListener_Creation(t *testing.T) {
 	mockEndpointMgr := newMockEndpointManager()
 	mockIface := &MockWGIfaceBind{endpointMgr: mockEndpointMgr}
@@ -211,9 +207,8 @@ func TestManager_BindMode(t *testing.T) {
 	require.NoError(t, err)
 
 	select {
-	case ev := <-mgr.OnActivityChan:
-		assert.Equal(t, cfg.PeerConnID, ev.PeerConnID, "Received peer connection ID should match")
-		assert.Nil(t, ev.FirstPacket, "Bind mode does not capture packets: reinjection is kernel-only")
+	case peerConnID := <-mgr.OnActivityChan:
+		assert.Equal(t, cfg.PeerConnID, peerConnID, "Received peer connection ID should match")
 	case <-time.After(2 * time.Second):
 		t.Fatal("timeout waiting for activity notification")
 	}
@@ -271,8 +266,8 @@ func TestManager_BindMode_MultiplePeers(t *testing.T) {
 	receivedPeers := make(map[peerid.ConnID]bool)
 	for i := 0; i < 2; i++ {
 		select {
-		case ev := <-mgr.OnActivityChan:
-			receivedPeers[ev.PeerConnID] = true
+		case peerConnID := <-mgr.OnActivityChan:
+			receivedPeers[peerConnID] = true
 		case <-time.After(2 * time.Second):
 			t.Fatal("timeout waiting for activity notifications")
 		}

client/internal/lazyconn/activity/listener_udp.go

@@ -3,13 +3,11 @@ package activity
 import (
 	"fmt"
 	"net"
-	"slices"
 	"sync"
 	"sync/atomic"
 
 	log "github.com/sirupsen/logrus"
 
-	"github.com/netbirdio/netbird/client/iface/bufsize"
 	"github.com/netbirdio/netbird/client/internal/lazyconn"
 )
 
@@ -22,8 +20,6 @@ type UDPListener struct {
 	done     sync.Mutex
 
 	isClosed atomic.Bool
-
-	capturedPacket []byte
 }
 
 // NewUDPListener creates a listener that detects activity via UDP socket reads.
@@ -50,13 +46,9 @@ func NewUDPListener(wgIface WgInterface, cfg lazyconn.PeerConfig) (*UDPListener,
 }
 
 // ReadPackets blocks reading from the UDP socket until activity is detected or the listener is closed.
-// The first packet that triggers activity is captured so it can be reinjected through the real
-// transport once it is established. Without this, kernel WireGuard's handshake initiation would be
-// dropped and WG would only retry after REKEY_TIMEOUT.
 func (d *UDPListener) ReadPackets() {
 	for {
-		buf := make([]byte, int(d.wgIface.MTU())+bufsize.WGBufferOverhead)
-		n, remoteAddr, err := d.conn.ReadFromUDP(buf)
+		n, remoteAddr, err := d.conn.ReadFromUDP(make([]byte, 1))
 		if err != nil {
 			if d.isClosed.Load() {
 				d.peerCfg.Log.Infof("exit from activity listener")
@@ -70,24 +62,20 @@ func (d *UDPListener) ReadPackets() {
 			d.peerCfg.Log.Warnf("received %d bytes from %s, too short", n, remoteAddr)
 			continue
 		}
-		d.capturedPacket = slices.Clone(buf[:n])
-		d.peerCfg.Log.Infof("activity detected, captured %d bytes for reinjection", n)
+		d.peerCfg.Log.Infof("activity detected")
 		break
 	}
 
-	// Leave the peer in place. ConfigureWGEndpoint will UpdatePeer with the real endpoint;
-	// removing the peer here wipes kernel WG's staged queue and drops the user packet that
-	// triggered activation.
+	d.peerCfg.Log.Debugf("removing lazy endpoint: %s", d.endpoint.String())
+	if err := d.wgIface.RemovePeer(d.peerCfg.PublicKey); err != nil {
+		d.peerCfg.Log.Errorf("failed to remove endpoint: %s", err)
+	}
+
+	// Ignore close error as it may return "use of closed network connection" if already closed.
 	_ = d.conn.Close()
 	d.done.Unlock()
 }
 
-// CapturedPacket returns the first packet that triggered activity, or nil if none was captured.
-// Safe to call after ReadPackets returns.
-func (d *UDPListener) CapturedPacket() []byte {
-	return d.capturedPacket
-}
-
 // Close stops the listener and cleans up resources.
 func (d *UDPListener) Close() {
 	d.peerCfg.Log.Infof("closing activity listener: %s", d.conn.LocalAddr().String())

client/internal/lazyconn/activity/manager.go

@@ -19,14 +19,6 @@ import (
 type listener interface {
 	ReadPackets()
 	Close()
-	CapturedPacket() []byte
-}
-
-// Event reports activity on a managed peer. FirstPacket is the bytes that triggered activation,
-// captured for reinjection through the real transport.
-type Event struct {
-	PeerConnID  peerid.ConnID
-	FirstPacket []byte
 }
 
 type WgInterface interface {
@@ -34,11 +26,10 @@ type WgInterface interface {
 	UpdatePeer(peerKey string, allowedIps []netip.Prefix, keepAlive time.Duration, endpoint *net.UDPAddr, preSharedKey *wgtypes.Key) error
 	IsUserspaceBind() bool
 	Address() wgaddr.Address
-	MTU() uint16
 }
 
 type Manager struct {
-	OnActivityChan chan Event
+	OnActivityChan chan peerid.ConnID
 
 	wgIface WgInterface
 
@@ -50,7 +41,7 @@ type Manager struct {
 
 func NewManager(wgIface WgInterface) *Manager {
 	m := &Manager{
-		OnActivityChan: make(chan Event, 1),
+		OnActivityChan: make(chan peerid.ConnID, 1),
 		wgIface:        wgIface,
 		peers:          make(map[peerid.ConnID]listener),
 		done:           make(chan struct{}),
@@ -125,12 +116,12 @@ func (m *Manager) waitForTraffic(l listener, peerConnID peerid.ConnID) {
 	delete(m.peers, peerConnID)
 	m.mu.Unlock()
 
-	m.notify(Event{PeerConnID: peerConnID, FirstPacket: l.CapturedPacket()})
+	m.notify(peerConnID)
 }
 
-func (m *Manager) notify(ev Event) {
+func (m *Manager) notify(peerConnID peerid.ConnID) {
 	select {
 	case <-m.done:
-	case m.OnActivityChan <- ev:
+	case m.OnActivityChan <- peerConnID:
 	}
 }

client/internal/lazyconn/activity/manager_test.go

@@ -1,7 +1,6 @@
 package activity
 
 import (
-	"bytes"
 	"net"
 	"net/netip"
 	"testing"
@@ -45,10 +44,6 @@ func (m MocWGIface) Address() wgaddr.Address {
 	}
 }
 
-func (m MocWGIface) MTU() uint16 {
-	return 1280
-}
-
 // GetPeerListener is a test helper to access listeners
 func (m *Manager) GetPeerListener(peerConnID peerid.ConnID) (listener, bool) {
 	m.mu.Lock()
@@ -91,15 +86,11 @@ func TestManager_MonitorPeerActivity(t *testing.T) {
 	}
 
 	select {
-	case ev := <-mgr.OnActivityChan:
-		if ev.PeerConnID != peerCfg1.PeerConnID {
-			t.Fatalf("unexpected peerConnID: %v", ev.PeerConnID)
-		}
-		if !bytes.Equal(ev.FirstPacket, []byte{0x01, 0x02, 0x03, 0x04, 0x05}) {
-			t.Fatalf("unexpected first packet: %v", ev.FirstPacket)
+	case peerConnID := <-mgr.OnActivityChan:
+		if peerConnID != peerCfg1.PeerConnID {
+			t.Fatalf("unexpected peerConnID: %v", peerConnID)
 		}
 	case <-time.After(1 * time.Second):
-		t.Fatal("timed out waiting for activity")
 	}
 }
 

client/internal/lazyconn/manager/manager.go

@@ -130,8 +130,8 @@ func (m *Manager) Start(ctx context.Context) {
 		select {
 		case <-ctx.Done():
 			return
-		case ev := <-m.activityManager.OnActivityChan:
-			m.onPeerActivity(ev)
+		case peerConnID := <-m.activityManager.OnActivityChan:
+			m.onPeerActivity(peerConnID)
 		case peerIDs := <-m.inactivityManager.InactivePeersChan():
 			m.onPeerInactivityTimedOut(peerIDs)
 		}
@@ -513,13 +513,13 @@ func (m *Manager) checkHaGroupActivity(haGroup route.HAUniqueID, peerID string,
 	return false
 }
 
-func (m *Manager) onPeerActivity(ev activity.Event) {
+func (m *Manager) onPeerActivity(peerConnID peerid.ConnID) {
 	m.managedPeersMu.Lock()
 	defer m.managedPeersMu.Unlock()
 
-	mp, ok := m.managedPeersByConnID[ev.PeerConnID]
+	mp, ok := m.managedPeersByConnID[peerConnID]
 	if !ok {
-		log.Errorf("peer not found by conn id: %v", ev.PeerConnID)
+		log.Errorf("peer not found by conn id: %v", peerConnID)
 		return
 	}
 
@@ -536,7 +536,7 @@ func (m *Manager) onPeerActivity(ev activity.Event) {
 
 	m.activateHAGroupPeers(mp.peerCfg)
 
-	m.peerStore.PeerConnOpenWithFirstPacket(m.engineCtx, mp.peerCfg.PublicKey, ev.FirstPacket)
+	m.peerStore.PeerConnOpen(m.engineCtx, mp.peerCfg.PublicKey)
 }
 
 func (m *Manager) onPeerInactivityTimedOut(peerIDs map[string]struct{}) {

client/internal/lazyconn/wgiface.go

@@ -17,5 +17,4 @@ type WGIface interface {
 	IsUserspaceBind() bool
 	Address() wgaddr.Address
 	LastActivities() map[string]monotime.Time
-	MTU() uint16
 }

client/internal/mapsync.go

@@ -0,0 +1,190 @@
+package internal
+
+import (
+	"context"
+	"sync"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+
+	mgmProto "github.com/netbirdio/netbird/shared/management/proto"
+)
+
+// mapStateManager is the single read/write point between the management stream
+// (writes) and the convergence loop (reads/applies).
+//
+// The stream calls SetTarget with the latest full SyncResponse — the complete
+// desired state. A single background goroutine (run) applies it to the engine in
+// bounded passes via apply() until converged, releasing syncMsgMux between passes
+// so other subsystems interleave. If a newer update arrives mid-flight, the loop
+// coalesces: it keeps converging toward the latest target and the intermediate one
+// is SKIPPED — never applied on its own (logged, no onConverged).
+//
+// Convergence is a single comparison: appliedGen == targetGen. targetGen
+// increments on every SetTarget (an internal generation counter, so it also covers
+// config-only updates that carry no network-map serial).
+//
+// onConverged fires once for each — and only each — map that is actually processed
+// (i.e. converged as the target). Skipped/superseded maps and dropped-on-error maps
+// do NOT fire it. So "sync finished in X" / RecordSyncDuration always corresponds
+// to a real, completed alignment.
+type mapStateManager struct {
+	// apply performs one bounded apply pass and reports whether more passes are needed.
+	// firstPass is true on the first pass of a given target, so the caller can run
+	// wholesale (firewall/routes/DNS/forward-rules) once per target and skip it on the
+	// re-runs that only drain the bounded peer batches. The manager owns this signal
+	// because it owns the convergence boundary; the engine need not track serials for it.
+	apply func(update *mgmProto.SyncResponse, firstPass bool) (bool, error)
+	// onConverged is called once per processed map, with the elapsed time since that
+	// map was received (for the sync-duration metric / "sync finished" log).
+	onConverged func(time.Duration)
+	// persist snapshots an update to disk for restore-on-restart. Called once per
+	// update received from management (in SetTarget), including ones later coalesced
+	// or skipped from apply, so the on-disk state mirrors what management last sent.
+	// The impl skips config-only updates (nil NetworkMap). May be nil.
+	persist func(*mgmProto.SyncResponse)
+
+	mu          sync.Mutex
+	target      *mgmProto.SyncResponse
+	targetGen   uint64
+	appliedGen  uint64
+	targetSetAt time.Time
+
+	wake chan struct{}
+}
+
+func newMapStateManager(apply func(update *mgmProto.SyncResponse, firstPass bool) (bool, error), persist func(*mgmProto.SyncResponse), onConverged func(time.Duration)) *mapStateManager {
+	return &mapStateManager{
+		apply:       apply,
+		persist:     persist,
+		onConverged: onConverged,
+		wake:        make(chan struct{}, 1),
+	}
+}
+
+// SetTarget records the latest update as the desired state and wakes the loop.
+// It returns immediately; convergence happens in the background. Serial-based
+// staleness of the network map is still enforced inside apply (updateNetworkMap).
+func (m *mapStateManager) SetTarget(update *mgmProto.SyncResponse) error {
+	m.mu.Lock()
+	// A target that has not settled yet (targetGen > appliedGen) is being superseded
+	// before it converged: we coalesce to the latest map and never apply this one on
+	// its own. It is SKIPPED — logged here, and it will not fire onConverged.
+	if m.target != nil && m.targetGen > m.appliedGen {
+		log.Debugf("sync map (gen %d) superseded before convergence, skipping", m.targetGen)
+	}
+	m.target = m.mergeTarget(m.target, update)
+	// Bump an internal generation counter, NOT the map serial: config-only updates
+	// (relay token rotation, STUN/TURN) arrive with NetworkMap == nil and carry no
+	// serial, yet must still be applied. Every SetTarget is therefore a distinct
+	// target regardless of payload. Map-serial staleness is enforced separately
+	// inside apply (updateNetworkMap).
+	m.targetGen++
+	m.targetSetAt = time.Now()
+	m.mu.Unlock()
+
+	select {
+	case m.wake <- struct{}{}:
+	default:
+	}
+
+	// Persist every update received from management — once per update (not per apply
+	// pass), and including ones that get coalesced/skipped from apply, so the on-disk
+	// state always reflects the latest map management sent. Done after waking the loop
+	// so convergence can start in parallel with the disk write. The persist impl skips
+	// config-only updates (nil NetworkMap).
+	if m.persist != nil {
+		m.persist(update)
+	}
+	return nil
+}
+
+// mergeTarget combines the currently pending target with a freshly received update
+// and returns the new desired state. It is called under m.mu from SetTarget and is
+// the single seam where the replace-vs-squash decision lives.
+//
+// Today management always sends a FULL map (the complete desired state), so the
+// update simply replaces whatever was pending — prev is ignored. When management
+// starts sending incremental/delta updates, squash `update` onto `prev` here; the
+// rest of the manager (generation tracking, convergence, signaling) is unaffected
+// because it already treats target as "the complete desired state, whatever it is".
+func (m *mapStateManager) mergeTarget(prev, update *mgmProto.SyncResponse) *mgmProto.SyncResponse {
+	return update
+}
+
+// run drives convergence until ctx is done. It is meant to run in its own goroutine.
+func (m *mapStateManager) run(ctx context.Context) {
+	// passGen is the generation of the most recent apply() call (0 = none). A pass is
+	// the first for its target when its generation differs from the previous one —
+	// true on a fresh target and on a coalesced switch to a newer target mid-flight.
+	var passGen uint64
+	for {
+		m.mu.Lock()
+		target, tg, ag := m.target, m.targetGen, m.appliedGen
+		m.mu.Unlock()
+
+		// Fully converged (or nothing yet): block until a new target arrives.
+		if target == nil || ag == tg {
+			select {
+			case <-ctx.Done():
+				return
+			case <-m.wake:
+				continue
+			}
+		}
+
+		firstPass := tg != passGen
+		passGen = tg
+		more, err := m.apply(target, firstPass)
+		if err != nil {
+			if ctx.Err() != nil {
+				return
+			}
+			// Log and DROP this target — do not retry it. A deterministic failure
+			// (e.g. a malformed peer in the map) would otherwise spin every pass
+			// making no progress. Management is the source of truth and re-delivers
+			// the full map on the next sync, so dropping is safe; peers already
+			// applied this convergence stay (idempotent diffs) and the remainder is
+			// reconciled by the next target. Mirrors the legacy handleSync path,
+			// where the apply error was logged by the gRPC client and the update
+			// dropped. No onConverged: this target did not converge.
+			log.Errorf("apply sync pass, dropping update: %v", err)
+			m.settle(tg, false)
+			continue
+		}
+
+		if more {
+			// keep converging the current target; syncMsgMux was released by apply
+			// between passes so other subsystems interleave.
+			continue
+		}
+
+		// This pass converged. Mark applied and signal this one map.
+		m.settle(tg, true)
+		// if a newer target arrived mid-pass, settle is a no-op (targetGen != tg) and
+		// ag<tg next iteration -> apply it; this generation was skipped (logged in
+		// SetTarget) and is not signaled.
+	}
+}
+
+// settle marks generation tg as processed so the loop goes idle instead of
+// re-applying the same target. It is a no-op when a newer target arrived during the
+// pass (targetGen != tg), leaving appliedGen behind so that target re-applies — the
+// just-finished generation was already counted as skipped.
+//
+// When signal is true (the pass converged) it fires onConverged once for this map;
+// when false (the target was dropped on error) it does not — the map did not converge.
+func (m *mapStateManager) settle(tg uint64, signal bool) {
+	m.mu.Lock()
+	if m.targetGen != tg {
+		m.mu.Unlock()
+		return
+	}
+	m.appliedGen = tg
+	setAt := m.targetSetAt
+	m.mu.Unlock()
+
+	if signal && m.onConverged != nil {
+		m.onConverged(time.Since(setAt))
+	}
+}

