Merge branch 'main' into userspace-router

The nhooyr.io/websocket package was renamed to github.com/coder/websocket when
the project was transferred to "coder" as the new maintainer.

Use the new import path and update go.mod and go.sum accordingly.

Signed-off-by: Christian Stewart <christian@aperture.us>

.devcontainer/Dockerfile

@@ -1,4 +1,4 @@
-FROM golang:1.21-bullseye
+FROM golang:1.23-bullseye
 
 RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
     && apt-get -y install --no-install-recommends\

.devcontainer/devcontainer.json

@@ -7,7 +7,7 @@
 	"features": {
 		"ghcr.io/devcontainers/features/docker-in-docker:2": {},
 		"ghcr.io/devcontainers/features/go:1": {
-			"version": "1.21"
+			"version": "1.23"
 		}
 	},
 	"workspaceFolder": "/workspaces/${localWorkspaceFolderBasename}",

.editorconfig

@@ -0,0 +1,8 @@
+root = true
+
+[*]
+end_of_line = lf
+insert_final_newline = true
+
+[*.go]
+indent_style = tab

.github/FUNDING.yml

@@ -0,0 +1,3 @@
+# These are supported funding model platforms
+
+github: [netbirdio]

.github/ISSUE_TEMPLATE/bug-issue-report.md

@@ -31,9 +31,14 @@ Please specify whether you use NetBird Cloud or self-host NetBird's control plan
 
 `netbird version`
 
-**NetBird status -d output:**
+**NetBird status -dA output:**
 
-If applicable, add the `netbird status -d' command output.
+If applicable, add the `netbird status -dA' command output.
+
+**Do you face any (non-mobile) client issues?**
+
+Please provide the file created by `netbird debug for 1m -AS`.
+We advise reviewing the anonymized files for any remaining PII.
 
 **Screenshots**
 

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

@@ -14,22 +14,24 @@ jobs:
   test:
     strategy:
       matrix:
-        store: ['jsonfile', 'sqlite']
+        store: ['sqlite']
     runs-on: macos-latest
     steps:
       - name: Install Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21.x"
+          go-version: "1.23.x"
+          cache: false
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Cache Go modules
-        uses: actions/cache@v3
+        uses: actions/cache@v4
         with:
           path: ~/go/pkg/mod
-          key: macos-go-${{ hashFiles('**/go.sum') }}
+          key: macos-gotest-${{ hashFiles('**/go.sum') }}
           restore-keys: |
+            macos-gotest-
             macos-go-
 
       - name: Install libpcap
@@ -42,4 +44,5 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} go test -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 ./...
+        run: NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -tags=devcert -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 $(go list ./... | grep -v /management)
+

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

@@ -0,0 +1,46 @@
+
+name: Test Code FreeBSD
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref || github.actor_id }}
+  cancel-in-progress: true
+
+jobs:
+  test:
+    runs-on: ubuntu-22.04
+    steps:
+      - uses: actions/checkout@v4
+      - name: Test in FreeBSD
+        id: test
+        uses: vmactions/freebsd-vm@v1
+        with:
+          usesh: true
+          copyback: false
+          release: "14.1"
+          prepare: |
+            pkg install -y go pkgconf xorg
+
+          # -x - to print all executed commands
+          # -e - to faile on first error
+          run: |
+            set -e -x
+            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/...
+            # NOTE: without -p1 `client/internal/dns` will fail because of `listen udp4 :33100: bind: address already in use`
+            time go test -timeout 8m -failfast -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 ./signal/...
+            time go test -timeout 1m -failfast ./util/...
+            time go test -timeout 1m -failfast ./version/...

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

@@ -11,29 +11,114 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  test:
-    strategy:
-      matrix:
-        arch: [ '386','amd64' ]
-        store: [ 'jsonfile', 'sqlite' ]
-    runs-on: ubuntu-latest
-    steps:
-      - name: Install Go
-        uses: actions/setup-go@v4
-        with:
-          go-version: "1.21.x"
+  build-cache:
+    runs-on: ubuntu-22.04
+    steps:        
+      - name: Checkout code
+        uses: actions/checkout@v4
 
+      - name: Install Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: "1.23.x"
+          cache: false
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV     
 
       - name: Cache Go modules
-        uses: actions/cache@v3
+        uses: actions/cache@v4
+        id: cache
         with:
-          path: ~/go/pkg/mod
-          key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
           restore-keys: |
-            ${{ runner.os }}-go-
+            ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
+            
+
+      - name: Install dependencies
+        if: steps.cache.outputs.cache-hit != 'true'
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
+
+      - name: Install 32-bit libpcap
+        if: steps.cache.outputs.cache-hit != 'true'
+        run: sudo dpkg --add-architecture i386 && sudo apt update && sudo apt-get install -y libpcap0.8-dev:i386
+
+      - name: Build client
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: client
+        run: CGO_ENABLED=1 go build .
+
+      - name: Build client 386
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: client
+        run: CGO_ENABLED=1 GOARCH=386 go build -o client-386 .
+
+      - name: Build management
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: management
+        run: CGO_ENABLED=1 go build .
+
+      - name: Build management 386
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: management
+        run: CGO_ENABLED=1 GOARCH=386 go build -o management-386 .
+
+      - name: Build signal
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: signal
+        run: CGO_ENABLED=1 go build .
+
+      - name: Build signal 386
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: signal
+        run: CGO_ENABLED=1 GOARCH=386 go build -o signal-386 .
+
+      - name: Build relay
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: relay
+        run: CGO_ENABLED=1 go build .
+
+      - name: Build relay 386
+        if: steps.cache.outputs.cache-hit != 'true'
+        working-directory: relay
+        run: CGO_ENABLED=1 GOARCH=386 go build -o relay-386 .
+
+  test:
+    needs: [build-cache]
+    strategy:
+      fail-fast: false
+      matrix:
+        arch: [ '386','amd64' ]
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Install Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: "1.23.x"
+          cache: false
 
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV     
+
+      - name: Cache Go modules
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-gotest-cache-
 
       - name: Install dependencies
         run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
@@ -49,26 +134,264 @@ jobs:
         run: git --no-pager diff --exit-code
 
       - name: Test
-        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} NETBIRD_STORE_ENGINE=${{ matrix.store }} go test -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 5m -p 1 ./...
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} CI=true go test -tags devcert -exec 'sudo' -timeout 10m -p 1 $(go list ./... | grep -v /management)
+
+  test_management:
+    needs: [ build-cache ]
+    strategy:
+      fail-fast: false
+      matrix:
+        arch: [ '386','amd64' ]
+        store: [ 'sqlite', 'postgres', 'mysql' ]
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Install Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: "1.23.x"
+          cache: false
+
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV     
+
+      - name: Cache Go modules
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-gotest-cache-
+
+      - name: Install dependencies
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
+
+      - name: Install 32-bit libpcap
+        if: matrix.arch == '386'
+        run: sudo dpkg --add-architecture i386 && sudo apt update && sudo apt-get install -y libpcap0.8-dev:i386
+
+      - name: Install modules
+        run: go mod tidy
+
+      - name: check git status
+        run: git --no-pager diff --exit-code
+
+      - name: Login to Docker hub
+        if: matrix.store == 'mysql' && (github.repository == github.head.repo.full_name || !github.head_ref)
+        uses: docker/login-action@v1
+        with:
+          username: ${{ secrets.DOCKER_USER }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: download mysql image
+        if: matrix.store == 'mysql'
+        run: docker pull mlsmaycon/warmed-mysql:8
+
+      - name: Test
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -tags=devcert -p 1 -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 10m $(go list ./... | grep /management)
+
+  benchmark:
+    needs: [ build-cache ]
+    strategy:
+      fail-fast: false
+      matrix:
+        arch: [ '386','amd64' ]
+        store: [ 'sqlite', 'postgres', 'mysql' ]
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Install Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: "1.23.x"
+          cache: false
+
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV     
+
+      - name: Cache Go modules
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-gotest-cache-
+
+      - name: Install dependencies
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
+
+      - name: Install 32-bit libpcap
+        if: matrix.arch == '386'
+        run: sudo dpkg --add-architecture i386 && sudo apt update && sudo apt-get install -y libpcap0.8-dev:i386
+
+      - name: Install modules
+        run: go mod tidy
+
+      - name: check git status
+        run: git --no-pager diff --exit-code
+
+      - name: Login to Docker hub
+        if: matrix.store == 'mysql' && (github.repository == github.head.repo.full_name || !github.head_ref)
+        uses: docker/login-action@v1
+        with:
+          username: ${{ secrets.DOCKER_USER }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: download mysql image
+        if: matrix.store == 'mysql'
+        run: docker pull mlsmaycon/warmed-mysql:8
+
+      - name: Test
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -tags devcert -run=^$ -bench=. -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 20m ./...
+
+  api_benchmark:
+    needs: [ build-cache ]
+    strategy:
+      fail-fast: false
+      matrix:
+        arch: [ '386','amd64' ]
+        store: [ 'sqlite', 'postgres' ]
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Install Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: "1.23.x"
+          cache: false
+
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV     
+
+      - name: Cache Go modules
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-gotest-cache-
+
+      - name: Install dependencies
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
+
+      - name: Install 32-bit libpcap
+        if: matrix.arch == '386'
+        run: sudo dpkg --add-architecture i386 && sudo apt update && sudo apt-get install -y libpcap0.8-dev:i386
+
+      - name: Install modules
+        run: go mod tidy
+
+      - name: check git status
+        run: git --no-pager diff --exit-code
+
+      - name: Login to Docker hub
+        if: matrix.store == 'mysql' && (github.repository == github.head.repo.full_name || !github.head_ref)
+        uses: docker/login-action@v1
+        with:
+          username: ${{ secrets.DOCKER_USER }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: download mysql image
+        if: matrix.store == 'mysql'
+        run: docker pull mlsmaycon/warmed-mysql:8
+        
+      - name: Test
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -run=^$ -tags=benchmark -bench=. -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 30m $(go list -tags=benchmark ./... | grep /management)
+
+  api_integration_test:
+    needs: [ build-cache ]
+    strategy:
+      fail-fast: false
+      matrix:
+        arch: [ '386','amd64' ]
+        store: [ 'sqlite', 'postgres']
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Install Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: "1.23.x"
+          cache: false
+
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV     
+
+      - name: Cache Go modules
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-gotest-cache-
+
+      - name: Install dependencies
+        run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
+
+      - name: Install 32-bit libpcap
+        if: matrix.arch == '386'
+        run: sudo dpkg --add-architecture i386 && sudo apt update && sudo apt-get install -y libpcap0.8-dev:i386
+
+      - name: Install modules
+        run: go mod tidy
+
+      - name: check git status
+        run: git --no-pager diff --exit-code
+
+      - name: Test
+        run: CGO_ENABLED=1 GOARCH=${{ matrix.arch }} NETBIRD_STORE_ENGINE=${{ matrix.store }} CI=true go test -tags=integration -p 1 -exec 'sudo --preserve-env=CI,NETBIRD_STORE_ENGINE' -timeout 30m $(go list -tags=integration ./... | grep /management)
 
   test_client_on_docker:
+    needs: [ build-cache ]
     runs-on: ubuntu-20.04
     steps:
       - name: Install Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21.x"
-
-      - name: Cache Go modules
-        uses: actions/cache@v3
-        with:
-          path: ~/go/pkg/mod
-          key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
-          restore-keys: |
-            ${{ runner.os }}-go-
+          go-version: "1.23.x"
+          cache: false
 
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $GITHUB_ENV     
+
+      - name: Cache Go modules
+        uses: actions/cache/restore@v4
+        with:
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-cache-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-gotest-cache-
 
       - name: Install dependencies
         run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev gcc-multilib libpcap-dev
@@ -79,14 +402,14 @@ jobs:
       - name: check git status
         run: git --no-pager diff --exit-code
 
-      - name: Generate Iface Test bin
-        run: CGO_ENABLED=0 go test -c -o iface-testing.bin ./iface/
-
       - name: Generate Shared Sock Test bin
         run: CGO_ENABLED=0 go test -c -o sharedsock-testing.bin ./sharedsock
 
       - name: Generate RouteManager Test bin
-        run: CGO_ENABLED=1 go test -c -o routemanager-testing.bin -tags netgo -ldflags '-w -extldflags "-static -ldbus-1 -lpcap"' ./client/internal/routemanager/...
+        run: CGO_ENABLED=0 go test -c -o routemanager-testing.bin  ./client/internal/routemanager
+
+      - name: Generate SystemOps Test bin
+        run: CGO_ENABLED=1 go test -c -o systemops-testing.bin -tags netgo -ldflags '-w -extldflags "-static -ldbus-1 -lpcap"' ./client/internal/routemanager/systemops
 
       - name: Generate nftables Manager Test bin
         run: CGO_ENABLED=0 go test -c -o nftablesmanager-testing.bin ./client/firewall/nftables/...
@@ -95,7 +418,7 @@ jobs:
         run: CGO_ENABLED=1 go test -c -o engine-testing.bin ./client/internal
 
       - name: Generate Peer Test bin
-        run: CGO_ENABLED=0 go test -c -o peer-testing.bin ./client/internal/peer/...
+        run: CGO_ENABLED=0 go test -c -o peer-testing.bin ./client/internal/peer/
 
       - run: chmod +x *testing.bin
 
@@ -103,11 +426,14 @@ jobs:
         run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/sharedsock --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/sharedsock-testing.bin -test.timeout 5m -test.parallel 1
 
       - name: Run Iface tests in docker
-        run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/iface --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/iface-testing.bin -test.timeout 5m -test.parallel 1
+        run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/netbird -v /tmp/cache:/tmp/cache -v /tmp/modcache:/tmp/modcache -w /netbird -e GOCACHE=/tmp/cache -e GOMODCACHE=/tmp/modcache -e CGO_ENABLED=0 golang:1.23-alpine go test  -test.timeout 5m -test.parallel 1 ./client/iface/...
 
       - name: Run RouteManager tests in docker
         run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/client/internal/routemanager --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/routemanager-testing.bin  -test.timeout 5m -test.parallel 1
 
+      - name: Run SystemOps tests in docker
+        run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/client/internal/routemanager/systemops --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/systemops-testing.bin  -test.timeout 5m -test.parallel 1
+
       - name: Run nftables Manager tests in docker
         run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/client/firewall --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/nftablesmanager-testing.bin  -test.timeout 5m -test.parallel 1
 
@@ -118,4 +444,4 @@ jobs:
         run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/client/internal -e NETBIRD_STORE_ENGINE="sqlite" --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/engine-testing.bin  -test.timeout 5m -test.parallel 1
 
       - name: Run Peer tests in docker
-        run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/client/internal/peer  --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/peer-testing.bin  -test.timeout 5m -test.parallel 1
+        run: docker run -t --cap-add=NET_ADMIN --privileged --rm -v $PWD:/ci -w /ci/client/internal/peer  --entrypoint /busybox/sh gcr.io/distroless/base:debug -c /ci/peer-testing.bin  -test.timeout 5m -test.parallel 1

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

@@ -17,13 +17,30 @@ jobs:
     runs-on: windows-latest
     steps:
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Install Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         id: go
         with:
-          go-version: "1.21.x"
+          go-version: "1.23.x"
+          cache: false
+
+      - name: Get Go environment
+        run: |
+          echo "cache=$(go env GOCACHE)" >> $env:GITHUB_ENV
+          echo "modcache=$(go env GOMODCACHE)" >> $env:GITHUB_ENV
+
+      - name: Cache Go modules
+        uses: actions/cache@v4
+        with:
+          path: |
+            ${{ env.cache }}
+            ${{ env.modcache }}
+          key: ${{ runner.os }}-gotest-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-gotest-
+            ${{ runner.os }}-go-
 
       - name: Download wintun
         uses: carlosperate/download-file-action@v2
@@ -42,11 +59,13 @@ jobs:
       - run: choco install -y sysinternals --ignore-checksums
       - run: choco install -y mingw
 
-      - run: PsExec64 -s -w ${{ github.workspace }} C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe env -w GOMODCACHE=C:\Users\runneradmin\go\pkg\mod
-      - run: PsExec64 -s -w ${{ github.workspace }} C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe env -w GOCACHE=C:\Users\runneradmin\AppData\Local\go-build
+      - run: PsExec64 -s -w ${{ github.workspace }} C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe env -w GOMODCACHE=${{ env.cache }}
+      - run: PsExec64 -s -w ${{ github.workspace }} C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe env -w GOCACHE=${{ env.modcache }}
+      - run: PsExec64 -s -w ${{ github.workspace }} C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe mod tidy
+      - run: echo "files=$(go list ./... | ForEach-Object { $_ } | Where-Object { $_ -notmatch '/management' })" >> $env:GITHUB_ENV
 
       - name: test
-        run: PsExec64 -s -w ${{ github.workspace }} cmd.exe /c "C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe test -timeout 10m -p 1 ./... > test-out.txt 2>&1"
+        run: PsExec64 -s -w ${{ github.workspace }} cmd.exe /c "C:\hostedtoolcache\windows\go\${{ steps.go.outputs.go-version }}\x64\bin\go.exe test -tags=devcert -timeout 10m -p 1 ${{ env.files }} > test-out.txt 2>&1"
       - name: test output
         if: ${{ always() }}
         run: Get-Content test-out.txt

.github/workflows/golangci-lint.yml

@@ -15,11 +15,11 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - name: codespell
         uses: codespell-project/actions-codespell@v2
         with:
-          ignore_words_list: erro,clienta
+          ignore_words_list: erro,clienta,hastable,iif,groupd,testin
           skip: go.mod,go.sum
           only_warn: 1
   golangci:
@@ -32,21 +32,21 @@ jobs:
     timeout-minutes: 15
     steps:
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - name: Check for duplicate constants
         if: matrix.os == 'ubuntu-latest'
         run: |
           ! awk '/const \(/,/)/{print $0}' management/server/activity/codes.go | grep -o '= [0-9]*' | sort | uniq -d | grep .
       - name: Install Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21.x"
+          go-version: "1.23.x"
           cache: false
       - name: Install dependencies
         if: matrix.os == 'ubuntu-latest'
         run: sudo apt update && sudo apt install -y -q libgtk-3-dev libayatana-appindicator3-dev libgl1-mesa-dev xorg-dev libpcap-dev
       - name: golangci-lint
-        uses: golangci/golangci-lint-action@v3
+        uses: golangci/golangci-lint-action@v4
         with:
           version: latest
-          args: --timeout=12m
+          args: --timeout=12m --out-format colored-line-number

.github/workflows/install-script-test.yml

@@ -13,6 +13,7 @@ concurrency:
 jobs:
   test-install-script:
     strategy:
+      fail-fast: false
       max-parallel: 2
       matrix:
         os: [ubuntu-latest, macos-latest]
@@ -21,7 +22,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     steps:
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: run install script
         env:

.github/workflows/mobile-build-validation.yml

@@ -15,30 +15,30 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - name: Install Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21.x"
+          go-version: "1.23.x"
       - name: Setup Android SDK
         uses: android-actions/setup-android@v3
         with:
           cmdline-tools-version: 8512546
       - name: Setup Java
-        uses: actions/setup-java@v3
+        uses: actions/setup-java@v4
         with:
           java-version: "11"
           distribution: "adopt"
       - name: NDK Cache
         id: ndk-cache
-        uses: actions/cache@v3
+        uses: actions/cache@v4
         with:
           path: /usr/local/lib/android/sdk/ndk
           key: ndk-cache-23.1.7779620
       - name: Setup NDK
         run: /usr/local/lib/android/sdk/cmdline-tools/7.0/bin/sdkmanager --install "ndk;23.1.7779620"
       - name: install gomobile
-        run: go install golang.org/x/mobile/cmd/gomobile@v0.0.0-20230531173138-3c911d8e3eda
+        run: go install golang.org/x/mobile/cmd/gomobile@v0.0.0-20240404231514-09dbf07665ed
       - name: gomobile init
         run: gomobile init
       - name: build android netbird lib
@@ -50,16 +50,16 @@ jobs:
     runs-on: macos-latest
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - name: Install Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21.x"
+          go-version: "1.23.x"
       - name: install gomobile
-        run: go install golang.org/x/mobile/cmd/gomobile@v0.0.0-20230531173138-3c911d8e3eda
+        run: go install golang.org/x/mobile/cmd/gomobile@v0.0.0-20240404231514-09dbf07665ed
       - name: gomobile init
         run: gomobile init
       - name: build iOS netbird lib
-        run: PATH=$PATH:$(go env GOPATH) gomobile bind -target=ios -bundleid=io.netbird.framework -ldflags="-X github.com/netbirdio/netbird/version.version=buildtest" -o $GITHUB_WORKSPACE/NetBirdSDK.xcframework $GITHUB_WORKSPACE/client/ios/NetBirdSDK
+        run: PATH=$PATH:$(go env GOPATH) gomobile bind -target=ios -bundleid=io.netbird.framework -ldflags="-X github.com/netbirdio/netbird/version.version=buildtest" -o ./NetBirdSDK.xcframework ./client/ios/NetBirdSDK
         env:
-          CGO_ENABLED: 0
+          CGO_ENABLED: 0

.github/workflows/release.yml

@@ -3,25 +3,16 @@ name: Release
 on:
   push:
     tags:
-      - 'v*'
+      - "v*"
     branches:
       - main
   pull_request:
-    paths:
-      - 'go.mod'
-      - 'go.sum'
-      - '.goreleaser.yml'
-      - '.goreleaser_ui.yaml'
-      - '.goreleaser_ui_darwin.yaml'
-      - '.github/workflows/release.yml'
-      - 'release_files/**'
-      - '**/Dockerfile'
-      - '**/Dockerfile.*'
-      - 'client/ui/**'
 
 env:
-  SIGN_PIPE_VER: "v0.0.11"
-  GORELEASER_VER: "v1.14.1"
+  SIGN_PIPE_VER: "v0.0.18"
+  GORELEASER_VER: "v2.3.2"
+  PRODUCT_NAME: "NetBird"
+  COPYRIGHT: "NetBird GmbH"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.ref }}-${{ github.head_ref || github.actor_id }}
@@ -29,26 +20,30 @@ concurrency:
 
 jobs:
   release:
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-22.04
     env:
       flags: ""
     steps:
+      - name: Parse semver string
+        id: semver_parser
+        uses: booxmedialtd/ws-action-parse-semver@v1
+        with:
+          input_string: ${{ (startsWith(github.ref, 'refs/tags/v') && github.ref) || 'refs/tags/v0.0.0' }}
+          version_extractor_regex: '\/v(.*)$'
+
       - if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
         run: echo "flags=--snapshot" >> $GITHUB_ENV
-      -
-        name: Checkout
-        uses: actions/checkout@v3
+      - name: Checkout
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0 # It is required for GoReleaser to work properly
-      -
-        name: Set up Go
-        uses: actions/setup-go@v4
+      - name: Set up Go
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21"
+          go-version: "1.23"
           cache: false
-      -
-        name: Cache Go modules
-        uses: actions/cache@v3
+      - name: Cache Go modules
+        uses: actions/cache@v4
         with:
           path: |
             ~/go/pkg/mod
@@ -56,76 +51,88 @@ jobs:
           key: ${{ runner.os }}-go-releaser-${{ hashFiles('**/go.sum') }}
           restore-keys: |
             ${{ runner.os }}-go-releaser-
-      -
-        name: Install modules
+      - name: Install modules
         run: go mod tidy
-      -
-        name: check git status
+      - name: check git status
         run: git --no-pager diff --exit-code
-      -
-        name: Set up QEMU
+      - name: Set up QEMU
         uses: docker/setup-qemu-action@v2
-      -
-        name: Set up Docker Buildx
+      - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v2
-      -
-        name: Login to Docker hub
+      - name: Login to Docker hub
         if: github.event_name != 'pull_request'
         uses: docker/login-action@v1
         with:
-          username: netbirdio
+          username: ${{ secrets.DOCKER_USER }}
           password: ${{ secrets.DOCKER_TOKEN }}
       - name: Install OS build dependencies
         run: sudo apt update && sudo apt install -y -q gcc-arm-linux-gnueabihf gcc-aarch64-linux-gnu
 
-      - name: Install rsrc
-        run: go install github.com/akavel/rsrc@v0.10.2
-      - name: Generate windows rsrc amd64
-        run: rsrc -arch amd64 -ico client/ui/netbird.ico -manifest client/manifest.xml -o client/resources_windows_amd64.syso
-      - name: Generate windows rsrc arm64
-        run: rsrc -arch arm64 -ico client/ui/netbird.ico -manifest client/manifest.xml -o client/resources_windows_arm64.syso
-      - name: Generate windows rsrc arm
-        run: rsrc -arch arm -ico client/ui/netbird.ico -manifest client/manifest.xml -o client/resources_windows_arm.syso
-      - name: Generate windows rsrc 386
-        run: rsrc -arch 386 -ico client/ui/netbird.ico -manifest client/manifest.xml -o client/resources_windows_386.syso
-      -
-        name: Run GoReleaser
+      - name: Install goversioninfo
+        run: go install github.com/josephspurrier/goversioninfo/cmd/goversioninfo@233067e
+      - name: Generate windows syso amd64
+        run: goversioninfo  -icon client/ui/netbird.ico -manifest client/manifest.xml -product-name ${{ env.PRODUCT_NAME }} -copyright "${{ env.COPYRIGHT }}" -ver-major ${{ steps.semver_parser.outputs.major }} -ver-minor ${{ steps.semver_parser.outputs.minor }} -ver-patch ${{ steps.semver_parser.outputs.patch }} -ver-build 0 -file-version ${{ steps.semver_parser.outputs.fullversion }}.0 -product-version ${{ steps.semver_parser.outputs.fullversion }}.0 -o client/resources_windows_amd64.syso
+      - name: Run GoReleaser
         uses: goreleaser/goreleaser-action@v4
         with:
           version: ${{ env.GORELEASER_VER }}
-          args: release --rm-dist ${{ env.flags }}
+          args: release --clean ${{ env.flags }}
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           HOMEBREW_TAP_GITHUB_TOKEN: ${{ secrets.HOMEBREW_TAP_GITHUB_TOKEN }}
           UPLOAD_DEBIAN_SECRET: ${{ secrets.PKG_UPLOAD_SECRET }}
           UPLOAD_YUM_SECRET: ${{ secrets.PKG_UPLOAD_SECRET }}
-      -
-        name: upload non tags for debug purposes
-        uses: actions/upload-artifact@v3
+      - name: upload non tags for debug purposes
+        uses: actions/upload-artifact@v4
         with:
           name: release
           path: dist/
           retention-days: 3
+      - name: upload linux packages
+        uses: actions/upload-artifact@v4
+        with:
+          name: linux-packages
+          path: dist/netbird_linux**
+          retention-days: 3
+      - name: upload windows packages
+        uses: actions/upload-artifact@v4
+        with:
+          name: windows-packages
+          path: dist/netbird_windows**
+          retention-days: 3
+      - name: upload macos packages
+        uses: actions/upload-artifact@v4
+        with:
+          name: macos-packages
+          path: dist/netbird_darwin**
+          retention-days: 3
 
   release_ui:
     runs-on: ubuntu-latest
     steps:
+      - name: Parse semver string
+        id: semver_parser
+        uses: booxmedialtd/ws-action-parse-semver@v1
+        with:
+          input_string: ${{ (startsWith(github.ref, 'refs/tags/v') && github.ref) || 'refs/tags/v0.0.0' }}
+          version_extractor_regex: '\/v(.*)$'
+
       - if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
         run: echo "flags=--snapshot" >> $GITHUB_ENV
       - name: Checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0 # It is required for GoReleaser to work properly
 
       - name: Set up Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21"
+          go-version: "1.23"
           cache: false
       - name: Cache Go modules
-        uses: actions/cache@v3
+        uses: actions/cache@v4
         with:
-          path: | 
+          path: |
             ~/go/pkg/mod
             ~/.cache/go-build
           key: ${{ runner.os }}-ui-go-releaser-${{ hashFiles('**/go.sum') }}
@@ -140,46 +147,44 @@ jobs:
 
       - name: Install dependencies
         run: sudo apt update && sudo apt install -y -q libappindicator3-dev gir1.2-appindicator3-0.1 libxxf86vm-dev gcc-mingw-w64-x86-64
-      - name: Install rsrc
-        run: go install github.com/akavel/rsrc@v0.10.2
-      - name: Generate windows rsrc
-        run: rsrc -arch amd64 -ico client/ui/netbird.ico -manifest client/ui/manifest.xml -o client/ui/resources_windows_amd64.syso
+      - name: Install goversioninfo
+        run: go install github.com/josephspurrier/goversioninfo/cmd/goversioninfo@233067e
+      - name: Generate windows syso amd64
+        run: goversioninfo -64 -icon client/ui/netbird.ico -manifest client/ui/manifest.xml -product-name ${{ env.PRODUCT_NAME }}-"UI" -copyright "${{ env.COPYRIGHT }}" -ver-major ${{ steps.semver_parser.outputs.major }} -ver-minor ${{ steps.semver_parser.outputs.minor }} -ver-patch ${{ steps.semver_parser.outputs.patch }} -ver-build 0 -file-version ${{ steps.semver_parser.outputs.fullversion }}.0 -product-version ${{ steps.semver_parser.outputs.fullversion }}.0 -o client/ui/resources_windows_amd64.syso
+
       - name: Run GoReleaser
         uses: goreleaser/goreleaser-action@v4
         with:
           version: ${{ env.GORELEASER_VER }}
-          args: release --config .goreleaser_ui.yaml --rm-dist ${{ env.flags }}
+          args: release --config .goreleaser_ui.yaml --clean ${{ env.flags }}
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           HOMEBREW_TAP_GITHUB_TOKEN: ${{ secrets.HOMEBREW_TAP_GITHUB_TOKEN }}
           UPLOAD_DEBIAN_SECRET: ${{ secrets.PKG_UPLOAD_SECRET }}
           UPLOAD_YUM_SECRET: ${{ secrets.PKG_UPLOAD_SECRET }}
       - name: upload non tags for debug purposes
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
           name: release-ui
           path: dist/
           retention-days: 3
 
   release_ui_darwin:
-    runs-on: macos-11
+    runs-on: macos-latest
     steps:
       - if: ${{ !startsWith(github.ref, 'refs/tags/v') }}
         run: echo "flags=--snapshot" >> $GITHUB_ENV
-      -
-        name: Checkout
-        uses: actions/checkout@v3
+      - name: Checkout
+        uses: actions/checkout@v4
         with:
           fetch-depth: 0 # It is required for GoReleaser to work properly
-      -
-        name: Set up Go
-        uses: actions/setup-go@v4
+      - name: Set up Go
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21"
+          go-version: "1.23"
           cache: false
-      -
-        name: Cache Go modules
-        uses: actions/cache@v3
+      - name: Cache Go modules
+        uses: actions/cache@v4
         with:
           path: |
             ~/go/pkg/mod
@@ -187,53 +192,35 @@ jobs:
           key: ${{ runner.os }}-ui-go-releaser-darwin-${{ hashFiles('**/go.sum') }}
           restore-keys: |
             ${{ runner.os }}-ui-go-releaser-darwin-
-      -
-        name: Install modules
+      - name: Install modules
         run: go mod tidy
-      - 
-        name: check git status
+      - name: check git status
         run: git --no-pager diff --exit-code
-      -
-        name: Run GoReleaser
+      - name: Run GoReleaser
         id: goreleaser
         uses: goreleaser/goreleaser-action@v4
         with:
           version: ${{ env.GORELEASER_VER }}
-          args: release --config .goreleaser_ui_darwin.yaml --rm-dist ${{ env.flags }}
+          args: release --config .goreleaser_ui_darwin.yaml --clean ${{ env.flags }}
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-      -
-        name: upload non tags for debug purposes
-        uses: actions/upload-artifact@v3
+      - name: upload non tags for debug purposes
+        uses: actions/upload-artifact@v4
         with:
           name: release-ui-darwin
           path: dist/
           retention-days: 3
 