client/internal/mapsync_test.go

@@ -0,0 +1,242 @@
+package internal
+
+import (
+	"context"
+	"errors"
+	"sync/atomic"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/require"
+
+	mgmProto "github.com/netbirdio/netbird/shared/management/proto"
+)
+
+// converges over the bounded passes (apply returns more until the 3rd pass),
+// fires onConverged exactly once, then blocks (no further apply) until a new target.
+func TestMapStateManager_ConvergesThenStops(t *testing.T) {
+	var passes int32
+	var firstPasses int32
+	converged := make(chan struct{}, 1)
+
+	apply := func(_ *mgmProto.SyncResponse, firstPass bool) (bool, error) {
+		n := atomic.AddInt32(&passes, 1)
+		if firstPass {
+			atomic.AddInt32(&firstPasses, 1)
+		}
+		return n < 3, nil // more on pass 1 and 2, converge on pass 3
+	}
+	m := newMapStateManager(apply, nil, func(time.Duration) { converged <- struct{}{} })
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+	go m.run(ctx)
+
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+
+	select {
+	case <-converged:
+	case <-time.After(2 * time.Second):
+		t.Fatal("manager did not converge")
+	}
+	require.EqualValues(t, 3, atomic.LoadInt32(&passes))
+	require.EqualValues(t, 1, atomic.LoadInt32(&firstPasses), "firstPass true only on pass 1, false on re-runs of the same target")
+
+	// once converged the loop blocks: no further apply calls
+	time.Sleep(100 * time.Millisecond)
+	require.EqualValues(t, 3, atomic.LoadInt32(&passes), "apply must not run after convergence")
+}
+
+// persist runs once per received update (not per apply pass), regardless of how many
+// bounded passes that target takes to converge.
+func TestMapStateManager_PersistsOncePerUpdate(t *testing.T) {
+	var passes, persists int32
+	converged := make(chan struct{}, 1)
+	apply := func(_ *mgmProto.SyncResponse, _ bool) (bool, error) {
+		n := atomic.AddInt32(&passes, 1)
+		return n < 3, nil // 3 passes for one target
+	}
+	persist := func(*mgmProto.SyncResponse) { atomic.AddInt32(&persists, 1) }
+	m := newMapStateManager(apply, persist, func(time.Duration) { converged <- struct{}{} })
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+	go m.run(ctx)
+
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+	select {
+	case <-converged:
+	case <-time.After(2 * time.Second):
+		t.Fatal("did not converge")
+	}
+	require.EqualValues(t, 3, atomic.LoadInt32(&passes))
+	require.EqualValues(t, 1, atomic.LoadInt32(&persists), "persist once per update, not per pass")
+}
+
+// every update received from management is persisted — even one that is coalesced /
+// skipped from apply before it ever converges.
+func TestMapStateManager_PersistsEveryUpdateIncludingSkipped(t *testing.T) {
+	release := make(chan struct{})
+	var persists int32
+	apply := func(_ *mgmProto.SyncResponse, _ bool) (bool, error) {
+		<-release // hold the first apply so the second update coalesces/skips
+		return false, nil
+	}
+	persist := func(*mgmProto.SyncResponse) { atomic.AddInt32(&persists, 1) }
+	m := newMapStateManager(apply, persist, nil)
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+	go m.run(ctx)
+
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{})) // map1 -> apply blocks
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{})) // map2 supersedes map1 (skipped from apply)
+	close(release)
+
+	// both updates persisted even though map1 is skipped from apply
+	require.Eventually(t, func() bool { return atomic.LoadInt32(&persists) == 2 }, 2*time.Second, 10*time.Millisecond)
+}
+
+// each map that is actually processed (converged before the next arrives) fires
+// onConverged exactly once — mirroring the legacy per-message handleSync timing.
+func TestMapStateManager_SignalsEachProcessedMap(t *testing.T) {
+	converged := make(chan struct{}, 8)
+	apply := func(_ *mgmProto.SyncResponse, _ bool) (bool, error) {
+		return false, nil // converge in one pass
+	}
+	m := newMapStateManager(apply, nil, func(time.Duration) { converged <- struct{}{} })
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+	go m.run(ctx)
+
+	const maps = 3
+	for i := 0; i < maps; i++ {
+		require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+		select { // wait for this map to converge before sending the next (no coalescing)
+		case <-converged:
+		case <-time.After(2 * time.Second):
+			t.Fatalf("map %d not signaled", i)
+		}
+	}
+
+	// no extra signals once the stream goes quiet
+	select {
+	case <-converged:
+		t.Fatal("unexpected extra onConverged")
+	case <-time.After(100 * time.Millisecond):
+	}
+}
+
+// a map superseded before it converges is skipped: only the latest (processed) map
+// fires onConverged, not the skipped one.
+func TestMapStateManager_SkippedMapNotSignaled(t *testing.T) {
+	release := make(chan struct{})
+	var applies, converged atomic.Int32
+	apply := func(_ *mgmProto.SyncResponse, _ bool) (bool, error) {
+		applies.Add(1)
+		<-release // hold the first apply in-flight so we can queue a newer target
+		return false, nil
+	}
+	m := newMapStateManager(apply, nil, func(time.Duration) { converged.Add(1) })
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+	go m.run(ctx)
+
+	// map1 is picked up; its apply blocks on release
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+	require.Eventually(t, func() bool { return applies.Load() >= 1 }, 2*time.Second, 5*time.Millisecond)
+
+	// map2 supersedes map1 before it settled -> map1 is skipped
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+	close(release) // let both applies proceed
+
+	// only the processed (latest) map signals; the skipped one does not
+	require.Eventually(t, func() bool { return converged.Load() == 1 }, 2*time.Second, 10*time.Millisecond)
+	time.Sleep(150 * time.Millisecond)
+	require.EqualValues(t, 1, converged.Load(), "skipped map must not fire onConverged")
+	require.EqualValues(t, 2, applies.Load(), "both targets entered apply (map1 once, map2 once)")
+}
+
+// an apply error drops the target: no retry of the same target, no onConverged,
+// the loop goes idle — and a fresh target is still applied afterwards.
+func TestMapStateManager_DropsTargetOnError(t *testing.T) {
+	applied := make(chan struct{}, 8)
+	var failNext atomic.Bool
+	failNext.Store(true)
+
+	apply := func(_ *mgmProto.SyncResponse, _ bool) (bool, error) {
+		applied <- struct{}{}
+		if failNext.Load() {
+			return false, errors.New("boom")
+		}
+		return false, nil // converge in one pass
+	}
+	var converged atomic.Int32
+	m := newMapStateManager(apply, nil, func(time.Duration) { converged.Add(1) })
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+	go m.run(ctx)
+
+	// first target errors -> applied once, then dropped (no retry, no onConverged)
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+	select {
+	case <-applied:
+	case <-time.After(2 * time.Second):
+		t.Fatal("errored target not applied")
+	}
+	select {
+	case <-applied:
+		t.Fatal("errored target must not be retried")
+	case <-time.After(150 * time.Millisecond):
+	}
+	require.EqualValues(t, 0, converged.Load(), "onConverged must not fire on error")
+
+	// a new target is still processed normally and converges
+	failNext.Store(false)
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+	select {
+	case <-applied:
+	case <-time.After(2 * time.Second):
+		t.Fatal("new target after error not applied")
+	}
+	require.Eventually(t, func() bool { return converged.Load() == 1 }, 2*time.Second, 10*time.Millisecond)
+}
+
+// a new target after convergence triggers a fresh apply; an idle (converged)
+// manager does not apply on its own.
+func TestMapStateManager_ReappliesOnNewTarget(t *testing.T) {
+	applied := make(chan struct{}, 8)
+	apply := func(_ *mgmProto.SyncResponse, _ bool) (bool, error) {
+		applied <- struct{}{}
+		return false, nil // converge in one pass
+	}
+	m := newMapStateManager(apply, nil, nil)
+
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+	go m.run(ctx)
+
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+	select {
+	case <-applied:
+	case <-time.After(2 * time.Second):
+		t.Fatal("first target not applied")
+	}
+
+	// converged → must stay idle (no spurious apply)
+	select {
+	case <-applied:
+		t.Fatal("unexpected apply while idle/converged")
+	case <-time.After(150 * time.Millisecond):
+	}
+
+	require.NoError(t, m.SetTarget(&mgmProto.SyncResponse{}))
+	select {
+	case <-applied:
+	case <-time.After(2 * time.Second):
+		t.Fatal("new target not applied")
+	}
+}

client/internal/peer/conn.go

@@ -6,7 +6,6 @@ import (
 	"net"
 	"net/netip"
 	"runtime"
-	"slices"
 	"sync"
 	"time"
 
@@ -137,46 +136,6 @@ type Conn struct {
 	// Connection stage timestamps for metrics
 	metricsRecorder MetricsRecorder
 	metricsStages   *MetricsStages
-
-	// pendingFirstPacket is the lazyconn-captured handshake init, replayed once the real
-	// transport is up.
-	pendingFirstPacket []byte
-}
-
-// SetPendingFirstPacket stashes a packet to be replayed once the connection is established.
-func (conn *Conn) SetPendingFirstPacket(b []byte) {
-	conn.mu.Lock()
-	conn.pendingFirstPacket = slices.Clone(b)
-	conn.mu.Unlock()
-}
-
-// injectPendingFirstPacket replays the captured handshake through the proxy if present, else
-// directly through the ICE conn. The packet is cleared only after a successful write, so a failed
-// or transport-less attempt leaves it available for a later reinjection. Caller must hold conn.mu.
-func (conn *Conn) injectPendingFirstPacket(proxy wgproxy.Proxy, directConn net.Conn) {
-	pkt := conn.pendingFirstPacket
-	if len(pkt) == 0 {
-		return
-	}
-
-	switch {
-	case proxy != nil:
-		if err := proxy.InjectPacket(pkt); err != nil {
-			conn.Log.Debugf("failed to reinject captured first packet via proxy: %v", err)
-			return
-		}
-	case directConn != nil:
-		if _, err := directConn.Write(pkt); err != nil {
-			conn.Log.Debugf("failed to reinject captured first packet via direct conn: %v", err)
-			return
-		}
-	default:
-		conn.Log.Debugf("no transport available to reinject captured first packet")
-		return
-	}
-
-	conn.pendingFirstPacket = nil
-	conn.Log.Debugf("reinjected captured first packet (%d bytes)", len(pkt))
 }
 
 // NewConn creates a new not opened Conn to the remote peer.
@@ -464,8 +423,6 @@ func (conn *Conn) onICEConnectionIsReady(priority conntype.ConnPriority, iceConn
 		conn.wgProxyRelay.RedirectAs(ep)
 	}
 
-	conn.injectPendingFirstPacket(wgProxy, iceConnInfo.RemoteConn)
-
 	conn.currentConnPriority = priority
 	conn.statusICE.SetConnected()
 	conn.updateIceState(iceConnInfo, updateTime)
@@ -589,8 +546,6 @@ func (conn *Conn) onRelayConnectionIsReady(rci RelayConnInfo) {
 
 	wgConfigWorkaround()
 
-	conn.injectPendingFirstPacket(wgProxy, nil)
-
 	conn.rosenpassRemoteKey = rci.rosenpassPubKey
 	conn.currentConnPriority = conntype.Relay
 	conn.statusRelay.SetConnected()

client/internal/peerstore/store.go

@@ -81,12 +81,6 @@ func (s *Store) PeerConn(pubKey string) (*peer.Conn, bool) {
 }
 
 func (s *Store) PeerConnOpen(ctx context.Context, pubKey string) {
-	s.PeerConnOpenWithFirstPacket(ctx, pubKey, nil)
-}
-
-// PeerConnOpenWithFirstPacket opens the peer connection and stashes a first packet to be
-// reinjected once the real transport is established.
-func (s *Store) PeerConnOpenWithFirstPacket(ctx context.Context, pubKey string, firstPacket []byte) {
 	s.peerConnsMu.RLock()
 	defer s.peerConnsMu.RUnlock()
 
@@ -94,15 +88,11 @@ func (s *Store) PeerConnOpenWithFirstPacket(ctx context.Context, pubKey string,
 	if !ok {
 		return
 	}
-	if len(firstPacket) > 0 {
-		p.SetPendingFirstPacket(firstPacket)
-	}
 	// this can be blocked because of the connect open limiter semaphore
 	if err := p.Open(ctx); err != nil {
 		p.Log.Errorf("failed to open peer connection: %v", err)
-		// Drop the stashed packet so a later open does not replay a stale handshake.
-		p.SetPendingFirstPacket(nil)
 	}
+
 }
 
 func (s *Store) PeerConnIdle(pubKey string) {

client/internal/routemanager/manager_test.go

@@ -1,3 +1,5 @@
+//go:build privileged
+
 package routemanager
 
 import (

client/internal/routemanager/systemops/rt_tables_linux_test.go

@@ -0,0 +1,69 @@
+//go:build linux && !android
+
+package systemops
+
+import (
+	"fmt"
+	"os"
+	"strings"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestEntryExists(t *testing.T) {
+	tempDir := t.TempDir()
+	tempFilePath := fmt.Sprintf("%s/rt_tables", tempDir)
+
+	content := []string{
+		"1000 reserved",
+		fmt.Sprintf("%d %s", NetbirdVPNTableID, NetbirdVPNTableName),
+		"9999 other_table",
+	}
+	require.NoError(t, os.WriteFile(tempFilePath, []byte(strings.Join(content, "\n")), 0644))
+
+	file, err := os.Open(tempFilePath)
+	require.NoError(t, err)
+	defer func() {
+		assert.NoError(t, file.Close())
+	}()
+
+	tests := []struct {
+		name        string
+		id          int
+		shouldExist bool
+		err         error
+	}{
+		{
+			name:        "ExistsWithNetbirdPrefix",
+			id:          7120,
+			shouldExist: true,
+			err:         nil,
+		},
+		{
+			name:        "ExistsWithDifferentName",
+			id:          1000,
+			shouldExist: true,
+			err:         ErrTableIDExists,
+		},
+		{
+			name:        "DoesNotExist",
+			id:          1234,
+			shouldExist: false,
+			err:         nil,
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			exists, err := entryExists(file, tc.id)
+			if tc.err != nil {
+				assert.ErrorIs(t, err, tc.err)
+			} else {
+				assert.NoError(t, err)
+			}
+			assert.Equal(t, tc.shouldExist, exists)
+		})
+	}
+}

client/internal/routemanager/systemops/systemops_bsd_privileged_test.go

@@ -0,0 +1,191 @@
+//go:build (darwin || dragonfly || freebsd || netbsd || openbsd) && privileged
+
+package systemops
+
+import (
+	"fmt"
+	"net"
+	"net/netip"
+	"os/exec"
+	"regexp"
+	"runtime"
+	"strings"
+	"sync"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func init() {
+	testCases = append(testCases, []testCase{
+		{
+			name:              "To more specific route without custom dialer via vpn",
+			expectedInterface: expectedVPNint,
+			dialer:            &net.Dialer{},
+			expectedPacket:    createPacketExpectation("100.64.0.1", 12345, "10.10.0.2", 53),
+		},
+	}...)
+}
+
+func TestConcurrentRoutes(t *testing.T) {
+	baseIP := netip.MustParseAddr("192.0.2.0")
+
+	var intf *net.Interface
+	var nexthop Nexthop
+
+	_, intf = setupDummyInterface(t)
+	nexthop = Nexthop{netip.Addr{}, intf}
+
+	r := New(nil, nil)
+
+	var wg sync.WaitGroup
+	for i := 0; i < 1024; i++ {
+		wg.Add(1)
+		go func(ip netip.Addr) {
+			defer wg.Done()
+			prefix := netip.PrefixFrom(ip, 32)
+			if err := r.addToRouteTable(prefix, nexthop); err != nil {
+				t.Errorf("Failed to add route for %s: %v", prefix, err)
+			}
+		}(baseIP)
+		baseIP = baseIP.Next()
+	}
+
+	wg.Wait()
+
+	baseIP = netip.MustParseAddr("192.0.2.0")
+
+	for i := 0; i < 1024; i++ {
+		wg.Add(1)
+		go func(ip netip.Addr) {
+			defer wg.Done()
+			prefix := netip.PrefixFrom(ip, 32)
+			if err := r.removeFromRouteTable(prefix, nexthop); err != nil {
+				t.Errorf("Failed to remove route for %s: %v", prefix, err)
+			}
+		}(baseIP)
+		baseIP = baseIP.Next()
+	}
+
+	wg.Wait()
+}
+
+func createAndSetupDummyInterface(t *testing.T, intf string, ipAddressCIDR string) string {
+	t.Helper()
+
+	if runtime.GOOS == "darwin" {
+		err := exec.Command("ifconfig", intf, "alias", ipAddressCIDR).Run()
+		require.NoError(t, err, "Failed to create loopback alias")
+
+		t.Cleanup(func() {
+			err := exec.Command("ifconfig", intf, ipAddressCIDR, "-alias").Run()
+			assert.NoError(t, err, "Failed to remove loopback alias")
+		})
+
+		return intf
+	}
+
+	prefix, err := netip.ParsePrefix(ipAddressCIDR)
+	require.NoError(t, err, "Failed to parse prefix")
+
+	netIntf, err := net.InterfaceByName(intf)
+	require.NoError(t, err, "Failed to get interface by name")
+
+	nexthop := Nexthop{netip.Addr{}, netIntf}
+
+	r := New(nil, nil)
+	err = r.addToRouteTable(prefix, nexthop)
+	require.NoError(t, err, "Failed to add route to table")
+
+	t.Cleanup(func() {
+		err := r.removeFromRouteTable(prefix, nexthop)
+		assert.NoError(t, err, "Failed to remove route from table")
+	})
+
+	return intf
+}
+
+func addDummyRoute(t *testing.T, dstCIDR string, gw netip.Addr, _ string) {
+	t.Helper()
+
+	var originalNexthop net.IP
+	if dstCIDR == "0.0.0.0/0" {
+		var err error
+		originalNexthop, err = fetchOriginalGateway()
+		if err != nil {
+			t.Logf("Failed to fetch original gateway: %v", err)
+		}
+
+		if output, err := exec.Command("route", "delete", "-net", dstCIDR).CombinedOutput(); err != nil {
+			t.Logf("Failed to delete route: %v, output: %s", err, output)
+		}
+	}
+
+	t.Cleanup(func() {
+		if originalNexthop != nil {
+			err := exec.Command("route", "add", "-net", dstCIDR, originalNexthop.String()).Run()
+			assert.NoError(t, err, "Failed to restore original route")
+		}
+	})
+
+	err := exec.Command("route", "add", "-net", dstCIDR, gw.String()).Run()
+	require.NoError(t, err, "Failed to add route")
+
+	t.Cleanup(func() {
+		err := exec.Command("route", "delete", "-net", dstCIDR).Run()
+		assert.NoError(t, err, "Failed to remove route")
+	})
+}
+
+func fetchOriginalGateway() (net.IP, error) {
+	output, err := exec.Command("route", "-n", "get", "default").CombinedOutput()
+	if err != nil {
+		return nil, err
+	}
+
+	matches := regexp.MustCompile(`gateway: (\S+)`).FindStringSubmatch(string(output))
+	if len(matches) == 0 {
+		return nil, fmt.Errorf("gateway not found")
+	}
+
+	return net.ParseIP(matches[1]), nil
+}
+
+// setupDummyInterface creates a dummy tun interface for FreeBSD route testing
+func setupDummyInterface(t *testing.T) (netip.Addr, *net.Interface) {
+	t.Helper()
+
+	if runtime.GOOS == "darwin" {
+		return netip.AddrFrom4([4]byte{192, 168, 1, 2}), &net.Interface{Name: "lo0"}
+	}
+
+	output, err := exec.Command("ifconfig", "tun", "create").CombinedOutput()
+	require.NoError(t, err, "Failed to create tun interface: %s", string(output))
+
+	tunName := strings.TrimSpace(string(output))
+
+	output, err = exec.Command("ifconfig", tunName, "192.168.1.1", "netmask", "255.255.0.0", "192.168.1.2", "up").CombinedOutput()
+	require.NoError(t, err, "Failed to configure tun interface: %s", string(output))
+
+	intf, err := net.InterfaceByName(tunName)
+	require.NoError(t, err, "Failed to get interface by name")
+
+	t.Cleanup(func() {
+		if err := exec.Command("ifconfig", tunName, "destroy").Run(); err != nil {
+			t.Logf("Failed to destroy tun interface %s: %v", tunName, err)
+		}
+	})
+
+	return netip.AddrFrom4([4]byte{192, 168, 1, 2}), intf
+}
+
+func setupDummyInterfacesAndRoutes(t *testing.T) {
+	t.Helper()
+
+	defaultDummy := createAndSetupDummyInterface(t, expectedExternalInt, "192.168.0.1/24")
+	addDummyRoute(t, "0.0.0.0/0", netip.AddrFrom4([4]byte{192, 168, 0, 1}), defaultDummy)
+
+	otherDummy := createAndSetupDummyInterface(t, expectedInternalInt, "192.168.1.1/24")
+	addDummyRoute(t, "10.0.0.0/8", netip.AddrFrom4([4]byte{192, 168, 1, 1}), otherDummy)
+}

client/internal/routemanager/systemops/systemops_bsd_test.go

@@ -3,79 +3,24 @@
 package systemops
 
 import (
-	"fmt"
-	"net"
-	"net/netip"
-	"os/exec"
-	"regexp"
-	"runtime"
-	"strings"
-	"sync"
 	"testing"
 
 	"github.com/stretchr/testify/assert"
-	"github.com/stretchr/testify/require"
 	"golang.org/x/net/route"
 )
 
+// Interface names used by the shared routing test fixtures. Kept untagged (no
+// privileged build tag) so the non-privileged test files in this package compile.
+//
+//nolint:unused // consumed by the privileged-tagged routing tests
 var expectedVPNint = "utun100"
+
+//nolint:unused // consumed by the privileged-tagged routing tests
 var expectedExternalInt = "lo0"
+
+//nolint:unused // consumed by the privileged-tagged routing tests
 var expectedInternalInt = "lo0"
 
-func init() {
-	testCases = append(testCases, []testCase{
-		{
-			name:              "To more specific route without custom dialer via vpn",
-			expectedInterface: expectedVPNint,
-			dialer:            &net.Dialer{},
-			expectedPacket:    createPacketExpectation("100.64.0.1", 12345, "10.10.0.2", 53),
-		},
-	}...)
-}
-
-func TestConcurrentRoutes(t *testing.T) {
-	baseIP := netip.MustParseAddr("192.0.2.0")
-
-	var intf *net.Interface
-	var nexthop Nexthop
-
-	_, intf = setupDummyInterface(t)
-	nexthop = Nexthop{netip.Addr{}, intf}
-
-	r := New(nil, nil)
-
-	var wg sync.WaitGroup
-	for i := 0; i < 1024; i++ {
-		wg.Add(1)
-		go func(ip netip.Addr) {
-			defer wg.Done()
-			prefix := netip.PrefixFrom(ip, 32)
-			if err := r.addToRouteTable(prefix, nexthop); err != nil {
-				t.Errorf("Failed to add route for %s: %v", prefix, err)
-			}
-		}(baseIP)
-		baseIP = baseIP.Next()
-	}
-
-	wg.Wait()
-
-	baseIP = netip.MustParseAddr("192.0.2.0")
-
-	for i := 0; i < 1024; i++ {
-		wg.Add(1)
-		go func(ip netip.Addr) {
-			defer wg.Done()
-			prefix := netip.PrefixFrom(ip, 32)
-			if err := r.removeFromRouteTable(prefix, nexthop); err != nil {
-				t.Errorf("Failed to remove route for %s: %v", prefix, err)
-			}
-		}(baseIP)
-		baseIP = baseIP.Next()
-	}
-
-	wg.Wait()
-}
-
 func TestBits(t *testing.T) {
 	tests := []struct {
 		name    string
@@ -122,122 +67,3 @@ func TestBits(t *testing.T) {
 		})
 	}
 }
-
-func createAndSetupDummyInterface(t *testing.T, intf string, ipAddressCIDR string) string {
-	t.Helper()
-
-	if runtime.GOOS == "darwin" {
-		err := exec.Command("ifconfig", intf, "alias", ipAddressCIDR).Run()
-		require.NoError(t, err, "Failed to create loopback alias")
-
-		t.Cleanup(func() {
-			err := exec.Command("ifconfig", intf, ipAddressCIDR, "-alias").Run()
-			assert.NoError(t, err, "Failed to remove loopback alias")
-		})
-
-		return intf
-	}
-
-	prefix, err := netip.ParsePrefix(ipAddressCIDR)
-	require.NoError(t, err, "Failed to parse prefix")
-
-	netIntf, err := net.InterfaceByName(intf)
-	require.NoError(t, err, "Failed to get interface by name")
-
-	nexthop := Nexthop{netip.Addr{}, netIntf}
-
-	r := New(nil, nil)
-	err = r.addToRouteTable(prefix, nexthop)
-	require.NoError(t, err, "Failed to add route to table")
-
-	t.Cleanup(func() {
-		err := r.removeFromRouteTable(prefix, nexthop)
-		assert.NoError(t, err, "Failed to remove route from table")
-	})
-
-	return intf
-}
-
-func addDummyRoute(t *testing.T, dstCIDR string, gw netip.Addr, _ string) {
-	t.Helper()
-
-	var originalNexthop net.IP
-	if dstCIDR == "0.0.0.0/0" {
-		var err error
-		originalNexthop, err = fetchOriginalGateway()
-		if err != nil {
-			t.Logf("Failed to fetch original gateway: %v", err)
-		}
-
-		if output, err := exec.Command("route", "delete", "-net", dstCIDR).CombinedOutput(); err != nil {
-			t.Logf("Failed to delete route: %v, output: %s", err, output)
-		}
-	}
-
-	t.Cleanup(func() {
-		if originalNexthop != nil {
-			err := exec.Command("route", "add", "-net", dstCIDR, originalNexthop.String()).Run()
-			assert.NoError(t, err, "Failed to restore original route")
-		}
-	})
-
-	err := exec.Command("route", "add", "-net", dstCIDR, gw.String()).Run()
-	require.NoError(t, err, "Failed to add route")
-
-	t.Cleanup(func() {
-		err := exec.Command("route", "delete", "-net", dstCIDR).Run()
-		assert.NoError(t, err, "Failed to remove route")
-	})
-}
-
-func fetchOriginalGateway() (net.IP, error) {
-	output, err := exec.Command("route", "-n", "get", "default").CombinedOutput()
-	if err != nil {
-		return nil, err
-	}
-
-	matches := regexp.MustCompile(`gateway: (\S+)`).FindStringSubmatch(string(output))
-	if len(matches) == 0 {
-		return nil, fmt.Errorf("gateway not found")
-	}
-
-	return net.ParseIP(matches[1]), nil
-}
-
-// setupDummyInterface creates a dummy tun interface for FreeBSD route testing
-func setupDummyInterface(t *testing.T) (netip.Addr, *net.Interface) {
-	t.Helper()
-
-	if runtime.GOOS == "darwin" {
-		return netip.AddrFrom4([4]byte{192, 168, 1, 2}), &net.Interface{Name: "lo0"}
-	}
-
-	output, err := exec.Command("ifconfig", "tun", "create").CombinedOutput()
-	require.NoError(t, err, "Failed to create tun interface: %s", string(output))
-
-	tunName := strings.TrimSpace(string(output))
-
-	output, err = exec.Command("ifconfig", tunName, "192.168.1.1", "netmask", "255.255.0.0", "192.168.1.2", "up").CombinedOutput()
-	require.NoError(t, err, "Failed to configure tun interface: %s", string(output))
-
-	intf, err := net.InterfaceByName(tunName)
-	require.NoError(t, err, "Failed to get interface by name")
-
-	t.Cleanup(func() {
-		if err := exec.Command("ifconfig", tunName, "destroy").Run(); err != nil {
-			t.Logf("Failed to destroy tun interface %s: %v", tunName, err)
-		}
-	})
-
-	return netip.AddrFrom4([4]byte{192, 168, 1, 2}), intf
-}
-
-func setupDummyInterfacesAndRoutes(t *testing.T) {
-	t.Helper()
-
-	defaultDummy := createAndSetupDummyInterface(t, expectedExternalInt, "192.168.0.1/24")
-	addDummyRoute(t, "0.0.0.0/0", netip.AddrFrom4([4]byte{192, 168, 0, 1}), defaultDummy)
-
-	otherDummy := createAndSetupDummyInterface(t, expectedInternalInt, "192.168.1.1/24")
-	addDummyRoute(t, "10.0.0.0/8", netip.AddrFrom4([4]byte{192, 168, 1, 1}), otherDummy)
-}

client/internal/routemanager/systemops/systemops_dialer_test.go

@@ -0,0 +1,17 @@
+//go:build !android && !ios
+
+package systemops
+
+import (
+	"context"
+	"net"
+)
+
+// dialer is shared by the per-platform routing test cases. Kept untagged (no
+// privileged build tag) so the non-privileged test files compile on every platform.
+//
+//nolint:unused // consumed by the privileged-tagged routing tests
+type dialer interface {
+	Dial(network, address string) (net.Conn, error)
+	DialContext(ctx context.Context, network, address string) (net.Conn, error)
+}

client/internal/routemanager/systemops/systemops_generic_test.go

@@ -1,4 +1,4 @@
-//go:build !android && !ios
+//go:build !android && !ios && privileged
 
 package systemops
 
@@ -26,11 +26,6 @@ import (
 	nbnet "github.com/netbirdio/netbird/client/net"
 )
 
-type dialer interface {
-	Dial(network, address string) (net.Conn, error)
-	DialContext(ctx context.Context, network, address string) (net.Conn, error)
-}
-
 func TestAddVPNRoute(t *testing.T) {
 	testCases := []struct {
 		name        string
@@ -515,125 +510,3 @@ func setupTestEnv(t *testing.T) {
 	// unique route in vpn table
 	setupRouteAndCleanup(t, r, netip.MustParsePrefix("172.16.0.0/12"), intf)
 }
-
-func TestIsVpnRoute(t *testing.T) {
-	tests := []struct {
-		name           string
-		addr           string
-		vpnRoutes      []string
-		localRoutes    []string
-		expectedVpn    bool
-		expectedPrefix netip.Prefix
-	}{
-		{
-			name:           "Match in VPN routes",
-			addr:           "192.168.1.1",
-			vpnRoutes:      []string{"192.168.1.0/24"},
-			localRoutes:    []string{"10.0.0.0/8"},
-			expectedVpn:    true,
-			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
-		},
-		{
-			name:           "Match in local routes",
-			addr:           "10.1.1.1",
-			vpnRoutes:      []string{"192.168.1.0/24"},
-			localRoutes:    []string{"10.0.0.0/8"},
-			expectedVpn:    false,
-			expectedPrefix: netip.MustParsePrefix("10.0.0.0/8"),
-		},
-		{
-			name:           "No match",
-			addr:           "172.16.0.1",
-			vpnRoutes:      []string{"192.168.1.0/24"},
-			localRoutes:    []string{"10.0.0.0/8"},
-			expectedVpn:    false,
-			expectedPrefix: netip.Prefix{},
-		},
-		{
-			name:           "Default route ignored",
-			addr:           "192.168.1.1",
-			vpnRoutes:      []string{"0.0.0.0/0", "192.168.1.0/24"},
-			localRoutes:    []string{"10.0.0.0/8"},
-			expectedVpn:    true,
-			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
-		},
-		{
-			name:           "Default route matches but ignored",
-			addr:           "172.16.1.1",
-			vpnRoutes:      []string{"0.0.0.0/0", "192.168.1.0/24"},
-			localRoutes:    []string{"10.0.0.0/8"},
-			expectedVpn:    false,
-			expectedPrefix: netip.Prefix{},
-		},
-		{
-			name:           "Longest prefix match local",
-			addr:           "192.168.1.1",
-			vpnRoutes:      []string{"192.168.0.0/16"},
-			localRoutes:    []string{"192.168.1.0/24"},
-			expectedVpn:    false,
-			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
-		},
-		{
-			name:           "Longest prefix match local multiple",
-			addr:           "192.168.0.1",
-			vpnRoutes:      []string{"192.168.0.0/16", "192.168.0.0/25", "192.168.0.0/27"},
-			localRoutes:    []string{"192.168.0.0/24", "192.168.0.0/26", "192.168.0.0/28"},
-			expectedVpn:    false,
-			expectedPrefix: netip.MustParsePrefix("192.168.0.0/28"),
-		},
-		{
-			name:           "Longest prefix match vpn",
-			addr:           "192.168.1.1",
-			vpnRoutes:      []string{"192.168.1.0/24"},
-			localRoutes:    []string{"192.168.0.0/16"},
-			expectedVpn:    true,
-			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
-		},
-		{
-			name:           "Longest prefix match vpn multiple",
-			addr:           "192.168.0.1",
-			vpnRoutes:      []string{"192.168.0.0/16", "192.168.0.0/25", "192.168.0.0/27"},
-			localRoutes:    []string{"192.168.0.0/24", "192.168.0.0/26"},
-			expectedVpn:    true,
-			expectedPrefix: netip.MustParsePrefix("192.168.0.0/27"),
-		},
-		{
-			name:           "Duplicate prefix in both",
-			addr:           "192.168.1.1",
-			vpnRoutes:      []string{"192.168.1.0/24"},
-			localRoutes:    []string{"192.168.1.0/24"},
-			expectedVpn:    false,
-			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
-		},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			addr, err := netip.ParseAddr(tt.addr)
-			if err != nil {
-				t.Fatalf("Failed to parse address %s: %v", tt.addr, err)
-			}
-
-			var vpnRoutes, localRoutes []netip.Prefix
-			for _, route := range tt.vpnRoutes {
-				prefix, err := netip.ParsePrefix(route)
-				if err != nil {
-					t.Fatalf("Failed to parse VPN route %s: %v", route, err)
-				}
-				vpnRoutes = append(vpnRoutes, prefix)
-			}
-
-			for _, route := range tt.localRoutes {
-				prefix, err := netip.ParsePrefix(route)
-				if err != nil {
-					t.Fatalf("Failed to parse local route %s: %v", route, err)
-				}
-				localRoutes = append(localRoutes, prefix)
-			}
-
-			isVpn, matchedPrefix := isVpnRoute(addr, vpnRoutes, localRoutes)
-			assert.Equal(t, tt.expectedVpn, isVpn, "isVpnRoute should return expectedVpn value")
-			assert.Equal(t, tt.expectedPrefix, matchedPrefix, "isVpnRoute should return expectedVpn prefix")
-		})
-	}
-}

client/internal/routemanager/systemops/systemops_isvpnroute_test.go

@@ -0,0 +1,132 @@
+//go:build !android && !ios
+
+package systemops
+
+import (
+	"net/netip"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestIsVpnRoute(t *testing.T) {
+	tests := []struct {
+		name           string
+		addr           string
+		vpnRoutes      []string
+		localRoutes    []string
+		expectedVpn    bool
+		expectedPrefix netip.Prefix
+	}{
+		{
+			name:           "Match in VPN routes",
+			addr:           "192.168.1.1",
+			vpnRoutes:      []string{"192.168.1.0/24"},
+			localRoutes:    []string{"10.0.0.0/8"},
+			expectedVpn:    true,
+			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
+		},
+		{
+			name:           "Match in local routes",
+			addr:           "10.1.1.1",
+			vpnRoutes:      []string{"192.168.1.0/24"},
+			localRoutes:    []string{"10.0.0.0/8"},
+			expectedVpn:    false,
+			expectedPrefix: netip.MustParsePrefix("10.0.0.0/8"),
+		},
+		{
+			name:           "No match",
+			addr:           "172.16.0.1",
+			vpnRoutes:      []string{"192.168.1.0/24"},
+			localRoutes:    []string{"10.0.0.0/8"},
+			expectedVpn:    false,
+			expectedPrefix: netip.Prefix{},
+		},
+		{
+			name:           "Default route ignored",
+			addr:           "192.168.1.1",
+			vpnRoutes:      []string{"0.0.0.0/0", "192.168.1.0/24"},
+			localRoutes:    []string{"10.0.0.0/8"},
+			expectedVpn:    true,
+			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
+		},
+		{
+			name:           "Default route matches but ignored",
+			addr:           "172.16.1.1",
+			vpnRoutes:      []string{"0.0.0.0/0", "192.168.1.0/24"},
+			localRoutes:    []string{"10.0.0.0/8"},
+			expectedVpn:    false,
+			expectedPrefix: netip.Prefix{},
+		},
+		{
+			name:           "Longest prefix match local",
+			addr:           "192.168.1.1",
+			vpnRoutes:      []string{"192.168.0.0/16"},
+			localRoutes:    []string{"192.168.1.0/24"},
+			expectedVpn:    false,
+			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
+		},
+		{
+			name:           "Longest prefix match local multiple",
+			addr:           "192.168.0.1",
+			vpnRoutes:      []string{"192.168.0.0/16", "192.168.0.0/25", "192.168.0.0/27"},
+			localRoutes:    []string{"192.168.0.0/24", "192.168.0.0/26", "192.168.0.0/28"},
+			expectedVpn:    false,
+			expectedPrefix: netip.MustParsePrefix("192.168.0.0/28"),
+		},
+		{
+			name:           "Longest prefix match vpn",
+			addr:           "192.168.1.1",
+			vpnRoutes:      []string{"192.168.1.0/24"},
+			localRoutes:    []string{"192.168.0.0/16"},
+			expectedVpn:    true,
+			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
+		},
+		{
+			name:           "Longest prefix match vpn multiple",
+			addr:           "192.168.0.1",
+			vpnRoutes:      []string{"192.168.0.0/16", "192.168.0.0/25", "192.168.0.0/27"},
+			localRoutes:    []string{"192.168.0.0/24", "192.168.0.0/26"},
+			expectedVpn:    true,
+			expectedPrefix: netip.MustParsePrefix("192.168.0.0/27"),
+		},
+		{
+			name:           "Duplicate prefix in both",
+			addr:           "192.168.1.1",
+			vpnRoutes:      []string{"192.168.1.0/24"},
+			localRoutes:    []string{"192.168.1.0/24"},
+			expectedVpn:    false,
+			expectedPrefix: netip.MustParsePrefix("192.168.1.0/24"),
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			addr, err := netip.ParseAddr(tt.addr)
+			if err != nil {
+				t.Fatalf("Failed to parse address %s: %v", tt.addr, err)
+			}
+
+			var vpnRoutes, localRoutes []netip.Prefix
+			for _, route := range tt.vpnRoutes {
+				prefix, err := netip.ParsePrefix(route)
+				if err != nil {
+					t.Fatalf("Failed to parse VPN route %s: %v", route, err)
+				}
+				vpnRoutes = append(vpnRoutes, prefix)
+			}
+
+			for _, route := range tt.localRoutes {
+				prefix, err := netip.ParsePrefix(route)
+				if err != nil {
+					t.Fatalf("Failed to parse local route %s: %v", route, err)
+				}
+				localRoutes = append(localRoutes, prefix)
+			}
+
+			isVpn, matchedPrefix := isVpnRoute(addr, vpnRoutes, localRoutes)
+			assert.Equal(t, tt.expectedVpn, isVpn, "isVpnRoute should return expectedVpn value")
+			assert.Equal(t, tt.expectedPrefix, matchedPrefix, "isVpnRoute should return expectedVpn prefix")
+		})
+	}
+}

client/internal/routemanager/systemops/systemops_linux_test.go

@@ -1,13 +1,10 @@
-//go:build !android
+//go:build linux && !android && privileged
 
 package systemops
 
 import (
 	"errors"
-	"fmt"
 	"net"
-	"os"
-	"strings"
 	"syscall"
 	"testing"
 