-  trigger_windows_signer:
+  trigger_signer:
     runs-on: ubuntu-latest
-    needs: [release,release_ui]
+    needs: [release, release_ui, release_ui_darwin]
     if: startsWith(github.ref, 'refs/tags/')
     steps:
-      - name: Trigger Windows binaries sign pipeline
+      - name: Trigger binaries sign pipelines
         uses: benc-uk/workflow-dispatch@v1
         with:
-          workflow: Sign windows bin and installer
+          workflow: Sign bin and installer
           repo: netbirdio/sign-pipelines
           ref: ${{ env.SIGN_PIPE_VER }}
           token: ${{ secrets.SIGN_GITHUB_TOKEN }}
-          inputs: '{ "tag": "${{ github.ref }}" }'
-
-  trigger_darwin_signer:
-    runs-on: ubuntu-latest
-    needs: [release,release_ui_darwin]
-    if: startsWith(github.ref, 'refs/tags/')
-    steps:
-      - name: Trigger Darwin App binaries sign pipeline
-        uses: benc-uk/workflow-dispatch@v1
-        with:
-          workflow: Sign darwin ui app with dispatch
-          repo: netbirdio/sign-pipelines
-          ref: ${{ env.SIGN_PIPE_VER }}
-          token: ${{ secrets.SIGN_GITHUB_TOKEN }}
-          inputs: '{ "tag": "${{ github.ref }}" }'
+          inputs: '{ "tag": "${{ github.ref }}", "skipRelease": false }'

.github/workflows/test-infrastructure-files.yml

@@ -18,7 +18,49 @@ concurrency:
 jobs:
   test-docker-compose:
     runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        store: [ 'sqlite', 'postgres', 'mysql' ]
+    services:
+      postgres:
+        image: ${{ (matrix.store == 'postgres') && 'postgres' || '' }}
+        env:
+          POSTGRES_USER: netbird
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_DB: netbird
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+        ports:
+          - 5432:5432
+      mysql:
+        image: ${{ (matrix.store == 'mysql') && 'mysql' || '' }}
+        env:
+          MYSQL_USER: netbird
+          MYSQL_PASSWORD: mysql
+          MYSQL_ROOT_PASSWORD: mysqlroot
+          MYSQL_DATABASE: netbird
+        options: >-
+          --health-cmd "mysqladmin ping --silent"
+          --health-interval 10s
+          --health-timeout 5s
+        ports:
+          - 3306:3306
     steps:
+      - name: Set Database Connection String
+        run: |
+          if [ "${{ matrix.store }}" == "postgres" ]; then
+            echo "NETBIRD_STORE_ENGINE_POSTGRES_DSN=host=$(hostname -I | awk '{print $1}') user=netbird password=postgres dbname=netbird port=5432" >> $GITHUB_ENV
+          else
+            echo "NETBIRD_STORE_ENGINE_POSTGRES_DSN==" >> $GITHUB_ENV
+          fi
+          if [ "${{ matrix.store }}" == "mysql" ]; then
+            echo "NETBIRD_STORE_ENGINE_MYSQL_DSN=netbird:mysql@tcp($(hostname -I | awk '{print $1}'):3306)/netbird" >> $GITHUB_ENV
+          else
+            echo "NETBIRD_STORE_ENGINE_MYSQL_DSN==" >> $GITHUB_ENV
+          fi
+
       - name: Install jq
         run: sudo apt-get install -y jq
 
@@ -26,12 +68,12 @@ jobs:
         run: sudo apt-get install -y curl
 
       - name: Install Go
-        uses: actions/setup-go@v4
+        uses: actions/setup-go@v5
         with:
-          go-version: "1.21.x"
+          go-version: "1.23.x"
 
       - name: Cache Go modules
-        uses: actions/cache@v3
+        uses: actions/cache@v4
         with:
           path: ~/go/pkg/mod
           key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
@@ -39,7 +81,7 @@ jobs:
             ${{ runner.os }}-go-
 
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: cp setup.env
         run: cp infrastructure_files/tests/setup.env infrastructure_files/
@@ -58,7 +100,9 @@ jobs:
           CI_NETBIRD_IDP_MGMT_CLIENT_ID: testing.client.id
           CI_NETBIRD_IDP_MGMT_CLIENT_SECRET: testing.client.secret
           CI_NETBIRD_AUTH_SUPPORTED_SCOPES: "openid profile email offline_access api email_verified"
-          CI_NETBIRD_STORE_CONFIG_ENGINE: "sqlite"
+          CI_NETBIRD_STORE_CONFIG_ENGINE: ${{ matrix.store }}
+          NETBIRD_STORE_ENGINE_POSTGRES_DSN: ${{ env.NETBIRD_STORE_ENGINE_POSTGRES_DSN }}
+          NETBIRD_STORE_ENGINE_MYSQL_DSN: ${{ env.NETBIRD_STORE_ENGINE_MYSQL_DSN }}
           CI_NETBIRD_MGMT_IDP_SIGNKEY_REFRESH: false
 
       - name: check values
@@ -85,7 +129,9 @@ jobs:
           CI_NETBIRD_IDP_MGMT_CLIENT_ID: testing.client.id
           CI_NETBIRD_IDP_MGMT_CLIENT_SECRET: testing.client.secret
           CI_NETBIRD_SIGNAL_PORT: 12345
-          CI_NETBIRD_STORE_CONFIG_ENGINE: "sqlite"
+          CI_NETBIRD_STORE_CONFIG_ENGINE: ${{ matrix.store }}
+          NETBIRD_STORE_ENGINE_POSTGRES_DSN: '${{ env.NETBIRD_STORE_ENGINE_POSTGRES_DSN }}$'
+          NETBIRD_STORE_ENGINE_MYSQL_DSN: '${{ env.NETBIRD_STORE_ENGINE_MYSQL_DSN }}$'
           CI_NETBIRD_MGMT_IDP_SIGNKEY_REFRESH: false
           CI_NETBIRD_TURN_EXTERNAL_IP: "1.2.3.4"
 
@@ -123,6 +169,15 @@ jobs:
           grep -A 10 PKCEAuthorizationFlow management.json | grep -A 10 ProviderConfig | grep Scope | grep "$CI_NETBIRD_AUTH_SUPPORTED_SCOPES"
           grep -A 10 PKCEAuthorizationFlow management.json | grep -A 10 ProviderConfig | grep -A 3 RedirectURLs | grep "http://localhost:53000"
           grep "external-ip" turnserver.conf | grep $CI_NETBIRD_TURN_EXTERNAL_IP
+          grep "NETBIRD_STORE_ENGINE_MYSQL_DSN=$NETBIRD_STORE_ENGINE_MYSQL_DSN" docker-compose.yml
+          grep NETBIRD_STORE_ENGINE_POSTGRES_DSN docker-compose.yml | egrep "$NETBIRD_STORE_ENGINE_POSTGRES_DSN"
+          # check relay values
+          grep "NB_EXPOSED_ADDRESS=$CI_NETBIRD_DOMAIN:33445" docker-compose.yml
+          grep "NB_LISTEN_ADDRESS=:33445" docker-compose.yml
+          grep '33445:33445' docker-compose.yml
+          grep -A 10 'relay:' docker-compose.yml | egrep 'NB_AUTH_SECRET=.+$'
+          grep -A 7 Relay management.json | grep "rel://$CI_NETBIRD_DOMAIN:33445"
+          grep -A 7 Relay management.json | egrep '"Secret": ".+"'
 
       - name: Install modules
         run: go mod tidy
@@ -148,26 +203,35 @@ jobs:
         run: |
           docker build -t netbirdio/signal:latest .
 
+      - name: Build relay binary
+        working-directory: relay
+        run: CGO_ENABLED=0 go build -o netbird-relay main.go
+
+      - name: Build relay docker image
+        working-directory: relay
+        run: |
+          docker build -t netbirdio/relay:latest .
+
       - name: run docker compose up
         working-directory: infrastructure_files/artifacts
         run: |
-          docker-compose up -d
+          docker compose up -d
           sleep 5
-          docker-compose ps
-          docker-compose logs --tail=20
+          docker compose ps
+          docker compose logs --tail=20
 
       - name: test running containers
         run: |
           count=$(docker compose ps --format json | jq '. | select(.Name | contains("artifacts")) | .State' | grep -c running)
-          test $count -eq 4
+          test $count -eq 5 || docker compose logs
         working-directory: infrastructure_files/artifacts
 
       - name: test geolocation databases
         working-directory: infrastructure_files/artifacts
         run: |
           sleep 30
-          docker compose exec management ls -l /var/lib/netbird/ | grep -i GeoLite2-City.mmdb
-          docker compose exec management ls -l /var/lib/netbird/ | grep -i geonames.db
+          docker compose exec management ls -l /var/lib/netbird/ | grep -i GeoLite2-City_[0-9]*.mmdb
+          docker compose exec management ls -l /var/lib/netbird/ | grep -i geonames_[0-9]*.db
 
   test-getting-started-script:
     runs-on: ubuntu-latest
@@ -176,36 +240,69 @@ jobs:
         run: sudo apt-get install -y jq
 
       - name: Checkout code
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
-      - name: run script
+      - name: run script with Zitadel PostgreSQL
         run: NETBIRD_DOMAIN=use-ip bash -x infrastructure_files/getting-started-with-zitadel.sh
 
-      - name: test Caddy file gen
+      - name: test Caddy file gen postgres
         run: test -f Caddyfile
-      - name: test docker-compose file gen
+
+      - name: test docker-compose file gen postgres
         run: test -f docker-compose.yml
-      - name: test management.json file gen
+
+      - name: test management.json file gen postgres
         run: test -f management.json
-      - name: test turnserver.conf file gen
-        run: | 
+
+      - name: test turnserver.conf file gen postgres
+        run: |
           set -x
           test -f turnserver.conf
           grep external-ip turnserver.conf
-      - name: test zitadel.env file gen
+
+      - name: test zitadel.env file gen postgres
         run: test -f zitadel.env
-      - name: test dashboard.env file gen
+
+      - name: test dashboard.env file gen postgres
         run: test -f dashboard.env
-  test-download-geolite2-script:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Install jq
-        run: sudo apt-get update && sudo apt-get install -y unzip sqlite3
-      - name: Checkout code
-        uses: actions/checkout@v3
-      - name: test script
-        run: bash -x infrastructure_files/download-geolite2.sh
-      - name: test mmdb file exists
-        run: test -f GeoLite2-City.mmdb
-      - name: test geonames file exists
-        run: test -f geonames.db
+
+      - name: test relay.env file gen postgres
+        run: test -f relay.env
+
+      - name: test zdb.env file gen postgres
+        run: test -f zdb.env
+
+      - name: Postgres run cleanup
+        run: |
+          docker compose down --volumes --rmi all
+          rm -rf docker-compose.yml Caddyfile zitadel.env dashboard.env machinekey/zitadel-admin-sa.token turnserver.conf management.json zdb.env
+
+      - name: run script with Zitadel CockroachDB
+        run: bash -x infrastructure_files/getting-started-with-zitadel.sh
+        env:
+          NETBIRD_DOMAIN: use-ip
+          ZITADEL_DATABASE: cockroach
+
+      - name: test Caddy file gen CockroachDB
+        run: test -f Caddyfile
+
+      - name: test docker-compose file gen CockroachDB
+        run: test -f docker-compose.yml
+
+      - name: test management.json file gen CockroachDB
+        run: test -f management.json
+
+      - name: test turnserver.conf file gen CockroachDB
+        run: |
+          set -x
+          test -f turnserver.conf
+          grep external-ip turnserver.conf
+
+      - name: test zitadel.env file gen CockroachDB
+        run: test -f zitadel.env
+
+      - name: test dashboard.env file gen CockroachDB
+        run: test -f dashboard.env
+
+      - name: test relay.env file gen CockroachDB
+        run: test -f relay.env

.gitignore

@@ -29,4 +29,3 @@ infrastructure_files/setup.env
 infrastructure_files/setup-*.env
 .vscode
 .DS_Store
-GeoLite2-City*

.golangci.yaml

@@ -130,3 +130,10 @@ issues:
     - path: mock\.go
       linters:
       - nilnil
+    # Exclude specific deprecation warnings for grpc methods
+    - linters:
+       - staticcheck
+      text: "grpc.DialContext is deprecated"
+    - linters:
+        - staticcheck
+      text: "grpc.WithBlock is deprecated"

.goreleaser.yaml

@@ -1,3 +1,5 @@
+version: 2
+
 project_name: netbird
 builds:
   - id: netbird
@@ -22,7 +24,7 @@ builds:
         goarch: 386
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
-    mod_timestamp: '{{ .CommitTimestamp }}'
+    mod_timestamp: "{{ .CommitTimestamp }}"
     tags:
       - load_wgnt_from_rsrc
 
@@ -42,19 +44,19 @@ builds:
       - softfloat
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
-    mod_timestamp: '{{ .CommitTimestamp }}'
+    mod_timestamp: "{{ .CommitTimestamp }}"
     tags:
       - load_wgnt_from_rsrc
 
   - id: netbird-mgmt
     dir: management
     env:
-    - CGO_ENABLED=1
-    - >-
-      {{- if eq .Runtime.Goos "linux" }}
-        {{- if eq .Arch "arm64"}}CC=aarch64-linux-gnu-gcc{{- end }}
-        {{- if eq .Arch "arm"}}CC=arm-linux-gnueabihf-gcc{{- end }}
-      {{- end }}
+      - CGO_ENABLED=1
+      - >-
+        {{- if eq .Runtime.Goos "linux" }}
+          {{- if eq .Arch "arm64"}}CC=aarch64-linux-gnu-gcc{{- end }}
+          {{- if eq .Arch "arm"}}CC=arm-linux-gnueabihf-gcc{{- end }}
+        {{- end }}
     binary: netbird-mgmt
     goos:
       - linux
@@ -64,7 +66,7 @@ builds:
       - arm
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
-    mod_timestamp: '{{ .CommitTimestamp }}'
+    mod_timestamp: "{{ .CommitTimestamp }}"
 
   - id: netbird-signal
     dir: signal
@@ -78,7 +80,24 @@ builds:
       - arm
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
-    mod_timestamp: '{{ .CommitTimestamp }}'
+    mod_timestamp: "{{ .CommitTimestamp }}"
+
+  - id: netbird-relay
+    dir: relay
+    env: [CGO_ENABLED=0]
+    binary: netbird-relay
+    goos:
+      - linux
+    goarch:
+      - amd64
+      - arm64
+      - arm
+    ldflags:
+      - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
+    mod_timestamp: "{{ .CommitTimestamp }}"
+
+universal_binaries:
+  - id: netbird
 
 archives:
   - builds:
@@ -86,7 +105,6 @@ archives:
       - netbird-static
 
 nfpms:
-
   - maintainer: Netbird <dev@netbird.io>
     description: Netbird client.
     homepage: https://netbird.io/
@@ -161,6 +179,97 @@ dockers:
       - "--label=org.opencontainers.image.revision={{.FullCommit}}"
       - "--label=org.opencontainers.image.version={{.Version}}"
       - "--label=maintainer=dev@netbird.io"
+
+  - image_templates:
+      - netbirdio/netbird:{{ .Version }}-rootless-amd64
+    ids:
+      - netbird
+    goarch: amd64
+    use: buildx
+    dockerfile: client/Dockerfile-rootless
+    build_flag_templates:
+      - "--platform=linux/amd64"
+      - "--label=org.opencontainers.image.created={{.Date}}"
+      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+      - "--label=maintainer=dev@netbird.io"
+  - image_templates:
+      - netbirdio/netbird:{{ .Version }}-rootless-arm64v8
+    ids:
+      - netbird
+    goarch: arm64
+    use: buildx
+    dockerfile: client/Dockerfile-rootless
+    build_flag_templates:
+      - "--platform=linux/arm64"
+      - "--label=org.opencontainers.image.created={{.Date}}"
+      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+      - "--label=maintainer=dev@netbird.io"
+  - image_templates:
+      - netbirdio/netbird:{{ .Version }}-rootless-arm
+    ids:
+      - netbird
+    goarch: arm
+    goarm: 6
+    use: buildx
+    dockerfile: client/Dockerfile-rootless
+    build_flag_templates:
+      - "--platform=linux/arm"
+      - "--label=org.opencontainers.image.created={{.Date}}"
+      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+      - "--label=maintainer=dev@netbird.io"
+
+  - image_templates:
+      - netbirdio/relay:{{ .Version }}-amd64
+    ids:
+      - netbird-relay
+    goarch: amd64
+    use: buildx
+    dockerfile: relay/Dockerfile
+    build_flag_templates:
+      - "--platform=linux/amd64"
+      - "--label=org.opencontainers.image.created={{.Date}}"
+      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=maintainer=dev@netbird.io"
+  - image_templates:
+      - netbirdio/relay:{{ .Version }}-arm64v8
+    ids:
+      - netbird-relay
+    goarch: arm64
+    use: buildx
+    dockerfile: relay/Dockerfile
+    build_flag_templates:
+      - "--platform=linux/arm64"
+      - "--label=org.opencontainers.image.created={{.Date}}"
+      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=maintainer=dev@netbird.io"
+  - image_templates:
+      - netbirdio/relay:{{ .Version }}-arm
+    ids:
+      - netbird-relay
+    goarch: arm
+    goarm: 6
+    use: buildx
+    dockerfile: relay/Dockerfile
+    build_flag_templates:
+      - "--platform=linux/arm"
+      - "--label=org.opencontainers.image.created={{.Date}}"
+      - "--label=org.opencontainers.image.title={{.ProjectName}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=org.opencontainers.image.revision={{.FullCommit}}"
+      - "--label=org.opencontainers.image.version={{.Version}}"
+      - "--label=maintainer=dev@netbird.io"
   - image_templates:
       - netbirdio/signal:{{ .Version }}-amd64
     ids:
@@ -313,6 +422,30 @@ docker_manifests:
       - netbirdio/netbird:{{ .Version }}-arm
       - netbirdio/netbird:{{ .Version }}-amd64
 
+  - name_template: netbirdio/netbird:{{ .Version }}-rootless
+    image_templates:
+      - netbirdio/netbird:{{ .Version }}-rootless-arm64v8
+      - netbirdio/netbird:{{ .Version }}-rootless-arm
+      - netbirdio/netbird:{{ .Version }}-rootless-amd64
+
+  - name_template: netbirdio/netbird:rootless-latest
+    image_templates:
+      - netbirdio/netbird:{{ .Version }}-rootless-arm64v8
+      - netbirdio/netbird:{{ .Version }}-rootless-arm
+      - netbirdio/netbird:{{ .Version }}-rootless-amd64
+
+  - name_template: netbirdio/relay:{{ .Version }}
+    image_templates:
+      - netbirdio/relay:{{ .Version }}-arm64v8
+      - netbirdio/relay:{{ .Version }}-arm
+      - netbirdio/relay:{{ .Version }}-amd64
+
+  - name_template: netbirdio/relay:latest
+    image_templates:
+      - netbirdio/relay:{{ .Version }}-arm64v8
+      - netbirdio/relay:{{ .Version }}-arm
+      - netbirdio/relay:{{ .Version }}-amd64
+
   - name_template: netbirdio/signal:{{ .Version }}
     image_templates:
       - netbirdio/signal:{{ .Version }}-arm64v8
@@ -344,10 +477,9 @@ docker_manifests:
       - netbirdio/management:{{ .Version }}-debug-amd64
 
 brews:
-  -
-    ids:
+  - ids:
       - default
-    tap:
+    repository:
       owner: netbirdio
       name: homebrew-tap
       token: "{{ .Env.HOMEBREW_TAP_GITHUB_TOKEN }}"
@@ -364,7 +496,7 @@ brews:
 uploads:
   - name: debian
     ids:
-    - netbird-deb
+      - netbird-deb
     mode: archive
     target: https://pkgs.wiretrustee.com/debian/pool/{{ .ArtifactName }};deb.distribution=stable;deb.component=main;deb.architecture={{ if .Arm }}armhf{{ else }}{{ .Arch }}{{ end }};deb.package=
     username: dev@wiretrustee.com
@@ -386,4 +518,4 @@ checksum:
 release:
   extra_files:
     - glob: ./infrastructure_files/getting-started-with-zitadel.sh
-    - glob: ./release_files/install.sh
+    - glob: ./release_files/install.sh

.goreleaser_ui.yaml

@@ -1,3 +1,5 @@
+version: 2
+
 project_name: netbird-ui
 builds:
   - id: netbird-ui
@@ -11,9 +13,7 @@ builds:
       - amd64
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
-    tags:
-      - legacy_appindicator
-    mod_timestamp: '{{ .CommitTimestamp }}'
+    mod_timestamp: "{{ .CommitTimestamp }}"
 
   - id: netbird-ui-windows
     dir: client/ui
@@ -28,7 +28,7 @@ builds:
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
       - -H windowsgui
-    mod_timestamp: '{{ .CommitTimestamp }}'
+    mod_timestamp: "{{ .CommitTimestamp }}"
 
 archives:
   - id: linux-arch
@@ -41,7 +41,6 @@ archives:
       - netbird-ui-windows
 
 nfpms:
-
   - maintainer: Netbird <dev@netbird.io>
     description: Netbird client UI.
     homepage: https://netbird.io/
@@ -79,7 +78,7 @@ nfpms:
 uploads:
   - name: debian
     ids:
-    - netbird-ui-deb
+      - netbird-ui-deb
     mode: archive
     target: https://pkgs.wiretrustee.com/debian/pool/{{ .ArtifactName }};deb.distribution=stable;deb.component=main;deb.architecture={{ if .Arm }}armhf{{ else }}{{ .Arch }}{{ end }};deb.package=
     username: dev@wiretrustee.com

.goreleaser_ui_darwin.yaml

@@ -1,10 +1,14 @@
+version: 2
+
 project_name: netbird-ui
 builds:
   - id: netbird-ui-darwin
     dir: client/ui
     binary: netbird-ui
-    env: [CGO_ENABLED=1]
-
+    env:
+      - CGO_ENABLED=1
+      - MACOSX_DEPLOYMENT_TARGET=11.0
+      - MACOS_DEPLOYMENT_TARGET=11.0
     goos:
       - darwin
     goarch:
@@ -15,10 +19,13 @@ builds:
       - softfloat
     ldflags:
       - -s -w -X github.com/netbirdio/netbird/version.version={{.Version}} -X main.commit={{.Commit}} -X main.date={{.CommitDate}} -X main.builtBy=goreleaser
-    mod_timestamp: '{{ .CommitTimestamp }}'
+    mod_timestamp: "{{ .CommitTimestamp }}"
     tags:
       - load_wgnt_from_rsrc
 
+universal_binaries:
+  - id: netbird-ui-darwin
+
 archives:
   - builds:
       - netbird-ui-darwin
@@ -26,4 +33,4 @@ archives:
 checksum:
   name_template: "{{ .ProjectName }}_darwin_checksums.txt"
 changelog:
-  skip: true
+  disable: true

AUTHORS

@@ -1,3 +1,3 @@
 Mikhail Bragin (https://github.com/braginini)
 Maycon Santos (https://github.com/mlsmaycon)
-Wiretrustee UG (haftungsbeschränkt)
+NetBird GmbH

CODE_OF_CONDUCT.md

@@ -5,7 +5,7 @@
 We as members, contributors, and leaders pledge to make participation in our
 community a harassment-free experience for everyone, regardless of age, body
 size, visible or invisible disability, ethnicity, sex characteristics, gender
-identity and expression, level of experience, education, socio-economic status,
+identity and expression, level of experience, education, socioeconomic status,
 nationality, personal appearance, race, caste, color, religion, or sexual
 identity and orientation.
 

CONTRIBUTING.md

@@ -96,7 +96,7 @@ They can be executed from the repository root before every push or PR:
 