@@ -18,10 +15,6 @@ import (
 	"github.com/netbirdio/netbird/client/internal/routemanager/vars"
 )
 
-var expectedVPNint = "wgtest0"
-var expectedExternalInt = "dummyext0"
-var expectedInternalInt = "dummyint0"
-
 func init() {
 	testCases = append(testCases, []testCase{
 		{
@@ -33,62 +26,6 @@ func init() {
 	}...)
 }
 
-func TestEntryExists(t *testing.T) {
-	tempDir := t.TempDir()
-	tempFilePath := fmt.Sprintf("%s/rt_tables", tempDir)
-
-	content := []string{
-		"1000 reserved",
-		fmt.Sprintf("%d %s", NetbirdVPNTableID, NetbirdVPNTableName),
-		"9999 other_table",
-	}
-	require.NoError(t, os.WriteFile(tempFilePath, []byte(strings.Join(content, "\n")), 0644))
-
-	file, err := os.Open(tempFilePath)
-	require.NoError(t, err)
-	defer func() {
-		assert.NoError(t, file.Close())
-	}()
-
-	tests := []struct {
-		name        string
-		id          int
-		shouldExist bool
-		err         error
-	}{
-		{
-			name:        "ExistsWithNetbirdPrefix",
-			id:          7120,
-			shouldExist: true,
-			err:         nil,
-		},
-		{
-			name:        "ExistsWithDifferentName",
-			id:          1000,
-			shouldExist: true,
-			err:         ErrTableIDExists,
-		},
-		{
-			name:        "DoesNotExist",
-			id:          1234,
-			shouldExist: false,
-			err:         nil,
-		},
-	}
-
-	for _, tc := range tests {
-		t.Run(tc.name, func(t *testing.T) {
-			exists, err := entryExists(file, tc.id)
-			if tc.err != nil {
-				assert.ErrorIs(t, err, tc.err)
-			} else {
-				assert.NoError(t, err)
-			}
-			assert.Equal(t, tc.shouldExist, exists)
-		})
-	}
-}
-
 func createAndSetupDummyInterface(t *testing.T, interfaceName, ipAddressCIDR string) string {
 	t.Helper()
 

client/internal/routemanager/systemops/systemops_routing_data_linux_test.go

@@ -0,0 +1,15 @@
+//go:build linux && !android
+
+package systemops
+
+// Interface names used by the shared routing test fixtures. Kept untagged (no
+// privileged build tag) so the non-privileged test files in this package compile.
+//
+//nolint:unused // consumed by the privileged-tagged routing tests
+var expectedVPNint = "wgtest0"
+
+//nolint:unused // consumed by the privileged-tagged routing tests
+var expectedExternalInt = "dummyext0"
+
+//nolint:unused // consumed by the privileged-tagged routing tests
+var expectedInternalInt = "dummyint0"

client/internal/routemanager/systemops/systemops_routing_data_test.go

@@ -0,0 +1,83 @@
+//go:build (linux && !android) || (darwin && !ios) || freebsd || openbsd || netbsd || dragonfly
+
+package systemops
+
+import (
+	"net"
+
+	nbnet "github.com/netbirdio/netbird/client/net"
+)
+
+// Shared, non-privileged routing test fixtures. The privileged TestRouting (and its
+// per-platform init() appenders) consume these; they live here so the unprivileged
+// BSD/darwin test files compile without the privileged build tag.
+
+type PacketExpectation struct {
+	SrcIP   net.IP
+	DstIP   net.IP
+	SrcPort int
+	DstPort int
+	UDP     bool
+	TCP     bool
+}
+
+//nolint:unused // consumed by the privileged-tagged routing tests
+type testCase struct {
+	name              string
+	expectedInterface string
+	dialer            dialer
+	expectedPacket    PacketExpectation
+}
+
+//nolint:unused // consumed by the privileged-tagged routing tests
+var testCases = []testCase{
+	{
+		name:              "To external host without custom dialer via vpn",
+		expectedInterface: expectedVPNint,
+		dialer:            &net.Dialer{},
+		expectedPacket:    createPacketExpectation("100.64.0.1", 12345, "192.0.2.1", 53),
+	},
+	{
+		name:              "To external host with custom dialer via physical interface",
+		expectedInterface: expectedExternalInt,
+		dialer:            nbnet.NewDialer(),
+		expectedPacket:    createPacketExpectation("192.168.0.1", 12345, "192.0.2.1", 53),
+	},
+
+	{
+		name:              "To duplicate internal route with custom dialer via physical interface",
+		expectedInterface: expectedInternalInt,
+		dialer:            nbnet.NewDialer(),
+		expectedPacket:    createPacketExpectation("192.168.1.1", 12345, "10.0.0.2", 53),
+	},
+	{
+		name:              "To duplicate internal route without custom dialer via physical interface", // local route takes precedence
+		expectedInterface: expectedInternalInt,
+		dialer:            &net.Dialer{},
+		expectedPacket:    createPacketExpectation("192.168.1.1", 12345, "10.0.0.2", 53),
+	},
+
+	{
+		name:              "To unique vpn route with custom dialer via physical interface",
+		expectedInterface: expectedExternalInt,
+		dialer:            nbnet.NewDialer(),
+		expectedPacket:    createPacketExpectation("192.168.0.1", 12345, "172.16.0.2", 53),
+	},
+	{
+		name:              "To unique vpn route without custom dialer via vpn",
+		expectedInterface: expectedVPNint,
+		dialer:            &net.Dialer{},
+		expectedPacket:    createPacketExpectation("100.64.0.1", 12345, "172.16.0.2", 53),
+	},
+}
+
+//nolint:unused // consumed by the privileged-tagged routing tests
+func createPacketExpectation(srcIP string, srcPort int, dstIP string, dstPort int) PacketExpectation {
+	return PacketExpectation{
+		SrcIP:   net.ParseIP(srcIP),
+		DstIP:   net.ParseIP(dstIP),
+		SrcPort: srcPort,
+		DstPort: dstPort,
+		UDP:     true,
+	}
+}

client/internal/routemanager/systemops/systemops_unix_test.go

@@ -1,4 +1,4 @@
-//go:build (linux && !android) || (darwin && !ios) || freebsd || openbsd || netbsd || dragonfly
+//go:build ((linux && !android) || (darwin && !ios) || freebsd || openbsd || netbsd || dragonfly) && privileged
 
 package systemops
 
@@ -20,63 +20,6 @@ import (
 	nbnet "github.com/netbirdio/netbird/client/net"
 )
 
-type PacketExpectation struct {
-	SrcIP   net.IP
-	DstIP   net.IP
-	SrcPort int
-	DstPort int
-	UDP     bool
-	TCP     bool
-}
-
-type testCase struct {
-	name              string
-	expectedInterface string
-	dialer            dialer
-	expectedPacket    PacketExpectation
-}
-
-var testCases = []testCase{
-	{
-		name:              "To external host without custom dialer via vpn",
-		expectedInterface: expectedVPNint,
-		dialer:            &net.Dialer{},
-		expectedPacket:    createPacketExpectation("100.64.0.1", 12345, "192.0.2.1", 53),
-	},
-	{
-		name:              "To external host with custom dialer via physical interface",
-		expectedInterface: expectedExternalInt,
-		dialer:            nbnet.NewDialer(),
-		expectedPacket:    createPacketExpectation("192.168.0.1", 12345, "192.0.2.1", 53),
-	},
-
-	{
-		name:              "To duplicate internal route with custom dialer via physical interface",
-		expectedInterface: expectedInternalInt,
-		dialer:            nbnet.NewDialer(),
-		expectedPacket:    createPacketExpectation("192.168.1.1", 12345, "10.0.0.2", 53),
-	},
-	{
-		name:              "To duplicate internal route without custom dialer via physical interface", // local route takes precedence
-		expectedInterface: expectedInternalInt,
-		dialer:            &net.Dialer{},
-		expectedPacket:    createPacketExpectation("192.168.1.1", 12345, "10.0.0.2", 53),
-	},
-
-	{
-		name:              "To unique vpn route with custom dialer via physical interface",
-		expectedInterface: expectedExternalInt,
-		dialer:            nbnet.NewDialer(),
-		expectedPacket:    createPacketExpectation("192.168.0.1", 12345, "172.16.0.2", 53),
-	},
-	{
-		name:              "To unique vpn route without custom dialer via vpn",
-		expectedInterface: expectedVPNint,
-		dialer:            &net.Dialer{},
-		expectedPacket:    createPacketExpectation("100.64.0.1", 12345, "172.16.0.2", 53),
-	},
-}
-
 func TestRouting(t *testing.T) {
 	nbnet.Init()
 	for _, tc := range testCases {
@@ -102,16 +45,6 @@ func TestRouting(t *testing.T) {
 	}
 }
 
-func createPacketExpectation(srcIP string, srcPort int, dstIP string, dstPort int) PacketExpectation {
-	return PacketExpectation{
-		SrcIP:   net.ParseIP(srcIP),
-		DstIP:   net.ParseIP(dstIP),
-		SrcPort: srcPort,
-		DstPort: dstPort,
-		UDP:     true,
-	}
-}
-
 func startPacketCapture(t *testing.T, intf, filter string) *pcap.Handle {
 	t.Helper()
 

client/internal/routemanager/systemops/systemops_windows_test.go

@@ -1,3 +1,5 @@
+//go:build windows && privileged
+
 package systemops
 
 import (

client/internal/routemanager/systemops/v6route_bsd_test.go

@@ -11,6 +11,8 @@ import (
 // ensureIPv6DefaultRoute installs an IPv6 default route via the loopback
 // interface so route lookups for global IPv6 prefixes resolve in environments
 // without v6 connectivity. If a default already exists it is left alone.
+//
+//nolint:unused // consumed by the privileged-tagged routing tests
 func ensureIPv6DefaultRoute(t *testing.T) {
 	t.Helper()
 

client/internal/routemanager/systemops/v6route_linux_test.go

@@ -1,4 +1,4 @@
-//go:build linux && !android
+//go:build linux && !android && privileged
 
 package systemops
 

client/internal/routemanager/systemops/v6route_windows_test.go

@@ -8,11 +8,14 @@ import (
 	"testing"
 )
 
+//nolint:unused // consumed by the privileged-tagged routing tests
 const loopbackIfaceWindows = "Loopback Pseudo-Interface 1"
 
 // ensureIPv6DefaultRoute installs an IPv6 default route via the loopback
 // interface so route lookups for global IPv6 prefixes resolve in environments
 // without v6 connectivity. If a default already exists it is left alone.
+//
+//nolint:unused // consumed by the privileged-tagged routing tests
 func ensureIPv6DefaultRoute(t *testing.T) {
 	t.Helper()
 

client/server/server_privileged_test.go

@@ -0,0 +1,235 @@
+//go:build privileged
+
+package server
+
+import (
+	"context"
+	"net"
+	"os/user"
+	"testing"
+	"time"
+
+	"github.com/golang/mock/gomock"
+	"github.com/stretchr/testify/require"
+	"go.opentelemetry.io/otel"
+
+	"github.com/netbirdio/netbird/management/server/integrations/integrated_validator/validator"
+
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/controller"
+	"github.com/netbirdio/netbird/management/internals/controllers/network_map/update_channel"
+	"github.com/netbirdio/netbird/management/internals/modules/peers"
+	"github.com/netbirdio/netbird/management/internals/modules/peers/ephemeral/manager"
+	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
+	"github.com/netbirdio/netbird/management/server/job"
+
+	"github.com/netbirdio/netbird/management/internals/server/config"
+	"github.com/netbirdio/netbird/management/server/groups"
+
+	log "github.com/sirupsen/logrus"
+	"google.golang.org/grpc"
+	"google.golang.org/grpc/keepalive"
+
+	"github.com/netbirdio/netbird/client/internal"
+	"github.com/netbirdio/netbird/client/internal/peer"
+	"github.com/netbirdio/netbird/client/internal/profilemanager"
+	"github.com/netbirdio/netbird/management/server"
+	"github.com/netbirdio/netbird/management/server/activity"
+	nbcache "github.com/netbirdio/netbird/management/server/cache"
+	"github.com/netbirdio/netbird/management/server/integrations/port_forwarding"
+	"github.com/netbirdio/netbird/management/server/permissions"
+	"github.com/netbirdio/netbird/management/server/settings"
+	"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/signal/proto"
+	signalServer "github.com/netbirdio/netbird/signal/server"
+)
+
+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,
+	}
+)
+
+// TestConnectWithRetryRuns checks that the connectWithRetry function runs and runs the retries according to the times specified via environment variables
+// we will use a management server started via to simulate the server and capture the number of retries
+func TestConnectWithRetryRuns(t *testing.T) {
+	// start the signal server
+	_, signalAddr, err := startSignal(t)
+	if err != nil {
+		t.Fatalf("failed to start signal server: %v", err)
+	}
+
+	counter := 0
+	// start the management server
+	_, mgmtAddr, err := startManagement(t, signalAddr, &counter)
+	if err != nil {
+		t.Fatalf("failed to start management server: %v", err)
+	}
+
+	ctx := internal.CtxInitState(context.Background())
+
+	ctx, cancel := context.WithDeadline(ctx, time.Now().Add(30*time.Second))
+	defer cancel()
+	// create new server
+	ic := profilemanager.ConfigInput{
+		ManagementURL: "http://" + mgmtAddr,
+		ConfigPath:    t.TempDir() + "/test-profile.json",
+	}
+
+	config, err := profilemanager.UpdateOrCreateConfig(ic)
+	if err != nil {
+		t.Fatalf("failed to create config: %v", err)
+	}
+
+	currUser, err := user.Current()
+	require.NoError(t, err)
+
+	pm := profilemanager.ServiceManager{}
+	err = pm.SetActiveProfileState(&profilemanager.ActiveProfileState{
+		ID:       "test-profile",
+		Username: currUser.Username,
+	})
+	if err != nil {
+		t.Fatalf("failed to set active profile state: %v", err)
+	}
+
+	s := New(ctx, "debug", "", false, false, false, false)
+
+	s.config = config
+
+	s.statusRecorder = peer.NewRecorder(config.ManagementURL.String())
+	t.Setenv(retryInitialIntervalVar, "1s")
+	t.Setenv(maxRetryIntervalVar, "2s")
+	t.Setenv(maxRetryTimeVar, "5s")
+	t.Setenv(retryMultiplierVar, "1")
+
+	s.connectWithRetryRuns(ctx, config, s.statusRecorder, nil, nil)
+	if counter < 3 {
+		t.Fatalf("expected counter > 2, got %d", counter)
+	}
+}
+
+type mockServer struct {
+	mgmtProto.ManagementServiceServer
+	counter *int
+}
+
+func (m *mockServer) Login(ctx context.Context, req *mgmtProto.EncryptedMessage) (*mgmtProto.EncryptedMessage, error) {
+	*m.counter++
+	return m.ManagementServiceServer.Login(ctx, req)
+}
+
+func startManagement(t *testing.T, signalAddr string, counter *int) (*grpc.Server, string, error) {
+	t.Helper()
+	dataDir := t.TempDir()
+
+	config := &config.Config{
+		Stuns:      []*config.Host{},
+		TURNConfig: &config.TURNConfig{},
+		Signal: &config.Host{
+			Proto: "http",
+			URI:   signalAddr,
+		},
+		Datadir:    dataDir,
+		HttpConfig: nil,
+	}
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	if err != nil {
+		return nil, "", err
+	}
+	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
+	store, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), "", config.Datadir)
+	if err != nil {
+		return nil, "", err
+	}
+	t.Cleanup(cleanUp)
+
+	eventStore := &activity.InMemoryEventStore{}
+	if err != nil {
+		return nil, "", err
+	}
+
+	ctrl := gomock.NewController(t)
+	t.Cleanup(ctrl.Finish)
+
+	permissionsManagerMock := permissions.NewMockManager(ctrl)
+	peersManager := peers.NewManager(store, permissionsManagerMock)
+	settingsManagerMock := settings.NewMockManager(ctrl)
+
+	jobManager := job.NewJobManager(nil, store, peersManager)
+
+	cacheStore, err := nbcache.NewStore(context.Background(), 100*time.Millisecond, 300*time.Millisecond, 100)
+	if err != nil {
+		return nil, "", err
+	}
+
+	ia, _ := validator.NewIntegratedValidator(context.Background(), peersManager, settingsManagerMock, eventStore, cacheStore)
+
+	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
+	require.NoError(t, err)
+
+	settingsMockManager := settings.NewMockManager(ctrl)
+	groupsManager := groups.NewManagerMock()
+
+	requestBuffer := server.NewAccountRequestBuffer(context.Background(), store)
+	peersUpdateManager := update_channel.NewPeersUpdateManager(metrics)
+	networkMapController := controller.NewController(context.Background(), store, metrics, peersUpdateManager, requestBuffer, server.MockIntegratedValidator{}, settingsMockManager, "netbird.selfhosted", port_forwarding.NewControllerMock(), manager.NewEphemeralManager(store, peersManager), config)
+	accountManager, err := server.BuildManager(context.Background(), config, store, networkMapController, jobManager, nil, "", eventStore, nil, false, ia, metrics, port_forwarding.NewControllerMock(), settingsMockManager, permissionsManagerMock, false, cacheStore)
+	if err != nil {
+		return nil, "", err
+	}
+
+	secretsManager, err := nbgrpc.NewTimeBasedAuthSecretsManager(peersUpdateManager, config.TURNConfig, config.Relay, settingsMockManager, groupsManager)
+	if err != nil {
+		return nil, "", err
+	}
+	mgmtServer, err := nbgrpc.NewServer(config, accountManager, settingsMockManager, jobManager, secretsManager, nil, nil, &server.MockIntegratedValidator{}, networkMapController, nil, nil)
+	if err != nil {
+		return nil, "", err
+	}
+	mock := &mockServer{
+		ManagementServiceServer: mgmtServer,
+		counter:                 counter,
+	}
+	mgmtProto.RegisterManagementServiceServer(s, mock)
+	go func() {
+		if err = s.Serve(lis); err != nil {
+			log.Fatalf("failed to serve: %v", err)
+		}
+	}()
+
+	return s, lis.Addr().String(), nil
+}
+
+func startSignal(t *testing.T) (*grpc.Server, string, error) {
+	t.Helper()
+
+	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
+
+	lis, err := net.Listen("tcp", "localhost:0")
+	if err != nil {
+		return nil, "", err
+	}
+
+	srv, err := signalServer.NewServer(context.Background(), otel.Meter(""))
+	require.NoError(t, err)
+	proto.RegisterSignalExchangeServer(s, srv)
+
+	go func() {
+		if err = s.Serve(lis); err != nil {
+			log.Fatalf("failed to serve: %v", err)
+		}
+	}()
+
+	return s, lis.Addr().String(), nil
+}

client/server/server_test.go

@@ -2,124 +2,22 @@ package server
 
 import (
 	"context"
-	"net"
 	"net/url"
 	"os/user"
 	"path/filepath"
 	"testing"
 	"time"
 