 **Goreleaser**
 ```shell
-goreleaser --snapshot --rm-dist
+goreleaser build --snapshot --clean
 ```
 **golangci-lint**
 ```shell

CONTRIBUTOR_LICENSE_AGREEMENT.md

@@ -3,10 +3,10 @@
 We are incredibly thankful for the contributions we receive from the community.
 We require our external contributors to sign a Contributor License Agreement ("CLA") in
 order to ensure that our projects remain licensed under Free and Open Source licenses such
-as BSD-3 while allowing Wiretrustee to build a sustainable business.
+as BSD-3 while allowing NetBird to build a sustainable business.
 
-Wiretrustee is committed to having a true Open Source Software ("OSS") license for
-our software. A CLA enables Wiretrustee to safely commercialize our products
+NetBird is committed to having a true Open Source Software ("OSS") license for
+our software. A CLA enables NetBird to safely commercialize our products
 while keeping a standard OSS license with all the rights that license grants to users: the
 ability to use the project in their own projects or businesses, to republish modified
 source, or to completely fork the project.
@@ -20,11 +20,11 @@ This is a human-readable summary of (and not a substitute for) the full agreemen
 This highlights only some of key terms of the CLA. It has no legal value and you should
 carefully review all the terms of the actual CLA before agreeing.
 
-<li>Grant of copyright license. You give Wiretrustee permission to use your copyrighted work
+<li>Grant of copyright license. You give NetBird permission to use your copyrighted work
 in commercial products.
 </li>
 
-<li>Grant of patent license. If your contributed work uses a patent, you give Wiretrustee a
+<li>Grant of patent license. If your contributed work uses a patent, you give NetBird a
 license to use that patent including within commercial products. You also agree that you
 have permission to grant this license.
 </li>
@@ -45,7 +45,7 @@ more.
 # Why require a CLA?
 
 Agreeing to a CLA explicitly states that you are entitled to provide a contribution, that you cannot withdraw permission
-to use your contribution at a later date, and that Wiretrustee has permission to use your contribution in our commercial
+to use your contribution at a later date, and that NetBird has permission to use your contribution in our commercial
 products.
 
 This removes any ambiguities or uncertainties caused by not having a CLA and allows users and customers to confidently
@@ -65,25 +65,25 @@ Follow the steps given by the bot to sign the CLA. This will require you to log
 information from your account) and to fill in a few additional details such as your name and email address. We will only
 use this information for CLA tracking; none of your submitted information will be used for marketing purposes.
 
-You only have to sign the CLA once. Once you've signed the CLA, future contributions to any Wiretrustee project will not
+You only have to sign the CLA once. Once you've signed the CLA, future contributions to any NetBird project will not
 require you to sign again.
 
 # Legal Terms and Agreement
 
-In order to clarify the intellectual property license granted with Contributions from any person or entity, Wiretrustee
-UG (haftungsbeschränkt) ("Wiretrustee") must have a Contributor License Agreement ("CLA") on file that has been signed
+In order to clarify the intellectual property license granted with Contributions from any person or entity, NetBird
+GmbH ("NetBird") must have a Contributor License Agreement ("CLA") on file that has been signed
 by each Contributor, indicating agreement to the license terms below. This license does not change your rights to use
 your own Contributions for any other purpose.
 
 You accept and agree to the following terms and conditions for Your present and future Contributions submitted to
-Wiretrustee. Except for the license granted herein to Wiretrustee and recipients of software distributed by Wiretrustee,
+NetBird. Except for the license granted herein to NetBird and recipients of software distributed by NetBird,
 You reserve all right, title, and interest in and to Your Contributions.
 
 1. Definitions.
 
     ```
    "You" (or "Your") shall mean the copyright owner or legal entity authorized by the copyright owner 
-   that is making this Agreement with Wiretrustee. For legal entities, the entity making a Contribution and all other 
+   that is making this Agreement with NetBird. For legal entities, the entity making a Contribution and all other 
    entities that control, are controlled by, or are under common control with that entity are considered 
    to be a single Contributor. For the purposes of this definition, "control" means (i) the power, direct or indirect,
     to cause the direction or management of such entity, whether by contract or otherwise, or (ii) ownership of fifty 
@@ -91,23 +91,23 @@ You reserve all right, title, and interest in and to Your Contributions.
    ```
    ```
     "Contribution" shall mean any original work of authorship, including any modifications or additions to 
-   an existing work, that is or previously has been intentionally submitted by You to Wiretrustee for inclusion in, 
-   or documentation of, any of the products owned or managed by Wiretrustee (the "Work"). 
+   an existing work, that is or previously has been intentionally submitted by You to NetBird for inclusion in, 
+   or documentation of, any of the products owned or managed by NetBird (the "Work"). 
    For the purposes of this definition, "submitted" means any form of electronic, verbal, or written communication 
-   sent to Wiretrustee or its representatives, including but not limited to communication on electronic mailing lists, 
+   sent to NetBird or its representatives, including but not limited to communication on electronic mailing lists, 
    source code control systems, and issue tracking systems that are managed by, or on behalf of, 
-   Wiretrustee for the purpose of discussing and improving the Work, but excluding communication that is conspicuously 
+   NetBird for the purpose of discussing and improving the Work, but excluding communication that is conspicuously 
    marked or otherwise designated in writing by You as "Not a Contribution."
    ```
 
-2. Grant of Copyright License. Subject to the terms and conditions of this Agreement, You hereby grant to Wiretrustee
-   and to recipients of software distributed by Wiretrustee a perpetual, worldwide, non-exclusive, no-charge,
+2. Grant of Copyright License. Subject to the terms and conditions of this Agreement, You hereby grant to NetBird
+   and to recipients of software distributed by NetBird a perpetual, worldwide, non-exclusive, no-charge,
    royalty-free, irrevocable copyright license to reproduce, prepare derivative works of, publicly display, publicly
    perform, sublicense, and distribute Your Contributions and such derivative works.
 
 
-3. Grant of Patent License. Subject to the terms and conditions of this Agreement, You hereby grant to Wiretrustee and
-   to recipients of software distributed by Wiretrustee a perpetual, worldwide, non-exclusive, no-charge, royalty-free,
+3. Grant of Patent License. Subject to the terms and conditions of this Agreement, You hereby grant to NetBird and
+   to recipients of software distributed by NetBird a perpetual, worldwide, non-exclusive, no-charge, royalty-free,
    irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import,
    and otherwise transfer the Work, where such license applies only to those patent claims licensable by You that are
    necessarily infringed by Your Contribution(s) alone or by combination of Your Contribution(s) with the Work to which
@@ -121,8 +121,8 @@ You reserve all right, title, and interest in and to Your Contributions.
    intellectual property that you create that includes your Contributions, you represent that you have received
    permission to make Contributions on behalf of that employer, that you will have received permission from your current
    and future employers for all future Contributions, that your applicable employer has waived such rights for all of
-   your current and future Contributions to Wiretrustee, or that your employer has executed a separate Corporate CLA
-   with Wiretrustee.
+   your current and future Contributions to NetBird, or that your employer has executed a separate Corporate CLA
+   with NetBird.
 
 
 5. You represent that each of Your Contributions is Your original creation (see section 7 for submissions on behalf of
@@ -138,11 +138,11 @@ You reserve all right, title, and interest in and to Your Contributions.
    MERCHANTABILITY, or FITNESS FOR A PARTICULAR PURPOSE.
 
 
-7. Should You wish to submit work that is not Your original creation, You may submit it to Wiretrustee separately from
+7. Should You wish to submit work that is not Your original creation, You may submit it to NetBird separately from
    any Contribution, identifying the complete details of its source and of any license or other restriction (including,
    but not limited to, related patents, trademarks, and license agreements) of which you are personally aware, and
    conspicuously marking the work as "Submitted on behalf of a third-party: [named here]".
 
 
-8. You agree to notify Wiretrustee of any facts or circumstances of which you become aware that would make these
-   representations inaccurate in any respect.
+8. You agree to notify NetBird of any facts or circumstances of which you become aware that would make these
+   representations inaccurate in any respect.

LICENSE

@@ -1,6 +1,6 @@
 BSD 3-Clause License
 
-Copyright (c) 2022 Wiretrustee UG (haftungsbeschränkt) & AUTHORS
+Copyright (c) 2022 NetBird GmbH & AUTHORS
 
 Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
 
@@ -10,4 +10,4 @@ Redistribution and use in source and binary forms, with or without modification,
 
 3. Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
 
-THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

README.md

@@ -1,22 +1,21 @@
-<p align="center">
- <strong>:hatching_chick: New Release! Device Posture Checks.</strong>
-  <a href="https://docs.netbird.io/how-to/manage-posture-checks">
-       Learn more
-     </a>   
-</p>
-<br/>
 <div align="center">
 <p align="center">
   <img width="234" src="docs/media/logo-full.png"/>
 </p>
   <p>
+   <a href="https://img.shields.io/badge/license-BSD--3-blue)">
+       <img src="https://sonarcloud.io/api/project_badges/measure?project=netbirdio_netbird&metric=alert_status" />
+     </a> 
      <a href="https://github.com/netbirdio/netbird/blob/main/LICENSE">
        <img src="https://img.shields.io/badge/license-BSD--3-blue" />
      </a> 
-   <a href="https://www.codacy.com/gh/netbirdio/netbird/dashboard?utm_source=github.com&amp;utm_medium=referral&amp;utm_content=netbirdio/netbird&amp;utm_campaign=Badge_Grade"><img src="https://app.codacy.com/project/badge/Grade/e3013d046aec44cdb7462c8673b00976"/></a>
     <br>
-    <a href="https://join.slack.com/t/netbirdio/shared_invite/zt-vrahf41g-ik1v7fV8du6t0RwxSrJ96A">
+    <a href="https://join.slack.com/t/netbirdio/shared_invite/zt-2utg2ncdz-W7LEB6toRBLE1Jca37dYpg">
         <img src="https://img.shields.io/badge/slack-@netbird-red.svg?logo=slack"/>
+     </a>  
+     <br>
+    <a href="https://gurubase.io/g/netbird">
+        <img src="https://img.shields.io/badge/Gurubase-Ask%20NetBird%20Guru-006BFF"/>
      </a>    
   </p>
 </div>
@@ -28,7 +27,7 @@
   <br/>
   See <a href="https://netbird.io/docs/">Documentation</a>
   <br/>
-   Join our <a href="https://join.slack.com/t/netbirdio/shared_invite/zt-vrahf41g-ik1v7fV8du6t0RwxSrJ96A">Slack channel</a>
+   Join our <a href="https://join.slack.com/t/netbirdio/shared_invite/zt-2utg2ncdz-W7LEB6toRBLE1Jca37dYpg">Slack channel</a>
   <br/>
  
 </strong>
@@ -44,8 +43,11 @@
 
 ### Open-Source Network Security in a Single Platform
 
-![image](https://github.com/netbirdio/netbird/assets/700848/c0d7bae4-3301-499a-bb4e-5e4a225bf35f)
 
+![netbird_2](https://github.com/netbirdio/netbird/assets/700848/46bc3b73-508d-4a0e-bb9a-f465d68646ab)
+
+### NetBird on Lawrence Systems (Video)
+[![Watch the video](https://img.youtube.com/vi/Kwrff6h0rEw/0.jpg)](https://www.youtube.com/watch?v=Kwrff6h0rEw)
 
 ### Key features
 
@@ -59,6 +61,7 @@
 |                                                                                                                              |                                                                                                          | <ul><li> - \[x] [Quantum-resistance with Rosenpass](https://netbird.io/knowledge-hub/the-first-quantum-resistant-mesh-vpn) </ul></li> |                                                                                                                                          | <ul><li> - \[x] OpenWRT </ul></li>                                                      |
 |                                                                                                                              |                                                                                                          | <ui><li> - \[x] [Periodic re-authentication](https://docs.netbird.io/how-to/enforce-periodic-user-authentication)</ul></li>           |                                                                                                                                          | <ul><li> - \[x] [Serverless](https://docs.netbird.io/how-to/netbird-on-faas) </ul></li> |
 |                                                                                                                              |                                                                                                          |                                                                                                                                       |                                                                                                                                          | <ul><li> - \[x] Docker </ul></li>                                                       |
+
 ### Quickstart with NetBird Cloud
 
 - Download and install NetBird at [https://app.netbird.io/install](https://app.netbird.io/install)

client/Dockerfile

@@ -1,5 +1,5 @@
-FROM alpine:3.18.5
+FROM alpine:3.21.0
 RUN apk add --no-cache ca-certificates iptables ip6tables
 ENV NB_FOREGROUND_MODE=true
-ENTRYPOINT [ "/go/bin/netbird","up"]
-COPY netbird /go/bin/netbird
+ENTRYPOINT [ "/usr/local/bin/netbird","up"]
+COPY netbird /usr/local/bin/netbird

client/Dockerfile-rootless

@@ -0,0 +1,17 @@
+FROM alpine:3.21.0
+
+COPY netbird /usr/local/bin/netbird
+
+RUN apk add --no-cache ca-certificates \
+    && adduser -D -h /var/lib/netbird netbird
+WORKDIR /var/lib/netbird
+USER netbird:netbird
+
+ENV NB_FOREGROUND_MODE=true
+ENV NB_USE_NETSTACK_MODE=true
+ENV NB_ENABLE_NETSTACK_LOCAL_FORWARDING=true
+ENV NB_CONFIG=config.json
+ENV NB_DAEMON_ADDR=unix://netbird.sock
+ENV NB_DISABLE_DNS=true
+
+ENTRYPOINT [ "/usr/local/bin/netbird", "up" ]

client/android/client.go

@@ -1,3 +1,5 @@
+//go:build android
+
 package android
 
 import (
@@ -6,6 +8,7 @@ import (
 
 	log "github.com/sirupsen/logrus"
 
+	"github.com/netbirdio/netbird/client/iface/device"
 	"github.com/netbirdio/netbird/client/internal"
 	"github.com/netbirdio/netbird/client/internal/dns"
 	"github.com/netbirdio/netbird/client/internal/listener"
@@ -13,7 +16,7 @@ import (
 	"github.com/netbirdio/netbird/client/internal/stdnet"
 	"github.com/netbirdio/netbird/client/system"
 	"github.com/netbirdio/netbird/formatter"
-	"github.com/netbirdio/netbird/iface"
+	"github.com/netbirdio/netbird/util/net"
 )
 
 // ConnectionListener export internal Listener for mobile
@@ -23,7 +26,7 @@ type ConnectionListener interface {
 
 // TunAdapter export internal TunAdapter for mobile
 type TunAdapter interface {
-	iface.TunAdapter
+	device.TunAdapter
 }
 
 // IFaceDiscover export internal IFaceDiscover for mobile
@@ -48,20 +51,23 @@ func init() {
 // Client struct manage the life circle of background service
 type Client struct {
 	cfgFile               string
-	tunAdapter            iface.TunAdapter
+	tunAdapter            device.TunAdapter
 	iFaceDiscover         IFaceDiscover
 	recorder              *peer.Status
 	ctxCancel             context.CancelFunc
 	ctxCancelLock         *sync.Mutex
 	deviceName            string
+	uiVersion             string
 	networkChangeListener listener.NetworkChangeListener
 }
 
 // NewClient instantiate a new Client
-func NewClient(cfgFile, deviceName string, tunAdapter TunAdapter, iFaceDiscover IFaceDiscover, networkChangeListener NetworkChangeListener) *Client {
+func NewClient(cfgFile, deviceName string, uiVersion string, tunAdapter TunAdapter, iFaceDiscover IFaceDiscover, networkChangeListener NetworkChangeListener) *Client {
+	net.SetAndroidProtectSocketFn(tunAdapter.ProtectSocket)
 	return &Client{
 		cfgFile:               cfgFile,
 		deviceName:            deviceName,
+		uiVersion:             uiVersion,
 		tunAdapter:            tunAdapter,
 		iFaceDiscover:         iFaceDiscover,
 		recorder:              peer.NewRecorder(""),
@@ -84,6 +90,9 @@ func (c *Client) Run(urlOpener URLOpener, dns *DNSList, dnsReadyListener DnsRead
 	var ctx context.Context
 	//nolint
 	ctxWithValues := context.WithValue(context.Background(), system.DeviceNameCtxKey, c.deviceName)
+	//nolint
+	ctxWithValues = context.WithValue(ctxWithValues, system.UiVersionCtxKey, c.uiVersion)
+
 	c.ctxCancelLock.Lock()
 	ctx, c.ctxCancel = context.WithCancel(ctxWithValues)
 	defer c.ctxCancel()
@@ -97,7 +106,8 @@ func (c *Client) Run(urlOpener URLOpener, dns *DNSList, dnsReadyListener DnsRead
 
 	// todo do not throw error in case of cancelled context
 	ctx = internal.CtxInitState(ctx)
-	return internal.RunClientMobile(ctx, cfg, c.recorder, c.tunAdapter, c.iFaceDiscover, c.networkChangeListener, dns.items, dnsReadyListener)
+	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder)
+	return connectClient.RunOnAndroid(c.tunAdapter, c.iFaceDiscover, c.networkChangeListener, dns.items, dnsReadyListener)
 }
 
 // RunWithoutLogin we apply this type of run function when the backed has been started without UI (i.e. after reboot).
@@ -122,7 +132,8 @@ func (c *Client) RunWithoutLogin(dns *DNSList, dnsReadyListener DnsReadyListener
 
 	// todo do not throw error in case of cancelled context
 	ctx = internal.CtxInitState(ctx)
-	return internal.RunClientMobile(ctx, cfg, c.recorder, c.tunAdapter, c.iFaceDiscover, c.networkChangeListener, dns.items, dnsReadyListener)
+	connectClient := internal.NewConnectClient(ctx, cfg, c.recorder)
+	return connectClient.RunOnAndroid(c.tunAdapter, c.iFaceDiscover, c.networkChangeListener, dns.items, dnsReadyListener)
 }
 
 // Stop the internal client and free the resources

client/android/login.go

@@ -84,7 +84,7 @@ func (a *Auth) SaveConfigIfSSOSupported(listener SSOListener) {
 func (a *Auth) saveConfigIfSSOSupported() (bool, error) {
 	supportsSSO := true
 	err := a.withBackOff(a.ctx, func() (err error) {
-		_, err = internal.GetPKCEAuthorizationFlowInfo(a.ctx, a.config.PrivateKey, a.config.ManagementURL)
+		_, err = internal.GetPKCEAuthorizationFlowInfo(a.ctx, a.config.PrivateKey, a.config.ManagementURL, nil)
 		if s, ok := gstatus.FromError(err); ok && (s.Code() == codes.NotFound || s.Code() == codes.Unimplemented) {
 			_, err = internal.GetDeviceAuthorizationFlowInfo(a.ctx, a.config.PrivateKey, a.config.ManagementURL)
 			s, ok := gstatus.FromError(err)
@@ -162,7 +162,7 @@ func (a *Auth) login(urlOpener URLOpener) error {
 
 	// check if we need to generate JWT token
 	err := a.withBackOff(a.ctx, func() (err error) {
-		needsLogin, err = internal.IsLoginRequired(a.ctx, a.config.PrivateKey, a.config.ManagementURL, a.config.SSHKey)
+		needsLogin, err = internal.IsLoginRequired(a.ctx, a.config)
 		return
 	})
 	if err != nil {

client/anonymize/anonymize.go

@@ -0,0 +1,240 @@
+package anonymize
+
+import (
+	"crypto/rand"
+	"fmt"
+	"math/big"
+	"net"
+	"net/netip"
+	"net/url"
+	"regexp"
+	"slices"
+	"strings"
+)
+
+const anonTLD = ".domain"
+
+type Anonymizer struct {
+	ipAnonymizer     map[netip.Addr]netip.Addr
+	domainAnonymizer map[string]string
+	currentAnonIPv4  netip.Addr
+	currentAnonIPv6  netip.Addr
+	startAnonIPv4    netip.Addr
+	startAnonIPv6    netip.Addr
+
+	domainKeyRegex *regexp.Regexp
+}
+
+func DefaultAddresses() (netip.Addr, netip.Addr) {
+	// 192.51.100.0, 100::
+	return netip.AddrFrom4([4]byte{198, 51, 100, 0}), netip.AddrFrom16([16]byte{0x01})
+}
+
+func NewAnonymizer(startIPv4, startIPv6 netip.Addr) *Anonymizer {
+	return &Anonymizer{
+		ipAnonymizer:     map[netip.Addr]netip.Addr{},
+		domainAnonymizer: map[string]string{},
+		currentAnonIPv4:  startIPv4,
+		currentAnonIPv6:  startIPv6,
+		startAnonIPv4:    startIPv4,
+		startAnonIPv6:    startIPv6,
+
+		domainKeyRegex: regexp.MustCompile(`\bdomain=([^\s,:"]+)`),
+	}
+}
+
+func (a *Anonymizer) AnonymizeIP(ip netip.Addr) netip.Addr {
+	if ip.IsLoopback() ||
+		ip.IsLinkLocalUnicast() ||
+		ip.IsLinkLocalMulticast() ||
+		ip.IsInterfaceLocalMulticast() ||
+		ip.IsPrivate() ||
+		ip.IsUnspecified() ||
+		ip.IsMulticast() ||
+		isWellKnown(ip) ||
+		a.isInAnonymizedRange(ip) {
+
+		return ip
+	}
+
+	if _, ok := a.ipAnonymizer[ip]; !ok {
+		if ip.Is4() {
+			a.ipAnonymizer[ip] = a.currentAnonIPv4
+			a.currentAnonIPv4 = a.currentAnonIPv4.Next()
+		} else {
+			a.ipAnonymizer[ip] = a.currentAnonIPv6
+			a.currentAnonIPv6 = a.currentAnonIPv6.Next()
+		}
+	}
+	return a.ipAnonymizer[ip]
+}
+
+// isInAnonymizedRange checks if an IP is within the range of already assigned anonymized IPs
+func (a *Anonymizer) isInAnonymizedRange(ip netip.Addr) bool {
+	if ip.Is4() && ip.Compare(a.startAnonIPv4) >= 0 && ip.Compare(a.currentAnonIPv4) <= 0 {
+		return true
+	} else if !ip.Is4() && ip.Compare(a.startAnonIPv6) >= 0 && ip.Compare(a.currentAnonIPv6) <= 0 {
+		return true
+	}
+	return false
+}
+
+func (a *Anonymizer) AnonymizeIPString(ip string) string {
+	addr, err := netip.ParseAddr(ip)
+	if err != nil {
+		return ip
+	}
+
+	return a.AnonymizeIP(addr).String()
+}
+
+func (a *Anonymizer) AnonymizeDomain(domain string) string {
+	baseDomain := domain
+	hasDot := strings.HasSuffix(domain, ".")
+	if hasDot {
+		baseDomain = domain[:len(domain)-1]
+	}
+
+	if strings.HasSuffix(baseDomain, "netbird.io") ||
+		strings.HasSuffix(baseDomain, "netbird.selfhosted") ||
+		strings.HasSuffix(baseDomain, "netbird.cloud") ||
+		strings.HasSuffix(baseDomain, "netbird.stage") ||
+		strings.HasSuffix(baseDomain, anonTLD) {
+		return domain
+	}
+
+	parts := strings.Split(baseDomain, ".")
+	if len(parts) < 2 {
+		return domain
+	}
+
+	baseForLookup := parts[len(parts)-2] + "." + parts[len(parts)-1]
+
+	anonymized, ok := a.domainAnonymizer[baseForLookup]
+	if !ok {
+		anonymizedBase := "anon-" + generateRandomString(5) + anonTLD
+		a.domainAnonymizer[baseForLookup] = anonymizedBase
+		anonymized = anonymizedBase
+	}
+
+	result := strings.Replace(baseDomain, baseForLookup, anonymized, 1)
+	if hasDot {
+		result += "."
+	}
+	return result
+}
+
+func (a *Anonymizer) AnonymizeURI(uri string) string {
+	u, err := url.Parse(uri)
+	if err != nil {
+		return uri
+	}
+
+	var anonymizedHost string
+	if u.Opaque != "" {
+		host, port, err := net.SplitHostPort(u.Opaque)
+		if err == nil {
+			anonymizedHost = fmt.Sprintf("%s:%s", a.AnonymizeDomain(host), port)
+		} else {
+			anonymizedHost = a.AnonymizeDomain(u.Opaque)
+		}
+		u.Opaque = anonymizedHost
+	} else if u.Host != "" {
+		host, port, err := net.SplitHostPort(u.Host)
+		if err == nil {
+			anonymizedHost = fmt.Sprintf("%s:%s", a.AnonymizeDomain(host), port)
+		} else {
+			anonymizedHost = a.AnonymizeDomain(u.Host)
+		}
+		u.Host = anonymizedHost
+	}
+	return u.String()
+}
+
+func (a *Anonymizer) AnonymizeString(str string) string {
+	ipv4Regex := regexp.MustCompile(`\b(?:[0-9]{1,3}\.){3}[0-9]{1,3}\b`)
+	ipv6Regex := regexp.MustCompile(`\b([0-9a-fA-F:]+:+[0-9a-fA-F]{0,4})(?:%[0-9a-zA-Z]+)?(?:\/[0-9]{1,3})?(?::[0-9]{1,5})?\b`)
+
+	str = ipv4Regex.ReplaceAllStringFunc(str, a.AnonymizeIPString)
+	str = ipv6Regex.ReplaceAllStringFunc(str, a.AnonymizeIPString)
+
+	for domain, anonDomain := range a.domainAnonymizer {
+		str = strings.ReplaceAll(str, domain, anonDomain)
+	}
+
+	str = a.AnonymizeSchemeURI(str)
+	str = a.AnonymizeDNSLogLine(str)
+
+	return str
+}
+
+// AnonymizeSchemeURI finds and anonymizes URIs with ws, wss, rel, rels, stun, stuns, turn, and turns schemes.
+func (a *Anonymizer) AnonymizeSchemeURI(text string) string {
+	re := regexp.MustCompile(`(?i)\b(wss?://|rels?://|stuns?:|turns?:|https?://)\S+\b`)
+
+	return re.ReplaceAllStringFunc(text, a.AnonymizeURI)
+}
+
+func (a *Anonymizer) AnonymizeDNSLogLine(logEntry string) string {
+	return a.domainKeyRegex.ReplaceAllStringFunc(logEntry, func(match string) string {
+		parts := strings.SplitN(match, "=", 2)
+		if len(parts) >= 2 {
+			domain := parts[1]
+			if strings.HasSuffix(domain, anonTLD) {
+				return match
+			}
+			return "domain=" + a.AnonymizeDomain(domain)
+		}
+		return match
+	})
+}
+
+// AnonymizeRoute anonymizes a route string by replacing IP addresses with anonymized versions and
+// domain names with random strings.
+func (a *Anonymizer) AnonymizeRoute(route string) string {
+	prefix, err := netip.ParsePrefix(route)
+	if err == nil {
+		ip := a.AnonymizeIPString(prefix.Addr().String())
+		return fmt.Sprintf("%s/%d", ip, prefix.Bits())
+	}
+	domains := strings.Split(route, ", ")
+	for i, domain := range domains {
+		domains[i] = a.AnonymizeDomain(domain)
+	}
+	return strings.Join(domains, ", ")
+}
+
+func isWellKnown(addr netip.Addr) bool {
+	wellKnown := []string{
+		"8.8.8.8", "8.8.4.4", // Google DNS IPv4
+		"2001:4860:4860::8888", "2001:4860:4860::8844", // Google DNS IPv6
+		"1.1.1.1", "1.0.0.1", // Cloudflare DNS IPv4
+		"2606:4700:4700::1111", "2606:4700:4700::1001", // Cloudflare DNS IPv6
+		"9.9.9.9", "149.112.112.112", // Quad9 DNS IPv4
+		"2620:fe::fe", "2620:fe::9", // Quad9 DNS IPv6
+
+		"128.0.0.0", "8000::", // 2nd split subnet for default routes
+	}
+
+	if slices.Contains(wellKnown, addr.String()) {
+		return true
+	}
+
+	cgnatRangeStart := netip.AddrFrom4([4]byte{100, 64, 0, 0})
+	cgnatRange := netip.PrefixFrom(cgnatRangeStart, 10)
+
+	return cgnatRange.Contains(addr)
+}
+
+func generateRandomString(length int) string {
+	const letters = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"
+	result := make([]byte, length)
+	for i := range result {
+		num, err := rand.Int(rand.Reader, big.NewInt(int64(len(letters))))
+		if err != nil {
+			continue
+		}
+		result[i] = letters[num.Int64()]
+	}
+	return string(result)
+}

client/anonymize/anonymize_test.go

@@ -0,0 +1,297 @@
+package anonymize_test
+
+import (
+	"net/netip"
+	"regexp"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	"github.com/netbirdio/netbird/client/anonymize"
+)
+
+func TestAnonymizeIP(t *testing.T) {
+	startIPv4 := netip.MustParseAddr("198.51.100.0")
+	startIPv6 := netip.MustParseAddr("100::")
+	anonymizer := anonymize.NewAnonymizer(startIPv4, startIPv6)
+
+	tests := []struct {
+		name   string
+		ip     string
+		expect string
+	}{
+		{"Well known", "8.8.8.8", "8.8.8.8"},
+		{"First Public IPv4", "1.2.3.4", "198.51.100.0"},
+		{"Second Public IPv4", "4.3.2.1", "198.51.100.1"},
+		{"Repeated IPv4", "1.2.3.4", "198.51.100.0"},
+		{"Private IPv4", "192.168.1.1", "192.168.1.1"},
+		{"First Public IPv6", "2607:f8b0:4005:805::200e", "100::"},
+		{"Second Public IPv6", "a::b", "100::1"},
+		{"Repeated IPv6", "2607:f8b0:4005:805::200e", "100::"},
+		{"Private IPv6", "fe80::1", "fe80::1"},
+		{"In Range IPv4", "198.51.100.2", "198.51.100.2"},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			ip := netip.MustParseAddr(tc.ip)
+			anonymizedIP := anonymizer.AnonymizeIP(ip)
+			if anonymizedIP.String() != tc.expect {
+				t.Errorf("%s: expected %s, got %s", tc.name, tc.expect, anonymizedIP)
+			}
+		})
+	}
+}
+
+func TestAnonymizeDNSLogLine(t *testing.T) {
+	anonymizer := anonymize.NewAnonymizer(netip.Addr{}, netip.Addr{})
+	tests := []struct {
+		name     string
+		input    string
+		original string
+		expect   string
+	}{
+		{
+			name:     "Basic domain with trailing content",
+			input:    "received DNS request for DNS forwarder: domain=example.com: something happened with code=123",