-	"github.com/golang/mock/gomock"
-	"github.com/stretchr/testify/require"
-	"go.opentelemetry.io/otel"
-
-	"github.com/netbirdio/netbird/management/server/integrations/integrated_validator/validator"
-
-	"github.com/netbirdio/netbird/management/internals/controllers/network_map/controller"
-	"github.com/netbirdio/netbird/management/internals/controllers/network_map/update_channel"
-	"github.com/netbirdio/netbird/management/internals/modules/peers"
-	"github.com/netbirdio/netbird/management/internals/modules/peers/ephemeral/manager"
-	nbgrpc "github.com/netbirdio/netbird/management/internals/shared/grpc"
-	"github.com/netbirdio/netbird/management/server/job"
-
-	"github.com/netbirdio/netbird/management/internals/server/config"
-	"github.com/netbirdio/netbird/management/server/groups"
-
 	log "github.com/sirupsen/logrus"
 	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
 	"google.golang.org/grpc"
-	"google.golang.org/grpc/keepalive"
 
 	"github.com/netbirdio/netbird/client/internal"
-	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/internal/profilemanager"
 	daemonProto "github.com/netbirdio/netbird/client/proto"
-	"github.com/netbirdio/netbird/management/server"
-	"github.com/netbirdio/netbird/management/server/activity"
-	nbcache "github.com/netbirdio/netbird/management/server/cache"
-	"github.com/netbirdio/netbird/management/server/integrations/port_forwarding"
-	"github.com/netbirdio/netbird/management/server/permissions"
-	"github.com/netbirdio/netbird/management/server/settings"
-	"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/signal/proto"
-	signalServer "github.com/netbirdio/netbird/signal/server"
 )
 
-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,
-	}
-)
-
-// TestConnectWithRetryRuns checks that the connectWithRetry function runs and runs the retries according to the times specified via environment variables
-// we will use a management server started via to simulate the server and capture the number of retries
-func TestConnectWithRetryRuns(t *testing.T) {
-	// start the signal server
-	_, signalAddr, err := startSignal(t)
-	if err != nil {
-		t.Fatalf("failed to start signal server: %v", err)
-	}
-
-	counter := 0
-	// start the management server
-	_, mgmtAddr, err := startManagement(t, signalAddr, &counter)
-	if err != nil {
-		t.Fatalf("failed to start management server: %v", err)
-	}
-
-	ctx := internal.CtxInitState(context.Background())
-
-	ctx, cancel := context.WithDeadline(ctx, time.Now().Add(30*time.Second))
-	defer cancel()
-	// create new server
-	ic := profilemanager.ConfigInput{
-		ManagementURL: "http://" + mgmtAddr,
-		ConfigPath:    t.TempDir() + "/test-profile.json",
-	}
-
-	config, err := profilemanager.UpdateOrCreateConfig(ic)
-	if err != nil {
-		t.Fatalf("failed to create config: %v", err)
-	}
-
-	currUser, err := user.Current()
-	require.NoError(t, err)
-
-	pm := profilemanager.ServiceManager{}
-	err = pm.SetActiveProfileState(&profilemanager.ActiveProfileState{
-		ID:       "test-profile",
-		Username: currUser.Username,
-	})
-	if err != nil {
-		t.Fatalf("failed to set active profile state: %v", err)
-	}
-
-	s := New(ctx, "debug", "", false, false, false, false)
-
-	s.config = config
-
-	s.statusRecorder = peer.NewRecorder(config.ManagementURL.String())
-	t.Setenv(retryInitialIntervalVar, "1s")
-	t.Setenv(maxRetryIntervalVar, "2s")
-	t.Setenv(maxRetryTimeVar, "5s")
-	t.Setenv(retryMultiplierVar, "1")
-
-	s.connectWithRetryRuns(ctx, config, s.statusRecorder, nil, nil)
-	if counter < 3 {
-		t.Fatalf("expected counter > 2, got %d", counter)
-	}
-}
-
 func TestServer_Up(t *testing.T) {
 	tempDir := t.TempDir()
 	origDefaultProfileDir := profilemanager.DefaultConfigPathDir
@@ -259,119 +157,3 @@ func TestServer_SubcribeEvents(t *testing.T) {
 
 	assert.NoError(t, err)
 }
-
-type mockServer struct {
-	mgmtProto.ManagementServiceServer
-	counter *int
-}
-
-func (m *mockServer) Login(ctx context.Context, req *mgmtProto.EncryptedMessage) (*mgmtProto.EncryptedMessage, error) {
-	*m.counter++
-	return m.ManagementServiceServer.Login(ctx, req)
-}
-
-func startManagement(t *testing.T, signalAddr string, counter *int) (*grpc.Server, string, error) {
-	t.Helper()
-	dataDir := t.TempDir()
-
-	config := &config.Config{
-		Stuns:      []*config.Host{},
-		TURNConfig: &config.TURNConfig{},
-		Signal: &config.Host{
-			Proto: "http",
-			URI:   signalAddr,
-		},
-		Datadir:    dataDir,
-		HttpConfig: nil,
-	}
-
-	lis, err := net.Listen("tcp", "localhost:0")
-	if err != nil {
-		return nil, "", err
-	}
-	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
-	store, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), "", config.Datadir)
-	if err != nil {
-		return nil, "", err
-	}
-	t.Cleanup(cleanUp)
-
-	eventStore := &activity.InMemoryEventStore{}
-	if err != nil {
-		return nil, "", err
-	}
-
-	ctrl := gomock.NewController(t)
-	t.Cleanup(ctrl.Finish)
-
-	permissionsManagerMock := permissions.NewMockManager(ctrl)
-	peersManager := peers.NewManager(store, permissionsManagerMock)
-	settingsManagerMock := settings.NewMockManager(ctrl)
-
-	jobManager := job.NewJobManager(nil, store, peersManager)
-
-	cacheStore, err := nbcache.NewStore(context.Background(), 100*time.Millisecond, 300*time.Millisecond, 100)
-	if err != nil {
-		return nil, "", err
-	}
-
-	ia, _ := validator.NewIntegratedValidator(context.Background(), peersManager, settingsManagerMock, eventStore, cacheStore)
-
-	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
-	require.NoError(t, err)
-
-	settingsMockManager := settings.NewMockManager(ctrl)
-	groupsManager := groups.NewManagerMock()
-
-	requestBuffer := server.NewAccountRequestBuffer(context.Background(), store)
-	peersUpdateManager := update_channel.NewPeersUpdateManager(metrics)
-	networkMapController := controller.NewController(context.Background(), store, metrics, peersUpdateManager, requestBuffer, server.MockIntegratedValidator{}, settingsMockManager, "netbird.selfhosted", port_forwarding.NewControllerMock(), manager.NewEphemeralManager(store, peersManager), config)
-	accountManager, err := server.BuildManager(context.Background(), config, store, networkMapController, jobManager, nil, "", eventStore, nil, false, ia, metrics, port_forwarding.NewControllerMock(), settingsMockManager, permissionsManagerMock, false, cacheStore)
-	if err != nil {
-		return nil, "", err
-	}
-
-	secretsManager, err := nbgrpc.NewTimeBasedAuthSecretsManager(peersUpdateManager, config.TURNConfig, config.Relay, settingsMockManager, groupsManager)
-	if err != nil {
-		return nil, "", err
-	}
-	mgmtServer, err := nbgrpc.NewServer(config, accountManager, settingsMockManager, jobManager, secretsManager, nil, nil, &server.MockIntegratedValidator{}, networkMapController, nil, nil)
-	if err != nil {
-		return nil, "", err
-	}
-	mock := &mockServer{
-		ManagementServiceServer: mgmtServer,
-		counter:                 counter,
-	}
-	mgmtProto.RegisterManagementServiceServer(s, mock)
-	go func() {
-		if err = s.Serve(lis); err != nil {
-			log.Fatalf("failed to serve: %v", err)
-		}
-	}()
-
-	return s, lis.Addr().String(), nil
-}
-
-func startSignal(t *testing.T) (*grpc.Server, string, error) {
-	t.Helper()
-
-	s := grpc.NewServer(grpc.KeepaliveEnforcementPolicy(kaep), grpc.KeepaliveParams(kasp))
-
-	lis, err := net.Listen("tcp", "localhost:0")
-	if err != nil {
-		log.Fatalf("failed to listen: %v", err)
-	}
-
-	srv, err := signalServer.NewServer(context.Background(), otel.Meter(""))
-	require.NoError(t, err)
-	proto.RegisterSignalExchangeServer(s, srv)
-
-	go func() {
-		if err = s.Serve(lis); err != nil {
-			log.Fatalf("failed to serve: %v", err)
-		}
-	}()
-
-	return s, lis.Addr().String(), nil
-}

client/ssh/client/client_privileged_test.go

@@ -0,0 +1,118 @@
+//go:build privileged
+
+package client
+
+import (
+	"context"
+	"errors"
+	"runtime"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	cryptossh "golang.org/x/crypto/ssh"
+
+	"github.com/netbirdio/netbird/client/ssh/testutil"
+)
+
+func TestSSHClient_CommandExecution(t *testing.T) {
+	if runtime.GOOS == "windows" && testutil.IsCI() {
+		t.Skip("Skipping Windows command execution tests in CI due to S4U authentication issues")
+	}
+
+	server, _, client := setupTestSSHServerAndClient(t)
+	defer func() {
+		err := server.Stop()
+		require.NoError(t, err)
+	}()
+	defer func() {
+		err := client.Close()
+		assert.NoError(t, err)
+	}()
+
+	ctx, cancel := context.WithTimeout(context.Background(), 3*time.Second)
+	defer cancel()
+
+	t.Run("ExecuteCommand captures output", func(t *testing.T) {
+		output, err := client.ExecuteCommand(ctx, "echo hello")
+		assert.NoError(t, err)
+		assert.Contains(t, string(output), "hello")
+	})
+
+	t.Run("ExecuteCommandWithIO streams output", func(t *testing.T) {
+		err := client.ExecuteCommandWithIO(ctx, "echo world")
+		assert.NoError(t, err)
+	})
+
+	t.Run("commands with flags work", func(t *testing.T) {
+		output, err := client.ExecuteCommand(ctx, "echo -n test_flag")
+		assert.NoError(t, err)
+		assert.Equal(t, "test_flag", strings.TrimSpace(string(output)))
+	})
+
+	t.Run("non-zero exit codes don't return errors", func(t *testing.T) {
+		var testCmd string
+		if runtime.GOOS == "windows" {
+			testCmd = "echo hello | Select-String notfound"
+		} else {
+			testCmd = "echo 'hello' | grep 'notfound'"
+		}
+		_, err := client.ExecuteCommand(ctx, testCmd)
+		assert.NoError(t, err)
+	})
+}
+
+func TestSSHClient_ContextCancellation(t *testing.T) {
+	server, serverAddr, _ := setupTestSSHServerAndClient(t)
+	defer func() {
+		err := server.Stop()
+		require.NoError(t, err)
+	}()
+
+	t.Run("connection with short timeout", func(t *testing.T) {
+		ctx, cancel := context.WithTimeout(context.Background(), 1*time.Millisecond)
+		defer cancel()
+
+		currentUser := testutil.GetTestUsername(t)
+		_, err := Dial(ctx, serverAddr, currentUser, DialOptions{
+			InsecureSkipVerify: true,
+		})
+		if err != nil {
+			// Check for actual timeout-related errors rather than string matching
+			assert.True(t,
+				errors.Is(err, context.DeadlineExceeded) ||
+					errors.Is(err, context.Canceled) ||
+					strings.Contains(err.Error(), "timeout"),
+				"Expected timeout-related error, got: %v", err)
+		}
+	})
+
+	t.Run("command execution cancellation", func(t *testing.T) {
+		ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+		defer cancel()
+		currentUser := testutil.GetTestUsername(t)
+		client, err := Dial(ctx, serverAddr, currentUser, DialOptions{
+			InsecureSkipVerify: true,
+		})
+		require.NoError(t, err)
+		defer func() {
+			if err := client.Close(); err != nil {
+				t.Logf("client close error: %v", err)
+			}
+		}()
+
+		cmdCtx, cmdCancel := context.WithTimeout(context.Background(), 100*time.Millisecond)
+		defer cmdCancel()
+
+		err = client.ExecuteCommandWithPTY(cmdCtx, "sleep 10")
+		if err != nil {
+			var exitMissingErr *cryptossh.ExitMissingError
+			isValidCancellation := errors.Is(err, context.DeadlineExceeded) ||
+				errors.Is(err, context.Canceled) ||
+				errors.As(err, &exitMissingErr)
+			assert.True(t, isValidCancellation, "Should handle command cancellation properly")
+		}
+	})
+}

client/ssh/client/client_test.go

@@ -15,7 +15,6 @@ import (
 
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
-	cryptossh "golang.org/x/crypto/ssh"
 
 	"github.com/netbirdio/netbird/client/ssh"
 	sshserver "github.com/netbirdio/netbird/client/ssh/server"
@@ -78,53 +77,6 @@ func TestSSHClient_DialWithKey(t *testing.T) {
 	assert.NotNil(t, client.client)
 }
 
-func TestSSHClient_CommandExecution(t *testing.T) {
-	if runtime.GOOS == "windows" && testutil.IsCI() {
-		t.Skip("Skipping Windows command execution tests in CI due to S4U authentication issues")
-	}
-
-	server, _, client := setupTestSSHServerAndClient(t)
-	defer func() {
-		err := server.Stop()
-		require.NoError(t, err)
-	}()
-	defer func() {
-		err := client.Close()
-		assert.NoError(t, err)
-	}()
-
-	ctx, cancel := context.WithTimeout(context.Background(), 3*time.Second)
-	defer cancel()
-
-	t.Run("ExecuteCommand captures output", func(t *testing.T) {
-		output, err := client.ExecuteCommand(ctx, "echo hello")
-		assert.NoError(t, err)
-		assert.Contains(t, string(output), "hello")
-	})
-
-	t.Run("ExecuteCommandWithIO streams output", func(t *testing.T) {
-		err := client.ExecuteCommandWithIO(ctx, "echo world")
-		assert.NoError(t, err)
-	})
-
-	t.Run("commands with flags work", func(t *testing.T) {
-		output, err := client.ExecuteCommand(ctx, "echo -n test_flag")
-		assert.NoError(t, err)
-		assert.Equal(t, "test_flag", strings.TrimSpace(string(output)))
-	})
-
-	t.Run("non-zero exit codes don't return errors", func(t *testing.T) {
-		var testCmd string
-		if runtime.GOOS == "windows" {
-			testCmd = "echo hello | Select-String notfound"
-		} else {
-			testCmd = "echo 'hello' | grep 'notfound'"
-		}
-		_, err := client.ExecuteCommand(ctx, testCmd)
-		assert.NoError(t, err)
-	})
-}
-
 func TestSSHClient_ConnectionHandling(t *testing.T) {
 	server, serverAddr, _ := setupTestSSHServerAndClient(t)
 	defer func() {
@@ -154,59 +106,6 @@ func TestSSHClient_ConnectionHandling(t *testing.T) {
 	}
 }
 
-func TestSSHClient_ContextCancellation(t *testing.T) {
-	server, serverAddr, _ := setupTestSSHServerAndClient(t)
-	defer func() {
-		err := server.Stop()
-		require.NoError(t, err)
-	}()
-
-	t.Run("connection with short timeout", func(t *testing.T) {
-		ctx, cancel := context.WithTimeout(context.Background(), 1*time.Millisecond)
-		defer cancel()
-
-		currentUser := testutil.GetTestUsername(t)
-		_, err := Dial(ctx, serverAddr, currentUser, DialOptions{
-			InsecureSkipVerify: true,
-		})
-		if err != nil {
-			// Check for actual timeout-related errors rather than string matching
-			assert.True(t,
-				errors.Is(err, context.DeadlineExceeded) ||
-					errors.Is(err, context.Canceled) ||
-					strings.Contains(err.Error(), "timeout"),
-				"Expected timeout-related error, got: %v", err)
-		}
-	})
-
-	t.Run("command execution cancellation", func(t *testing.T) {
-		ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
-		defer cancel()
-		currentUser := testutil.GetTestUsername(t)
-		client, err := Dial(ctx, serverAddr, currentUser, DialOptions{
-			InsecureSkipVerify: true,
-		})
-		require.NoError(t, err)
-		defer func() {
-			if err := client.Close(); err != nil {
-				t.Logf("client close error: %v", err)
-			}
-		}()
-
-		cmdCtx, cmdCancel := context.WithTimeout(context.Background(), 100*time.Millisecond)
-		defer cmdCancel()
-
-		err = client.ExecuteCommandWithPTY(cmdCtx, "sleep 10")
-		if err != nil {
-			var exitMissingErr *cryptossh.ExitMissingError
-			isValidCancellation := errors.Is(err, context.DeadlineExceeded) ||
-				errors.Is(err, context.Canceled) ||
-				errors.As(err, &exitMissingErr)
-			assert.True(t, isValidCancellation, "Should handle command cancellation properly")
-		}
-	})
-}
-
 func TestSSHClient_NoAuthMode(t *testing.T) {
 	hostKey, err := ssh.GeneratePrivateKey(ssh.ED25519)
 	require.NoError(t, err)

client/ssh/proxy/proxy_privileged_test.go

@@ -0,0 +1,423 @@
+//go:build privileged
+
+package proxy
+
+import (
+	"bytes"
+	"context"
+	"crypto/rand"
+	"crypto/rsa"
+	"encoding/base64"
+	"encoding/json"
+	"io"
+	"math/big"
+	"net"
+	"net/http"
+	"net/http/httptest"
+	"os"
+	"runtime"
+	"strconv"
+	"testing"
+	"time"
+
+	"github.com/golang-jwt/jwt/v5"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	cryptossh "golang.org/x/crypto/ssh"
+
+	nbssh "github.com/netbirdio/netbird/client/ssh"
+	sshauth "github.com/netbirdio/netbird/client/ssh/auth"
+	"github.com/netbirdio/netbird/client/ssh/server"
+	"github.com/netbirdio/netbird/client/ssh/testutil"
+	nbjwt "github.com/netbirdio/netbird/shared/auth/jwt"
+	sshuserhash "github.com/netbirdio/netbird/shared/sshauth"
+)
+
+func (m *mockDaemon) setJWTToken(token string) {
+	m.impl.jwtToken = token
+}
+
+func TestSSHProxy_Connect(t *testing.T) {
+	if testing.Short() {
+		t.Skip("Skipping integration test in short mode")
+	}
+
+	// TODO: Windows test times out - user switching and command execution tested on Linux
+	if runtime.GOOS == "windows" {
+		t.Skip("Skipping on Windows - covered by Linux tests")
+	}
+
+	const (
+		issuer   = "https://test-issuer.example.com"
+		audience = "test-audience"
+	)
+
+	jwksServer, privateKey, jwksURL := setupJWKSServer(t)
+	defer jwksServer.Close()
+
+	hostKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
+	require.NoError(t, err)
+	hostPubKey, err := nbssh.GeneratePublicKey(hostKey)
+	require.NoError(t, err)
+
+	serverConfig := &server.Config{
+		HostKeyPEM: hostKey,
+		JWT: &server.JWTConfig{
+			Issuer:       issuer,
+			Audiences:    []string{audience},
+			KeysLocation: jwksURL,
+		},
+	}
+	sshServer := server.New(serverConfig)
+	sshServer.SetAllowRootLogin(true)
+
+	// Configure SSH authorization for the test user
+	testUsername := testutil.GetTestUsername(t)
+	testJWTUser := "test-username"
+	testUserHash, err := sshuserhash.HashUserID(testJWTUser)
+	require.NoError(t, err)
+
+	authConfig := &sshauth.Config{
+		UserIDClaim:     sshauth.DefaultUserIDClaim,
+		AuthorizedUsers: []sshuserhash.UserIDHash{testUserHash},
+		MachineUsers: map[string][]uint32{
+			testUsername: {0}, // Index 0 in AuthorizedUsers
+		},
+	}
+	sshServer.UpdateSSHAuth(authConfig)
+
+	sshServerAddr := server.StartTestServer(t, sshServer)
+	defer func() { _ = sshServer.Stop() }()
+
+	mockDaemon := startMockDaemon(t)
+	defer mockDaemon.stop()
+
+	host, portStr, err := net.SplitHostPort(sshServerAddr)
+	require.NoError(t, err)
+	port, err := strconv.Atoi(portStr)
+	require.NoError(t, err)
+
+	mockDaemon.setHostKey(host, hostPubKey)
+
+	validToken := generateValidJWT(t, privateKey, issuer, audience, testJWTUser)
+	mockDaemon.setJWTToken(validToken)
+
+	proxyInstance, err := New(mockDaemon.addr, host, port, io.Discard, nil)
+	require.NoError(t, err)
+
+	clientConn, proxyConn := net.Pipe()
+	defer func() { _ = clientConn.Close() }()
+
+	origStdin := os.Stdin
+	origStdout := os.Stdout
+	defer func() {
+		os.Stdin = origStdin
+		os.Stdout = origStdout
+	}()
+
+	stdinReader, stdinWriter, err := os.Pipe()
+	require.NoError(t, err)
+	stdoutReader, stdoutWriter, err := os.Pipe()
+	require.NoError(t, err)
+
+	os.Stdin = stdinReader
+	os.Stdout = stdoutWriter
+
+	go func() {
+		_, _ = io.Copy(stdinWriter, proxyConn)
+	}()
+	go func() {
+		_, _ = io.Copy(proxyConn, stdoutReader)
+	}()
+
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	connectErrCh := make(chan error, 1)
+	go func() {
+		connectErrCh <- proxyInstance.Connect(ctx)
+	}()
+
+	sshConfig := &cryptossh.ClientConfig{
+		User:            testutil.GetTestUsername(t),
+		Auth:            []cryptossh.AuthMethod{},
+		HostKeyCallback: cryptossh.InsecureIgnoreHostKey(),
+		Timeout:         3 * time.Second,
+	}
+
+	sshClientConn, chans, reqs, err := cryptossh.NewClientConn(clientConn, "test", sshConfig)
+	require.NoError(t, err, "Should connect to proxy server")
+	defer func() { _ = sshClientConn.Close() }()
+
+	sshClient := cryptossh.NewClient(sshClientConn, chans, reqs)
+
+	session, err := sshClient.NewSession()
+	require.NoError(t, err, "Should create session through full proxy to backend")
+
+	outputCh := make(chan []byte, 1)
+	errCh := make(chan error, 1)
+	go func() {
+		output, err := session.Output("echo hello-from-proxy")
+		outputCh <- output
+		errCh <- err
+	}()
+
+	select {
+	case output := <-outputCh:
+		err := <-errCh
+		require.NoError(t, err, "Command should execute successfully through proxy")
+		assert.Contains(t, string(output), "hello-from-proxy", "Should receive command output through proxy")
+	case <-time.After(3 * time.Second):
+		t.Fatal("Command execution timed out")
+	}
+
+	_ = session.Close()
+	_ = sshClient.Close()
+	_ = clientConn.Close()
+	cancel()
+}
+
+// TestSSHProxy_CommandQuoting verifies that the proxy preserves shell quoting
+// when forwarding commands to the backend. This is critical for tools like
+// Ansible that send commands such as:
+//
+//	/bin/sh -c '( umask 77 && mkdir -p ... ) && sleep 0'
+//
+// The single quotes must be preserved so the backend shell receives the
+// subshell expression as a single argument to -c.
+func TestSSHProxy_CommandQuoting(t *testing.T) {
+	if testing.Short() {
+		t.Skip("Skipping integration test in short mode")
+	}
+
+	sshClient, cleanup := setupProxySSHClient(t)
+	defer cleanup()
+
+	// These commands simulate what the SSH protocol delivers as exec payloads.
+	// When a user types: ssh host '/bin/sh -c "( echo hello )"'
+	// the local shell strips the outer single quotes, and the SSH exec request
+	// contains the raw string: /bin/sh -c "( echo hello )"
+	//
+	// The proxy must forward this string verbatim. Using session.Command()
+	// (shlex.Split + strings.Join) strips the inner double quotes, breaking
+	// the command on the backend.
+	tests := []struct {
+		name    string
+		command string
+		expect  string
+	}{
+		{
+			name:    "subshell_in_double_quotes",
+			command: `/bin/sh -c "( echo from-subshell ) && echo outer"`,
+			expect:  "from-subshell\nouter\n",
+		},
+		{
+			name:    "printf_with_special_chars",
+			command: `/bin/sh -c "printf '%s\n' 'hello world'"`,
+			expect:  "hello world\n",
+		},
+		{
+			name:    "nested_command_substitution",
+			command: `/bin/sh -c "echo $(echo nested)"`,
+			expect:  "nested\n",
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			session, err := sshClient.NewSession()
+			require.NoError(t, err)
+			defer func() { _ = session.Close() }()
+
+			var stderrBuf bytes.Buffer
+			session.Stderr = &stderrBuf
+
+			outputCh := make(chan []byte, 1)
+			errCh := make(chan error, 1)
+			go func() {
+				output, err := session.Output(tc.command)
+				outputCh <- output
+				errCh <- err
+			}()
+
+			select {
+			case output := <-outputCh:
+				err := <-errCh
+				if stderrBuf.Len() > 0 {
+					t.Logf("stderr: %s", stderrBuf.String())
+				}
+				require.NoError(t, err, "command should succeed: %s", tc.command)
+				assert.Equal(t, tc.expect, string(output), "output mismatch for: %s", tc.command)
+			case <-time.After(5 * time.Second):
+				t.Fatalf("command timed out: %s", tc.command)
+			}
+		})
+	}
+}
+
+// setupProxySSHClient creates a full proxy test environment and returns
+// an SSH client connected through the proxy to a backend NetBird SSH server.
+func setupProxySSHClient(t *testing.T) (*cryptossh.Client, func()) {
+	t.Helper()
+
+	const (
+		issuer   = "https://test-issuer.example.com"
+		audience = "test-audience"
+	)
+
+	jwksServer, privateKey, jwksURL := setupJWKSServer(t)
+
+	hostKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
+	require.NoError(t, err)
+	hostPubKey, err := nbssh.GeneratePublicKey(hostKey)
+	require.NoError(t, err)
+
+	serverConfig := &server.Config{
+		HostKeyPEM: hostKey,
+		JWT: &server.JWTConfig{
+			Issuer:       issuer,
+			Audiences:    []string{audience},
+			KeysLocation: jwksURL,
+		},
+	}
+	sshServer := server.New(serverConfig)
+	sshServer.SetAllowRootLogin(true)
+
+	testUsername := testutil.GetTestUsername(t)
+	testJWTUser := "test-username"
+	testUserHash, err := sshuserhash.HashUserID(testJWTUser)
+	require.NoError(t, err)
+
+	authConfig := &sshauth.Config{
+		UserIDClaim:     sshauth.DefaultUserIDClaim,
+		AuthorizedUsers: []sshuserhash.UserIDHash{testUserHash},
+		MachineUsers: map[string][]uint32{
+			testUsername: {0},
+		},
+	}
+	sshServer.UpdateSSHAuth(authConfig)
+
+	sshServerAddr := server.StartTestServer(t, sshServer)
+
+	mockDaemon := startMockDaemon(t)
+
+	host, portStr, err := net.SplitHostPort(sshServerAddr)
+	require.NoError(t, err)
+	port, err := strconv.Atoi(portStr)
+	require.NoError(t, err)
+
+	mockDaemon.setHostKey(host, hostPubKey)
+
+	validToken := generateValidJWT(t, privateKey, issuer, audience, testJWTUser)
+	mockDaemon.setJWTToken(validToken)
+
+	proxyInstance, err := New(mockDaemon.addr, host, port, io.Discard, nil)
+	require.NoError(t, err)
+
+	origStdin := os.Stdin
+	origStdout := os.Stdout
+
+	stdinReader, stdinWriter, err := os.Pipe()
+	require.NoError(t, err)
+	stdoutReader, stdoutWriter, err := os.Pipe()
+	require.NoError(t, err)
+
+	os.Stdin = stdinReader
+	os.Stdout = stdoutWriter
+
+	clientConn, proxyConn := net.Pipe()
+
+	go func() { _, _ = io.Copy(stdinWriter, proxyConn) }()
+	go func() { _, _ = io.Copy(proxyConn, stdoutReader) }()
+
+	ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
+
+	go func() {
+		_ = proxyInstance.Connect(ctx)
+	}()
+
+	sshConfig := &cryptossh.ClientConfig{
+		User:            testutil.GetTestUsername(t),
+		Auth:            []cryptossh.AuthMethod{},
+		HostKeyCallback: cryptossh.InsecureIgnoreHostKey(),
+		Timeout:         5 * time.Second,
+	}
+
+	sshClientConn, chans, reqs, err := cryptossh.NewClientConn(clientConn, "test", sshConfig)
+	require.NoError(t, err)
+
+	client := cryptossh.NewClient(sshClientConn, chans, reqs)
+
+	cleanupFn := func() {
+		_ = client.Close()
+		_ = clientConn.Close()
+		cancel()
+		os.Stdin = origStdin
+		os.Stdout = origStdout
+		_ = sshServer.Stop()
+		mockDaemon.stop()
+		jwksServer.Close()
+	}
+
+	return client, cleanupFn
+}
+
+func setupJWKSServer(t *testing.T) (*httptest.Server, *rsa.PrivateKey, string) {
+	t.Helper()
+	privateKey, jwksJSON := generateTestJWKS(t)
+
+	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		if _, err := w.Write(jwksJSON); err != nil {
+			http.Error(w, err.Error(), http.StatusInternalServerError)
+		}
+	}))
+
+	return server, privateKey, server.URL
+}
+
+func generateTestJWKS(t *testing.T) (*rsa.PrivateKey, []byte) {
+	t.Helper()
+	privateKey, err := rsa.GenerateKey(rand.Reader, 2048)
+	require.NoError(t, err)
+
+	publicKey := &privateKey.PublicKey
+	n := publicKey.N.Bytes()
+	e := publicKey.E
+
+	jwk := nbjwt.JSONWebKey{
+		Kty: "RSA",
+		Kid: "test-key-id",
+		Use: "sig",
+		N:   base64.RawURLEncoding.EncodeToString(n),
+		E:   base64.RawURLEncoding.EncodeToString(big.NewInt(int64(e)).Bytes()),
+	}
+
+	jwks := nbjwt.Jwks{
+		Keys: []nbjwt.JSONWebKey{jwk},
+	}
+
+	jwksJSON, err := json.Marshal(jwks)
+	require.NoError(t, err)
+
+	return privateKey, jwksJSON
+}
+
+func generateValidJWT(t *testing.T, privateKey *rsa.PrivateKey, issuer, audience string, user string) string {
+	t.Helper()
+	claims := jwt.MapClaims{
+		"iss": issuer,
+		"aud": audience,
+		"sub": user,
+		"exp": time.Now().Add(time.Hour).Unix(),
+		"iat": time.Now().Unix(),
+	}
+
+	token := jwt.NewWithClaims(jwt.SigningMethodRS256, claims)
+	token.Header["kid"] = "test-key-id"
+
+	tokenString, err := token.SignedString(privateKey)
+	require.NoError(t, err)
+
+	return tokenString
+}

client/ssh/proxy/proxy_test.go

@@ -1,25 +1,12 @@
 package proxy
 
 import (
-	"bytes"
 	"context"
-	"crypto/rand"
-	"crypto/rsa"
-	"encoding/base64"
-	"encoding/json"
 	"fmt"
-	"io"
-	"math/big"
 	"net"
-	"net/http"
-	"net/http/httptest"
 	"os"
-	"runtime"
-	"strconv"
 	"testing"
-	"time"
 
-	"github.com/golang-jwt/jwt/v5"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 	cryptossh "golang.org/x/crypto/ssh"
@@ -28,11 +15,7 @@ import (
 
 	"github.com/netbirdio/netbird/client/proto"
 	nbssh "github.com/netbirdio/netbird/client/ssh"
-	sshauth "github.com/netbirdio/netbird/client/ssh/auth"
-	"github.com/netbirdio/netbird/client/ssh/server"
 	"github.com/netbirdio/netbird/client/ssh/testutil"
-	nbjwt "github.com/netbirdio/netbird/shared/auth/jwt"
-	sshuserhash "github.com/netbirdio/netbird/shared/sshauth"
 )
 
 func TestMain(m *testing.M) {
@@ -106,331 +89,6 @@ func TestSSHProxy_verifyHostKey(t *testing.T) {
 	})
 }
 
-func TestSSHProxy_Connect(t *testing.T) {
-	if testing.Short() {
-		t.Skip("Skipping integration test in short mode")
-	}
-
-	// TODO: Windows test times out - user switching and command execution tested on Linux
-	if runtime.GOOS == "windows" {
-		t.Skip("Skipping on Windows - covered by Linux tests")
-	}
-
-	const (
-		issuer   = "https://test-issuer.example.com"
-		audience = "test-audience"
-	)
-
-	jwksServer, privateKey, jwksURL := setupJWKSServer(t)
-	defer jwksServer.Close()
-
-	hostKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
-	require.NoError(t, err)
-	hostPubKey, err := nbssh.GeneratePublicKey(hostKey)
-	require.NoError(t, err)
-
-	serverConfig := &server.Config{
-		HostKeyPEM: hostKey,
-		JWT: &server.JWTConfig{
-			Issuer:       issuer,
-			Audiences:    []string{audience},
-			KeysLocation: jwksURL,
-		},
-	}
-	sshServer := server.New(serverConfig)
-	sshServer.SetAllowRootLogin(true)
-
-	// Configure SSH authorization for the test user
-	testUsername := testutil.GetTestUsername(t)
-	testJWTUser := "test-username"
-	testUserHash, err := sshuserhash.HashUserID(testJWTUser)
-	require.NoError(t, err)
-
-	authConfig := &sshauth.Config{
-		UserIDClaim:     sshauth.DefaultUserIDClaim,
-		AuthorizedUsers: []sshuserhash.UserIDHash{testUserHash},
-		MachineUsers: map[string][]uint32{
-			testUsername: {0}, // Index 0 in AuthorizedUsers
-		},
-	}
-	sshServer.UpdateSSHAuth(authConfig)
-
-	sshServerAddr := server.StartTestServer(t, sshServer)
-	defer func() { _ = sshServer.Stop() }()
-
-	mockDaemon := startMockDaemon(t)
-	defer mockDaemon.stop()
-
-	host, portStr, err := net.SplitHostPort(sshServerAddr)
-	require.NoError(t, err)
-	port, err := strconv.Atoi(portStr)
-	require.NoError(t, err)
-
-	mockDaemon.setHostKey(host, hostPubKey)
-
-	validToken := generateValidJWT(t, privateKey, issuer, audience, testJWTUser)
-	mockDaemon.setJWTToken(validToken)
-
-	proxyInstance, err := New(mockDaemon.addr, host, port, io.Discard, nil)
-	require.NoError(t, err)
-
-	clientConn, proxyConn := net.Pipe()
-	defer func() { _ = clientConn.Close() }()
-
-	origStdin := os.Stdin
-	origStdout := os.Stdout
-	defer func() {
-		os.Stdin = origStdin
-		os.Stdout = origStdout
-	}()
-
-	stdinReader, stdinWriter, err := os.Pipe()
-	require.NoError(t, err)
-	stdoutReader, stdoutWriter, err := os.Pipe()
-	require.NoError(t, err)
-
-	os.Stdin = stdinReader
-	os.Stdout = stdoutWriter
-
-	go func() {
-		_, _ = io.Copy(stdinWriter, proxyConn)
-	}()
-	go func() {
-		_, _ = io.Copy(proxyConn, stdoutReader)
-	}()
-
-	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
-	defer cancel()
-
-	connectErrCh := make(chan error, 1)
-	go func() {
-		connectErrCh <- proxyInstance.Connect(ctx)
-	}()
-
-	sshConfig := &cryptossh.ClientConfig{
-		User:            testutil.GetTestUsername(t),
-		Auth:            []cryptossh.AuthMethod{},
-		HostKeyCallback: cryptossh.InsecureIgnoreHostKey(),
-		Timeout:         3 * time.Second,
-	}
-
-	sshClientConn, chans, reqs, err := cryptossh.NewClientConn(clientConn, "test", sshConfig)
-	require.NoError(t, err, "Should connect to proxy server")
-	defer func() { _ = sshClientConn.Close() }()
-
-	sshClient := cryptossh.NewClient(sshClientConn, chans, reqs)
-
-	session, err := sshClient.NewSession()
-	require.NoError(t, err, "Should create session through full proxy to backend")
-
-	outputCh := make(chan []byte, 1)
-	errCh := make(chan error, 1)
-	go func() {
-		output, err := session.Output("echo hello-from-proxy")
-		outputCh <- output
-		errCh <- err
-	}()
-
-	select {
-	case output := <-outputCh:
-		err := <-errCh
-		require.NoError(t, err, "Command should execute successfully through proxy")
-		assert.Contains(t, string(output), "hello-from-proxy", "Should receive command output through proxy")
-	case <-time.After(3 * time.Second):
-		t.Fatal("Command execution timed out")
-	}
-
-	_ = session.Close()
-	_ = sshClient.Close()
-	_ = clientConn.Close()
-	cancel()
-}
-
-// TestSSHProxy_CommandQuoting verifies that the proxy preserves shell quoting
-// when forwarding commands to the backend. This is critical for tools like
-// Ansible that send commands such as:
-//
-//	/bin/sh -c '( umask 77 && mkdir -p ... ) && sleep 0'
-//
-// The single quotes must be preserved so the backend shell receives the
-// subshell expression as a single argument to -c.
-func TestSSHProxy_CommandQuoting(t *testing.T) {
-	if testing.Short() {
-		t.Skip("Skipping integration test in short mode")
-	}
-
-	sshClient, cleanup := setupProxySSHClient(t)
-	defer cleanup()
-
-	// These commands simulate what the SSH protocol delivers as exec payloads.
-	// When a user types: ssh host '/bin/sh -c "( echo hello )"'
-	// the local shell strips the outer single quotes, and the SSH exec request
-	// contains the raw string: /bin/sh -c "( echo hello )"
-	//
-	// The proxy must forward this string verbatim. Using session.Command()
-	// (shlex.Split + strings.Join) strips the inner double quotes, breaking
-	// the command on the backend.
-	tests := []struct {
-		name    string
-		command string
-		expect  string
-	}{
-		{
-			name:    "subshell_in_double_quotes",
-			command: `/bin/sh -c "( echo from-subshell ) && echo outer"`,
-			expect:  "from-subshell\nouter\n",
-		},
-		{
-			name:    "printf_with_special_chars",
-			command: `/bin/sh -c "printf '%s\n' 'hello world'"`,
-			expect:  "hello world\n",
-		},
-		{
-			name:    "nested_command_substitution",
-			command: `/bin/sh -c "echo $(echo nested)"`,
-			expect:  "nested\n",
-		},
-	}
-
-	for _, tc := range tests {
-		t.Run(tc.name, func(t *testing.T) {
-			session, err := sshClient.NewSession()
-			require.NoError(t, err)
-			defer func() { _ = session.Close() }()
-
-			var stderrBuf bytes.Buffer
-			session.Stderr = &stderrBuf
-
-			outputCh := make(chan []byte, 1)
-			errCh := make(chan error, 1)
-			go func() {
-				output, err := session.Output(tc.command)
-				outputCh <- output
-				errCh <- err
-			}()
-
-			select {
-			case output := <-outputCh:
-				err := <-errCh
-				if stderrBuf.Len() > 0 {
-					t.Logf("stderr: %s", stderrBuf.String())
-				}
-				require.NoError(t, err, "command should succeed: %s", tc.command)
-				assert.Equal(t, tc.expect, string(output), "output mismatch for: %s", tc.command)
-			case <-time.After(5 * time.Second):
-				t.Fatalf("command timed out: %s", tc.command)
-			}
-		})
-	}
-}
-
-// setupProxySSHClient creates a full proxy test environment and returns
-// an SSH client connected through the proxy to a backend NetBird SSH server.
-func setupProxySSHClient(t *testing.T) (*cryptossh.Client, func()) {
-	t.Helper()
-
-	const (
-		issuer   = "https://test-issuer.example.com"
-		audience = "test-audience"
-	)
-
-	jwksServer, privateKey, jwksURL := setupJWKSServer(t)
-
-	hostKey, err := nbssh.GeneratePrivateKey(nbssh.ED25519)
-	require.NoError(t, err)
-	hostPubKey, err := nbssh.GeneratePublicKey(hostKey)
-	require.NoError(t, err)
-
-	serverConfig := &server.Config{
-		HostKeyPEM: hostKey,
-		JWT: &server.JWTConfig{
-			Issuer:       issuer,
-			Audiences:    []string{audience},
-			KeysLocation: jwksURL,
-		},
-	}
-	sshServer := server.New(serverConfig)
-	sshServer.SetAllowRootLogin(true)
-
-	testUsername := testutil.GetTestUsername(t)
-	testJWTUser := "test-username"
-	testUserHash, err := sshuserhash.HashUserID(testJWTUser)
-	require.NoError(t, err)
-
-	authConfig := &sshauth.Config{
-		UserIDClaim:     sshauth.DefaultUserIDClaim,
-		AuthorizedUsers: []sshuserhash.UserIDHash{testUserHash},
-		MachineUsers: map[string][]uint32{
-			testUsername: {0},
-		},
-	}
-	sshServer.UpdateSSHAuth(authConfig)
-
-	sshServerAddr := server.StartTestServer(t, sshServer)
-
-	mockDaemon := startMockDaemon(t)
-
-	host, portStr, err := net.SplitHostPort(sshServerAddr)
-	require.NoError(t, err)
-	port, err := strconv.Atoi(portStr)
-	require.NoError(t, err)
-
-	mockDaemon.setHostKey(host, hostPubKey)
-
-	validToken := generateValidJWT(t, privateKey, issuer, audience, testJWTUser)
-	mockDaemon.setJWTToken(validToken)
-
-	proxyInstance, err := New(mockDaemon.addr, host, port, io.Discard, nil)
-	require.NoError(t, err)
-
-	origStdin := os.Stdin
-	origStdout := os.Stdout
-
-	stdinReader, stdinWriter, err := os.Pipe()
-	require.NoError(t, err)
-	stdoutReader, stdoutWriter, err := os.Pipe()
-	require.NoError(t, err)
-
-	os.Stdin = stdinReader
-	os.Stdout = stdoutWriter
-
-	clientConn, proxyConn := net.Pipe()
-
-	go func() { _, _ = io.Copy(stdinWriter, proxyConn) }()
-	go func() { _, _ = io.Copy(proxyConn, stdoutReader) }()
-
-	ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
-
-	go func() {
-		_ = proxyInstance.Connect(ctx)
-	}()
-
-	sshConfig := &cryptossh.ClientConfig{
-		User:            testutil.GetTestUsername(t),
-		Auth:            []cryptossh.AuthMethod{},
-		HostKeyCallback: cryptossh.InsecureIgnoreHostKey(),
-		Timeout:         5 * time.Second,
-	}
-
-	sshClientConn, chans, reqs, err := cryptossh.NewClientConn(clientConn, "test", sshConfig)
-	require.NoError(t, err)
-
-	client := cryptossh.NewClient(sshClientConn, chans, reqs)
-
-	cleanupFn := func() {
-		_ = client.Close()
-		_ = clientConn.Close()
-		cancel()
-		os.Stdin = origStdin
-		os.Stdout = origStdout
-		_ = sshServer.Stop()
-		mockDaemon.stop()
-		jwksServer.Close()
-	}
-
-	return client, cleanupFn
-}
-
 type mockDaemonServer struct {
 	proto.UnimplementedDaemonServiceServer
 	hostKeys map[string][]byte
@@ -492,10 +150,6 @@ func (m *mockDaemon) setHostKey(addr string, pubKey []byte) {
 	m.impl.hostKeys[addr] = pubKey
 }
 