+			original: "example.com",
+			expect:   `received DNS request for DNS forwarder: domain=anon-[a-zA-Z0-9]+\.domain: something happened with code=123`,
+		},
+		{
+			name:     "Domain with trailing dot",
+			input:    "domain=example.com. processing request with status=pending",
+			original: "example.com",
+			expect:   `domain=anon-[a-zA-Z0-9]+\.domain\. processing request with status=pending`,
+		},
+		{
+			name:     "Multiple domains in log",
+			input:    "forward domain=first.com status=ok, redirect to domain=second.com port=443",
+			original: "first.com", // testing just one is sufficient as AnonymizeDomain is tested separately
+			expect:   `forward domain=anon-[a-zA-Z0-9]+\.domain status=ok, redirect to domain=anon-[a-zA-Z0-9]+\.domain port=443`,
+		},
+		{
+			name:     "Already anonymized domain",
+			input:    "got request domain=anon-xyz123.domain from=client1 to=server2",
+			original: "", // nothing should be anonymized
+			expect:   `got request domain=anon-xyz123\.domain from=client1 to=server2`,
+		},
+		{
+			name:     "Subdomain with trailing dot",
+			input:    "domain=sub.example.com. next_hop=10.0.0.1 proto=udp",
+			original: "example.com",
+			expect:   `domain=sub\.anon-[a-zA-Z0-9]+\.domain\. next_hop=10\.0\.0\.1 proto=udp`,
+		},
+		{
+			name:     "Handler chain pattern log",
+			input:    "pattern: domain=example.com. original: domain=*.example.com. wildcard=true priority=100",
+			original: "example.com",
+			expect:   `pattern: domain=anon-[a-zA-Z0-9]+\.domain\. original: domain=\*\.anon-[a-zA-Z0-9]+\.domain\. wildcard=true priority=100`,
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			result := anonymizer.AnonymizeDNSLogLine(tc.input)
+			if tc.original != "" {
+				assert.NotContains(t, result, tc.original)
+			}
+			assert.Regexp(t, tc.expect, result)
+		})
+	}
+}
+
+func TestAnonymizeDomain(t *testing.T) {
+	anonymizer := anonymize.NewAnonymizer(netip.Addr{}, netip.Addr{})
+	tests := []struct {
+		name            string
+		domain          string
+		expectPattern   string
+		shouldAnonymize bool
+	}{
+		{
+			"General Domain",
+			"example.com",
+			`^anon-[a-zA-Z0-9]+\.domain$`,
+			true,
+		},
+		{
+			"Domain with Trailing Dot",
+			"example.com.",
+			`^anon-[a-zA-Z0-9]+\.domain.$`,
+			true,
+		},
+		{
+			"Subdomain",
+			"sub.example.com",
+			`^sub\.anon-[a-zA-Z0-9]+\.domain$`,
+			true,
+		},
+		{
+			"Subdomain with Trailing Dot",
+			"sub.example.com.",
+			`^sub\.anon-[a-zA-Z0-9]+\.domain.$`,
+			true,
+		},
+		{
+			"Protected Domain",
+			"netbird.io",
+			`^netbird\.io$`,
+			false,
+		},
+		{
+			"Protected Domain with Trailing Dot",
+			"netbird.io.",
+			`^netbird\.io.$`,
+			false,
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			result := anonymizer.AnonymizeDomain(tc.domain)
+			if tc.shouldAnonymize {
+				assert.Regexp(t, tc.expectPattern, result, "The anonymized domain should match the expected pattern")
+				assert.NotContains(t, result, tc.domain, "The original domain should not be present in the result")
+			} else {
+				assert.Equal(t, tc.domain, result, "Protected domains should not be anonymized")
+			}
+		})
+	}
+}
+
+func TestAnonymizeURI(t *testing.T) {
+	anonymizer := anonymize.NewAnonymizer(netip.Addr{}, netip.Addr{})
+	tests := []struct {
+		name  string
+		uri   string
+		regex string
+	}{
+		{
+			"HTTP URI with Port",
+			"http://example.com:80/path",
+			`^http://anon-[a-zA-Z0-9]+\.domain:80/path$`,
+		},
+		{
+			"HTTP URI without Port",
+			"http://example.com/path",
+			`^http://anon-[a-zA-Z0-9]+\.domain/path$`,
+		},
+		{
+			"Opaque URI with Port",
+			"stun:example.com:80?transport=udp",
+			`^stun:anon-[a-zA-Z0-9]+\.domain:80\?transport=udp$`,
+		},
+		{
+			"Opaque URI without Port",
+			"stun:example.com?transport=udp",
+			`^stun:anon-[a-zA-Z0-9]+\.domain\?transport=udp$`,
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			result := anonymizer.AnonymizeURI(tc.uri)
+			assert.Regexp(t, regexp.MustCompile(tc.regex), result, "URI should match expected pattern")
+			require.NotContains(t, result, "example.com", "Original domain should not be present")
+		})
+	}
+}
+
+func TestAnonymizeSchemeURI(t *testing.T) {
+	anonymizer := anonymize.NewAnonymizer(netip.Addr{}, netip.Addr{})
+	tests := []struct {
+		name   string
+		input  string
+		expect string
+	}{
+		{"STUN URI in text", "Connection made via stun:example.com", `Connection made via stun:anon-[a-zA-Z0-9]+\.domain`},
+		{"STUNS URI in message", "Secure connection to stuns:example.com:443", `Secure connection to stuns:anon-[a-zA-Z0-9]+\.domain:443`},
+		{"TURN URI in log", "Failed attempt turn:some.example.com:3478?transport=tcp: retrying", `Failed attempt turn:some.anon-[a-zA-Z0-9]+\.domain:3478\?transport=tcp: retrying`},
+		{"TURNS URI in message", "Secure connection to turns:example.com:5349", `Secure connection to turns:anon-[a-zA-Z0-9]+\.domain:5349`},
+		{"HTTP URI in text", "Visit http://example.com for more", `Visit http://anon-[a-zA-Z0-9]+\.domain for more`},
+		{"HTTPS URI in CAPS", "Visit HTTPS://example.com for more", `Visit https://anon-[a-zA-Z0-9]+\.domain for more`},
+		{"HTTPS URI in message", "Visit https://example.com for more", `Visit https://anon-[a-zA-Z0-9]+\.domain for more`},
+		{"WS URI in log", "Connection established to ws://example.com:8080", `Connection established to ws://anon-[a-zA-Z0-9]+\.domain:8080`},
+		{"WSS URI in message", "Secure connection to wss://example.com", `Secure connection to wss://anon-[a-zA-Z0-9]+\.domain`},
+		{"Rel URI in text", "Relaying to rel://example.com", `Relaying to rel://anon-[a-zA-Z0-9]+\.domain`},
+		{"Rels URI in message", "Relaying to rels://example.com", `Relaying to rels://anon-[a-zA-Z0-9]+\.domain`},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			result := anonymizer.AnonymizeSchemeURI(tc.input)
+			assert.Regexp(t, tc.expect, result, "The anonymized output should match expected pattern")
+			require.NotContains(t, result, "example.com", "Original domain should not be present")
+		})
+	}
+}
+
+func TestAnonymizString_MemorizedDomain(t *testing.T) {
+	anonymizer := anonymize.NewAnonymizer(netip.Addr{}, netip.Addr{})
+	domain := "example.com"
+	anonymizedDomain := anonymizer.AnonymizeDomain(domain)
+
+	sampleString := "This is a test string including the domain example.com which should be anonymized."
+
+	firstPassResult := anonymizer.AnonymizeString(sampleString)
+	secondPassResult := anonymizer.AnonymizeString(firstPassResult)
+
+	assert.Contains(t, firstPassResult, anonymizedDomain, "The domain should be anonymized in the first pass")
+	assert.NotContains(t, firstPassResult, domain, "The original domain should not appear in the first pass output")
+
+	assert.Equal(t, firstPassResult, secondPassResult, "The second pass should not further anonymize the string")
+}
+
+func TestAnonymizeString_DoubleURI(t *testing.T) {
+	anonymizer := anonymize.NewAnonymizer(netip.Addr{}, netip.Addr{})
+	domain := "example.com"
+	anonymizedDomain := anonymizer.AnonymizeDomain(domain)
+
+	sampleString := "Check out our site at https://example.com for more info."
+
+	firstPassResult := anonymizer.AnonymizeString(sampleString)
+	secondPassResult := anonymizer.AnonymizeString(firstPassResult)
+
+	assert.Contains(t, firstPassResult, "https://"+anonymizedDomain, "The URI should be anonymized in the first pass")
+	assert.NotContains(t, firstPassResult, "https://example.com", "The original URI should not appear in the first pass output")
+
+	assert.Equal(t, firstPassResult, secondPassResult, "The second pass should not further anonymize the URI")
+}
+
+func TestAnonymizeString_IPAddresses(t *testing.T) {
+	anonymizer := anonymize.NewAnonymizer(anonymize.DefaultAddresses())
+	tests := []struct {
+		name   string
+		input  string
+		expect string
+	}{
+		{
+			name:   "IPv4 Address",
+			input:  "Error occurred at IP 122.138.1.1",
+			expect: "Error occurred at IP 198.51.100.0",
+		},
+		{
+			name:   "IPv6 Address",
+			input:  "Access attempted from 2001:db8::ff00:42",
+			expect: "Access attempted from 100::",
+		},
+		{
+			name:   "IPv6 Address with Port",
+			input:  "Access attempted from [2001:db8::ff00:42]:8080",
+			expect: "Access attempted from [100::]:8080",
+		},
+		{
+			name:   "Both IPv4 and IPv6",
+			input:  "IPv4: 142.108.0.1 and IPv6: 2001:db8::ff00:43",
+			expect: "IPv4: 198.51.100.1 and IPv6: 100::1",
+		},
+	}
+
+	for _, tc := range tests {
+		t.Run(tc.name, func(t *testing.T) {
+			result := anonymizer.AnonymizeString(tc.input)
+			assert.Equal(t, tc.expect, result, "IP addresses should be anonymized correctly")
+		})
+	}
+}

client/cmd/debug.go

@@ -0,0 +1,325 @@
+package cmd
+
+import (
+	"context"
+	"fmt"
+	"strings"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+	"github.com/spf13/cobra"
+	"google.golang.org/grpc/status"
+
+	"github.com/netbirdio/netbird/client/internal"
+	"github.com/netbirdio/netbird/client/proto"
+	"github.com/netbirdio/netbird/client/server"
+)
+
+const errCloseConnection = "Failed to close connection: %v"
+
+var debugCmd = &cobra.Command{
+	Use:   "debug",
+	Short: "Debugging commands",
+	Long:  "Provides commands for debugging and logging control within the Netbird daemon.",
+}
+
+var debugBundleCmd = &cobra.Command{
+	Use:     "bundle",
+	Example: "  netbird debug bundle",
+	Short:   "Create a debug bundle",
+	Long:    "Generates a compressed archive of the daemon's logs and status for debugging purposes.",
+	RunE:    debugBundle,
+}
+
+var logCmd = &cobra.Command{
+	Use:   "log",
+	Short: "Manage logging for the Netbird daemon",
+	Long:  `Commands to manage logging settings for the Netbird daemon, including ICE, gRPC, and general log levels.`,
+}
+
+var logLevelCmd = &cobra.Command{
+	Use:   "level <level>",
+	Short: "Set the logging level for this session",
+	Long: `Sets the logging level for the current session. This setting is temporary and will revert to the default on daemon restart.
+Available log levels are:
+  panic:   for panic level, highest level of severity
+  fatal:   for fatal level errors that cause the program to exit
+  error:   for error conditions
+  warn:    for warning conditions
+  info:    for informational messages
+  debug:   for debug-level messages
+  trace:   for trace-level messages, which include more fine-grained information than debug`,
+	Args: cobra.ExactArgs(1),
+	RunE: setLogLevel,
+}
+
+var forCmd = &cobra.Command{
+	Use:     "for <time>",
+	Short:   "Run debug logs for a specified duration and create a debug bundle",
+	Long:    `Sets the logging level to trace, runs for the specified duration, and then generates a debug bundle.`,
+	Example: "  netbird debug for 5m",
+	Args:    cobra.ExactArgs(1),
+	RunE:    runForDuration,
+}
+
+var persistenceCmd = &cobra.Command{
+	Use:     "persistence [on|off]",
+	Short:   "Set network map memory persistence",
+	Long:    `Configure whether the latest network map should persist in memory. When enabled, the last known network map will be kept in memory.`,
+	Example: "  netbird debug persistence on",
+	Args:    cobra.ExactArgs(1),
+	RunE:    setNetworkMapPersistence,
+}
+
+func debugBundle(cmd *cobra.Command, _ []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.DebugBundle(cmd.Context(), &proto.DebugBundleRequest{
+		Anonymize:  anonymizeFlag,
+		Status:     getStatusOutput(cmd),
+		SystemInfo: debugSystemInfoFlag,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to bundle debug: %v", status.Convert(err).Message())
+	}
+
+	cmd.Println(resp.GetPath())
+
+	return nil
+}
+
+func setLogLevel(cmd *cobra.Command, args []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	client := proto.NewDaemonServiceClient(conn)
+	level := server.ParseLogLevel(args[0])
+	if level == proto.LogLevel_UNKNOWN {
+		return fmt.Errorf("unknown log level: %s. Available levels are: panic, fatal, error, warn, info, debug, trace\n", args[0])
+	}
+
+	_, err = client.SetLogLevel(cmd.Context(), &proto.SetLogLevelRequest{
+		Level: level,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to set log level: %v", status.Convert(err).Message())
+	}
+
+	cmd.Println("Log level set successfully to", args[0])
+	return nil
+}
+
+func runForDuration(cmd *cobra.Command, args []string) error {
+	duration, err := time.ParseDuration(args[0])
+	if err != nil {
+		return fmt.Errorf("invalid duration format: %v", err)
+	}
+
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	client := proto.NewDaemonServiceClient(conn)
+
+	stat, err := client.Status(cmd.Context(), &proto.StatusRequest{})
+	if err != nil {
+		return fmt.Errorf("failed to get status: %v", status.Convert(err).Message())
+	}
+
+	stateWasDown := stat.Status != string(internal.StatusConnected) && stat.Status != string(internal.StatusConnecting)
+
+	initialLogLevel, err := client.GetLogLevel(cmd.Context(), &proto.GetLogLevelRequest{})
+	if err != nil {
+		return fmt.Errorf("failed to get log level: %v", status.Convert(err).Message())
+	}
+
+	if stateWasDown {
+		if _, err := client.Up(cmd.Context(), &proto.UpRequest{}); err != nil {
+			return fmt.Errorf("failed to up: %v", status.Convert(err).Message())
+		}
+		cmd.Println("Netbird up")
+		time.Sleep(time.Second * 10)
+	}
+
+	initialLevelTrace := initialLogLevel.GetLevel() >= proto.LogLevel_TRACE
+	if !initialLevelTrace {
+		_, err = client.SetLogLevel(cmd.Context(), &proto.SetLogLevelRequest{
+			Level: proto.LogLevel_TRACE,
+		})
+		if err != nil {
+			return fmt.Errorf("failed to set log level to TRACE: %v", status.Convert(err).Message())
+		}
+		cmd.Println("Log level set to trace.")
+	}
+
+	if _, err := client.Down(cmd.Context(), &proto.DownRequest{}); err != nil {
+		return fmt.Errorf("failed to down: %v", status.Convert(err).Message())
+	}
+	cmd.Println("Netbird down")
+
+	time.Sleep(1 * time.Second)
+
+	// Enable network map persistence before bringing the service up
+	if _, err := client.SetNetworkMapPersistence(cmd.Context(), &proto.SetNetworkMapPersistenceRequest{
+		Enabled: true,
+	}); err != nil {
+		return fmt.Errorf("failed to enable network map persistence: %v", status.Convert(err).Message())
+	}
+
+	if _, err := client.Up(cmd.Context(), &proto.UpRequest{}); err != nil {
+		return fmt.Errorf("failed to up: %v", status.Convert(err).Message())
+	}
+	cmd.Println("Netbird up")
+
+	time.Sleep(3 * time.Second)
+
+	headerPostUp := fmt.Sprintf("----- Netbird post-up - Timestamp: %s", time.Now().Format(time.RFC3339))
+	statusOutput := fmt.Sprintf("%s\n%s", headerPostUp, getStatusOutput(cmd))
+
+	if waitErr := waitForDurationOrCancel(cmd.Context(), duration, cmd); waitErr != nil {
+		return waitErr
+	}
+	cmd.Println("\nDuration completed")
+
+	cmd.Println("Creating debug bundle...")
+
+	headerPreDown := fmt.Sprintf("----- Netbird pre-down - Timestamp: %s - Duration: %s", time.Now().Format(time.RFC3339), duration)
+	statusOutput = fmt.Sprintf("%s\n%s\n%s", statusOutput, headerPreDown, getStatusOutput(cmd))
+
+	resp, err := client.DebugBundle(cmd.Context(), &proto.DebugBundleRequest{
+		Anonymize:  anonymizeFlag,
+		Status:     statusOutput,
+		SystemInfo: debugSystemInfoFlag,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to bundle debug: %v", status.Convert(err).Message())
+	}
+
+	// Disable network map persistence after creating the debug bundle
+	if _, err := client.SetNetworkMapPersistence(cmd.Context(), &proto.SetNetworkMapPersistenceRequest{
+		Enabled: false,
+	}); err != nil {
+		return fmt.Errorf("failed to disable network map persistence: %v", status.Convert(err).Message())
+	}
+
+	if stateWasDown {
+		if _, err := client.Down(cmd.Context(), &proto.DownRequest{}); err != nil {
+			return fmt.Errorf("failed to down: %v", status.Convert(err).Message())
+		}
+		cmd.Println("Netbird down")
+	}
+
+	if !initialLevelTrace {
+		if _, err := client.SetLogLevel(cmd.Context(), &proto.SetLogLevelRequest{Level: initialLogLevel.GetLevel()}); err != nil {
+			return fmt.Errorf("failed to restore log level: %v", status.Convert(err).Message())
+		}
+		cmd.Println("Log level restored to", initialLogLevel.GetLevel())
+	}
+
+	cmd.Println(resp.GetPath())
+
+	return nil
+}
+
+func setNetworkMapPersistence(cmd *cobra.Command, args []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	persistence := strings.ToLower(args[0])
+	if persistence != "on" && persistence != "off" {
+		return fmt.Errorf("invalid persistence value: %s. Use 'on' or 'off'", args[0])
+	}
+
+	client := proto.NewDaemonServiceClient(conn)
+	_, err = client.SetNetworkMapPersistence(cmd.Context(), &proto.SetNetworkMapPersistenceRequest{
+		Enabled: persistence == "on",
+	})
+	if err != nil {
+		return fmt.Errorf("failed to set network map persistence: %v", status.Convert(err).Message())
+	}
+
+	cmd.Printf("Network map persistence set to: %s\n", persistence)
+	return nil
+}
+
+func getStatusOutput(cmd *cobra.Command) string {
+	var statusOutputString string
+	statusResp, err := getStatus(cmd.Context())
+	if err != nil {
+		cmd.PrintErrf("Failed to get status: %v\n", err)
+	} else {
+		statusOutputString = parseToFullDetailSummary(convertToStatusOutputOverview(statusResp))
+	}
+	return statusOutputString
+}
+
+func waitForDurationOrCancel(ctx context.Context, duration time.Duration, cmd *cobra.Command) error {
+	ticker := time.NewTicker(1 * time.Second)
+	defer ticker.Stop()
+
+	startTime := time.Now()
+
+	done := make(chan struct{})
+	go func() {
+		defer close(done)
+		for {
+			select {
+			case <-ctx.Done():
+				return
+			case <-ticker.C:
+				elapsed := time.Since(startTime)
+				if elapsed >= duration {
+					return
+				}
+				remaining := duration - elapsed
+				cmd.Printf("\rRemaining time: %s", formatDuration(remaining))
+			}
+		}
+	}()
+
+	select {
+	case <-ctx.Done():
+		return ctx.Err()
+	case <-done:
+		return nil
+	}
+}
+
+func formatDuration(d time.Duration) string {
+	d = d.Round(time.Second)
+	h := d / time.Hour
+	d %= time.Hour
+	m := d / time.Minute
+	d %= time.Minute
+	s := d / time.Second
+	return fmt.Sprintf("%02d:%02d:%02d", h, m, s)
+}

client/cmd/down.go

@@ -2,9 +2,10 @@ package cmd
 
 import (
 	"context"
-	"github.com/netbirdio/netbird/util"
 	"time"
 
+	"github.com/netbirdio/netbird/util"
+
 	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 
@@ -25,7 +26,7 @@ var downCmd = &cobra.Command{
 			return err
 		}
 
-		ctx, cancel := context.WithTimeout(context.Background(), time.Second*3)
+		ctx, cancel := context.WithTimeout(context.Background(), time.Second*7)
 		defer cancel()
 
 		conn, err := DialClientGRPCServer(ctx, daemonAddr)
@@ -41,6 +42,8 @@ var downCmd = &cobra.Command{
 			log.Errorf("call service down method: %v", err)
 			return err
 		}
+
+		cmd.Println("Disconnected")
 		return nil
 	},
 }

client/cmd/login.go

@@ -39,6 +39,11 @@ var loginCmd = &cobra.Command{
 			ctx = context.WithValue(ctx, system.DeviceNameCtxKey, hostName)
 		}
 
+		providedSetupKey, err := getSetupKey()
+		if err != nil {
+			return err
+		}
+
 		// workaround to run without service
 		if logFile == "console" {
 			err = handleRebrand(cmd)
@@ -62,7 +67,7 @@ var loginCmd = &cobra.Command{
 
 			config, _ = internal.UpdateOldManagementURL(ctx, config, configPath)
 
-			err = foregroundLogin(ctx, cmd, config, setupKey)
+			err = foregroundLogin(ctx, cmd, config, providedSetupKey)
 			if err != nil {
 				return fmt.Errorf("foreground login failed: %v", err)
 			}
@@ -81,7 +86,7 @@ var loginCmd = &cobra.Command{
 		client := proto.NewDaemonServiceClient(conn)
 
 		loginRequest := proto.LoginRequest{
-			SetupKey:             setupKey,
+			SetupKey:             providedSetupKey,
 			ManagementUrl:        managementURL,
 			IsLinuxDesktopClient: isLinuxRunningDesktop(),
 			Hostname:             hostName,

client/cmd/login_test.go

@@ -5,8 +5,8 @@ import (
 	"strings"
 	"testing"
 
+	"github.com/netbirdio/netbird/client/iface"
 	"github.com/netbirdio/netbird/client/internal"
-	"github.com/netbirdio/netbird/iface"
 	"github.com/netbirdio/netbird/util"
 )
 

client/cmd/networks.go

@@ -0,0 +1,173 @@
+package cmd
+
+import (
+	"fmt"
+	"strings"
+
+	"github.com/spf13/cobra"
+	"google.golang.org/grpc/status"
+
+	"github.com/netbirdio/netbird/client/proto"
+)
+
+var appendFlag bool
+
+var networksCMD = &cobra.Command{
+	Use:     "networks",
+	Aliases: []string{"routes"},
+	Short:   "Manage networks",
+	Long:    `Commands to list, select, or deselect networks. Replaces the "routes" command.`,
+}
+
+var routesListCmd = &cobra.Command{
+	Use:     "list",
+	Aliases: []string{"ls"},
+	Short:   "List networks",
+	Example: "  netbird networks list",
+	Long:    "List all available network routes.",
+	RunE:    networksList,
+}
+
+var routesSelectCmd = &cobra.Command{
+	Use:     "select network...|all",
+	Short:   "Select network",
+	Long:    "Select a list of networks by identifiers or 'all' to clear all selections and to accept all (including new) networks.\nDefault mode is replace, use -a to append to already selected networks.",
+	Example: "  netbird networks select all\n  netbird networks select route1 route2\n  netbird routes select -a route3",
+	Args:    cobra.MinimumNArgs(1),
+	RunE:    networksSelect,
+}
+
+var routesDeselectCmd = &cobra.Command{
+	Use:     "deselect network...|all",
+	Short:   "Deselect networks",
+	Long:    "Deselect previously selected networks by identifiers or 'all' to disable accepting any networks.",
+	Example: "  netbird networks deselect all\n  netbird networks deselect route1 route2",
+	Args:    cobra.MinimumNArgs(1),
+	RunE:    networksDeselect,
+}
+
+func init() {
+	routesSelectCmd.PersistentFlags().BoolVarP(&appendFlag, "append", "a", false, "Append to current network selection instead of replacing")
+}
+
+func networksList(cmd *cobra.Command, _ []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.ListNetworks(cmd.Context(), &proto.ListNetworksRequest{})
+	if err != nil {
+		return fmt.Errorf("failed to list network: %v", status.Convert(err).Message())
+	}
+
+	if len(resp.Routes) == 0 {
+		cmd.Println("No networks available.")
+		return nil
+	}
+
+	printNetworks(cmd, resp)
+
+	return nil
+}
+
+func printNetworks(cmd *cobra.Command, resp *proto.ListNetworksResponse) {
+	cmd.Println("Available Networks:")
+	for _, route := range resp.Routes {
+		printNetwork(cmd, route)
+	}
+}
+
+func printNetwork(cmd *cobra.Command, route *proto.Network) {
+	selectedStatus := getSelectedStatus(route)
+	domains := route.GetDomains()
+
+	if len(domains) > 0 {
+		printDomainRoute(cmd, route, domains, selectedStatus)
+	} else {
+		printNetworkRoute(cmd, route, selectedStatus)
+	}
+}
+
+func getSelectedStatus(route *proto.Network) string {
+	if route.GetSelected() {
+		return "Selected"
+	}
+	return "Not Selected"
+}
+
+func printDomainRoute(cmd *cobra.Command, route *proto.Network, domains []string, selectedStatus string) {
+	cmd.Printf("\n  - ID: %s\n    Domains: %s\n    Status: %s\n", route.GetID(), strings.Join(domains, ", "), selectedStatus)
+	resolvedIPs := route.GetResolvedIPs()
+
+	if len(resolvedIPs) > 0 {
+		printResolvedIPs(cmd, domains, resolvedIPs)
+	} else {
+		cmd.Printf("    Resolved IPs: -\n")
+	}
+}
+
+func printNetworkRoute(cmd *cobra.Command, route *proto.Network, selectedStatus string) {
+	cmd.Printf("\n  - ID: %s\n    Network: %s\n    Status: %s\n", route.GetID(), route.GetRange(), selectedStatus)
+}
+
+func printResolvedIPs(cmd *cobra.Command, _ []string, resolvedIPs map[string]*proto.IPList) {
+	cmd.Printf("    Resolved IPs:\n")
+	for resolvedDomain, ipList := range resolvedIPs {
+		cmd.Printf("      [%s]: %s\n", resolvedDomain, strings.Join(ipList.GetIps(), ", "))
+	}
+}
+
+func networksSelect(cmd *cobra.Command, args []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	req := &proto.SelectNetworksRequest{
+		NetworkIDs: args,
+	}
+
+	if len(args) == 1 && args[0] == "all" {
+		req.All = true
+	} else if appendFlag {
+		req.Append = true
+	}
+
+	if _, err := client.SelectNetworks(cmd.Context(), req); err != nil {
+		return fmt.Errorf("failed to select networks: %v", status.Convert(err).Message())
+	}
+
+	cmd.Println("Networks selected successfully.")
+
+	return nil
+}
+
+func networksDeselect(cmd *cobra.Command, args []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	req := &proto.SelectNetworksRequest{
+		NetworkIDs: args,
+	}
+
+	if len(args) == 1 && args[0] == "all" {
+		req.All = true
+	}
+
+	if _, err := client.DeselectNetworks(cmd.Context(), req); err != nil {
+		return fmt.Errorf("failed to deselect networks: %v", status.Convert(err).Message())
+	}
+
+	cmd.Println("Networks deselected successfully.")
+
+	return nil
+}

client/cmd/pprof.go

@@ -0,0 +1,33 @@
+//go:build pprof
+// +build pprof
+
+package cmd
+
+import (
+	"net/http"
+	_ "net/http/pprof"
+	"os"
+
+	log "github.com/sirupsen/logrus"
+)
+
+func init() {
+	addr := pprofAddr()
+	go pprof(addr)
+}
+
+func pprofAddr() string {
+	listenAddr := os.Getenv("NB_PPROF_ADDR")
+	if listenAddr == "" {
+		return "localhost:6969"
+	}
+
+	return listenAddr
+}
+
+func pprof(listenAddr string) {
+	log.Infof("listening pprof on: %s\n", listenAddr)
+	if err := http.ListenAndServe(listenAddr, nil); err != nil {
+		log.Fatalf("Failed to start pprof: %v", err)
+	}
+}

client/cmd/root.go

@@ -32,9 +32,13 @@ const (
 	preSharedKeyFlag        = "preshared-key"
 	interfaceNameFlag       = "interface-name"
 	wireguardPortFlag       = "wireguard-port"
+	networkMonitorFlag      = "network-monitor"
 	disableAutoConnectFlag  = "disable-auto-connect"
 	serverSSHAllowedFlag    = "allow-server-ssh"
 	extraIFaceBlackListFlag = "extra-iface-blacklist"
+	dnsRouteIntervalFlag    = "dns-router-interval"
+	systemInfoFlag          = "system-info"
+	blockLANAccessFlag      = "block-lan-access"
 )
 
 var (
@@ -53,6 +57,7 @@ var (
 	managementURL           string
 	adminURL                string
 	setupKey                string
+	setupKeyPath            string
 	hostName                string
 	preSharedKey            string
 	natExternalIPs          []string
@@ -62,10 +67,16 @@ var (
 	serverSSHAllowed        bool
 	interfaceName           string
 	wireguardPort           uint16
+	networkMonitor          bool
 	serviceName             string
 	autoConnectDisabled     bool
 	extraIFaceBlackList     []string
-	rootCmd                 = &cobra.Command{
+	anonymizeFlag           bool
+	debugSystemInfoFlag     bool
+	dnsRouteInterval        time.Duration
+	blockLANAccess          bool
+
+	rootCmd = &cobra.Command{
 		Use:          "netbird",
 		Short:        "",
 		Long:         "",
@@ -85,12 +96,15 @@ func init() {
 	oldDefaultConfigPathDir = "/etc/wiretrustee/"
 	oldDefaultLogFileDir = "/var/log/wiretrustee/"
 
-	if runtime.GOOS == "windows" {
+	switch runtime.GOOS {
+	case "windows":
 		defaultConfigPathDir = os.Getenv("PROGRAMDATA") + "\\Netbird\\"
 		defaultLogFileDir = os.Getenv("PROGRAMDATA") + "\\Netbird\\"
 
 		oldDefaultConfigPathDir = os.Getenv("PROGRAMDATA") + "\\Wiretrustee\\"
 		oldDefaultLogFileDir = os.Getenv("PROGRAMDATA") + "\\Wiretrustee\\"
+	case "freebsd":
+		defaultConfigPathDir = "/var/db/netbird/"
 	}
 
 	defaultConfigPath = defaultConfigPathDir + "config.json"
@@ -115,10 +129,14 @@ func init() {
 	rootCmd.PersistentFlags().StringVarP(&serviceName, "service", "s", defaultServiceName, "Netbird system service name")
 	rootCmd.PersistentFlags().StringVarP(&configPath, "config", "c", defaultConfigPath, "Netbird config file location")
 	rootCmd.PersistentFlags().StringVarP(&logLevel, "log-level", "l", "info", "sets Netbird log level")
-	rootCmd.PersistentFlags().StringVar(&logFile, "log-file", defaultLogFile, "sets Netbird log path. If console is specified the log will be output to stdout")
+	rootCmd.PersistentFlags().StringVar(&logFile, "log-file", defaultLogFile, "sets Netbird log path. If console is specified the log will be output to stdout. If syslog is specified the log will be sent to syslog daemon.")
 	rootCmd.PersistentFlags().StringVarP(&setupKey, "setup-key", "k", "", "Setup key obtained from the Management Service Dashboard (used to register peer)")
+	rootCmd.PersistentFlags().StringVar(&setupKeyPath, "setup-key-file", "", "The path to a setup key obtained from the Management Service Dashboard (used to register peer) This is ignored if the setup-key flag is provided.")
+	rootCmd.MarkFlagsMutuallyExclusive("setup-key", "setup-key-file")
 	rootCmd.PersistentFlags().StringVar(&preSharedKey, preSharedKeyFlag, "", "Sets Wireguard PreSharedKey property. If set, then only peers that have the same key can communicate.")
 	rootCmd.PersistentFlags().StringVarP(&hostName, "hostname", "n", "", "Sets a custom hostname for the device")
+	rootCmd.PersistentFlags().BoolVarP(&anonymizeFlag, "anonymize", "A", false, "anonymize IP addresses and non-netbird.io domains in logs and status output")
+
 	rootCmd.AddCommand(serviceCmd)
 	rootCmd.AddCommand(upCmd)
 	rootCmd.AddCommand(downCmd)
@@ -126,8 +144,21 @@ func init() {
 	rootCmd.AddCommand(loginCmd)
 	rootCmd.AddCommand(versionCmd)
 	rootCmd.AddCommand(sshCmd)
+	rootCmd.AddCommand(networksCMD)
+	rootCmd.AddCommand(debugCmd)
+
 	serviceCmd.AddCommand(runCmd, startCmd, stopCmd, restartCmd) // service control commands are subcommands of service
 	serviceCmd.AddCommand(installCmd, uninstallCmd)              // service installer commands are subcommands of service
+
+	networksCMD.AddCommand(routesListCmd)
+	networksCMD.AddCommand(routesSelectCmd, routesDeselectCmd)
+
+	debugCmd.AddCommand(debugBundleCmd)
+	debugCmd.AddCommand(logCmd)
+	logCmd.AddCommand(logLevelCmd)
+	debugCmd.AddCommand(forCmd)
+	debugCmd.AddCommand(persistenceCmd)
+
 	upCmd.PersistentFlags().StringSliceVar(&natExternalIPs, externalIPMapFlag, nil,
 		`Sets external IPs maps between local addresses and interfaces.`+
 			`You can specify a comma-separated list with a single IP and IP/IP or IP/Interface Name. `+
@@ -145,6 +176,8 @@ func init() {
 	upCmd.PersistentFlags().BoolVar(&rosenpassPermissive, rosenpassPermissiveFlag, false, "[Experimental] Enable Rosenpass in permissive mode to allow this peer to accept WireGuard connections without requiring Rosenpass functionality from peers that do not have Rosenpass enabled.")
 	upCmd.PersistentFlags().BoolVar(&serverSSHAllowed, serverSSHAllowedFlag, false, "Allow SSH server on peer. If enabled, the SSH server will be permitted")
 	upCmd.PersistentFlags().BoolVar(&autoConnectDisabled, disableAutoConnectFlag, false, "Disables auto-connect feature. If enabled, then the client won't connect automatically when the service starts.")
+
+	debugCmd.PersistentFlags().BoolVarP(&debugSystemInfoFlag, systemInfoFlag, "S", false, "Adds system information to the debug bundle")
 }
 
 // SetupCloseHandler handles SIGTERM signal and exits with success
@@ -226,6 +259,21 @@ var CLIBackOffSettings = &backoff.ExponentialBackOff{
 	Clock:               backoff.SystemClock,
 }
 
+func getSetupKey() (string, error) {
+	if setupKeyPath != "" && setupKey == "" {
+		return getSetupKeyFromFile(setupKeyPath)
+	}
+	return setupKey, nil
+}
+
+func getSetupKeyFromFile(setupKeyPath string) (string, error) {
+	data, err := os.ReadFile(setupKeyPath)
+	if err != nil {
+		return "", fmt.Errorf("failed to read setup key file: %v", err)
+	}
+	return strings.TrimSpace(string(data)), nil
+}
+
 func handleRebrand(cmd *cobra.Command) error {
 	var err error
 	if logFile == defaultLogFile {
@@ -335,3 +383,17 @@ func migrateToNetbird(oldPath, newPath string) bool {
 
 	return true
 }
+
+func getClient(cmd *cobra.Command) (*grpc.ClientConn, error) {
+	SetFlagsFromEnvVars(rootCmd)
+	cmd.SetOut(cmd.OutOrStdout())
+
+	conn, err := DialClientGRPCServer(cmd.Context(), daemonAddr)
+	if err != nil {
+		return nil, fmt.Errorf("failed to connect to daemon error: %v\n"+
+			"If the daemon is not running please run: "+
+			"\nnetbird service install \nnetbird service start\n", err)
+	}
+
+	return conn, nil
+}

client/cmd/root_test.go

@@ -4,6 +4,10 @@ import (
 	"fmt"
 	"io"
 	"testing"
+
+	"github.com/spf13/cobra"
+
+	"github.com/netbirdio/netbird/client/iface"
 )
 
 func TestInitCommands(t *testing.T) {
@@ -34,3 +38,44 @@ func TestInitCommands(t *testing.T) {
 		})
 	}
 }
+
+func TestSetFlagsFromEnvVars(t *testing.T) {
+	var cmd = &cobra.Command{
+		Use:          "netbird",
+		Long:         "test",
+		SilenceUsage: true,
+		Run: func(cmd *cobra.Command, args []string) {
+			SetFlagsFromEnvVars(cmd)
+		},
+	}
+
+	cmd.PersistentFlags().StringSliceVar(&natExternalIPs, externalIPMapFlag, nil,
+		`comma separated list of external IPs to map to the Wireguard interface`)
+	cmd.PersistentFlags().StringVar(&interfaceName, interfaceNameFlag, iface.WgInterfaceDefault, "Wireguard interface name")
+	cmd.PersistentFlags().BoolVar(&rosenpassEnabled, enableRosenpassFlag, false, "Enable Rosenpass feature Rosenpass.")
+	cmd.PersistentFlags().Uint16Var(&wireguardPort, wireguardPortFlag, iface.DefaultWgPort, "Wireguard interface listening port")
+
+	t.Setenv("NB_EXTERNAL_IP_MAP", "abc,dec")
+	t.Setenv("NB_INTERFACE_NAME", "test-name")
+	t.Setenv("NB_ENABLE_ROSENPASS", "true")
+	t.Setenv("NB_WIREGUARD_PORT", "10000")
+	err := cmd.Execute()
+	if err != nil {
+		t.Fatalf("expected no error while running netbird command, got %v", err)
+	}
+	if len(natExternalIPs) != 2 {
+		t.Errorf("expected 2 external ips, got %d", len(natExternalIPs))
+	}
+	if natExternalIPs[0] != "abc" || natExternalIPs[1] != "dec" {
+		t.Errorf("expected abc,dec, got %s,%s", natExternalIPs[0], natExternalIPs[1])
+	}
+	if interfaceName != "test-name" {
+		t.Errorf("expected test-name, got %s", interfaceName)
+	}
+	if !rosenpassEnabled {
+		t.Errorf("expected rosenpassEnabled to be true, got false")
+	}
+	if wireguardPort != 10000 {
+		t.Errorf("expected wireguardPort to be 10000, got %d", wireguardPort)
+	}
+}

client/cmd/service.go

@@ -2,18 +2,23 @@ package cmd
 
 import (
 	"context"
+	"sync"
+
 	"github.com/kardianos/service"
 	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 	"google.golang.org/grpc"
 
 	"github.com/netbirdio/netbird/client/internal"
+	"github.com/netbirdio/netbird/client/server"
 )
 
 type program struct {
-	ctx    context.Context
-	cancel context.CancelFunc
-	serv   *grpc.Server
+	ctx              context.Context
+	cancel           context.CancelFunc
+	serv             *grpc.Server
+	serverInstance   *server.Server
+	serverInstanceMu sync.Mutex
 }
 
 func newProgram(ctx context.Context, cancel context.CancelFunc) *program {

client/cmd/service_controller.go

@@ -61,6 +61,10 @@ func (p *program) Start(svc service.Service) error {
 		}
 		proto.RegisterDaemonServiceServer(p.serv, serverInstance)
 
+		p.serverInstanceMu.Lock()
+		p.serverInstance = serverInstance
+		p.serverInstanceMu.Unlock()
+
 		log.Printf("started daemon server: %v", split[1])
 		if err := p.serv.Serve(listen); err != nil {
 			log.Errorf("failed to serve daemon requests: %v", err)
@@ -70,6 +74,16 @@ func (p *program) Start(svc service.Service) error {
 }
 
 func (p *program) Stop(srv service.Service) error {
+	p.serverInstanceMu.Lock()
+	if p.serverInstance != nil {
+		in := new(proto.DownRequest)
+		_, err := p.serverInstance.Down(p.ctx, in)
+		if err != nil {
+			log.Errorf("failed to stop daemon: %v", err)
+		}
+	}
+	p.serverInstanceMu.Unlock()
+
 	p.cancel()
 
 	if p.serv != nil {

client/cmd/service_installer.go

@@ -31,6 +31,8 @@ var installCmd = &cobra.Command{
 			configPath,
 			"--log-level",
 			logLevel,
+			"--daemon-addr",
+			daemonAddr,
 		}
 
 		if managementURL != "" {

client/cmd/ssh.go

@@ -9,7 +9,6 @@ import (
 	"strings"
 	"syscall"
 
-	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 
 	"github.com/netbirdio/netbird/client/internal"
@@ -24,7 +23,7 @@ var (
 )
 
 var sshCmd = &cobra.Command{
-	Use: "ssh",
+	Use: "ssh [user@]host",
 	Args: func(cmd *cobra.Command, args []string) error {
 		if len(args) < 1 {
 			return errors.New("requires a host argument")
@@ -73,7 +72,7 @@ var sshCmd = &cobra.Command{
 		go func() {
 			// blocking
 			if err := runSSH(sshctx, host, []byte(config.SSHKey), cmd); err != nil {
-				log.Debug(err)
+				cmd.Printf("Error: %v\n", err)
 				os.Exit(1)
 			}
 			cancel()
@@ -94,7 +93,7 @@ func runSSH(ctx context.Context, addr string, pemKey []byte, cmd *cobra.Command)
 	if err != nil {
 		cmd.Printf("Error: %v\n", err)
 		cmd.Printf("Couldn't connect. Please check the connection status or if the ssh server is enabled on the other peer" +
-			"You can verify the connection by running:\n\n" +
+			"\nYou can verify the connection by running:\n\n" +
 			" netbird status\n\n")
 		return err
 	}

client/cmd/state.go

@@ -0,0 +1,181 @@
+package cmd
+
+import (
+	"fmt"
+
+	log "github.com/sirupsen/logrus"
+	"github.com/spf13/cobra"
+	"google.golang.org/grpc/status"
+
+	"github.com/netbirdio/netbird/client/proto"
+)
+
+var (
+	allFlag bool
+)
+
+var stateCmd = &cobra.Command{
+	Use:   "state",
+	Short: "Manage daemon state",
+	Long:  "Provides commands for managing and inspecting the Netbird daemon state.",
+}
+
+var stateListCmd = &cobra.Command{
+	Use:     "list",
+	Aliases: []string{"ls"},
+	Short:   "List all stored states",
+	Long:    "Lists all registered states with their status and basic information.",
+	Example: "  netbird state list",
+	RunE:    stateList,
+}
+
+var stateCleanCmd = &cobra.Command{
+	Use:   "clean [state-name]",
+	Short: "Clean stored states",
+	Long: `Clean specific state or all states. The daemon must not be running.
+This will perform cleanup operations and remove the state.`,
+	Example: `  netbird state clean dns_state
+  netbird state clean --all`,
+	RunE: stateClean,
+	PreRunE: func(cmd *cobra.Command, args []string) error {
+		// Check mutual exclusivity between --all flag and state-name argument
+		if allFlag && len(args) > 0 {
+			return fmt.Errorf("cannot specify both --all flag and state name")
+		}
+		if !allFlag && len(args) != 1 {
+			return fmt.Errorf("requires a state name argument or --all flag")
+		}
+		return nil
+	},
+}
+
+var stateDeleteCmd = &cobra.Command{
+	Use:   "delete [state-name]",
+	Short: "Delete stored states",
+	Long: `Delete specific state or all states from storage. The daemon must not be running.
+This will remove the state without performing any cleanup operations.`,
+	Example: `  netbird state delete dns_state
+  netbird state delete --all`,
+	RunE: stateDelete,
+	PreRunE: func(cmd *cobra.Command, args []string) error {
+		// Check mutual exclusivity between --all flag and state-name argument
+		if allFlag && len(args) > 0 {
+			return fmt.Errorf("cannot specify both --all flag and state name")
+		}
+		if !allFlag && len(args) != 1 {
+			return fmt.Errorf("requires a state name argument or --all flag")
+		}
+		return nil
+	},
+}
+
+func init() {
+	rootCmd.AddCommand(stateCmd)
+	stateCmd.AddCommand(stateListCmd, stateCleanCmd, stateDeleteCmd)
+
+	stateCleanCmd.Flags().BoolVarP(&allFlag, "all", "a", false, "Clean all states")
+	stateDeleteCmd.Flags().BoolVarP(&allFlag, "all", "a", false, "Delete all states")
+}
+
+func stateList(cmd *cobra.Command, _ []string) error {
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.ListStates(cmd.Context(), &proto.ListStatesRequest{})
+	if err != nil {
+		return fmt.Errorf("failed to list states: %v", status.Convert(err).Message())
+	}
+
+	cmd.Printf("\nStored states:\n\n")
+	for _, state := range resp.States {
+		cmd.Printf("- %s\n", state.Name)
+	}
+
+	return nil
+}
+
+func stateClean(cmd *cobra.Command, args []string) error {
+	var stateName string
+	if !allFlag {
+		stateName = args[0]
+	}
+
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.CleanState(cmd.Context(), &proto.CleanStateRequest{
+		StateName: stateName,
+		All:       allFlag,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to clean state: %v", status.Convert(err).Message())
+	}
+
+	if resp.CleanedStates == 0 {
+		cmd.Println("No states were cleaned")
+		return nil
+	}
+
+	if allFlag {
+		cmd.Printf("Successfully cleaned %d states\n", resp.CleanedStates)
+	} else {
+		cmd.Printf("Successfully cleaned state %q\n", stateName)
+	}
+
+	return nil
+}
+
+func stateDelete(cmd *cobra.Command, args []string) error {
+	var stateName string
+	if !allFlag {
+		stateName = args[0]
+	}
+
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			log.Errorf(errCloseConnection, err)
+		}
+	}()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.DeleteState(cmd.Context(), &proto.DeleteStateRequest{
+		StateName: stateName,
+		All:       allFlag,
+	})
+	if err != nil {
+		return fmt.Errorf("failed to delete state: %v", status.Convert(err).Message())
+	}
+
+	if resp.DeletedStates == 0 {
+		cmd.Println("No states were deleted")
+		return nil
+	}
+
+	if allFlag {
+		cmd.Printf("Successfully deleted %d states\n", resp.DeletedStates)
+	} else {
+		cmd.Printf("Successfully deleted state %q\n", stateName)
+	}
+
+	return nil
+}

client/cmd/status.go

@@ -6,6 +6,8 @@ import (
 	"fmt"
 	"net"
 	"net/netip"
+	"os"
+	"runtime"
 	"sort"
 	"strings"
 	"time"
@@ -14,6 +16,7 @@ import (
 	"google.golang.org/grpc/status"
 	"gopkg.in/yaml.v3"
 
+	"github.com/netbirdio/netbird/client/anonymize"
 	"github.com/netbirdio/netbird/client/internal"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/proto"
@@ -28,15 +31,16 @@ type peerStateDetailOutput struct {
 	Status                 string           `json:"status" yaml:"status"`
 	LastStatusUpdate       time.Time        `json:"lastStatusUpdate" yaml:"lastStatusUpdate"`
 	ConnType               string           `json:"connectionType" yaml:"connectionType"`
-	Direct                 bool             `json:"direct" yaml:"direct"`
 	IceCandidateType       iceCandidateType `json:"iceCandidateType" yaml:"iceCandidateType"`
 	IceCandidateEndpoint   iceCandidateType `json:"iceCandidateEndpoint" yaml:"iceCandidateEndpoint"`
+	RelayAddress           string           `json:"relayAddress" yaml:"relayAddress"`
 	LastWireguardHandshake time.Time        `json:"lastWireguardHandshake" yaml:"lastWireguardHandshake"`
 	TransferReceived       int64            `json:"transferReceived" yaml:"transferReceived"`
 	TransferSent           int64            `json:"transferSent" yaml:"transferSent"`
 	Latency                time.Duration    `json:"latency" yaml:"latency"`
 	RosenpassEnabled       bool             `json:"quantumResistance" yaml:"quantumResistance"`
 	Routes                 []string         `json:"routes" yaml:"routes"`
+	Networks               []string         `json:"networks" yaml:"networks"`
 }
 
 type peersStateOutput struct {
@@ -95,6 +99,7 @@ type statusOutputOverview struct {
 	RosenpassEnabled    bool                       `json:"quantumResistance" yaml:"quantumResistance"`
 	RosenpassPermissive bool                       `json:"quantumResistancePermissive" yaml:"quantumResistancePermissive"`
 	Routes              []string                   `json:"routes" yaml:"routes"`
+	Networks            []string                   `json:"networks" yaml:"networks"`
 	NSServerGroups      []nsServerGroupStateOutput `json:"dnsServers" yaml:"dnsServers"`
 }
 
@@ -144,9 +149,9 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 		return fmt.Errorf("failed initializing log %v", err)
 	}
 
-	ctx := internal.CtxInitState(context.Background())
+	ctx := internal.CtxInitState(cmd.Context())
 
-	resp, err := getStatus(ctx, cmd)
+	resp, err := getStatus(ctx)
 	if err != nil {
 		return err
 	}
@@ -191,7 +196,7 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 	return nil
 }
 
-func getStatus(ctx context.Context, cmd *cobra.Command) (*proto.StatusResponse, error) {
+func getStatus(ctx context.Context) (*proto.StatusResponse, error) {
 	conn, err := DialClientGRPCServer(ctx, daemonAddr)
 	if err != nil {
 		return nil, fmt.Errorf("failed to connect to daemon error: %v\n"+
@@ -200,7 +205,7 @@ func getStatus(ctx context.Context, cmd *cobra.Command) (*proto.StatusResponse,
 	}
 	defer conn.Close()
 
-	resp, err := proto.NewDaemonServiceClient(conn).Status(cmd.Context(), &proto.StatusRequest{GetFullPeerStatus: true})
+	resp, err := proto.NewDaemonServiceClient(conn).Status(ctx, &proto.StatusRequest{GetFullPeerStatus: true})
 	if err != nil {
 		return nil, fmt.Errorf("status failed: %v", status.Convert(err).Message())
 	}
@@ -279,10 +284,16 @@ func convertToStatusOutputOverview(resp *proto.StatusResponse) statusOutputOverv
 		FQDN:                pbFullStatus.GetLocalPeerState().GetFqdn(),
 		RosenpassEnabled:    pbFullStatus.GetLocalPeerState().GetRosenpassEnabled(),
 		RosenpassPermissive: pbFullStatus.GetLocalPeerState().GetRosenpassPermissive(),
-		Routes:              pbFullStatus.GetLocalPeerState().GetRoutes(),
+		Routes:              pbFullStatus.GetLocalPeerState().GetNetworks(),
+		Networks:            pbFullStatus.GetLocalPeerState().GetNetworks(),
 		NSServerGroups:      mapNSGroups(pbFullStatus.GetDnsServers()),
 	}
 
+	if anonymizeFlag {
+		anonymizer := anonymize.NewAnonymizer(anonymize.DefaultAddresses())
+		anonymizeOverview(anonymizer, &overview)
+	}
+
 	return overview
 }
 
@@ -327,16 +338,18 @@ func mapNSGroups(servers []*proto.NSGroupState) []nsServerGroupStateOutput {
 
 func mapPeers(peers []*proto.PeerState) peersStateOutput {
 	var peersStateDetail []peerStateDetailOutput
-	localICE := ""
-	remoteICE := ""
-	localICEEndpoint := ""
-	remoteICEEndpoint := ""
-	connType := ""
 	peersConnected := 0
-	lastHandshake := time.Time{}
-	transferReceived := int64(0)
-	transferSent := int64(0)
 	for _, pbPeerState := range peers {
+		localICE := ""
+		remoteICE := ""
+		localICEEndpoint := ""
+		remoteICEEndpoint := ""
+		relayServerAddress := ""
+		connType := ""
+		lastHandshake := time.Time{}
+		transferReceived := int64(0)
+		transferSent := int64(0)
+
 		isPeerConnected := pbPeerState.ConnStatus == peer.StatusConnected.String()
 		if skipDetailByFilters(pbPeerState, isPeerConnected) {
 			continue
@@ -352,6 +365,7 @@ func mapPeers(peers []*proto.PeerState) peersStateOutput {
 			if pbPeerState.Relayed {
 				connType = "Relayed"
 			}
+			relayServerAddress = pbPeerState.GetRelayAddress()
 			lastHandshake = pbPeerState.GetLastWireguardHandshake().AsTime().Local()
 			transferReceived = pbPeerState.GetBytesRx()
 			transferSent = pbPeerState.GetBytesTx()
@@ -364,7 +378,6 @@ func mapPeers(peers []*proto.PeerState) peersStateOutput {
 			Status:           pbPeerState.GetConnStatus(),
 			LastStatusUpdate: timeLocal,
 			ConnType:         connType,
-			Direct:           pbPeerState.GetDirect(),
 			IceCandidateType: iceCandidateType{
 				Local:  localICE,
 				Remote: remoteICE,
@@ -373,13 +386,15 @@ func mapPeers(peers []*proto.PeerState) peersStateOutput {
 				Local:  localICEEndpoint,
 				Remote: remoteICEEndpoint,
 			},
+			RelayAddress:           relayServerAddress,
 			FQDN:                   pbPeerState.GetFqdn(),
 			LastWireguardHandshake: lastHandshake,
 			TransferReceived:       transferReceived,
 			TransferSent:           transferSent,
 			Latency:                pbPeerState.GetLatency().AsDuration(),
 			RosenpassEnabled:       pbPeerState.GetRosenpassEnabled(),
-			Routes:                 pbPeerState.GetRoutes(),
+			Routes:                 pbPeerState.GetNetworks(),
+			Networks:               pbPeerState.GetNetworks(),
 		}
 
 		peersStateDetail = append(peersStateDetail, peerState)
@@ -480,10 +495,10 @@ func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays
 		relaysString = fmt.Sprintf("%d/%d Available", overview.Relays.Available, overview.Relays.Total)
 	}
 
-	routes := "-"
-	if len(overview.Routes) > 0 {
-		sort.Strings(overview.Routes)
-		routes = strings.Join(overview.Routes, ", ")
+	networks := "-"
+	if len(overview.Networks) > 0 {
+		sort.Strings(overview.Networks)
+		networks = strings.Join(overview.Networks, ", ")
 	}
 
 	var dnsServersString string
@@ -525,8 +540,16 @@ func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays
 
 	peersCountString := fmt.Sprintf("%d/%d Connected", overview.Peers.Connected, overview.Peers.Total)
 
+	goos := runtime.GOOS
+	goarch := runtime.GOARCH
+	goarm := ""
+	if goarch == "arm" {
+		goarm = fmt.Sprintf(" (ARMv%s)", os.Getenv("GOARM"))
+	}
+
 	summary := fmt.Sprintf(
-		"Daemon version: %s\n"+
+		"OS: %s\n"+
+			"Daemon version: %s\n"+
 			"CLI version: %s\n"+
 			"Management: %s\n"+
 			"Signal: %s\n"+
@@ -537,7 +560,9 @@ func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays
 			"Interface type: %s\n"+
 			"Quantum resistance: %s\n"+
 			"Routes: %s\n"+
+			"Networks: %s\n"+
 			"Peers count: %s\n",
+		fmt.Sprintf("%s/%s%s", goos, goarch, goarm),
 		overview.DaemonVersion,
 		version.NetbirdVersion(),
 		managementConnString,
@@ -548,7 +573,8 @@ func parseGeneralSummary(overview statusOutputOverview, showURL bool, showRelays
 		interfaceIP,
 		interfaceTypeString,
 		rosenpassEnabledStatus,
-		routes,
+		networks,
+		networks,
 		peersCountString,
 	)
 	return summary
@@ -593,15 +619,6 @@ func parsePeers(peers peersStateOutput, rosenpassEnabled, rosenpassPermissive bo
 		if peerState.IceCandidateEndpoint.Remote != "" {
 			remoteICEEndpoint = peerState.IceCandidateEndpoint.Remote
 		}
-		lastStatusUpdate := "-"
-		if !peerState.LastStatusUpdate.IsZero() {
-			lastStatusUpdate = peerState.LastStatusUpdate.Format("2006-01-02 15:04:05")
-		}
-
-		lastWireGuardHandshake := "-"
-		if !peerState.LastWireguardHandshake.IsZero() && peerState.LastWireguardHandshake != time.Unix(0, 0) {
-			lastWireGuardHandshake = peerState.LastWireguardHandshake.Format("2006-01-02 15:04:05")
-		}
 
 		rosenpassEnabledStatus := "false"
 		if rosenpassEnabled {
@@ -620,10 +637,10 @@ func parsePeers(peers peersStateOutput, rosenpassEnabled, rosenpassPermissive bo
 			}
 		}
 
-		routes := "-"
-		if len(peerState.Routes) > 0 {
-			sort.Strings(peerState.Routes)
-			routes = strings.Join(peerState.Routes, ", ")
+		networks := "-"
+		if len(peerState.Networks) > 0 {
+			sort.Strings(peerState.Networks)
+			networks = strings.Join(peerState.Networks, ", ")
 		}
 
 		peerString := fmt.Sprintf(
@@ -633,31 +650,33 @@ func parsePeers(peers peersStateOutput, rosenpassEnabled, rosenpassPermissive bo
 				"  Status: %s\n"+
 				"  -- detail --\n"+
 				"  Connection type: %s\n"+
-				"  Direct: %t\n"+
 				"  ICE candidate (Local/Remote): %s/%s\n"+
 				"  ICE candidate endpoints (Local/Remote): %s/%s\n"+
+				"  Relay server address: %s\n"+
 				"  Last connection update: %s\n"+
 				"  Last WireGuard handshake: %s\n"+
 				"  Transfer status (received/sent) %s/%s\n"+
 				"  Quantum resistance: %s\n"+
 				"  Routes: %s\n"+
+				"  Networks: %s\n"+
 				"  Latency: %s\n",
 			peerState.FQDN,
 			peerState.IP,
 			peerState.PubKey,
 			peerState.Status,
 			peerState.ConnType,
-			peerState.Direct,
 			localICE,
 			remoteICE,
 			localICEEndpoint,
 			remoteICEEndpoint,
-			lastStatusUpdate,
-			lastWireGuardHandshake,
+			peerState.RelayAddress,
+			timeAgo(peerState.LastStatusUpdate),
+			timeAgo(peerState.LastWireguardHandshake),
 			toIEC(peerState.TransferReceived),
 			toIEC(peerState.TransferSent),
 			rosenpassEnabledStatus,
-			routes,
+			networks,
+			networks,
 			peerState.Latency.String(),
 		)
 
@@ -669,7 +688,7 @@ func parsePeers(peers peersStateOutput, rosenpassEnabled, rosenpassPermissive bo
 func skipDetailByFilters(peerState *proto.PeerState, isConnected bool) bool {
 	statusEval := false
 	ipEval := false
-	nameEval := false
+	nameEval := true
 
 	if statusFilter != "" {
 		lowerStatusFilter := strings.ToLower(statusFilter)
@@ -689,11 +708,13 @@ func skipDetailByFilters(peerState *proto.PeerState, isConnected bool) bool {
 
 	if len(prefixNamesFilter) > 0 {
 		for prefixNameFilter := range prefixNamesFilterMap {
-			if !strings.HasPrefix(peerState.Fqdn, prefixNameFilter) {
-				nameEval = true
+			if strings.HasPrefix(peerState.Fqdn, prefixNameFilter) {
+				nameEval = false
 				break
 			}
 		}
+	} else {
+		nameEval = false
 	}
 
 	return statusEval || ipEval || nameEval
@@ -722,3 +743,136 @@ func countEnabled(dnsServers []nsServerGroupStateOutput) int {
 	}
 	return count
 }
+
+// timeAgo returns a string representing the duration since the provided time in a human-readable format.
+func timeAgo(t time.Time) string {
+	if t.IsZero() || t.Equal(time.Unix(0, 0)) {
+		return "-"
+	}
+	duration := time.Since(t)
+	switch {
+	case duration < time.Second:
+		return "Now"
+	case duration < time.Minute:
+		seconds := int(duration.Seconds())
+		if seconds == 1 {
+			return "1 second ago"
+		}
+		return fmt.Sprintf("%d seconds ago", seconds)
+	case duration < time.Hour:
+		minutes := int(duration.Minutes())
+		seconds := int(duration.Seconds()) % 60
+		if minutes == 1 {
+			if seconds == 1 {
+				return "1 minute, 1 second ago"
+			} else if seconds > 0 {
+				return fmt.Sprintf("1 minute, %d seconds ago", seconds)
+			}
+			return "1 minute ago"
+		}
+		if seconds > 0 {
+			return fmt.Sprintf("%d minutes, %d seconds ago", minutes, seconds)
+		}
+		return fmt.Sprintf("%d minutes ago", minutes)
+	case duration < 24*time.Hour:
+		hours := int(duration.Hours())
+		minutes := int(duration.Minutes()) % 60
+		if hours == 1 {
+			if minutes == 1 {
+				return "1 hour, 1 minute ago"
+			} else if minutes > 0 {
+				return fmt.Sprintf("1 hour, %d minutes ago", minutes)
+			}
+			return "1 hour ago"
+		}
+		if minutes > 0 {
+			return fmt.Sprintf("%d hours, %d minutes ago", hours, minutes)
+		}
+		return fmt.Sprintf("%d hours ago", hours)
+	}
+
+	days := int(duration.Hours()) / 24
+	hours := int(duration.Hours()) % 24
+	if days == 1 {
+		if hours == 1 {
+			return "1 day, 1 hour ago"
+		} else if hours > 0 {
+			return fmt.Sprintf("1 day, %d hours ago", hours)
+		}
+		return "1 day ago"
+	}
+	if hours > 0 {
+		return fmt.Sprintf("%d days, %d hours ago", days, hours)
+	}
+	return fmt.Sprintf("%d days ago", days)
+}
+
+func anonymizePeerDetail(a *anonymize.Anonymizer, peer *peerStateDetailOutput) {
+	peer.FQDN = a.AnonymizeDomain(peer.FQDN)
+	if localIP, port, err := net.SplitHostPort(peer.IceCandidateEndpoint.Local); err == nil {
+		peer.IceCandidateEndpoint.Local = fmt.Sprintf("%s:%s", a.AnonymizeIPString(localIP), port)
+	}
+	if remoteIP, port, err := net.SplitHostPort(peer.IceCandidateEndpoint.Remote); err == nil {
+		peer.IceCandidateEndpoint.Remote = fmt.Sprintf("%s:%s", a.AnonymizeIPString(remoteIP), port)
+	}
+
+	peer.RelayAddress = a.AnonymizeURI(peer.RelayAddress)
+
+	for i, route := range peer.Networks {
+		peer.Networks[i] = a.AnonymizeIPString(route)
+	}
+
+	for i, route := range peer.Networks {
+		peer.Networks[i] = a.AnonymizeRoute(route)
+	}
+
+	for i, route := range peer.Routes {
+		peer.Routes[i] = a.AnonymizeIPString(route)
+	}
+
+	for i, route := range peer.Routes {
+		peer.Routes[i] = a.AnonymizeRoute(route)
+	}
+}
+
+func anonymizeOverview(a *anonymize.Anonymizer, overview *statusOutputOverview) {
+	for i, peer := range overview.Peers.Details {
+		peer := peer
+		anonymizePeerDetail(a, &peer)
+		overview.Peers.Details[i] = peer
+	}
+
+	overview.ManagementState.URL = a.AnonymizeURI(overview.ManagementState.URL)
+	overview.ManagementState.Error = a.AnonymizeString(overview.ManagementState.Error)
+	overview.SignalState.URL = a.AnonymizeURI(overview.SignalState.URL)
+	overview.SignalState.Error = a.AnonymizeString(overview.SignalState.Error)
+
+	overview.IP = a.AnonymizeIPString(overview.IP)
+	for i, detail := range overview.Relays.Details {
+		detail.URI = a.AnonymizeURI(detail.URI)
+		detail.Error = a.AnonymizeString(detail.Error)
+		overview.Relays.Details[i] = detail
+	}
+
+	for i, nsGroup := range overview.NSServerGroups {
+		for j, domain := range nsGroup.Domains {
+			overview.NSServerGroups[i].Domains[j] = a.AnonymizeDomain(domain)
+		}
+		for j, ns := range nsGroup.Servers {
+			host, port, err := net.SplitHostPort(ns)
+			if err == nil {
+				overview.NSServerGroups[i].Servers[j] = fmt.Sprintf("%s:%s", a.AnonymizeIPString(host), port)
+			}
+		}
+	}
+
+	for i, route := range overview.Networks {
+		overview.Networks[i] = a.AnonymizeRoute(route)
+	}
+
+	for i, route := range overview.Routes {
+		overview.Routes[i] = a.AnonymizeRoute(route)
+	}
+
+	overview.FQDN = a.AnonymizeDomain(overview.FQDN)
+}

client/cmd/status_test.go

@@ -3,6 +3,8 @@ package cmd
 import (
 	"bytes"
 	"encoding/json"
+	"fmt"
+	"runtime"
 	"testing"
 	"time"
 
@@ -35,7 +37,6 @@ var resp = &proto.StatusResponse{
 				ConnStatus:                 "Connected",
 				ConnStatusUpdate:           timestamppb.New(time.Date(2001, time.Month(1), 1, 1, 1, 1, 0, time.UTC)),
 				Relayed:                    false,
-				Direct:                     true,
 				LocalIceCandidateType:      "",
 				RemoteIceCandidateType:     "",
 				LocalIceCandidateEndpoint:  "",
@@ -43,7 +44,7 @@ var resp = &proto.StatusResponse{
 				LastWireguardHandshake:     timestamppb.New(time.Date(2001, time.Month(1), 1, 1, 1, 2, 0, time.UTC)),