-func (m *mockDaemon) setJWTToken(token string) {
-	m.impl.jwtToken = token
-}
-
 func (m *mockDaemon) stop() {
 	if m.server != nil {
 		m.server.Stop()
@@ -508,63 +162,3 @@ func mustParsePublicKey(t *testing.T, pubKeyBytes []byte) cryptossh.PublicKey {
 	require.NoError(t, err)
 	return pubKey
 }
-
-func setupJWKSServer(t *testing.T) (*httptest.Server, *rsa.PrivateKey, string) {
-	t.Helper()
-	privateKey, jwksJSON := generateTestJWKS(t)
-
-	server := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
-		w.Header().Set("Content-Type", "application/json")
-		if _, err := w.Write(jwksJSON); err != nil {
-			http.Error(w, err.Error(), http.StatusInternalServerError)
-		}
-	}))
-
-	return server, privateKey, server.URL
-}
-
-func generateTestJWKS(t *testing.T) (*rsa.PrivateKey, []byte) {
-	t.Helper()
-	privateKey, err := rsa.GenerateKey(rand.Reader, 2048)
-	require.NoError(t, err)
-
-	publicKey := &privateKey.PublicKey
-	n := publicKey.N.Bytes()
-	e := publicKey.E
-
-	jwk := nbjwt.JSONWebKey{
-		Kty: "RSA",
-		Kid: "test-key-id",
-		Use: "sig",
-		N:   base64.RawURLEncoding.EncodeToString(n),
-		E:   base64.RawURLEncoding.EncodeToString(big.NewInt(int64(e)).Bytes()),
-	}
-
-	jwks := nbjwt.Jwks{
-		Keys: []nbjwt.JSONWebKey{jwk},
-	}
-
-	jwksJSON, err := json.Marshal(jwks)
-	require.NoError(t, err)
-
-	return privateKey, jwksJSON
-}
-
-func generateValidJWT(t *testing.T, privateKey *rsa.PrivateKey, issuer, audience string, user string) string {
-	t.Helper()
-	claims := jwt.MapClaims{
-		"iss": issuer,
-		"aud": audience,
-		"sub": user,
-		"exp": time.Now().Add(time.Hour).Unix(),
-		"iat": time.Now().Unix(),
-	}
-
-	token := jwt.NewWithClaims(jwt.SigningMethodRS256, claims)
-	token.Header["kid"] = "test-key-id"
-
-	tokenString, err := token.SignedString(privateKey)
-	require.NoError(t, err)
-
-	return tokenString
-}

client/ssh/server/executor_unix_privileged_test.go

@@ -0,0 +1,66 @@
+//go:build unix && privileged
+
+package server
+
+import (
+	"context"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestPrivilegeDropper_CreateExecutorCommand(t *testing.T) {
+	pd := NewPrivilegeDropper()
+
+	config := ExecutorConfig{
+		UID:        1000,
+		GID:        1000,
+		Groups:     []uint32{1000, 1001},
+		WorkingDir: "/home/testuser",
+		Shell:      "/bin/bash",
+		Command:    "ls -la",
+	}
+
+	cmd, err := pd.CreateExecutorCommand(context.Background(), config)
+	require.NoError(t, err)
+	require.NotNil(t, cmd)
+
+	// Verify the command is calling netbird ssh exec
+	assert.Contains(t, cmd.Args, "ssh")
+	assert.Contains(t, cmd.Args, "exec")
+	assert.Contains(t, cmd.Args, "--uid")
+	assert.Contains(t, cmd.Args, "1000")
+	assert.Contains(t, cmd.Args, "--gid")
+	assert.Contains(t, cmd.Args, "1000")
+	assert.Contains(t, cmd.Args, "--groups")
+	assert.Contains(t, cmd.Args, "1000")
+	assert.Contains(t, cmd.Args, "1001")
+	assert.Contains(t, cmd.Args, "--working-dir")
+	assert.Contains(t, cmd.Args, "/home/testuser")
+	assert.Contains(t, cmd.Args, "--shell")
+	assert.Contains(t, cmd.Args, "/bin/bash")
+	assert.Contains(t, cmd.Args, "--cmd")
+	assert.Contains(t, cmd.Args, "ls -la")
+}
+
+func TestPrivilegeDropper_CreateExecutorCommandInteractive(t *testing.T) {
+	pd := NewPrivilegeDropper()
+
+	config := ExecutorConfig{
+		UID:        1000,
+		GID:        1000,
+		Groups:     []uint32{1000},
+		WorkingDir: "/home/testuser",
+		Shell:      "/bin/bash",
+		Command:    "",
+	}
+
+	cmd, err := pd.CreateExecutorCommand(context.Background(), config)
+	require.NoError(t, err)
+	require.NotNil(t, cmd)
+
+	// Verify no command mode (command is empty so no --cmd flag)
+	assert.NotContains(t, cmd.Args, "--cmd")
+	assert.NotContains(t, cmd.Args, "--interactive")
+}

client/ssh/server/executor_unix_test.go

@@ -73,61 +73,6 @@ func TestPrivilegeDropper_ValidatePrivileges(t *testing.T) {
 	}
 }
 
-func TestPrivilegeDropper_CreateExecutorCommand(t *testing.T) {
-	pd := NewPrivilegeDropper()
-
-	config := ExecutorConfig{
-		UID:        1000,
-		GID:        1000,
-		Groups:     []uint32{1000, 1001},
-		WorkingDir: "/home/testuser",
-		Shell:      "/bin/bash",
-		Command:    "ls -la",
-	}
-
-	cmd, err := pd.CreateExecutorCommand(context.Background(), config)
-	require.NoError(t, err)
-	require.NotNil(t, cmd)
-
-	// Verify the command is calling netbird ssh exec
-	assert.Contains(t, cmd.Args, "ssh")
-	assert.Contains(t, cmd.Args, "exec")
-	assert.Contains(t, cmd.Args, "--uid")
-	assert.Contains(t, cmd.Args, "1000")
-	assert.Contains(t, cmd.Args, "--gid")
-	assert.Contains(t, cmd.Args, "1000")
-	assert.Contains(t, cmd.Args, "--groups")
-	assert.Contains(t, cmd.Args, "1000")
-	assert.Contains(t, cmd.Args, "1001")
-	assert.Contains(t, cmd.Args, "--working-dir")
-	assert.Contains(t, cmd.Args, "/home/testuser")
-	assert.Contains(t, cmd.Args, "--shell")
-	assert.Contains(t, cmd.Args, "/bin/bash")
-	assert.Contains(t, cmd.Args, "--cmd")
-	assert.Contains(t, cmd.Args, "ls -la")
-}
-
-func TestPrivilegeDropper_CreateExecutorCommandInteractive(t *testing.T) {
-	pd := NewPrivilegeDropper()
-
-	config := ExecutorConfig{
-		UID:        1000,
-		GID:        1000,
-		Groups:     []uint32{1000},
-		WorkingDir: "/home/testuser",
-		Shell:      "/bin/bash",
-		Command:    "",
-	}
-
-	cmd, err := pd.CreateExecutorCommand(context.Background(), config)
-	require.NoError(t, err)
-	require.NotNil(t, cmd)
-
-	// Verify no command mode (command is empty so no --cmd flag)
-	assert.NotContains(t, cmd.Args, "--cmd")
-	assert.NotContains(t, cmd.Args, "--interactive")
-}
-
 // TestPrivilegeDropper_ActualPrivilegeDrop tests actual privilege dropping
 // This test requires root privileges and will be skipped if not running as root
 func TestPrivilegeDropper_ActualPrivilegeDrop(t *testing.T) {

client/testutil/privileged/runner_test.go

@@ -0,0 +1,196 @@
+//go:build privileged && (linux || darwin)
+
+// Package privileged provides a self-hosting harness that runs the repo's
+// privileged-tagged test suite inside a --privileged --cap-add=NET_ADMIN
+// container, so developers can exercise the root/system-mutating tests on a
+// non-root host with a single `go test` invocation.
+package privileged
+
+import (
+	"bytes"
+	"context"
+	"fmt"
+	"os"
+	"os/exec"
+	"path/filepath"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/moby/moby/api/types/container"
+	"github.com/ory/dockertest/v4"
+)
+
+// containerImage / containerTag match the image used by the CI privileged job
+// (.github/workflows/golang-test-linux.yml, test_client_on_docker).
+const (
+	containerImage = "golang"
+	containerTag   = "1.25-alpine"
+)
+
+const (
+	containerWorkdir    = "/app"
+	containerGoCache    = "/root/.cache/go-build"
+	containerGoModCache = "/go/pkg/mod"
+)
+
+// alpinePackages are the build/runtime deps the privileged tests need, mirroring
+// the CI container setup.
+const alpinePackages = "ca-certificates iptables ip6tables dbus dbus-dev libpcap-dev build-base"
+
+// privilegedTestPackages is the package list the suite runs, excluding the
+// server-side trees and UI/upload helpers, matching the CI Docker job's filter.
+const privilegedTestPackages = `go list -buildvcs=false ./... | grep -v -e /management -e /signal -e /relay -e /proxy -e /combined -e /client/ui -e /upload-server`
+
+// testWriter forwards container output to the test log line by line.
+type testWriter struct{ t *testing.T }
+
+func (w testWriter) Write(p []byte) (int, error) {
+	for _, line := range strings.Split(strings.TrimRight(string(p), "\n"), "\n") {
+		w.t.Log(line)
+	}
+	return len(p), nil
+}
+
+// TestRunPrivilegedSuiteInDocker spins up a privileged container, mounts the repo,
+// and runs `go test -tags 'devcert privileged'` inside it. When already running
+// inside that container (DOCKER_CI=true) it returns immediately so the real
+// privileged tests in the suite execute in place instead of recursing.
+func TestRunPrivilegedSuiteInDocker(t *testing.T) {
+	if os.Getenv("DOCKER_CI") == "true" {
+		t.Skip("inside privileged container, skipping container spawn; privileged tests run in place")
+	}
+
+	repoRoot, err := findRepoRoot()
+	if err != nil {
+		t.Fatalf("locate repo root: %v", err)
+	}
+	goCache, goModCache := hostGoCaches(t)
+
+	// dockertest reads DOCKER_HOST; point it at the active context's socket when
+	// the default one is absent (macOS Docker Desktop, Colima, OrbStack).
+	if host := dockerHost(); host != "" {
+		t.Setenv("DOCKER_HOST", host)
+	}
+
+	// NewPoolT registers container cleanup via t.Cleanup automatically.
+	pool := dockertest.NewPoolT(t, "", dockertest.WithMaxWait(30*time.Minute))
+
+	// Keep the container alive so the suite runs via Exec, which yields a clean
+	// exit code (the v4 Resource API exposes no container wait/exit-code).
+	resource := pool.RunT(t, containerImage,
+		dockertest.WithTag(containerTag),
+		dockertest.WithWorkingDir(containerWorkdir),
+		dockertest.WithMounts([]string{
+			repoRoot + ":" + containerWorkdir,
+			goCache + ":" + containerGoCache,
+			goModCache + ":" + containerGoModCache,
+		}),
+		dockertest.WithEnv([]string{
+			"CGO_ENABLED=1",
+			"CI=true",
+			"DOCKER_CI=true",
+			"CONTAINER=true",
+			"GOCACHE=" + containerGoCache,
+			"GOMODCACHE=" + containerGoModCache,
+		}),
+		dockertest.WithCmd([]string{"sleep", "infinity"}),
+		dockertest.WithHostConfig(func(hc *container.HostConfig) {
+			hc.Privileged = true
+			hc.CapAdd = []string{"NET_ADMIN"}
+		}),
+		dockertest.WithoutReuse(),
+	)
+
+	ctx, cancel := context.WithTimeout(context.Background(), 30*time.Minute)
+	defer cancel()
+
+	result, err := resource.Exec(ctx, []string{"sh", "-c", buildTestScript()})
+	if err != nil {
+		t.Fatalf("run privileged suite in container: %v", err)
+	}
+
+	w := testWriter{t}
+	_, _ = w.Write([]byte(result.StdOut))
+	_, _ = w.Write([]byte(result.StdErr))
+
+	if result.ExitCode != 0 {
+		t.Fatalf("privileged test suite failed in container (exit code %d)", result.ExitCode)
+	}
+}
+
+// findRepoRoot walks up from the test's working directory to the module root.
+func findRepoRoot() (string, error) {
+	dir, err := os.Getwd()
+	if err != nil {
+		return "", err
+	}
+	for {
+		if _, statErr := os.Stat(filepath.Join(dir, "go.mod")); statErr == nil {
+			return dir, nil
+		}
+		parent := filepath.Dir(dir)
+		if parent == dir {
+			return "", fmt.Errorf("go.mod not found above %s", dir)
+		}
+		dir = parent
+	}
+}
+
+// dockerHost returns a DOCKER_HOST override when the default socket is missing.
+// An empty result means the caller should leave DOCKER_HOST untouched (it is
+// already set, or the default unix socket exists). When neither is present
+// (common on macOS Docker Desktop, Colima and OrbStack, which use a per-user
+// socket), it resolves the active docker context's endpoint.
+func dockerHost() string {
+	if os.Getenv("DOCKER_HOST") != "" {
+		return ""
+	}
+	if _, err := os.Stat("/var/run/docker.sock"); err == nil {
+		return ""
+	}
+
+	out, err := exec.Command("docker", "context", "inspect", "-f", "{{.Endpoints.docker.Host}}").Output()
+	if err != nil {
+		return ""
+	}
+	return strings.TrimSpace(string(out))
+}
+
+// hostGoCaches resolves the host GOCACHE/GOMODCACHE so the container reuses the
+// existing build/module cache for speed.
+func hostGoCaches(t *testing.T) (string, string) {
+	t.Helper()
+	return goEnv(t, "GOCACHE"), goEnv(t, "GOMODCACHE")
+}
+
+func goEnv(t *testing.T, key string) string {
+	t.Helper()
+	var out bytes.Buffer
+	cmd := exec.Command("go", "env", key)
+	cmd.Stdout = &out
+	if err := cmd.Run(); err != nil {
+		t.Fatalf("go env %s: %v", key, err)
+	}
+	return strings.TrimSpace(out.String())
+}
+
+// buildTestScript builds the in-container command. PRIV_PKGS overrides the package
+// list (default: the full filtered set); PRIV_RUN adds a -run test-name filter.
+// Both empty reproduces the full privileged suite.
+func buildTestScript() string {
+	pkgs := privilegedTestPackages + " | xargs"
+	if p := os.Getenv("PRIV_PKGS"); p != "" {
+		pkgs = "echo " + p + " | xargs"
+	}
+
+	runFilter := ""
+	if r := os.Getenv("PRIV_RUN"); r != "" {
+		runFilter = "-run '" + r + "' "
+	}
+
+	return fmt.Sprintf(
+		"apk update >/dev/null && apk add --no-cache %s >/dev/null && %s go test -buildvcs=false -tags 'devcert privileged' %s-v -timeout 20m -p 1",
+		alpinePackages, pkgs, runFilter,
+	)
+}