 				BytesRx:                    200,
 				BytesTx:                    100,
-				Routes: []string{
+				Networks: []string{
 					"10.1.0.0/24",
 				},
 				Latency: durationpb.New(time.Duration(10000000)),
@@ -55,7 +56,6 @@ var resp = &proto.StatusResponse{
 				ConnStatus:                 "Connected",
 				ConnStatusUpdate:           timestamppb.New(time.Date(2002, time.Month(2), 2, 2, 2, 2, 0, time.UTC)),
 				Relayed:                    true,
-				Direct:                     false,
 				LocalIceCandidateType:      "relay",
 				RemoteIceCandidateType:     "prflx",
 				LocalIceCandidateEndpoint:  "10.0.0.1:10001",
@@ -93,7 +93,7 @@ var resp = &proto.StatusResponse{
 			PubKey:          "Some-Pub-Key",
 			KernelInterface: true,
 			Fqdn:            "some-localhost.awesome-domain.com",
-			Routes: []string{
+			Networks: []string{
 				"10.10.0.0/24",
 			},
 		},
@@ -135,7 +135,6 @@ var overview = statusOutputOverview{
 				Status:           "Connected",
 				LastStatusUpdate: time.Date(2001, 1, 1, 1, 1, 1, 0, time.UTC),
 				ConnType:         "P2P",
-				Direct:           true,
 				IceCandidateType: iceCandidateType{
 					Local:  "",
 					Remote: "",
@@ -150,6 +149,9 @@ var overview = statusOutputOverview{
 				Routes: []string{
 					"10.1.0.0/24",
 				},
+				Networks: []string{
+					"10.1.0.0/24",
+				},
 				Latency: time.Duration(10000000),
 			},
 			{
@@ -159,7 +161,6 @@ var overview = statusOutputOverview{
 				Status:           "Connected",
 				LastStatusUpdate: time.Date(2002, 2, 2, 2, 2, 2, 0, time.UTC),
 				ConnType:         "Relayed",
-				Direct:           false,
 				IceCandidateType: iceCandidateType{
 					Local:  "relay",
 					Remote: "prflx",
@@ -232,6 +233,9 @@ var overview = statusOutputOverview{
 	Routes: []string{
 		"10.10.0.0/24",
 	},
+	Networks: []string{
+		"10.10.0.0/24",
+	},
 }
 
 func TestConversionFromFullStatusToOutputOverview(t *testing.T) {
@@ -281,7 +285,6 @@ func TestParsingToJSON(t *testing.T) {
                 "status": "Connected",
                 "lastStatusUpdate": "2001-01-01T01:01:01Z",
                 "connectionType": "P2P",
-                "direct": true,
                 "iceCandidateType": {
                   "local": "",
                   "remote": ""
@@ -290,6 +293,7 @@ func TestParsingToJSON(t *testing.T) {
                   "local": "",
                   "remote": ""
                 },
+				"relayAddress": "",
                 "lastWireguardHandshake": "2001-01-01T01:01:02Z",
                 "transferReceived": 200,
                 "transferSent": 100,
@@ -297,6 +301,9 @@ func TestParsingToJSON(t *testing.T) {
                 "quantumResistance": false,
                 "routes": [
                   "10.1.0.0/24"
+                ],
+                "networks": [
+                  "10.1.0.0/24"
                 ]
               },
               {
@@ -306,7 +313,6 @@ func TestParsingToJSON(t *testing.T) {
                 "status": "Connected",
                 "lastStatusUpdate": "2002-02-02T02:02:02Z",
                 "connectionType": "Relayed",
-                "direct": false,
                 "iceCandidateType": {
                   "local": "relay",
                   "remote": "prflx"
@@ -315,12 +321,14 @@ func TestParsingToJSON(t *testing.T) {
                   "local": "10.0.0.1:10001",
                   "remote": "10.0.10.1:10002"
                 },
+				"relayAddress": "",
                 "lastWireguardHandshake": "2002-02-02T02:02:03Z",
                 "transferReceived": 2000,
                 "transferSent": 1000,
 				"latency": 10000000,
                 "quantumResistance": false,
-                "routes": null
+                "routes": null,
+                "networks": null
               }
             ]
           },
@@ -361,6 +369,9 @@ func TestParsingToJSON(t *testing.T) {
           "routes": [
             "10.10.0.0/24"
           ],
+          "networks": [
+            "10.10.0.0/24"
+          ],
           "dnsServers": [
             {
               "servers": [
@@ -406,13 +417,13 @@ func TestParsingToYAML(t *testing.T) {
           status: Connected
           lastStatusUpdate: 2001-01-01T01:01:01Z
           connectionType: P2P
-          direct: true
           iceCandidateType:
             local: ""
             remote: ""
           iceCandidateEndpoint:
             local: ""
             remote: ""
+          relayAddress: ""
           lastWireguardHandshake: 2001-01-01T01:01:02Z
           transferReceived: 200
           transferSent: 100
@@ -420,25 +431,28 @@ func TestParsingToYAML(t *testing.T) {
           quantumResistance: false
           routes:
             - 10.1.0.0/24
+          networks:
+            - 10.1.0.0/24
         - fqdn: peer-2.awesome-domain.com
           netbirdIp: 192.168.178.102
           publicKey: Pubkey2
           status: Connected
           lastStatusUpdate: 2002-02-02T02:02:02Z
           connectionType: Relayed
-          direct: false
           iceCandidateType:
             local: relay
             remote: prflx
           iceCandidateEndpoint:
             local: 10.0.0.1:10001
             remote: 10.0.10.1:10002
+          relayAddress: ""
           lastWireguardHandshake: 2002-02-02T02:02:03Z
           transferReceived: 2000
           transferSent: 1000
           latency: 10ms
           quantumResistance: false
           routes: []
+          networks: []
 cliVersion: development
 daemonVersion: 0.14.1
 management:
@@ -467,6 +481,8 @@ quantumResistance: false
 quantumResistancePermissive: false
 routes:
     - 10.10.0.0/24
+networks:
+    - 10.10.0.0/24
 dnsServers:
     - servers:
         - 8.8.8.8:53
@@ -487,9 +503,15 @@ dnsServers:
 }
 
 func TestParsingToDetail(t *testing.T) {
+	// Calculate time ago based on the fixture dates
+	lastConnectionUpdate1 := timeAgo(overview.Peers.Details[0].LastStatusUpdate)
+	lastHandshake1 := timeAgo(overview.Peers.Details[0].LastWireguardHandshake)
+	lastConnectionUpdate2 := timeAgo(overview.Peers.Details[1].LastStatusUpdate)
+	lastHandshake2 := timeAgo(overview.Peers.Details[1].LastWireguardHandshake)
+
 	detail := parseToFullDetailSummary(overview)
 
-	expectedDetail :=
+	expectedDetail := fmt.Sprintf(
 		`Peers detail:
  peer-1.awesome-domain.com:
   NetBird IP: 192.168.178.101
@@ -497,14 +519,15 @@ func TestParsingToDetail(t *testing.T) {
   Status: Connected
   -- detail --
   Connection type: P2P
-  Direct: true
   ICE candidate (Local/Remote): -/-
   ICE candidate endpoints (Local/Remote): -/-
-  Last connection update: 2001-01-01 01:01:01
-  Last WireGuard handshake: 2001-01-01 01:01:02
+  Relay server address: 
+  Last connection update: %s
+  Last WireGuard handshake: %s
   Transfer status (received/sent) 200 B/100 B
   Quantum resistance: false
   Routes: 10.1.0.0/24
+  Networks: 10.1.0.0/24
   Latency: 10ms
 
  peer-2.awesome-domain.com:
@@ -513,18 +536,20 @@ func TestParsingToDetail(t *testing.T) {
   Status: Connected
   -- detail --
   Connection type: Relayed
-  Direct: false
   ICE candidate (Local/Remote): relay/prflx
   ICE candidate endpoints (Local/Remote): 10.0.0.1:10001/10.0.10.1:10002
-  Last connection update: 2002-02-02 02:02:02
-  Last WireGuard handshake: 2002-02-02 02:02:03
+  Relay server address: 
+  Last connection update: %s
+  Last WireGuard handshake: %s
   Transfer status (received/sent) 2.0 KiB/1000 B
   Quantum resistance: false
   Routes: -
+  Networks: -
   Latency: 10ms
 
+OS: %s/%s
 Daemon version: 0.14.1
-CLI version: development
+CLI version: %s
 Management: Connected to my-awesome-management.com:443
 Signal: Connected to my-awesome-signal.com:443
 Relays: 
@@ -538,8 +563,9 @@ NetBird IP: 192.168.178.100/16
 Interface type: Kernel
 Quantum resistance: false
 Routes: 10.10.0.0/24
+Networks: 10.10.0.0/24
 Peers count: 2/2 Connected
-`
+`, lastConnectionUpdate1, lastHandshake1, lastConnectionUpdate2, lastHandshake2, runtime.GOOS, runtime.GOARCH, overview.CliVersion)
 
 	assert.Equal(t, expectedDetail, detail)
 }
@@ -547,8 +573,8 @@ Peers count: 2/2 Connected
 func TestParsingToShortVersion(t *testing.T) {
 	shortVersion := parseGeneralSummary(overview, false, false, false)
 
-	expectedString :=
-		`Daemon version: 0.14.1
+	expectedString := fmt.Sprintf("OS: %s/%s", runtime.GOOS, runtime.GOARCH) + `
+Daemon version: 0.14.1
 CLI version: development
 Management: Connected
 Signal: Connected
@@ -559,6 +585,7 @@ NetBird IP: 192.168.178.100/16
 Interface type: Kernel
 Quantum resistance: false
 Routes: 10.10.0.0/24
+Networks: 10.10.0.0/24
 Peers count: 2/2 Connected
 `
 
@@ -572,3 +599,31 @@ func TestParsingOfIP(t *testing.T) {
 
 	assert.Equal(t, "192.168.178.123\n", parsedIP)
 }
+
+func TestTimeAgo(t *testing.T) {
+	now := time.Now()
+
+	cases := []struct {
+		name     string
+		input    time.Time
+		expected string
+	}{
+		{"Now", now, "Now"},
+		{"Seconds ago", now.Add(-10 * time.Second), "10 seconds ago"},
+		{"One minute ago", now.Add(-1 * time.Minute), "1 minute ago"},
+		{"Minutes and seconds ago", now.Add(-(1*time.Minute + 30*time.Second)), "1 minute, 30 seconds ago"},
+		{"One hour ago", now.Add(-1 * time.Hour), "1 hour ago"},
+		{"Hours and minutes ago", now.Add(-(2*time.Hour + 15*time.Minute)), "2 hours, 15 minutes ago"},
+		{"One day ago", now.Add(-24 * time.Hour), "1 day ago"},
+		{"Multiple days ago", now.Add(-(72*time.Hour + 20*time.Minute)), "3 days ago"},
+		{"Zero time", time.Time{}, "-"},
+		{"Unix zero time", time.Unix(0, 0), "-"},
+	}
+
+	for _, tc := range cases {
+		t.Run(tc.name, func(t *testing.T) {
+			result := timeAgo(tc.input)
+			assert.Equal(t, tc.expected, result, "Failed %s", tc.name)
+		})
+	}
+}

client/cmd/system.go

@@ -0,0 +1,31 @@
+package cmd
+
+// Flag constants for system configuration
+const (
+	disableClientRoutesFlag = "disable-client-routes"
+	disableServerRoutesFlag = "disable-server-routes"
+	disableDNSFlag          = "disable-dns"
+	disableFirewallFlag     = "disable-firewall"
+)
+
+var (
+	disableClientRoutes bool
+	disableServerRoutes bool
+	disableDNS          bool
+	disableFirewall     bool
+)
+
+func init() {
+	// Add system flags to upCmd
+	upCmd.PersistentFlags().BoolVar(&disableClientRoutes, disableClientRoutesFlag, false,
+		"Disable client routes. If enabled, the client won't process client routes received from the management service.")
+
+	upCmd.PersistentFlags().BoolVar(&disableServerRoutes, disableServerRoutesFlag, false,
+		"Disable server routes. If enabled, the client won't act as a router for server routes received from the management service.")
+
+	upCmd.PersistentFlags().BoolVar(&disableDNS, disableDNSFlag, false,
+		"Disable DNS. If enabled, the client won't configure DNS settings.")
+
+	upCmd.PersistentFlags().BoolVar(&disableFirewall, disableFirewallFlag, false,
+		"Disable firewall configuration. If enabled, the client won't modify firewall rules.")
+}

client/cmd/testutil_test.go

@@ -3,17 +3,23 @@ package cmd
 import (
 	"context"
 	"net"
-	"path/filepath"
 	"testing"
 	"time"
 
+	"github.com/stretchr/testify/require"
+	"go.opentelemetry.io/otel"
+
 	"github.com/netbirdio/netbird/management/server/activity"
+	"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/util"
 
 	"google.golang.org/grpc"
 
 	"github.com/netbirdio/management-integrations/integrations"
+
 	clientProto "github.com/netbirdio/netbird/client/proto"
 	client "github.com/netbirdio/netbird/client/server"
 	mgmtProto "github.com/netbirdio/netbird/management/proto"
@@ -29,18 +35,12 @@ func startTestingServices(t *testing.T) string {
 	if err != nil {
 		t.Fatal(err)
 	}
-	testDir := t.TempDir()
-	config.Datadir = testDir
-	err = util.CopyFileContents("../testdata/store.json", filepath.Join(testDir, "store.json"))
-	if err != nil {
-		t.Fatal(err)
-	}
 
 	_, signalLis := startSignal(t)
 	signalAddr := signalLis.Addr().String()
 	config.Signal.URI = signalAddr
 
-	_, mgmLis := startManagement(t, config)
+	_, mgmLis := startManagement(t, config, "../testdata/store.sql")
 	mgmAddr := mgmLis.Addr().String()
 	return mgmAddr
 }
@@ -52,7 +52,10 @@ func startSignal(t *testing.T) (*grpc.Server, net.Listener) {
 		t.Fatal(err)
 	}
 	s := grpc.NewServer()
-	sigProto.RegisterSignalExchangeServer(s, sig.NewServer())
+	srv, err := sig.NewServer(context.Background(), otel.Meter(""))
+	require.NoError(t, err)
+
+	sigProto.RegisterSignalExchangeServer(s, srv)
 	go func() {
 		if err := s.Serve(lis); err != nil {
 			panic(err)
@@ -62,30 +65,37 @@ func startSignal(t *testing.T) (*grpc.Server, net.Listener) {
 	return s, lis
 }
 
-func startManagement(t *testing.T, config *mgmt.Config) (*grpc.Server, net.Listener) {
+func startManagement(t *testing.T, config *mgmt.Config, testFile string) (*grpc.Server, net.Listener) {
 	t.Helper()
+
 	lis, err := net.Listen("tcp", ":0")
 	if err != nil {
 		t.Fatal(err)
 	}
 	s := grpc.NewServer()
-	store, err := mgmt.NewStoreFromJson(config.Datadir, nil)
+	store, cleanUp, err := store.NewTestStoreFromSQL(context.Background(), testFile, t.TempDir())
 	if err != nil {
 		t.Fatal(err)
 	}
+	t.Cleanup(cleanUp)
 
 	peersUpdateManager := mgmt.NewPeersUpdateManager(nil)
 	eventStore := &activity.InMemoryEventStore{}
 	if err != nil {
 		return nil, nil
 	}
-	iv, _ := integrations.NewIntegratedValidator(eventStore)
-	accountManager, err := mgmt.BuildManager(store, peersUpdateManager, nil, "", "netbird.selfhosted", eventStore, nil, false, iv)
+	iv, _ := integrations.NewIntegratedValidator(context.Background(), eventStore)
+
+	metrics, err := telemetry.NewDefaultAppMetrics(context.Background())
+	require.NoError(t, err)
+
+	accountManager, err := mgmt.BuildManager(context.Background(), store, peersUpdateManager, nil, "", "netbird.selfhosted", eventStore, nil, false, iv, metrics)
 	if err != nil {
 		t.Fatal(err)
 	}
-	turnManager := mgmt.NewTimeBasedAuthSecretsManager(peersUpdateManager, config.TURNConfig)
-	mgmtServer, err := mgmt.NewServer(config, accountManager, peersUpdateManager, turnManager, nil, nil)
+
+	secretsManager := mgmt.NewTimeBasedAuthSecretsManager(peersUpdateManager, config.TURNConfig, config.Relay)
+	mgmtServer, err := mgmt.NewServer(context.Background(), config, accountManager, settings.NewManager(store), peersUpdateManager, secretsManager, nil, nil)
 	if err != nil {
 		t.Fatal(err)
 	}
@@ -100,7 +110,7 @@ func startManagement(t *testing.T, config *mgmt.Config) (*grpc.Server, net.Liste
 }
 
 func startClientDaemon(
-	t *testing.T, ctx context.Context, managementURL, configPath string,
+	t *testing.T, ctx context.Context, _, configPath string,
 ) (*grpc.Server, net.Listener) {
 	t.Helper()
 	lis, err := net.Listen("tcp", "127.0.0.1:0")

client/cmd/trace.go

@@ -0,0 +1,137 @@
+package cmd
+
+import (
+	"fmt"
+	"math/rand"
+	"strings"
+
+	"github.com/spf13/cobra"
+	"google.golang.org/grpc/status"
+
+	"github.com/netbirdio/netbird/client/proto"
+)
+
+var traceCmd = &cobra.Command{
+	Use:   "trace <direction> <source-ip> <dest-ip>",
+	Short: "Trace a packet through the firewall",
+	Example: `
+  netbird debug trace in 192.168.1.10 10.10.0.2 -p tcp --sport 12345 --dport 443 --syn --ack
+  netbird debug trace out 10.10.0.1 8.8.8.8 -p udp  --dport 53
+  netbird debug trace in 10.10.0.2 10.10.0.1 -p icmp --type 8 --code 0
+  netbird debug trace in 100.64.1.1 self -p tcp --dport 80`,
+	Args: cobra.ExactArgs(3),
+	RunE: tracePacket,
+}
+
+func init() {
+	debugCmd.AddCommand(traceCmd)
+
+	traceCmd.Flags().StringP("protocol", "p", "tcp", "Protocol (tcp/udp/icmp)")
+	traceCmd.Flags().Uint16("sport", 0, "Source port")
+	traceCmd.Flags().Uint16("dport", 0, "Destination port")
+	traceCmd.Flags().Uint8("icmp-type", 0, "ICMP type")
+	traceCmd.Flags().Uint8("icmp-code", 0, "ICMP code")
+	traceCmd.Flags().Bool("syn", false, "TCP SYN flag")
+	traceCmd.Flags().Bool("ack", false, "TCP ACK flag")
+	traceCmd.Flags().Bool("fin", false, "TCP FIN flag")
+	traceCmd.Flags().Bool("rst", false, "TCP RST flag")
+	traceCmd.Flags().Bool("psh", false, "TCP PSH flag")
+	traceCmd.Flags().Bool("urg", false, "TCP URG flag")
+}
+
+func tracePacket(cmd *cobra.Command, args []string) error {
+	direction := strings.ToLower(args[0])
+	if direction != "in" && direction != "out" {
+		return fmt.Errorf("invalid direction: use 'in' or 'out'")
+	}
+
+	protocol := cmd.Flag("protocol").Value.String()
+	if protocol != "tcp" && protocol != "udp" && protocol != "icmp" {
+		return fmt.Errorf("invalid protocol: use tcp/udp/icmp")
+	}
+
+	sport, err := cmd.Flags().GetUint16("sport")
+	if err != nil {
+		return fmt.Errorf("invalid source port: %v", err)
+	}
+	dport, err := cmd.Flags().GetUint16("dport")
+	if err != nil {
+		return fmt.Errorf("invalid destination port: %v", err)
+	}
+
+	// For TCP/UDP, generate random ephemeral port (49152-65535) if not specified
+	if protocol != "icmp" {
+		if sport == 0 {
+			sport = uint16(rand.Intn(16383) + 49152)
+		}
+		if dport == 0 {
+			dport = uint16(rand.Intn(16383) + 49152)
+		}
+	}
+
+	var tcpFlags *proto.TCPFlags
+	if protocol == "tcp" {
+		syn, _ := cmd.Flags().GetBool("syn")
+		ack, _ := cmd.Flags().GetBool("ack")
+		fin, _ := cmd.Flags().GetBool("fin")
+		rst, _ := cmd.Flags().GetBool("rst")
+		psh, _ := cmd.Flags().GetBool("psh")
+		urg, _ := cmd.Flags().GetBool("urg")
+
+		tcpFlags = &proto.TCPFlags{
+			Syn: syn,
+			Ack: ack,
+			Fin: fin,
+			Rst: rst,
+			Psh: psh,
+			Urg: urg,
+		}
+	}
+
+	icmpType, _ := cmd.Flags().GetUint32("icmp-type")
+	icmpCode, _ := cmd.Flags().GetUint32("icmp-code")
+
+	conn, err := getClient(cmd)
+	if err != nil {
+		return err
+	}
+	defer conn.Close()
+
+	client := proto.NewDaemonServiceClient(conn)
+	resp, err := client.TracePacket(cmd.Context(), &proto.TracePacketRequest{
+		SourceIp:        args[1],
+		DestinationIp:   args[2],
+		Protocol:        protocol,
+		SourcePort:      uint32(sport),
+		DestinationPort: uint32(dport),
+		Direction:       direction,
+		TcpFlags:        tcpFlags,
+		IcmpType:        &icmpType,
+		IcmpCode:        &icmpCode,
+	})
+	if err != nil {
+		return fmt.Errorf("trace failed: %v", status.Convert(err).Message())
+	}
+
+	printTrace(cmd, args[1], args[2], protocol, sport, dport, resp)
+	return nil
+}
+
+func printTrace(cmd *cobra.Command, src, dst, proto string, sport, dport uint16, resp *proto.TracePacketResponse) {
+	cmd.Printf("Packet trace %s:%d -> %s:%d (%s)\n\n", src, sport, dst, dport, strings.ToUpper(proto))
+
+	for _, stage := range resp.Stages {
+		if stage.ForwardingDetails != nil {
+			cmd.Printf("%s: %s [%s]\n", stage.Name, stage.Message, *stage.ForwardingDetails)
+		} else {
+			cmd.Printf("%s: %s\n", stage.Name, stage.Message)
+		}
+	}
+
+	disposition := map[bool]string{
+		true:  "\033[32mALLOWED\033[0m", // Green
+		false: "\033[31mDENIED\033[0m",  // Red
+	}[resp.FinalDisposition]
+
+	cmd.Printf("\nFinal disposition: %s\n", disposition)
+}

client/cmd/up.go

@@ -7,17 +7,19 @@ import (
 	"net/netip"
 	"runtime"
 	"strings"
+	"time"
 
 	log "github.com/sirupsen/logrus"
 	"github.com/spf13/cobra"
 	"google.golang.org/grpc/codes"
 	gstatus "google.golang.org/grpc/status"
+	"google.golang.org/protobuf/types/known/durationpb"
 
+	"github.com/netbirdio/netbird/client/iface"
 	"github.com/netbirdio/netbird/client/internal"
 	"github.com/netbirdio/netbird/client/internal/peer"
 	"github.com/netbirdio/netbird/client/proto"
 	"github.com/netbirdio/netbird/client/system"
-	"github.com/netbirdio/netbird/iface"
 	"github.com/netbirdio/netbird/util"
 )
 
@@ -40,7 +42,13 @@ func init() {
 	upCmd.PersistentFlags().BoolVarP(&foregroundMode, "foreground-mode", "F", false, "start service in foreground")
 	upCmd.PersistentFlags().StringVar(&interfaceName, interfaceNameFlag, iface.WgInterfaceDefault, "Wireguard interface name")
 	upCmd.PersistentFlags().Uint16Var(&wireguardPort, wireguardPortFlag, iface.DefaultWgPort, "Wireguard interface listening port")
+	upCmd.PersistentFlags().BoolVarP(&networkMonitor, networkMonitorFlag, "N", networkMonitor,
+		`Manage network monitoring. Defaults to true on Windows and macOS, false on Linux. `+
+			`E.g. --network-monitor=false to disable or --network-monitor=true to enable.`,
+	)
 	upCmd.PersistentFlags().StringSliceVar(&extraIFaceBlackList, extraIFaceBlackListFlag, nil, "Extra list of default interfaces to ignore for listening")
+	upCmd.PersistentFlags().DurationVar(&dnsRouteInterval, dnsRouteIntervalFlag, time.Minute, "DNS route update interval")
+	upCmd.PersistentFlags().BoolVar(&blockLANAccess, blockLANAccessFlag, false, "Block access to local networks (LAN) when using this peer as a router or exit node")
 }
 
 func upFunc(cmd *cobra.Command, args []string) error {
@@ -116,6 +124,10 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command) error {
 		ic.WireguardPort = &p
 	}
 
+	if cmd.Flag(networkMonitorFlag).Changed {
+		ic.NetworkMonitor = &networkMonitor
+	}
+
 	if rootCmd.PersistentFlags().Changed(preSharedKeyFlag) {
 		ic.PreSharedKey = &preSharedKey
 	}
@@ -132,6 +144,32 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command) error {
 		}
 	}
 
+	if cmd.Flag(dnsRouteIntervalFlag).Changed {
+		ic.DNSRouteInterval = &dnsRouteInterval
+	}
+
+	if cmd.Flag(disableClientRoutesFlag).Changed {
+		ic.DisableClientRoutes = &disableClientRoutes
+	}
+	if cmd.Flag(disableServerRoutesFlag).Changed {
+		ic.DisableServerRoutes = &disableServerRoutes
+	}
+	if cmd.Flag(disableDNSFlag).Changed {
+		ic.DisableDNS = &disableDNS
+	}
+	if cmd.Flag(disableFirewallFlag).Changed {
+		ic.DisableFirewall = &disableFirewall
+	}
+
+	if cmd.Flag(blockLANAccessFlag).Changed {
+		ic.BlockLANAccess = &blockLANAccess
+	}
+
+	providedSetupKey, err := getSetupKey()
+	if err != nil {
+		return err
+	}
+
 	config, err := internal.UpdateOrCreateConfig(ic)
 	if err != nil {
 		return fmt.Errorf("get config file: %v", err)
@@ -139,7 +177,7 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command) error {
 
 	config, _ = internal.UpdateOldManagementURL(ctx, config, configPath)
 
-	err = foregroundLogin(ctx, cmd, config, setupKey)
+	err = foregroundLogin(ctx, cmd, config, providedSetupKey)
 	if err != nil {
 		return fmt.Errorf("foreground login failed: %v", err)
 	}
@@ -147,7 +185,12 @@ func runInForegroundMode(ctx context.Context, cmd *cobra.Command) error {
 	var cancel context.CancelFunc
 	ctx, cancel = context.WithCancel(ctx)
 	SetupCloseHandler(ctx, cancel)
-	return internal.RunClient(ctx, config, peer.NewRecorder(config.ManagementURL.String()))
+
+	r := peer.NewRecorder(config.ManagementURL.String())
+	r.GetFullStatus()
+
+	connectClient := internal.NewConnectClient(ctx, config, r)
+	return connectClient.Run(nil)
 }
 
 func runInDaemonMode(ctx context.Context, cmd *cobra.Command) error {
@@ -182,8 +225,13 @@ func runInDaemonMode(ctx context.Context, cmd *cobra.Command) error {
 		return nil
 	}
 
+	providedSetupKey, err := getSetupKey()
+	if err != nil {
+		return err
+	}
+
 	loginRequest := proto.LoginRequest{
-		SetupKey:             setupKey,
+		SetupKey:             providedSetupKey,
 		ManagementUrl:        managementURL,
 		AdminURL:             adminURL,
 		NatExternalIPs:       natExternalIPs,
@@ -226,6 +274,31 @@ func runInDaemonMode(ctx context.Context, cmd *cobra.Command) error {
 		loginRequest.WireguardPort = &wp
 	}
 
+	if cmd.Flag(networkMonitorFlag).Changed {
+		loginRequest.NetworkMonitor = &networkMonitor
+	}
+
+	if cmd.Flag(dnsRouteIntervalFlag).Changed {
+		loginRequest.DnsRouteInterval = durationpb.New(dnsRouteInterval)
+	}
+
+	if cmd.Flag(disableClientRoutesFlag).Changed {
+		loginRequest.DisableClientRoutes = &disableClientRoutes
+	}
+	if cmd.Flag(disableServerRoutesFlag).Changed {
+		loginRequest.DisableServerRoutes = &disableServerRoutes
+	}
+	if cmd.Flag(disableDNSFlag).Changed {
+		loginRequest.DisableDns = &disableDNS
+	}
+	if cmd.Flag(disableFirewallFlag).Changed {
+		loginRequest.DisableFirewall = &disableFirewall
+	}
+
+	if cmd.Flag(blockLANAccessFlag).Changed {
+		loginRequest.BlockLanAccess = &blockLANAccess
+	}
+
 	var loginErr error
 
 	var loginResp *proto.LoginResponse

client/cmd/up_daemon_test.go

@@ -2,6 +2,7 @@ package cmd
 
 import (
 	"context"
+	"os"
 	"testing"
 	"time"
 
@@ -40,6 +41,36 @@ func TestUpDaemon(t *testing.T) {
 		return
 	}
 
+	// Test the setup-key-file flag.
+	tempFile, err := os.CreateTemp("", "setup-key")
+	if err != nil {
+		t.Errorf("could not create temp file, got error %v", err)
+		return
+	}
+	defer os.Remove(tempFile.Name())
+	if _, err := tempFile.Write([]byte("A2C8E62B-38F5-4553-B31E-DD66C696CEBB")); err != nil {
+		t.Errorf("could not write to temp file, got error %v", err)
+		return
+	}
+	if err := tempFile.Close(); err != nil {
+		t.Errorf("unable to close file, got error %v", err)
+	}
+	rootCmd.SetArgs([]string{
+		"login",
+		"--daemon-addr", "tcp://" + cliAddr,
+		"--setup-key-file", tempFile.Name(),
+		"--log-file", "",
+	})
+	if err := rootCmd.Execute(); err != nil {
+		t.Errorf("expected no error while running up command, got %v", err)
+		return
+	}
+	time.Sleep(time.Second * 3)
+	if status, err := state.Status(); err != nil && status != internal.StatusIdle {
+		t.Errorf("wrong status after login: %s, %v", internal.StatusIdle, err)
+		return
+	}
+
 	rootCmd.SetArgs([]string{
 		"up",
 		"--daemon-addr", "tcp://" + cliAddr,

client/configs/configs.go

@@ -0,0 +1,24 @@
+package configs
+
+import (
+	"os"
+	"path/filepath"
+	"runtime"
+)
+
+var StateDir string
+
+func init() {
+	StateDir = os.Getenv("NB_STATE_DIR")
+	if StateDir != "" {
+		return
+	}
+	switch runtime.GOOS {
+	case "windows":
+		StateDir = filepath.Join(os.Getenv("PROGRAMDATA"), "Netbird")
+	case "darwin", "linux":
+		StateDir = "/var/lib/netbird"
+	case "freebsd", "openbsd", "netbsd", "dragonfly":
+		StateDir = "/var/db/netbird"
+	}
+}

client/errors/errors.go

@@ -0,0 +1,30 @@
+package errors
+
+import (
+	"fmt"
+	"strings"
+
+	"github.com/hashicorp/go-multierror"
+)
+
+func formatError(es []error) string {
+	if len(es) == 1 {
+		return fmt.Sprintf("1 error occurred:\n\t* %s", es[0])
+	}
+
+	points := make([]string, len(es))
+	for i, err := range es {
+		points[i] = fmt.Sprintf("* %s", err)
+	}
+
+	return fmt.Sprintf(
+		"%d errors occurred:\n\t%s",
+		len(es), strings.Join(points, "\n\t"))
+}
+
+func FormatErrorOrNil(err *multierror.Error) error {
+	if err != nil {
+		err.ErrorFormat = formatError
+	}
+	return err.ErrorOrNil()
+}

client/firewall/create.go

@@ -3,7 +3,6 @@
 package firewall
 
 import (
-	"context"
 	"fmt"
 	"runtime"
 
@@ -11,16 +10,17 @@ import (
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
 	"github.com/netbirdio/netbird/client/firewall/uspfilter"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
 )
 
 // NewFirewall creates a firewall manager instance
-func NewFirewall(context context.Context, iface IFaceMapper) (firewall.Manager, error) {
+func NewFirewall(iface IFaceMapper, _ *statemanager.Manager, disableServerRoutes bool) (firewall.Manager, error) {
 	if !iface.IsUserspaceBind() {
 		return nil, fmt.Errorf("not implemented for this OS: %s", runtime.GOOS)
 	}
 
 	// use userspace packet filtering firewall
-	fm, err := uspfilter.Create(iface)
+	fm, err := uspfilter.Create(iface, disableServerRoutes)
 	if err != nil {
 		return nil, err
 	}

client/firewall/create_linux.go

@@ -3,7 +3,7 @@
 package firewall
 
 import (
-	"context"
+	"errors"
 	"fmt"
 	"os"
 
@@ -15,6 +15,7 @@ import (
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
 	nbnftables "github.com/netbirdio/netbird/client/firewall/nftables"
 	"github.com/netbirdio/netbird/client/firewall/uspfilter"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
 )
 
 const (
@@ -32,69 +33,122 @@ const SKIP_NFTABLES_ENV = "NB_SKIP_NFTABLES_CHECK"
 // FWType is the type for the firewall type
 type FWType int
 
-func NewFirewall(context context.Context, iface IFaceMapper) (firewall.Manager, error) {
+func NewFirewall(iface IFaceMapper, stateManager *statemanager.Manager, disableServerRoutes bool) (firewall.Manager, error) {
 	// on the linux system we try to user nftables or iptables
 	// in any case, because we need to allow netbird interface traffic
 	// so we use AllowNetbird traffic from these firewall managers
 	// for the userspace packet filtering firewall
-	var fm firewall.Manager
-	var errFw error
+	fm, err := createNativeFirewall(iface, stateManager, disableServerRoutes)
 
-	switch check() {
-	case IPTABLES:
-		log.Debug("creating an iptables firewall manager")