docs/testing-privileged.md

@@ -0,0 +1,78 @@
+# Privileged tests
+
+Some tests in this repo need `root` or mutate host network state: they create
+TUN/WireGuard interfaces, open netlink/raw sockets, run eBPF programs, or shell
+out to `ip`/`iptables`/`nft`/`ifconfig`/`route`. Running them on a developer
+machine would require `sudo` and could leave stray interfaces or routes behind.
+
+These tests are gated behind the **`privileged` build tag** so the default test
+run is host-safe.
+
+## Running tests
+
+```bash
+# Host-safe: excludes privileged tests. Runs as a normal user, no sudo.
+make test-unit
+# equivalently:
+go test -tags devcert ./...
+
+# Privileged suite: runs the privileged-tagged tests inside a
+# --privileged --cap-add=NET_ADMIN container (requires Docker).
+make test-privileged
+
+# Narrow the container run to a single test / package:
+PRIV_RUN=TestNftablesManager PRIV_PKGS=./client/firewall/nftables/... make test-privileged
+```
+
+`PRIV_RUN` adds a `-run` test-name filter and `PRIV_PKGS` overrides the package
+list; both are optional and default to the full privileged suite.
+
+`make test-privileged` invokes the `ory/dockertest` harness in
+`client/testutil/privileged/`. The harness:
+
+1. Skips immediately when it detects it is already inside the container
+   (`DOCKER_CI=true`), so the privileged tests run in place instead of recursing.
+2. Otherwise spins up a `golang:1.25-alpine` container (matching CI),
+   bind-mounts the repo and the host Go build/module caches, installs the
+   required packages, and runs `go test -tags 'devcert privileged'` over the
+   client packages.
+3. Streams the container's output to the test log and fails if the suite fails.
+
+## Adding a privileged test
+
+A test is privileged if it does any of:
+
+- creates a real interface via `iface.NewWGIFace(...).Create()`,
+- opens a netlink or raw socket that hard-fails without `CAP_NET_ADMIN`,
+- runs an eBPF program (`ebpf.*.Listen()`),
+- shells out to `ip`, `iptables`, `nft`, `ifconfig`, or `route` to change state.
+
+Add the tag to the **top** of the file, combined with any existing platform
+constraint:
+
+```go
+//go:build privileged && linux
+
+package foo
+```
+
+If a file mixes privileged and pure-logic tests, **split it**: keep the pure
+tests (and any shared data — type/var declarations, table-driven `testCases`,
+helper interfaces) in an untagged file, and move the privileged tests into a
+`*_privileged_test.go` file with the tag. Shared declarations must stay untagged,
+otherwise the unprivileged files in the package will not compile.
+
+Always verify both build modes compile on every target platform:
+
+```bash
+go vet -tags devcert ./...
+go vet -tags 'devcert privileged' ./...
+```
+
+## CI
+
+- The `Client / Unit` job runs `go test -tags devcert` with **no** `sudo` — only
+  host-safe tests.
+- The `Client (Docker) / Unit` job runs `go test -tags 'devcert privileged'`
+  inside a `--privileged --cap-add=NET_ADMIN` container, which is where the
+  privileged tests actually execute.

go.mod

@@ -78,10 +78,12 @@ require (
 	github.com/mdp/qrterminal/v3 v3.2.1
 	github.com/miekg/dns v1.1.72
 	github.com/mitchellh/hashstructure/v2 v2.0.2
+	github.com/moby/moby/api v1.54.1
 	github.com/netbirdio/management-integrations/integrations v0.0.0-20260416123949-2355d972be42
 	github.com/netbirdio/signal-dispatcher/dispatcher v0.0.0-20250805121659-6b4ac470ca45
 	github.com/oapi-codegen/runtime v1.1.2
 	github.com/okta/okta-sdk-golang/v2 v2.18.0
+	github.com/ory/dockertest/v4 v4.0.0
 	github.com/oschwald/maxminddb-golang v1.12.0
 	github.com/patrickmn/go-cache v2.1.0+incompatible
 	github.com/petermattis/goid v0.0.0-20250303134427-723919f7f203
@@ -145,7 +147,7 @@ require (
 	dario.cat/mergo v1.0.1 // indirect
 	filippo.io/edwards25519 v1.1.1 // indirect
 	github.com/AppsFlyer/go-sundheit v0.6.0 // indirect
-	github.com/Azure/go-ansiterm v0.0.0-20230124172434-306776ec8161 // indirect
+	github.com/Azure/go-ansiterm v0.0.0-20250102033503-faa5f7b0171c // indirect
 	github.com/Azure/go-ntlmssp v0.1.0 // indirect
 	github.com/BurntSushi/toml v1.5.0 // indirect
 	github.com/Masterminds/goutils v1.1.1 // indirect
@@ -177,6 +179,8 @@ require (
 	github.com/caddyserver/zerossl v0.1.3 // indirect
 	github.com/cenkalti/backoff/v5 v5.0.3 // indirect
 	github.com/cespare/xxhash/v2 v2.3.0 // indirect
+	github.com/containerd/errdefs v1.0.0 // indirect
+	github.com/containerd/errdefs/pkg v0.3.0 // indirect
 	github.com/containerd/log v0.1.0 // indirect
 	github.com/containerd/platforms v0.2.1 // indirect
 	github.com/cpuguy83/dockercfg v0.3.2 // indirect
@@ -271,11 +275,12 @@ require (
 	github.com/mitchellh/mapstructure v1.5.0 // indirect
 	github.com/mitchellh/reflectwalk v1.0.2 // indirect
 	github.com/moby/docker-image-spec v1.3.1 // indirect
+	github.com/moby/moby/client v0.4.0 // indirect
 	github.com/moby/patternmatcher v0.6.0 // indirect
 	github.com/moby/sys/sequential v0.5.0 // indirect
 	github.com/moby/sys/user v0.3.0 // indirect
 	github.com/moby/sys/userns v0.1.0 // indirect
-	github.com/moby/term v0.5.0 // indirect
+	github.com/moby/term v0.5.2 // indirect
 	github.com/morikuni/aec v1.0.0 // indirect
 	github.com/munnerz/goautoneg v0.0.0-20191010083416-a7dc8b61c822 // indirect
 	github.com/nfnt/resize v0.0.0-20180221191011-83c6a9932646 // indirect

go.sum

@@ -23,8 +23,8 @@ github.com/AdaLogics/go-fuzz-headers v0.0.0-20230811130428-ced1acdcaa24 h1:bvDV9
 github.com/AdaLogics/go-fuzz-headers v0.0.0-20230811130428-ced1acdcaa24/go.mod h1:8o94RPi1/7XTJvwPpRSzSUedZrtlirdB3r9Z20bi2f8=
 github.com/AppsFlyer/go-sundheit v0.6.0 h1:d2hBvCjBSb2lUsEWGfPigr4MCOt04sxB+Rppl0yUMSk=
 github.com/AppsFlyer/go-sundheit v0.6.0/go.mod h1:LDdBHD6tQBtmHsdW+i1GwdTt6Wqc0qazf5ZEJVTbTME=
-github.com/Azure/go-ansiterm v0.0.0-20230124172434-306776ec8161 h1:L/gRVlceqvL25UVaW/CKtUDjefjrs0SPonmDGUVOYP0=
-github.com/Azure/go-ansiterm v0.0.0-20230124172434-306776ec8161/go.mod h1:xomTg63KZ2rFqZQzSB4Vz2SUXa1BpHTVz9L5PTmPC4E=
+github.com/Azure/go-ansiterm v0.0.0-20250102033503-faa5f7b0171c h1:udKWzYgxTojEKWjV8V+WSxDXJ4NFATAsZjh8iIbsQIg=
+github.com/Azure/go-ansiterm v0.0.0-20250102033503-faa5f7b0171c/go.mod h1:xomTg63KZ2rFqZQzSB4Vz2SUXa1BpHTVz9L5PTmPC4E=
 github.com/Azure/go-ntlmssp v0.1.0 h1:DjFo6YtWzNqNvQdrwEyr/e4nhU3vRiwenz5QX7sFz+A=
 github.com/Azure/go-ntlmssp v0.1.0/go.mod h1:NYqdhxd/8aAct/s4qSYZEerdPuH1liG2/X9DiVTbhpk=
 github.com/BurntSushi/toml v1.5.0 h1:W5quZX/G/csjUnuI8SUYlsHs9M38FC7znL0lIO+DvMg=
@@ -117,6 +117,10 @@ github.com/cilium/ebpf v0.19.0 h1:Ro/rE64RmFBeA9FGjcTc+KmCeY6jXmryu6FfnzPRIao=
 github.com/cilium/ebpf v0.19.0/go.mod h1:fLCgMo3l8tZmAdM3B2XqdFzXBpwkcSTroaVqN08OWVY=
 github.com/coder/websocket v1.8.14 h1:9L0p0iKiNOibykf283eHkKUHHrpG7f65OE3BhhO7v9g=
 github.com/coder/websocket v1.8.14/go.mod h1:NX3SzP+inril6yawo5CQXx8+fk145lPDC6pumgx0mVg=
+github.com/containerd/errdefs v1.0.0 h1:tg5yIfIlQIrxYtu9ajqY42W3lpS19XqdxRQeEwYG8PI=
+github.com/containerd/errdefs v1.0.0/go.mod h1:+YBYIdtsnF4Iw6nWZhJcqGSg/dwvV7tyJ/kCkyJ2k+M=
+github.com/containerd/errdefs/pkg v0.3.0 h1:9IKJ06FvyNlexW690DXuQNx2KA2cUJXx151Xdx3ZPPE=
+github.com/containerd/errdefs/pkg v0.3.0/go.mod h1:NJw6s9HwNuRhnjJhM7pylWwMyAkmCQvQ4GpJHEqRLVk=
 github.com/containerd/log v0.1.0 h1:TCJt7ioM2cr/tfR8GPbGf9/VRAX8D2B4PjzCpfX540I=
 github.com/containerd/log v0.1.0/go.mod h1:VRRf09a7mHDIRezVKTRCrOq78v577GXq3bSa3EhrzVo=
 github.com/containerd/platforms v0.2.1 h1:zvwtM3rz2YHPQsF2CHYM8+KtB5dvhISiXh5ZpSBQv6A=
@@ -480,6 +484,10 @@ github.com/mitchellh/reflectwalk v1.0.2 h1:G2LzWKi524PWgd3mLHV8Y5k7s6XUvT0Gef6zx
 github.com/mitchellh/reflectwalk v1.0.2/go.mod h1:mSTlrgnPZtwu0c4WaC2kGObEpuNDbx0jmZXqmk4esnw=
 github.com/moby/docker-image-spec v1.3.1 h1:jMKff3w6PgbfSa69GfNg+zN/XLhfXJGnEx3Nl2EsFP0=
 github.com/moby/docker-image-spec v1.3.1/go.mod h1:eKmb5VW8vQEh/BAr2yvVNvuiJuY6UIocYsFu/DxxRpo=
+github.com/moby/moby/api v1.54.1 h1:TqVzuJkOLsgLDDwNLmYqACUuTehOHRGKiPhvH8V3Nn4=
+github.com/moby/moby/api v1.54.1/go.mod h1:+RQ6wluLwtYaTd1WnPLykIDPekkuyD/ROWQClE83pzs=
+github.com/moby/moby/client v0.4.0 h1:S+2XegzHQrrvTCvF6s5HFzcrywWQmuVnhOXe2kiWjIw=
+github.com/moby/moby/client v0.4.0/go.mod h1:QWPbvWchQbxBNdaLSpoKpCdf5E+WxFAgNHogCWDoa7g=
 github.com/moby/patternmatcher v0.6.0 h1:GmP9lR19aU5GqSSFko+5pRqHi+Ohk1O69aFiKkVGiPk=
 github.com/moby/patternmatcher v0.6.0/go.mod h1:hDPoyOpDY7OrrMDLaYoY3hf52gNCR/YOUYxkhApJIxc=
 github.com/moby/sys/sequential v0.5.0 h1:OPvI35Lzn9K04PBbCLW0g4LcFAJgHsvXsRyewg5lXtc=
@@ -488,8 +496,8 @@ github.com/moby/sys/user v0.3.0 h1:9ni5DlcW5an3SvRSx4MouotOygvzaXbaSrc/wGDFWPo=
 github.com/moby/sys/user v0.3.0/go.mod h1:bG+tYYYJgaMtRKgEmuueC0hJEAZWwtIbZTB+85uoHjs=
 github.com/moby/sys/userns v0.1.0 h1:tVLXkFOxVu9A64/yh59slHVv9ahO9UIev4JZusOLG/g=
 github.com/moby/sys/userns v0.1.0/go.mod h1:IHUYgu/kao6N8YZlp9Cf444ySSvCmDlmzUcYfDHOl28=
-github.com/moby/term v0.5.0 h1:xt8Q1nalod/v7BqbG21f8mQPqH+xAaC9C3N3wfWbVP0=
-github.com/moby/term v0.5.0/go.mod h1:8FzsFHVUBGZdbDsJw/ot+X+d5HLUbvklYLJ9uGfcI3Y=
+github.com/moby/term v0.5.2 h1:6qk3FJAFDs6i/q3W/pQ97SX192qKfZgGjCQqfCJkgzQ=
+github.com/moby/term v0.5.2/go.mod h1:d3djjFCrjnB+fl8NJux+EJzu0msscUP+f8it8hPkFLc=
 github.com/modern-go/concurrent v0.0.0-20180228061459-e0a39a4cb421/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
 github.com/modern-go/reflect2 v0.0.0-20180701023420-4b7aa43c6742/go.mod h1:bx2lNnkwVCuqBIxFjflWJWanXIb3RllmbCylyMrvgv0=
 github.com/morikuni/aec v1.0.0 h1:nP9CBfwrvYnBRgY6qfDQkygYDmYwOilePFkwzv4dU8A=
@@ -542,6 +550,8 @@ github.com/opencontainers/go-digest v1.0.0 h1:apOUWs51W5PlhuyGyz9FCeeBIOUDA/6nW8
 github.com/opencontainers/go-digest v1.0.0/go.mod h1:0JzlMkj0TRzQZfJkVvzbP0HBR3IKzErnv2BNG4W4MAM=
 github.com/opencontainers/image-spec v1.1.1 h1:y0fUlFfIZhPF1W537XOLg0/fcx6zcHCJwooC2xJA040=
 github.com/opencontainers/image-spec v1.1.1/go.mod h1:qpqAh3Dmcf36wStyyWU+kCeDgrGnAve2nCC8+7h8Q0M=
+github.com/ory/dockertest/v4 v4.0.0 h1:i19aFsO/VXE0VrMk4ifnKW4G/KIJ93PCjLOslxXoPME=
+github.com/ory/dockertest/v4 v4.0.0/go.mod h1:b5Ofu8VIxWNhXFvQcLu17pRNQdoUBKtXBW74G4Ygzx8=
 github.com/oschwald/maxminddb-golang v1.12.0 h1:9FnTOD0YOhP7DGxGsq4glzpGy5+w7pq50AS6wALUMYs=
 github.com/oschwald/maxminddb-golang v1.12.0/go.mod h1:q0Nob5lTCqyQ8WT6FYgS1L7PXKVVbgiymefNwIjPzgY=
 github.com/patrickmn/go-cache v2.1.0+incompatible h1:HRMgzkcYKYpi3C8ajMPV8OFXaaRUnok+kx1WdO15EQc=
@@ -973,11 +983,13 @@ gorm.io/driver/sqlite v1.5.7/go.mod h1:U+J8craQU6Fzkcvu8oLeAQmi50TkwPEhHDEjQZXDa
 gorm.io/gorm v1.25.7/go.mod h1:hbnx/Oo0ChWMn1BIhpy1oYozzpM15i4YPuHDmfYtwg8=
 gorm.io/gorm v1.25.12 h1:I0u8i2hWQItBq1WfE0o2+WuL9+8L21K9e2HHSTE/0f8=
 gorm.io/gorm v1.25.12/go.mod h1:xh7N7RHfYlNc5EmcI/El95gXusucDrQnHXe0+CgWcLQ=
-gotest.tools/v3 v3.5.1 h1:EENdUnS3pdur5nybKYIh2Vfgc8IUNBjxDPSjtiJcOzU=
-gotest.tools/v3 v3.5.1/go.mod h1:isy3WKz7GK6uNw/sbHzfKBLvlvXwUyV06n6brMxxopU=
+gotest.tools/v3 v3.5.2 h1:7koQfIKdy+I8UTetycgUqXWSDwpgv193Ka+qRsmBY8Q=
+gotest.tools/v3 v3.5.2/go.mod h1:LtdLGcnqToBH83WByAAi/wiwSFCArdFIUV/xxN4pcjA=
 gvisor.dev/gvisor v0.0.0-20260219192049-0f2374377e89 h1:mGJaeA61P8dEHTqdvAgc70ZIV3QoUoJcXCRyyjO26OA=
 gvisor.dev/gvisor v0.0.0-20260219192049-0f2374377e89/go.mod h1:QkHjoMIBaYtpVufgwv3keYAbln78mBoCuShZrPrer1Q=
 howett.net/plist v1.0.1 h1:37GdZ8tP09Q35o9ych3ehygcsL+HqKSwzctveSlarvM=
 howett.net/plist v1.0.1/go.mod h1:lqaXoTrLY4hg8tnEzNru53gicrbv7rrk+2xJA/7hw9g=
+pgregory.net/rapid v1.2.0 h1:keKAYRcjm+e1F0oAuU5F5+YPAWcyxNNRK2wud503Gnk=
+pgregory.net/rapid v1.2.0/go.mod h1:PY5XlDGj0+V1FCq0o192FdRhpKHGTRIWBgqjDBTrq04=
 rsc.io/qr v0.2.0 h1:6vBLea5/NRMVTz8V66gipeLycZMl/+UlFmk8DvqQ6WY=
 rsc.io/qr v0.2.0/go.mod h1:IF+uZjkb9fqyeF/4tlBoynqmQxUoPfWEKh921coOuXs=

sharedsock/sock_linux_test.go

@@ -1,3 +1,5 @@
+//go:build privileged
+
 package sharedsock
 
 import (