-		fm, errFw = nbiptables.Create(context, iface)
-		if errFw != nil {
-			log.Errorf("failed to create iptables manager: %s", errFw)
-		}
-	case NFTABLES:
-		log.Debug("creating an nftables firewall manager")
-		fm, errFw = nbnftables.Create(context, iface)
-		if errFw != nil {
-			log.Errorf("failed to create nftables manager: %s", errFw)
-		}
-	default:
-		errFw = fmt.Errorf("no firewall manager found")
-		log.Debug("no firewall manager found, try to use userspace packet filtering firewall")
+	if !iface.IsUserspaceBind() {
+		return fm, err
 	}
 
-	if iface.IsUserspaceBind() {
-		var errUsp error
-		if errFw == nil {
-			fm, errUsp = uspfilter.CreateWithNativeFirewall(iface, fm)
-		} else {
-			fm, errUsp = uspfilter.Create(iface)
-		}
-		if errUsp != nil {
-			log.Debugf("failed to create userspace filtering firewall: %s", errUsp)
-			return nil, errUsp
-		}
+	if err != nil {
+		log.Warnf("failed to create native firewall: %v. Proceeding with userspace", err)
+	}
+	return createUserspaceFirewall(iface, fm, disableServerRoutes)
+}
 
-		if err := fm.AllowNetbird(); err != nil {
-			log.Errorf("failed to allow netbird interface traffic: %v", err)
-		}
-		return fm, nil
+func createNativeFirewall(iface IFaceMapper, stateManager *statemanager.Manager, routes bool) (firewall.Manager, error) {
+	fm, err := createFW(iface)
+	if err != nil {
+		return nil, fmt.Errorf("create firewall: %s", err)
 	}
 
-	if errFw != nil {
-		return nil, errFw
+	if err = fm.Init(stateManager); err != nil {
+		return nil, fmt.Errorf("init firewall: %s", err)
 	}
 
 	return fm, nil
 }
 
-// check returns the firewall type based on common lib checks. It returns UNKNOWN if no firewall is found.
-func check() FWType {
-	nf := nftables.Conn{}
-	if _, err := nf.ListChains(); err == nil && os.Getenv(SKIP_NFTABLES_ENV) != "true" {
-		return NFTABLES
+func createFW(iface IFaceMapper) (firewall.Manager, error) {
+	switch check() {
+	case IPTABLES:
+		log.Info("creating an iptables firewall manager")
+		return nbiptables.Create(iface)
+	case NFTABLES:
+		log.Info("creating an nftables firewall manager")
+		return nbnftables.Create(iface)
+	default:
+		log.Info("no firewall manager found, trying to use userspace packet filtering firewall")
+		return nil, errors.New("no firewall manager found")
+	}
+}
+
+func createUserspaceFirewall(iface IFaceMapper, fm firewall.Manager, disableServerRoutes bool) (firewall.Manager, error) {
+	var errUsp error
+	if fm != nil {
+		fm, errUsp = uspfilter.CreateWithNativeFirewall(iface, fm, disableServerRoutes)
+	} else {
+		fm, errUsp = uspfilter.Create(iface, disableServerRoutes)
 	}
 
-	ip, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
-	if err != nil {
-		return UNKNOWN
+	if errUsp != nil {
+		return nil, fmt.Errorf("create userspace firewall: %s", errUsp)
 	}
-	if isIptablesClientAvailable(ip) {
+
+	if err := fm.AllowNetbird(); err != nil {
+		log.Errorf("failed to allow netbird interface traffic: %v", err)
+	}
+	return fm, nil
+}
+
+// check returns the firewall type based on common lib checks. It returns UNKNOWN if no firewall is found.
+func check() FWType {
+	useIPTABLES := false
+	var iptablesChains []string
+	ip, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
+	if err == nil && isIptablesClientAvailable(ip) {
+		major, minor, _ := ip.GetIptablesVersion()
+		// use iptables when its version is lower than 1.8.0 which doesn't work well with our nftables manager
+		if major < 1 || (major == 1 && minor < 8) {
+			return IPTABLES
+		}
+
+		useIPTABLES = true
+
+		iptablesChains, err = ip.ListChains("filter")
+		if err != nil {
+			log.Errorf("failed to list iptables chains: %s", err)
+			useIPTABLES = false
+		}
+	}
+
+	nf := nftables.Conn{}
+	if chains, err := nf.ListChains(); err == nil && os.Getenv(SKIP_NFTABLES_ENV) != "true" {
+		if !useIPTABLES {
+			return NFTABLES
+		}
+
+		// search for chains where table is filter
+		// if we find one, we assume that nftables manager can be used with iptables
+		for _, chain := range chains {
+			if chain.Table.Name == "filter" {
+				return NFTABLES
+			}
+		}
+
+		// check tables for the following constraints:
+		// 1. there is no chain in nftables for the filter table and there is at least one chain in iptables, we assume that nftables manager can not be used
+		// 2. there is no tables or more than one table, we assume that nftables manager can be used
+		// 3. there is only one table and its name is filter, we assume that nftables manager can not be used, since there was no chain in it
+		// 4. if we find an error we log and continue with iptables check
+		nbTablesList, err := nf.ListTables()
+		switch {
+		case err == nil && len(iptablesChains) > 0:
+			return IPTABLES
+		case err == nil && len(nbTablesList) != 1:
+			return NFTABLES
+		case err == nil && len(nbTablesList) == 1 && nbTablesList[0].Name == "filter":
+			return IPTABLES
+		case err != nil:
+			log.Errorf("failed to list nftables tables on fw manager discovery: %s", err)
+		}
+	}
+
+	if useIPTABLES {
 		return IPTABLES
 	}
 

client/firewall/iface.go

@@ -1,11 +1,17 @@
 package firewall
 
-import "github.com/netbirdio/netbird/iface"
+import (
+	wgdevice "golang.zx2c4.com/wireguard/device"
+
+	"github.com/netbirdio/netbird/client/iface/device"
+)
 
 // IFaceMapper defines subset methods of interface required for manager
 type IFaceMapper interface {
 	Name() string
-	Address() iface.WGAddress
+	Address() device.WGAddress
 	IsUserspaceBind() bool
-	SetFilter(iface.PacketFilter) error
+	SetFilter(device.PacketFilter) error
+	GetDevice() *device.FilteredDevice
+	GetWGDevice() *wgdevice.Device
 }

client/firewall/iptables/acl_linux.go

@@ -3,7 +3,7 @@ package iptables
 import (
 	"fmt"
 	"net"
-	"strconv"
+	"slices"
 
 	"github.com/coreos/go-iptables/iptables"
 	"github.com/google/uuid"
@@ -11,83 +11,94 @@ import (
 	log "github.com/sirupsen/logrus"
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
+	nbnet "github.com/netbirdio/netbird/util/net"
 )
 
 const (
 	tableName = "filter"
 
 	// rules chains contains the effective ACL rules
-	chainNameInputRules  = "NETBIRD-ACL-INPUT"
-	chainNameOutputRules = "NETBIRD-ACL-OUTPUT"
-
-	postRoutingMark = "0x000007e4"
+	chainNameInputRules = "NETBIRD-ACL-INPUT"
 )
 
-type aclManager struct {
-	iptablesClient      *iptables.IPTables
-	wgIface             iFaceMapper
-	routeingFwChainName string
+type aclEntries map[string][][]string
 
-	entries    map[string][][]string
-	ipsetStore *ipsetStore
+type entry struct {
+	spec     []string
+	position int
 }
 
-func newAclManager(iptablesClient *iptables.IPTables, wgIface iFaceMapper, routeingFwChainName string) (*aclManager, error) {
+type aclManager struct {
+	iptablesClient     *iptables.IPTables
+	wgIface            iFaceMapper
+	routingFwChainName string
+
+	entries         aclEntries
+	optionalEntries map[string][]entry
+	ipsetStore      *ipsetStore
+
+	stateManager *statemanager.Manager
+}
+
+func newAclManager(iptablesClient *iptables.IPTables, wgIface iFaceMapper, routingFwChainName string) (*aclManager, error) {
 	m := &aclManager{
-		iptablesClient:      iptablesClient,
-		wgIface:             wgIface,
-		routeingFwChainName: routeingFwChainName,
+		iptablesClient:     iptablesClient,
+		wgIface:            wgIface,
+		routingFwChainName: routingFwChainName,
 
-		entries:    make(map[string][][]string),
-		ipsetStore: newIpsetStore(),
+		entries:         make(map[string][][]string),
+		optionalEntries: make(map[string][]entry),
+		ipsetStore:      newIpsetStore(),
 	}
 
-	err := ipset.Init()
-	if err != nil {
-		return nil, fmt.Errorf("failed to init ipset: %w", err)
+	if err := ipset.Init(); err != nil {
+		return nil, fmt.Errorf("init ipset: %w", err)
 	}
 
-	m.seedInitialEntries()
-
-	err = m.cleanChains()
-	if err != nil {
-		return nil, err
-	}
-
-	err = m.createDefaultChains()
-	if err != nil {
-		return nil, err
-	}
 	return m, nil
 }
 
-func (m *aclManager) AddFiltering(
+func (m *aclManager) init(stateManager *statemanager.Manager) error {
+	m.stateManager = stateManager
+
+	m.seedInitialEntries()
+	m.seedInitialOptionalEntries()
+
+	if err := m.cleanChains(); err != nil {
+		return fmt.Errorf("clean chains: %w", err)
+	}
+
+	if err := m.createDefaultChains(); err != nil {
+		return fmt.Errorf("create default chains: %w", err)
+	}
+
+	m.updateState()
+
+	return nil
+}
+
+func (m *aclManager) AddPeerFiltering(
 	ip net.IP,
 	protocol firewall.Protocol,
 	sPort *firewall.Port,
 	dPort *firewall.Port,
-	direction firewall.RuleDirection,
 	action firewall.Action,
 	ipsetName string,
 ) ([]firewall.Rule, error) {
-	var dPortVal, sPortVal string
-	if dPort != nil && dPort.Values != nil {
-		// TODO: we support only one port per rule in current implementation of ACLs
-		dPortVal = strconv.Itoa(dPort.Values[0])
-	}
-	if sPort != nil && sPort.Values != nil {
-		sPortVal = strconv.Itoa(sPort.Values[0])
-	}
+	chain := chainNameInputRules
 
-	var chain string
-	if direction == firewall.RuleDirectionOUT {
-		chain = chainNameOutputRules
-	} else {
-		chain = chainNameInputRules
-	}
+	ipsetName = transformIPsetName(ipsetName, sPort, dPort)
+	specs := filterRuleSpecs(ip, string(protocol), sPort, dPort, action, ipsetName)
 
-	ipsetName = transformIPsetName(ipsetName, sPortVal, dPortVal)
-	specs := filterRuleSpecs(ip, string(protocol), sPortVal, dPortVal, direction, action, ipsetName)
+	mangleSpecs := slices.Clone(specs)
+	mangleSpecs = append(mangleSpecs,
+		"-i", m.wgIface.Name(),
+		"-m", "addrtype", "--dst-type", "LOCAL",
+		"-j", "MARK", "--set-xmark", fmt.Sprintf("%#x", nbnet.PreroutingFwmarkRedirected),
+	)
+
+	specs = append(specs, "-j", actionToStr(action))
 	if ipsetName != "" {
 		if ipList, ipsetExists := m.ipsetStore.ipset(ipsetName); ipsetExists {
 			if err := ipset.Add(ipsetName, ip.String()); err != nil {
@@ -119,7 +130,7 @@ func (m *aclManager) AddFiltering(
 		m.ipsetStore.addIpList(ipsetName, ipList)
 	}
 
-	ok, err := m.iptablesClient.Exists("filter", chain, specs...)
+	ok, err := m.iptablesClient.Exists(tableFilter, chain, specs...)
 	if err != nil {
 		return nil, fmt.Errorf("failed to check rule: %w", err)
 	}
@@ -127,40 +138,36 @@ func (m *aclManager) AddFiltering(
 		return nil, fmt.Errorf("rule already exists")
 	}
 
-	if err := m.iptablesClient.Insert("filter", chain, 1, specs...); err != nil {
+	if err := m.iptablesClient.Append(tableFilter, chain, specs...); err != nil {
 		return nil, err
 	}
 
+	if err := m.iptablesClient.Append(tableMangle, chainRTPRE, mangleSpecs...); err != nil {
+		log.Errorf("failed to add mangle rule: %v", err)
+		mangleSpecs = nil
+	}
+
 	rule := &Rule{
-		ruleID:    uuid.New().String(),
-		specs:     specs,
-		ipsetName: ipsetName,
-		ip:        ip.String(),
-		chain:     chain,
+		ruleID:      uuid.New().String(),
+		specs:       specs,
+		mangleSpecs: mangleSpecs,
+		ipsetName:   ipsetName,
+		ip:          ip.String(),
+		chain:       chain,
 	}
 
-	if !shouldAddToPrerouting(protocol, dPort, direction) {
-		return []firewall.Rule{rule}, nil
-	}
+	m.updateState()
 
-	rulePrerouting, err := m.addPreroutingFilter(ipsetName, string(protocol), dPortVal, ip)
-	if err != nil {
-		return []firewall.Rule{rule}, err
-	}
-	return []firewall.Rule{rule, rulePrerouting}, nil
+	return []firewall.Rule{rule}, nil
 }
 
-// DeleteRule from the firewall by rule definition
-func (m *aclManager) DeleteRule(rule firewall.Rule) error {
+// DeletePeerRule from the firewall by rule definition
+func (m *aclManager) DeletePeerRule(rule firewall.Rule) error {
 	r, ok := rule.(*Rule)
 	if !ok {
 		return fmt.Errorf("invalid rule type")
 	}
 
-	if r.chain == "PREROUTING" {
-		goto DELETERULE
-	}
-
 	if ipsetList, ok := m.ipsetStore.ipset(r.ipsetName); ok {
 		// delete IP from ruleset IPs list and ipset
 		if _, ok := ipsetList.ips[r.ip]; ok {
@@ -185,86 +192,34 @@ func (m *aclManager) DeleteRule(rule firewall.Rule) error {
 		}
 	}
 
-DELETERULE:
-	var table string
-	if r.chain == "PREROUTING" {
-		table = "mangle"
-	} else {
-		table = "filter"
+	if err := m.iptablesClient.Delete(tableName, r.chain, r.specs...); err != nil {
+		return fmt.Errorf("failed to delete rule: %s, %v: %w", r.chain, r.specs, err)
 	}
-	err := m.iptablesClient.Delete(table, r.chain, r.specs...)
-	if err != nil {
-		log.Debugf("failed to delete rule, %s, %v: %s", r.chain, r.specs, err)
+
+	if r.mangleSpecs != nil {
+		if err := m.iptablesClient.Delete(tableMangle, chainRTPRE, r.mangleSpecs...); err != nil {
+			log.Errorf("failed to delete mangle rule: %v", err)
+		}
 	}
-	return err
+
+	m.updateState()
+
+	return nil
 }
 
 func (m *aclManager) Reset() error {
-	return m.cleanChains()
-}
-
-func (m *aclManager) addPreroutingFilter(ipsetName string, protocol string, port string, ip net.IP) (*Rule, error) {
-	var src []string
-	if ipsetName != "" {
-		src = []string{"-m", "set", "--set", ipsetName, "src"}
-	} else {
-		src = []string{"-s", ip.String()}
-	}
-	specs := []string{
-		"-d", m.wgIface.Address().IP.String(),
-		"-p", protocol,
-		"--dport", port,
-		"-j", "MARK", "--set-mark", postRoutingMark,
+	if err := m.cleanChains(); err != nil {
+		return fmt.Errorf("clean chains: %w", err)
 	}
 
-	specs = append(src, specs...)
+	m.updateState()
 
-	ok, err := m.iptablesClient.Exists("mangle", "PREROUTING", specs...)
-	if err != nil {
-		return nil, fmt.Errorf("failed to check rule: %w", err)
-	}
-	if ok {
-		return nil, fmt.Errorf("rule already exists")
-	}
-
-	if err := m.iptablesClient.Insert("mangle", "PREROUTING", 1, specs...); err != nil {
-		return nil, err
-	}
-
-	rule := &Rule{
-		ruleID:    uuid.New().String(),
-		specs:     specs,
-		ipsetName: ipsetName,
-		ip:        ip.String(),
-		chain:     "PREROUTING",
-	}
-	return rule, nil
+	return nil
 }
 
 // todo write less destructive cleanup mechanism
 func (m *aclManager) cleanChains() error {
-	ok, err := m.iptablesClient.ChainExists(tableName, chainNameOutputRules)
-	if err != nil {
-		log.Debugf("failed to list chains: %s", err)
-		return err
-	}
-	if ok {
-		rules := m.entries["OUTPUT"]
-		for _, rule := range rules {
-			err := m.iptablesClient.DeleteIfExists(tableName, "OUTPUT", rule...)
-			if err != nil {
-				log.Errorf("failed to delete rule: %v, %s", rule, err)
-			}
-		}
-
-		err = m.iptablesClient.ClearAndDeleteChain(tableName, chainNameOutputRules)
-		if err != nil {
-			log.Debugf("failed to clear and delete %s chain: %s", chainNameOutputRules, err)
-			return err
-		}
-	}
-
-	ok, err = m.iptablesClient.ChainExists(tableName, chainNameInputRules)
+	ok, err := m.iptablesClient.ChainExists(tableName, chainNameInputRules)
 	if err != nil {
 		log.Debugf("failed to list chains: %s", err)
 		return err
@@ -293,8 +248,7 @@ func (m *aclManager) cleanChains() error {
 
 	ok, err = m.iptablesClient.ChainExists("mangle", "PREROUTING")
 	if err != nil {
-		log.Debugf("failed to list chains: %s", err)
-		return err
+		return fmt.Errorf("list chains: %w", err)
 	}
 	if ok {
 		for _, rule := range m.entries["PREROUTING"] {
@@ -303,11 +257,6 @@ func (m *aclManager) cleanChains() error {
 				log.Errorf("failed to delete rule: %v, %s", rule, err)
 			}
 		}
-		err = m.iptablesClient.ClearChain("mangle", "PREROUTING")
-		if err != nil {
-			log.Debugf("failed to clear %s chain: %s", "PREROUTING", err)
-			return err
-		}
 	}
 
 	for _, ipsetName := range m.ipsetStore.ipsetNames() {
@@ -330,109 +279,106 @@ func (m *aclManager) createDefaultChains() error {
 		return err
 	}
 
-	// chain netbird-acl-output-rules
-	if err := m.iptablesClient.NewChain(tableName, chainNameOutputRules); err != nil {
-		log.Debugf("failed to create '%s' chain: %s", chainNameOutputRules, err)
-		return err
-	}
-
 	for chainName, rules := range m.entries {
 		for _, rule := range rules {
-			if chainName == "FORWARD" {
-				// position 2 because we add it after router's, jump rule
-				if err := m.iptablesClient.InsertUnique(tableName, "FORWARD", 2, rule...); err != nil {
-					log.Debugf("failed to create input chain jump rule: %s", err)
-					return err
-				}
-			} else {
-				if err := m.iptablesClient.AppendUnique(tableName, chainName, rule...); err != nil {
-					log.Debugf("failed to create input chain jump rule: %s", err)
-					return err
-				}
+			if err := m.iptablesClient.InsertUnique(tableName, chainName, 1, rule...); err != nil {
+				log.Debugf("failed to create input chain jump rule: %s", err)
+				return err
 			}
 		}
 	}
 
+	for chainName, entries := range m.optionalEntries {
+		for _, entry := range entries {
+			if err := m.iptablesClient.InsertUnique(tableName, chainName, entry.position, entry.spec...); err != nil {
+				log.Errorf("failed to insert optional entry %v: %v", entry.spec, err)
+				continue
+			}
+			m.entries[chainName] = append(m.entries[chainName], entry.spec)
+		}
+	}
+	clear(m.optionalEntries)
+
 	return nil
 }
 
+// seedInitialEntries adds default rules to the entries map, rules are inserted on pos 1, hence the order is reversed.
+// We want to make sure our traffic is not dropped by existing rules.
+
+// The existing FORWARD rules/policies decide outbound traffic towards our interface.
+// In case the FORWARD policy is set to "drop", we add an established/related rule to allow return traffic for the inbound rule.
 func (m *aclManager) seedInitialEntries() {
-	m.appendToEntries("INPUT",
-		[]string{"-i", m.wgIface.Name(), "!", "-s", m.wgIface.Address().String(), "-d", m.wgIface.Address().String(), "-j", "ACCEPT"})
-
-	m.appendToEntries("INPUT",
-		[]string{"-i", m.wgIface.Name(), "-s", m.wgIface.Address().String(), "!", "-d", m.wgIface.Address().String(), "-j", "ACCEPT"})
-
-	m.appendToEntries("INPUT",
-		[]string{"-i", m.wgIface.Name(), "-s", m.wgIface.Address().String(), "-d", m.wgIface.Address().String(), "-j", chainNameInputRules})
+	established := getConntrackEstablished()
 
 	m.appendToEntries("INPUT", []string{"-i", m.wgIface.Name(), "-j", "DROP"})
-
-	m.appendToEntries("OUTPUT",
-		[]string{"-o", m.wgIface.Name(), "!", "-s", m.wgIface.Address().String(), "-d", m.wgIface.Address().String(), "-j", "ACCEPT"})
-
-	m.appendToEntries("OUTPUT",
-		[]string{"-o", m.wgIface.Name(), "-s", m.wgIface.Address().String(), "!", "-d", m.wgIface.Address().String(), "-j", "ACCEPT"})
-
-	m.appendToEntries("OUTPUT",
-		[]string{"-o", m.wgIface.Name(), "-s", m.wgIface.Address().String(), "-d", m.wgIface.Address().String(), "-j", chainNameOutputRules})
-
-	m.appendToEntries("OUTPUT", []string{"-o", m.wgIface.Name(), "-j", "DROP"})
+	m.appendToEntries("INPUT", []string{"-i", m.wgIface.Name(), "-j", chainNameInputRules})
+	m.appendToEntries("INPUT", append([]string{"-i", m.wgIface.Name()}, established...))
 
 	m.appendToEntries("FORWARD", []string{"-i", m.wgIface.Name(), "-j", "DROP"})
-	m.appendToEntries("FORWARD", []string{"-i", m.wgIface.Name(), "-j", chainNameInputRules})
-	m.appendToEntries("FORWARD",
-		[]string{"-o", m.wgIface.Name(), "-m", "mark", "--mark", postRoutingMark, "-j", "ACCEPT"})
-	m.appendToEntries("FORWARD",
-		[]string{"-i", m.wgIface.Name(), "-m", "mark", "--mark", postRoutingMark, "-j", "ACCEPT"})
-	m.appendToEntries("FORWARD", []string{"-o", m.wgIface.Name(), "-j", m.routeingFwChainName})
-	m.appendToEntries("FORWARD", []string{"-i", m.wgIface.Name(), "-j", m.routeingFwChainName})
+	m.appendToEntries("FORWARD", []string{"-i", m.wgIface.Name(), "-j", m.routingFwChainName})
+	m.appendToEntries("FORWARD", append([]string{"-o", m.wgIface.Name()}, established...))
+}
 
-	m.appendToEntries("PREROUTING",
-		[]string{"-t", "mangle", "-i", m.wgIface.Name(), "!", "-s", m.wgIface.Address().String(), "-d", m.wgIface.Address().IP.String(), "-m", "mark", "--mark", postRoutingMark})
+func (m *aclManager) seedInitialOptionalEntries() {
+	m.optionalEntries["FORWARD"] = []entry{
+		{
+			spec:     []string{"-m", "mark", "--mark", fmt.Sprintf("%#x", nbnet.PreroutingFwmarkRedirected), "-j", "ACCEPT"},
+			position: 2,
+		},
+	}
 }
 
 func (m *aclManager) appendToEntries(chainName string, spec []string) {
 	m.entries[chainName] = append(m.entries[chainName], spec)
 }
 
+func (m *aclManager) updateState() {
+	if m.stateManager == nil {
+		return
+	}
+
+	var currentState *ShutdownState
+	if existing := m.stateManager.GetState(currentState); existing != nil {
+		if existingState, ok := existing.(*ShutdownState); ok {
+			currentState = existingState
+		}
+	}
+	if currentState == nil {
+		currentState = &ShutdownState{}
+	}
+
+	currentState.Lock()
+	defer currentState.Unlock()
+
+	currentState.ACLEntries = m.entries
+	currentState.ACLIPsetStore = m.ipsetStore
+
+	if err := m.stateManager.UpdateState(currentState); err != nil {
+		log.Errorf("failed to update state: %v", err)
+	}
+}
+
 // filterRuleSpecs returns the specs of a filtering rule
-func filterRuleSpecs(
-	ip net.IP, protocol string, sPort, dPort string, direction firewall.RuleDirection, action firewall.Action, ipsetName string,
-) (specs []string) {
+func filterRuleSpecs(ip net.IP, protocol string, sPort, dPort *firewall.Port, action firewall.Action, ipsetName string) (specs []string) {
 	matchByIP := true
 	// don't use IP matching if IP is ip 0.0.0.0
 	if ip.String() == "0.0.0.0" {
 		matchByIP = false
 	}
-	switch direction {
-	case firewall.RuleDirectionIN:
-		if matchByIP {
-			if ipsetName != "" {
-				specs = append(specs, "-m", "set", "--set", ipsetName, "src")
-			} else {
-				specs = append(specs, "-s", ip.String())
-			}
-		}
-	case firewall.RuleDirectionOUT:
-		if matchByIP {
-			if ipsetName != "" {
-				specs = append(specs, "-m", "set", "--set", ipsetName, "dst")
-			} else {
-				specs = append(specs, "-d", ip.String())
-			}
+
+	if matchByIP {
+		if ipsetName != "" {
+			specs = append(specs, "-m", "set", "--set", ipsetName, "src")
+		} else {
+			specs = append(specs, "-s", ip.String())
 		}
 	}
 	if protocol != "all" {
 		specs = append(specs, "-p", protocol)
 	}
-	if sPort != "" {
-		specs = append(specs, "--sport", sPort)
-	}
-	if dPort != "" {
-		specs = append(specs, "--dport", dPort)
-	}
-	return append(specs, "-j", actionToStr(action))
+	specs = append(specs, applyPort("--sport", sPort)...)
+	specs = append(specs, applyPort("--dport", dPort)...)
+	return specs
 }
 
 func actionToStr(action firewall.Action) string {
@@ -442,32 +388,17 @@ func actionToStr(action firewall.Action) string {
 	return "DROP"
 }
 
-func transformIPsetName(ipsetName string, sPort, dPort string) string {
+func transformIPsetName(ipsetName string, sPort, dPort *firewall.Port) string {
 	switch {
 	case ipsetName == "":
 		return ""
-	case sPort != "" && dPort != "":
+	case sPort != nil && dPort != nil:
 		return ipsetName + "-sport-dport"
-	case sPort != "":
+	case sPort != nil:
 		return ipsetName + "-sport"
-	case dPort != "":
+	case dPort != nil:
 		return ipsetName + "-dport"
 	default:
 		return ipsetName
 	}
 }
-
-func shouldAddToPrerouting(proto firewall.Protocol, dPort *firewall.Port, direction firewall.RuleDirection) bool {
-	if proto == "all" {
-		return false
-	}
-
-	if direction != firewall.RuleDirectionIN {
-		return false
-	}
-
-	if dPort == nil {
-		return false
-	}
-	return true
-}

client/firewall/iptables/manager_linux.go

@@ -4,13 +4,17 @@ import (
 	"context"
 	"fmt"
 	"net"
+	"net/netip"
 	"sync"
 
 	"github.com/coreos/go-iptables/iptables"
+	"github.com/hashicorp/go-multierror"
 	log "github.com/sirupsen/logrus"
 
+	nberrors "github.com/netbirdio/netbird/client/errors"
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
-	"github.com/netbirdio/netbird/iface"
+	"github.com/netbirdio/netbird/client/iface"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
 )
 
 // Manager of iptables firewall
@@ -21,7 +25,7 @@ type Manager struct {
 
 	ipv4Client *iptables.IPTables
 	aclMgr     *aclManager
-	router     *routerManager
+	router     *router
 }
 
 // iFaceMapper defines subset methods of interface required for manager
@@ -32,10 +36,10 @@ type iFaceMapper interface {
 }
 
 // Create iptables firewall manager
-func Create(context context.Context, wgIface iFaceMapper) (*Manager, error) {
+func Create(wgIface iFaceMapper) (*Manager, error) {
 	iptablesClient, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
 	if err != nil {
-		return nil, fmt.Errorf("iptables is not installed in the system or not supported")
+		return nil, fmt.Errorf("init iptables: %w", err)
 	}
 
 	m := &Manager{
@@ -43,80 +47,146 @@ func Create(context context.Context, wgIface iFaceMapper) (*Manager, error) {
 		ipv4Client: iptablesClient,
 	}
 
-	m.router, err = newRouterManager(context, iptablesClient)
+	m.router, err = newRouter(iptablesClient, wgIface)
 	if err != nil {
-		log.Debugf("failed to initialize route related chains: %s", err)
-		return nil, err
+		return nil, fmt.Errorf("create router: %w", err)
 	}
-	m.aclMgr, err = newAclManager(iptablesClient, wgIface, m.router.RouteingFwChainName())
+
+	m.aclMgr, err = newAclManager(iptablesClient, wgIface, chainRTFWD)
 	if err != nil {
-		log.Debugf("failed to initialize ACL manager: %s", err)
-		return nil, err
+		return nil, fmt.Errorf("create acl manager: %w", err)
 	}
 
 	return m, nil
 }
 
-// AddFiltering rule to the firewall
+func (m *Manager) Init(stateManager *statemanager.Manager) error {
+	state := &ShutdownState{
+		InterfaceState: &InterfaceState{
+			NameStr:       m.wgIface.Name(),
+			WGAddress:     m.wgIface.Address(),
+			UserspaceBind: m.wgIface.IsUserspaceBind(),
+		},
+	}
+	stateManager.RegisterState(state)
+	if err := stateManager.UpdateState(state); err != nil {
+		log.Errorf("failed to update state: %v", err)
+	}
+
+	if err := m.router.init(stateManager); err != nil {
+		return fmt.Errorf("router init: %w", err)
+	}
+
+	if err := m.aclMgr.init(stateManager); err != nil {
+		// TODO: cleanup router
+		return fmt.Errorf("acl manager init: %w", err)
+	}
+
+	// persist early to ensure cleanup of chains
+	go func() {
+		if err := stateManager.PersistState(context.Background()); err != nil {
+			log.Errorf("failed to persist state: %v", err)
+		}
+	}()
+
+	return nil
+}
+
+// AddPeerFiltering adds a rule to the firewall
 //
 // Comment will be ignored because some system this feature is not supported
-func (m *Manager) AddFiltering(
+func (m *Manager) AddPeerFiltering(
 	ip net.IP,
 	protocol firewall.Protocol,
 	sPort *firewall.Port,
 	dPort *firewall.Port,
-	direction firewall.RuleDirection,
 	action firewall.Action,
 	ipsetName string,
-	comment string,
+	_ string,
 ) ([]firewall.Rule, error) {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	return m.aclMgr.AddFiltering(ip, protocol, sPort, dPort, direction, action, ipsetName)
+	return m.aclMgr.AddPeerFiltering(ip, protocol, sPort, dPort, action, ipsetName)
 }
 
-// DeleteRule from the firewall by rule definition
-func (m *Manager) DeleteRule(rule firewall.Rule) error {
+func (m *Manager) AddRouteFiltering(
+	sources []netip.Prefix,
+	destination netip.Prefix,
+	proto firewall.Protocol,
+	sPort *firewall.Port,
+	dPort *firewall.Port,
+	action firewall.Action,
+) (firewall.Rule, error) {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	return m.aclMgr.DeleteRule(rule)
+	if !destination.Addr().Is4() {
+		return nil, fmt.Errorf("unsupported IP version: %s", destination.Addr().String())
+	}
+
+	return m.router.AddRouteFiltering(sources, destination, proto, sPort, dPort, action)
+}
+
+// DeletePeerRule from the firewall by rule definition
+func (m *Manager) DeletePeerRule(rule firewall.Rule) error {
+	m.mutex.Lock()
+	defer m.mutex.Unlock()
+
+	return m.aclMgr.DeletePeerRule(rule)
+}
+
+func (m *Manager) DeleteRouteRule(rule firewall.Rule) error {
+	m.mutex.Lock()
+	defer m.mutex.Unlock()
+
+	return m.router.DeleteRouteRule(rule)
 }
 
 func (m *Manager) IsServerRouteSupported() bool {
 	return true
 }
 
-func (m *Manager) InsertRoutingRules(pair firewall.RouterPair) error {
+func (m *Manager) AddNatRule(pair firewall.RouterPair) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	return m.router.InsertRoutingRules(pair)
+	return m.router.AddNatRule(pair)
 }
 
-func (m *Manager) RemoveRoutingRules(pair firewall.RouterPair) error {
+func (m *Manager) RemoveNatRule(pair firewall.RouterPair) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	return m.router.RemoveRoutingRules(pair)
+	return m.router.RemoveNatRule(pair)
+}
+
+func (m *Manager) SetLegacyManagement(isLegacy bool) error {
+	return firewall.SetLegacyManagement(m.router, isLegacy)
 }
 
 // Reset firewall to the default state
-func (m *Manager) Reset() error {
+func (m *Manager) Reset(stateManager *statemanager.Manager) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	errAcl := m.aclMgr.Reset()
-	if errAcl != nil {
-		log.Errorf("failed to clean up ACL rules from firewall: %s", errAcl)
+	var merr *multierror.Error
+
+	if err := m.aclMgr.Reset(); err != nil {
+		merr = multierror.Append(merr, fmt.Errorf("reset acl manager: %w", err))
 	}
-	errMgr := m.router.Reset()
-	if errMgr != nil {
-		log.Errorf("failed to clean up router rules from firewall: %s", errMgr)
-		return errMgr
+	if err := m.router.Reset(); err != nil {
+		merr = multierror.Append(merr, fmt.Errorf("reset router: %w", err))
 	}
-	return errAcl
+
+	// attempt to delete state only if all other operations succeeded
+	if merr == nil {
+		if err := stateManager.DeleteState(&ShutdownState{}); err != nil {
+			merr = multierror.Append(merr, fmt.Errorf("delete state: %w", err))
+		}
+	}
+
+	return nberrors.FormatErrorOrNil(merr)
 }
 
 // AllowNetbird allows netbird interface traffic
@@ -125,31 +195,29 @@ func (m *Manager) AllowNetbird() error {
 		return nil
 	}
 
-	_, err := m.AddFiltering(
-		net.ParseIP("0.0.0.0"),
+	_, err := m.AddPeerFiltering(
+		net.IP{0, 0, 0, 0},
 		"all",
 		nil,
 		nil,
-		firewall.RuleDirectionIN,
 		firewall.ActionAccept,
 		"",
 		"",
 	)
 	if err != nil {
-		return fmt.Errorf("failed to allow netbird interface traffic: %w", err)
+		return fmt.Errorf("allow netbird interface traffic: %w", err)
 	}
-	_, err = m.AddFiltering(
-		net.ParseIP("0.0.0.0"),
-		"all",
-		nil,
-		nil,
-		firewall.RuleDirectionOUT,
-		firewall.ActionAccept,
-		"",
-		"",
-	)
-	return err
+	return nil
 }
 
 // Flush doesn't need to be implemented for this manager
 func (m *Manager) Flush() error { return nil }
+
+// SetLogLevel sets the log level for the firewall manager
+func (m *Manager) SetLogLevel(log.Level) {
+	// not supported
+}
+
+func getConntrackEstablished() []string {
+	return []string{"-m", "conntrack", "--ctstate", "RELATED,ESTABLISHED", "-j", "ACCEPT"}
+}

client/firewall/iptables/manager_linux_test.go

@@ -1,7 +1,6 @@
 package iptables
 
 import (
-	"context"
 	"fmt"
 	"net"
 	"testing"
@@ -11,9 +10,24 @@ import (
 	"github.com/stretchr/testify/require"
 
 	fw "github.com/netbirdio/netbird/client/firewall/manager"
-	"github.com/netbirdio/netbird/iface"
+	"github.com/netbirdio/netbird/client/iface"
 )
 
+var ifaceMock = &iFaceMock{
+	NameFunc: func() string {
+		return "lo"
+	},
+	AddressFunc: func() iface.WGAddress {
+		return iface.WGAddress{
+			IP: net.ParseIP("10.20.0.1"),
+			Network: &net.IPNet{
+				IP:   net.ParseIP("10.20.0.0"),
+				Mask: net.IPv4Mask(255, 255, 255, 0),
+			},
+		}
+	},
+}
+
 // iFaceMapper defines subset methods of interface required for manager
 type iFaceMock struct {
 	NameFunc    func() string
@@ -40,55 +54,28 @@ func TestIptablesManager(t *testing.T) {
 	ipv4Client, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
 	require.NoError(t, err)
 
-	mock := &iFaceMock{
-		NameFunc: func() string {
-			return "lo"
-		},
-		AddressFunc: func() iface.WGAddress {
-			return iface.WGAddress{
-				IP: net.ParseIP("10.20.0.1"),
-				Network: &net.IPNet{
-					IP:   net.ParseIP("10.20.0.0"),
-					Mask: net.IPv4Mask(255, 255, 255, 0),
-				},
-			}
-		},
-	}
-
 	// just check on the local interface
-	manager, err := Create(context.Background(), mock)
+	manager, err := Create(ifaceMock)
 	require.NoError(t, err)
+	require.NoError(t, manager.Init(nil))
 
 	time.Sleep(time.Second)
 
 	defer func() {
-		err := manager.Reset()
+		err := manager.Reset(nil)
 		require.NoError(t, err, "clear the manager state")
 
 		time.Sleep(time.Second)
 	}()
 
-	var rule1 []fw.Rule
-	t.Run("add first rule", func(t *testing.T) {
-		ip := net.ParseIP("10.20.0.2")
-		port := &fw.Port{Values: []int{8080}}
-		rule1, err = manager.AddFiltering(ip, "tcp", nil, port, fw.RuleDirectionOUT, fw.ActionAccept, "", "accept HTTP traffic")
-		require.NoError(t, err, "failed to add rule")
-
-		for _, r := range rule1 {
-			checkRuleSpecs(t, ipv4Client, chainNameOutputRules, true, r.(*Rule).specs...)
-		}
-
-	})
-
 	var rule2 []fw.Rule
 	t.Run("add second rule", func(t *testing.T) {
 		ip := net.ParseIP("10.20.0.3")
 		port := &fw.Port{
-			Values: []int{8043: 8046},
+			IsRange: true,
+			Values:  []uint16{8043, 8046},
 		}
-		rule2, err = manager.AddFiltering(
-			ip, "tcp", port, nil, fw.RuleDirectionIN, fw.ActionAccept, "", "accept HTTPS traffic from ports range")
+		rule2, err = manager.AddPeerFiltering(ip, "tcp", port, nil, fw.ActionAccept, "", "accept HTTPS traffic from ports range")
 		require.NoError(t, err, "failed to add rule")
 
 		for _, r := range rule2 {
@@ -97,18 +84,9 @@ func TestIptablesManager(t *testing.T) {
 		}
 	})
 
-	t.Run("delete first rule", func(t *testing.T) {
-		for _, r := range rule1 {
-			err := manager.DeleteRule(r)
-			require.NoError(t, err, "failed to delete rule")
-
-			checkRuleSpecs(t, ipv4Client, chainNameOutputRules, false, r.(*Rule).specs...)
-		}
-	})
-
 	t.Run("delete second rule", func(t *testing.T) {
 		for _, r := range rule2 {
-			err := manager.DeleteRule(r)
+			err := manager.DeletePeerRule(r)
 			require.NoError(t, err, "failed to delete rule")
 		}
 
@@ -118,11 +96,11 @@ func TestIptablesManager(t *testing.T) {
 	t.Run("reset check", func(t *testing.T) {
 		// add second rule
 		ip := net.ParseIP("10.20.0.3")
-		port := &fw.Port{Values: []int{5353}}
-		_, err = manager.AddFiltering(ip, "udp", nil, port, fw.RuleDirectionOUT, fw.ActionAccept, "", "accept Fake DNS traffic")
+		port := &fw.Port{Values: []uint16{5353}}
+		_, err = manager.AddPeerFiltering(ip, "udp", nil, port, fw.ActionAccept, "", "accept Fake DNS traffic")
 		require.NoError(t, err, "failed to add rule")
 
-		err = manager.Reset()
+		err = manager.Reset(nil)
 		require.NoError(t, err, "failed to reset")
 
 		ok, err := ipv4Client.ChainExists("filter", chainNameInputRules)
@@ -135,9 +113,6 @@ func TestIptablesManager(t *testing.T) {
 }
 
 func TestIptablesManagerIPSet(t *testing.T) {
-	ipv4Client, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
-	require.NoError(t, err)
-
 	mock := &iFaceMock{
 		NameFunc: func() string {
 			return "lo"
@@ -154,45 +129,26 @@ func TestIptablesManagerIPSet(t *testing.T) {
 	}
 
 	// just check on the local interface
-	manager, err := Create(context.Background(), mock)
+	manager, err := Create(mock)
 	require.NoError(t, err)
+	require.NoError(t, manager.Init(nil))
 
 	time.Sleep(time.Second)
 
 	defer func() {
-		err := manager.Reset()
+		err := manager.Reset(nil)
 		require.NoError(t, err, "clear the manager state")
 
 		time.Sleep(time.Second)
 	}()
 
-	var rule1 []fw.Rule
-	t.Run("add first rule with set", func(t *testing.T) {
-		ip := net.ParseIP("10.20.0.2")
-		port := &fw.Port{Values: []int{8080}}
-		rule1, err = manager.AddFiltering(
-			ip, "tcp", nil, port, fw.RuleDirectionOUT,
-			fw.ActionAccept, "default", "accept HTTP traffic",
-		)
-		require.NoError(t, err, "failed to add rule")
-
-		for _, r := range rule1 {
-			checkRuleSpecs(t, ipv4Client, chainNameOutputRules, true, r.(*Rule).specs...)
-			require.Equal(t, r.(*Rule).ipsetName, "default-dport", "ipset name must be set")
-			require.Equal(t, r.(*Rule).ip, "10.20.0.2", "ipset IP must be set")
-		}
-	})
-
 	var rule2 []fw.Rule
 	t.Run("add second rule", func(t *testing.T) {
 		ip := net.ParseIP("10.20.0.3")
 		port := &fw.Port{
-			Values: []int{443},
+			Values: []uint16{443},
 		}
-		rule2, err = manager.AddFiltering(
-			ip, "tcp", port, nil, fw.RuleDirectionIN, fw.ActionAccept,
-			"default", "accept HTTPS traffic from ports range",
-		)
+		rule2, err = manager.AddPeerFiltering(ip, "tcp", port, nil, fw.ActionAccept, "default", "accept HTTPS traffic from ports range")
 		for _, r := range rule2 {
 			require.NoError(t, err, "failed to add rule")
 			require.Equal(t, r.(*Rule).ipsetName, "default-sport", "ipset name must be set")
@@ -200,18 +156,9 @@ func TestIptablesManagerIPSet(t *testing.T) {
 		}
 	})
 
-	t.Run("delete first rule", func(t *testing.T) {
-		for _, r := range rule1 {
-			err := manager.DeleteRule(r)
-			require.NoError(t, err, "failed to delete rule")
-
-			require.NotContains(t, manager.aclMgr.ipsetStore.ipsets, r.(*Rule).ruleID, "rule must be removed form the ruleset index")
-		}
-	})
-
 	t.Run("delete second rule", func(t *testing.T) {
 		for _, r := range rule2 {
-			err := manager.DeleteRule(r)
+			err := manager.DeletePeerRule(r)
 			require.NoError(t, err, "failed to delete rule")
 
 			require.Empty(t, manager.aclMgr.ipsetStore.ipsets, "rulesets index after removed second rule must be empty")
@@ -219,7 +166,7 @@ func TestIptablesManagerIPSet(t *testing.T) {
 	})
 
 	t.Run("reset check", func(t *testing.T) {
-		err = manager.Reset()
+		err = manager.Reset(nil)
 		require.NoError(t, err, "failed to reset")
 	})
 }
@@ -251,12 +198,13 @@ func TestIptablesCreatePerformance(t *testing.T) {
 	for _, testMax := range []int{10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000} {
 		t.Run(fmt.Sprintf("Testing %d rules", testMax), func(t *testing.T) {
 			// just check on the local interface
-			manager, err := Create(context.Background(), mock)
+			manager, err := Create(mock)
 			require.NoError(t, err)
+			require.NoError(t, manager.Init(nil))
 			time.Sleep(time.Second)
 
 			defer func() {
-				err := manager.Reset()
+				err := manager.Reset(nil)
 				require.NoError(t, err, "clear the manager state")
 
 				time.Sleep(time.Second)
@@ -267,12 +215,8 @@ func TestIptablesCreatePerformance(t *testing.T) {
 			ip := net.ParseIP("10.20.0.100")
 			start := time.Now()
 			for i := 0; i < testMax; i++ {
-				port := &fw.Port{Values: []int{1000 + i}}
-				if i%2 == 0 {
-					_, err = manager.AddFiltering(ip, "tcp", nil, port, fw.RuleDirectionOUT, fw.ActionAccept, "", "accept HTTP traffic")
-				} else {
-					_, err = manager.AddFiltering(ip, "tcp", nil, port, fw.RuleDirectionIN, fw.ActionAccept, "", "accept HTTP traffic")
-				}
+				port := &fw.Port{Values: []uint16{uint16(1000 + i)}}
+				_, err = manager.AddPeerFiltering(ip, "tcp", nil, port, fw.ActionAccept, "", "accept HTTP traffic")
 
 				require.NoError(t, err, "failed to add rule")
 			}

client/firewall/iptables/router_linux.go

@@ -3,338 +3,606 @@
 package iptables
 
 import (
-	"context"
 	"fmt"
+	"net/netip"
+	"strconv"
 	"strings"
 
 	"github.com/coreos/go-iptables/iptables"
+	"github.com/hashicorp/go-multierror"
+	"github.com/nadoo/ipset"
 	log "github.com/sirupsen/logrus"
 
+	nberrors "github.com/netbirdio/netbird/client/errors"
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
-)
-
-const (
-	Ipv4Forwarding = "netbird-rt-forwarding"
-	ipv4Nat        = "netbird-rt-nat"
+	"github.com/netbirdio/netbird/client/internal/acl/id"
+	"github.com/netbirdio/netbird/client/internal/routemanager/refcounter"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
+	nbnet "github.com/netbirdio/netbird/util/net"
 )
 
 // constants needed to manage and create iptable rules
 const (
 	tableFilter             = "filter"
 	tableNat                = "nat"
-	chainFORWARD            = "FORWARD"
+	tableMangle             = "mangle"
 	chainPOSTROUTING        = "POSTROUTING"
+	chainPREROUTING         = "PREROUTING"
 	chainRTNAT              = "NETBIRD-RT-NAT"
 	chainRTFWD              = "NETBIRD-RT-FWD"
+	chainRTPRE              = "NETBIRD-RT-PRE"
 	routingFinalForwardJump = "ACCEPT"
 	routingFinalNatJump     = "MASQUERADE"
+
+	jumpPre  = "jump-pre"
+	jumpNat  = "jump-nat"
+	matchSet = "--match-set"
 )
 
-type routerManager struct {
-	ctx            context.Context
-	stop           context.CancelFunc
-	iptablesClient *iptables.IPTables
-	rules          map[string][]string
+type routeFilteringRuleParams struct {
+	Sources     []netip.Prefix
+	Destination netip.Prefix
+	Proto       firewall.Protocol
+	SPort       *firewall.Port
+	DPort       *firewall.Port
+	Direction   firewall.RuleDirection
+	Action      firewall.Action
+	SetName     string
 }
 
-func newRouterManager(parentCtx context.Context, iptablesClient *iptables.IPTables) (*routerManager, error) {
-	ctx, cancel := context.WithCancel(parentCtx)
-	m := &routerManager{
-		ctx:            ctx,
-		stop:           cancel,
+type routeRules map[string][]string
+
+type ipsetCounter = refcounter.Counter[string, []netip.Prefix, struct{}]
+
+type router struct {
+	iptablesClient   *iptables.IPTables
+	rules            routeRules
+	ipsetCounter     *ipsetCounter
+	wgIface          iFaceMapper
+	legacyManagement bool
+
+	stateManager *statemanager.Manager
+}
+
+func newRouter(iptablesClient *iptables.IPTables, wgIface iFaceMapper) (*router, error) {
+	r := &router{
 		iptablesClient: iptablesClient,
 		rules:          make(map[string][]string),
+		wgIface:        wgIface,
 	}
 
-	err := m.cleanUpDefaultForwardRules()
-	if err != nil {
-		log.Errorf("failed to cleanup routing rules: %s", err)
-		return nil, err
+	r.ipsetCounter = refcounter.New(
+		func(name string, sources []netip.Prefix) (struct{}, error) {
+			return struct{}{}, r.createIpSet(name, sources)
+		},
+		func(name string, _ struct{}) error {
+			return r.deleteIpSet(name)
+		},
+	)
+
+	if err := ipset.Init(); err != nil {
+		return nil, fmt.Errorf("init ipset: %w", err)
 	}
-	err = m.createContainers()
-	if err != nil {
-		log.Errorf("failed to create containers for route: %s", err)
-	}
-	return m, err
+
+	return r, nil
 }
 
-// InsertRoutingRules inserts an iptables rule pair to the forwarding chain and if enabled, to the nat chain
-func (i *routerManager) InsertRoutingRules(pair firewall.RouterPair) error {
-	err := i.insertRoutingRule(firewall.ForwardingFormat, tableFilter, chainRTFWD, routingFinalForwardJump, pair)
-	if err != nil {
-		return err
+func (r *router) init(stateManager *statemanager.Manager) error {
+	r.stateManager = stateManager
+
+	if err := r.cleanUpDefaultForwardRules(); err != nil {
+		log.Errorf("failed to clean up rules from FORWARD chain: %s", err)
+	}
+
+	if err := r.createContainers(); err != nil {
+		return fmt.Errorf("create containers: %w", err)
+	}
+
+	r.updateState()
+
+	return nil
+}
+
+func (r *router) AddRouteFiltering(
+	sources []netip.Prefix,
+	destination netip.Prefix,
+	proto firewall.Protocol,
+	sPort *firewall.Port,
+	dPort *firewall.Port,
+	action firewall.Action,
+) (firewall.Rule, error) {
+	ruleKey := id.GenerateRouteRuleKey(sources, destination, proto, sPort, dPort, action)
+	if _, ok := r.rules[string(ruleKey)]; ok {
+		return ruleKey, nil
+	}
+
+	var setName string
+	if len(sources) > 1 {
+		setName = firewall.GenerateSetName(sources)
+		if _, err := r.ipsetCounter.Increment(setName, sources); err != nil {
+			return nil, fmt.Errorf("create or get ipset: %w", err)
+		}
+	}
+
+	params := routeFilteringRuleParams{
+		Sources:     sources,
+		Destination: destination,
+		Proto:       proto,
+		SPort:       sPort,
+		DPort:       dPort,
+		Action:      action,
+		SetName:     setName,
+	}
+
+	rule := genRouteFilteringRuleSpec(params)
+	// Insert DROP rules at the beginning, append ACCEPT rules at the end
+	var err error
+	if action == firewall.ActionDrop {
+		// after the established rule
+		err = r.iptablesClient.Insert(tableFilter, chainRTFWD, 2, rule...)
+	} else {
+		err = r.iptablesClient.Append(tableFilter, chainRTFWD, rule...)
 	}
 
-	err = i.insertRoutingRule(firewall.InForwardingFormat, tableFilter, chainRTFWD, routingFinalForwardJump, firewall.GetInPair(pair))
 	if err != nil {
-		return err
+		return nil, fmt.Errorf("add route rule: %v", err)
+	}
+
+	r.rules[string(ruleKey)] = rule
+
+	r.updateState()
+
+	return ruleKey, nil
+}
+
+func (r *router) DeleteRouteRule(rule firewall.Rule) error {
+	ruleKey := rule.GetRuleID()
+
+	if rule, exists := r.rules[ruleKey]; exists {
+		setName := r.findSetNameInRule(rule)
+
+		if err := r.iptablesClient.Delete(tableFilter, chainRTFWD, rule...); err != nil {
+			return fmt.Errorf("delete route rule: %v", err)
+		}
+		delete(r.rules, ruleKey)
+
+		if setName != "" {
+			if _, err := r.ipsetCounter.Decrement(setName); err != nil {
+				return fmt.Errorf("failed to remove ipset: %w", err)
+			}
+		}
+	} else {
+		log.Debugf("route rule %s not found", ruleKey)
+	}
+
+	r.updateState()
+
+	return nil
+}
+
+func (r *router) findSetNameInRule(rule []string) string {
+	for i, arg := range rule {
+		if arg == "-m" && i+3 < len(rule) && rule[i+1] == "set" && rule[i+2] == matchSet {
+			return rule[i+3]
+		}
+	}
+	return ""
+}
+
+func (r *router) createIpSet(setName string, sources []netip.Prefix) error {
+	if err := ipset.Create(setName, ipset.OptTimeout(0)); err != nil {
+		return fmt.Errorf("create set %s: %w", setName, err)
+	}
+
+	for _, prefix := range sources {
+		if err := ipset.AddPrefix(setName, prefix); err != nil {
+			return fmt.Errorf("add element to set %s: %w", setName, err)
+		}
+	}
+
+	return nil
+}
+
+func (r *router) deleteIpSet(setName string) error {
+	if err := ipset.Destroy(setName); err != nil {
+		return fmt.Errorf("destroy set %s: %w", setName, err)
+	}
+	return nil
+}
+
+// AddNatRule inserts an iptables rule pair into the nat chain
+func (r *router) AddNatRule(pair firewall.RouterPair) error {
+	if r.legacyManagement {
+		log.Warnf("This peer is connected to a NetBird Management service with an older version. Allowing all traffic for %s", pair.Destination)
+		if err := r.addLegacyRouteRule(pair); err != nil {
+			return fmt.Errorf("add legacy routing rule: %w", err)
+		}
 	}
 
 	if !pair.Masquerade {
 		return nil
 	}
 
-	err = i.insertRoutingRule(firewall.NatFormat, tableNat, chainRTNAT, routingFinalNatJump, pair)
-	if err != nil {
-		return err
+	if err := r.addNatRule(pair); err != nil {
+		return fmt.Errorf("add nat rule: %w", err)
 	}
 
-	err = i.insertRoutingRule(firewall.InNatFormat, tableNat, chainRTNAT, routingFinalNatJump, firewall.GetInPair(pair))
-	if err != nil {
-		return err
+	if err := r.addNatRule(firewall.GetInversePair(pair)); err != nil {
+		return fmt.Errorf("add inverse nat rule: %w", err)
 	}
 
+	r.updateState()
+
 	return nil
 }
 
-// insertRoutingRule inserts an iptable rule
-func (i *routerManager) insertRoutingRule(keyFormat, table, chain, jump string, pair firewall.RouterPair) error {
-	var err error
+// RemoveNatRule removes an iptables rule pair from forwarding and nat chains
+func (r *router) RemoveNatRule(pair firewall.RouterPair) error {
+	if err := r.removeNatRule(pair); err != nil {
+		return fmt.Errorf("remove nat rule: %w", err)
+	}
 
-	ruleKey := firewall.GenKey(keyFormat, pair.ID)
-	rule := genRuleSpec(jump, ruleKey, pair.Source, pair.Destination)
-	existingRule, found := i.rules[ruleKey]
-	if found {
-		err = i.iptablesClient.DeleteIfExists(table, chain, existingRule...)
-		if err != nil {
-			return fmt.Errorf("error while removing existing %s rule for %s: %v", getIptablesRuleType(table), pair.Destination, err)
+	if err := r.removeNatRule(firewall.GetInversePair(pair)); err != nil {
+		return fmt.Errorf("remove inverse nat rule: %w", err)
+	}
+
+	if err := r.removeLegacyRouteRule(pair); err != nil {
+		return fmt.Errorf("remove legacy routing rule: %w", err)
+	}
+
+	r.updateState()
+
+	return nil
+}
+
+// addLegacyRouteRule adds a legacy routing rule for mgmt servers pre route acls
+func (r *router) addLegacyRouteRule(pair firewall.RouterPair) error {
+	ruleKey := firewall.GenKey(firewall.ForwardingFormat, pair)
+
+	if err := r.removeLegacyRouteRule(pair); err != nil {
+		return err
+	}
+
+	rule := []string{"-s", pair.Source.String(), "-d", pair.Destination.String(), "-j", routingFinalForwardJump}
+	if err := r.iptablesClient.Append(tableFilter, chainRTFWD, rule...); err != nil {
+		return fmt.Errorf("add legacy forwarding rule %s -> %s: %v", pair.Source, pair.Destination, err)
+	}
+
+	r.rules[ruleKey] = rule
+
+	return nil
+}
+
+func (r *router) removeLegacyRouteRule(pair firewall.RouterPair) error {
+	ruleKey := firewall.GenKey(firewall.ForwardingFormat, pair)
+
+	if rule, exists := r.rules[ruleKey]; exists {
+		if err := r.iptablesClient.DeleteIfExists(tableFilter, chainRTFWD, rule...); err != nil {
+			return fmt.Errorf("remove legacy forwarding rule %s -> %s: %v", pair.Source, pair.Destination, err)
 		}
-		delete(i.rules, ruleKey)
-	}
-	err = i.iptablesClient.Insert(table, chain, 1, rule...)
-	if err != nil {
-		return fmt.Errorf("error while adding new %s rule for %s: %v", getIptablesRuleType(table), pair.Destination, err)
-	}
-
-	i.rules[ruleKey] = rule
-
-	return nil
-}
-
-// RemoveRoutingRules removes an iptables rule pair from forwarding and nat chains
-func (i *routerManager) RemoveRoutingRules(pair firewall.RouterPair) error {
-	err := i.removeRoutingRule(firewall.ForwardingFormat, tableFilter, chainRTFWD, pair)
-	if err != nil {
-		return err
-	}
-
-	err = i.removeRoutingRule(firewall.InForwardingFormat, tableFilter, chainRTFWD, firewall.GetInPair(pair))
-	if err != nil {
-		return err
-	}
-
-	if !pair.Masquerade {
-		return nil
-	}
-
-	err = i.removeRoutingRule(firewall.NatFormat, tableNat, chainRTNAT, pair)
-	if err != nil {
-		return err
-	}
-
-	err = i.removeRoutingRule(firewall.InNatFormat, tableNat, chainRTNAT, firewall.GetInPair(pair))
-	if err != nil {
-		return err
+		delete(r.rules, ruleKey)
+	} else {
+		log.Debugf("legacy forwarding rule %s not found", ruleKey)
 	}
 
 	return nil
 }
 
-func (i *routerManager) removeRoutingRule(keyFormat, table, chain string, pair firewall.RouterPair) error {
-	var err error
+// GetLegacyManagement returns the current legacy management mode
+func (r *router) GetLegacyManagement() bool {
+	return r.legacyManagement
+}
 
-	ruleKey := firewall.GenKey(keyFormat, pair.ID)
-	existingRule, found := i.rules[ruleKey]
-	if found {
-		err = i.iptablesClient.DeleteIfExists(table, chain, existingRule...)
-		if err != nil {
-			return fmt.Errorf("error while removing existing %s rule for %s: %v", getIptablesRuleType(table), pair.Destination, err)
+// SetLegacyManagement sets the route manager to use legacy management mode
+func (r *router) SetLegacyManagement(isLegacy bool) {
+	r.legacyManagement = isLegacy
+}
+
+// RemoveAllLegacyRouteRules removes all legacy routing rules for mgmt servers pre route acls
+func (r *router) RemoveAllLegacyRouteRules() error {
+	var merr *multierror.Error
+	for k, rule := range r.rules {
+		if !strings.HasPrefix(k, firewall.ForwardingFormatPrefix) {
+			continue
+		}
+		if err := r.iptablesClient.DeleteIfExists(tableFilter, chainRTFWD, rule...); err != nil {
+			merr = multierror.Append(merr, fmt.Errorf("remove legacy forwarding rule: %v", err))
+		} else {
+			delete(r.rules, k)
 		}
 	}
-	delete(i.rules, ruleKey)
 
-	return nil
+	r.updateState()
+
+	return nberrors.FormatErrorOrNil(merr)
 }
 
-func (i *routerManager) RouteingFwChainName() string {
-	return chainRTFWD
-}
-
-func (i *routerManager) Reset() error {
-	err := i.cleanUpDefaultForwardRules()
-	if err != nil {
-		return err
+func (r *router) Reset() error {
+	var merr *multierror.Error
+	if err := r.cleanUpDefaultForwardRules(); err != nil {
+		merr = multierror.Append(merr, err)
 	}
-	i.rules = make(map[string][]string)
-	return nil
+	r.rules = make(map[string][]string)
+
+	if err := r.ipsetCounter.Flush(); err != nil {
+		merr = multierror.Append(merr, err)
+	}
+
+	r.updateState()
+
+	return nberrors.FormatErrorOrNil(merr)
 }
 
-func (i *routerManager) cleanUpDefaultForwardRules() error {
-	err := i.cleanJumpRules()
-	if err != nil {
-		return err
+func (r *router) cleanUpDefaultForwardRules() error {
+	if err := r.cleanJumpRules(); err != nil {
+		return fmt.Errorf("clean jump rules: %w", err)
 	}
 
 	log.Debug("flushing routing related tables")
-	ok, err := i.iptablesClient.ChainExists(tableFilter, chainRTFWD)
-	if err != nil {
-		log.Errorf("failed check chain %s,error: %v", chainRTFWD, err)
-		return err
-	} else if ok {
-		err = i.iptablesClient.ClearAndDeleteChain(tableFilter, chainRTFWD)
+	for _, chainInfo := range []struct {
+		chain string
+		table string
+	}{
+		{chainRTFWD, tableFilter},
+		{chainRTNAT, tableNat},
+		{chainRTPRE, tableMangle},
+	} {
+		ok, err := r.iptablesClient.ChainExists(chainInfo.table, chainInfo.chain)
 		if err != nil {
-			log.Errorf("failed cleaning chain %s,error: %v", chainRTFWD, err)
-			return err
+			return fmt.Errorf("check chain %s in table %s: %w", chainInfo.chain, chainInfo.table, err)
+		} else if ok {
+			if err = r.iptablesClient.ClearAndDeleteChain(chainInfo.table, chainInfo.chain); err != nil {
+				return fmt.Errorf("clear and delete chain %s in table %s: %w", chainInfo.chain, chainInfo.table, err)
+			}
 		}
 	}
 
-	ok, err = i.iptablesClient.ChainExists(tableNat, chainRTNAT)
+	return nil
+}
+
+func (r *router) createContainers() error {
+	for _, chainInfo := range []struct {
+		chain string
+		table string
+	}{
+		{chainRTFWD, tableFilter},
+		{chainRTPRE, tableMangle},
+		{chainRTNAT, tableNat},
+	} {
+		if err := r.createAndSetupChain(chainInfo.chain); err != nil {
+			return fmt.Errorf("create chain %s in table %s: %w", chainInfo.chain, chainInfo.table, err)
+		}
+	}
+
+	if err := r.insertEstablishedRule(chainRTFWD); err != nil {
+		return fmt.Errorf("insert established rule: %w", err)
+	}
+
+	if err := r.addPostroutingRules(); err != nil {
+		return fmt.Errorf("add static nat rules: %w", err)
+	}
+
+	if err := r.addJumpRules(); err != nil {
+		return fmt.Errorf("add jump rules: %w", err)
+	}
+
+	return nil
+}
+
+func (r *router) addPostroutingRules() error {
+	// First rule for outbound masquerade
+	rule1 := []string{
+		"-m", "mark", "--mark", fmt.Sprintf("%#x", nbnet.PreroutingFwmarkMasquerade),
+		"!", "-o", "lo",
+		"-j", routingFinalNatJump,
+	}
+	if err := r.iptablesClient.Append(tableNat, chainRTNAT, rule1...); err != nil {
+		return fmt.Errorf("add outbound masquerade rule: %v", err)
+	}
+	r.rules["static-nat-outbound"] = rule1
+
+	// Second rule for return traffic masquerade
+	rule2 := []string{
+		"-m", "mark", "--mark", fmt.Sprintf("%#x", nbnet.PreroutingFwmarkMasqueradeReturn),
+		"-o", r.wgIface.Name(),
+		"-j", routingFinalNatJump,
+	}
+	if err := r.iptablesClient.Append(tableNat, chainRTNAT, rule2...); err != nil {
+		return fmt.Errorf("add return masquerade rule: %v", err)
+	}
+	r.rules["static-nat-return"] = rule2
+
+	return nil
+}
+
+func (r *router) createAndSetupChain(chain string) error {
+	table := r.getTableForChain(chain)
+
+	if err := r.iptablesClient.NewChain(table, chain); err != nil {
+		return fmt.Errorf("failed creating chain %s, error: %v", chain, err)
+	}
+
+	return nil
+}
+
+func (r *router) getTableForChain(chain string) string {
+	switch chain {
+	case chainRTNAT:
+		return tableNat
+	case chainRTPRE:
+		return tableMangle
+	default:
+		return tableFilter
+	}
+}
+
+func (r *router) insertEstablishedRule(chain string) error {
+	establishedRule := getConntrackEstablished()
+
+	err := r.iptablesClient.Insert(tableFilter, chain, 1, establishedRule...)
 	if err != nil {
-		log.Errorf("failed check chain %s,error: %v", chainRTNAT, err)
-		return err
-	} else if ok {
-		err = i.iptablesClient.ClearAndDeleteChain(tableNat, chainRTNAT)
-		if err != nil {
-			log.Errorf("failed cleaning chain %s,error: %v", chainRTNAT, err)
-			return err
+		return fmt.Errorf("failed to insert established rule: %v", err)
+	}
+
+	ruleKey := "established-" + chain
+	r.rules[ruleKey] = establishedRule
+
+	return nil
+}
+
+func (r *router) addJumpRules() error {
+	// Jump to NAT chain
+	natRule := []string{"-j", chainRTNAT}
+	if err := r.iptablesClient.Insert(tableNat, chainPOSTROUTING, 1, natRule...); err != nil {
+		return fmt.Errorf("add nat jump rule: %v", err)
+	}
+	r.rules[jumpNat] = natRule
+
+	// Jump to prerouting chain
+	preRule := []string{"-j", chainRTPRE}
+	if err := r.iptablesClient.Insert(tableMangle, chainPREROUTING, 1, preRule...); err != nil {
+		return fmt.Errorf("add prerouting jump rule: %v", err)
+	}
+	r.rules[jumpPre] = preRule
+
+	return nil
+}
+
+func (r *router) cleanJumpRules() error {
+	for _, ruleKey := range []string{jumpNat, jumpPre} {
+		if rule, exists := r.rules[ruleKey]; exists {
+			table := tableNat
+			chain := chainPOSTROUTING
+			if ruleKey == jumpPre {
+				table = tableMangle
+				chain = chainPREROUTING
+			}
+
+			if err := r.iptablesClient.DeleteIfExists(table, chain, rule...); err != nil {
+				return fmt.Errorf("delete rule from chain %s in table %s, err: %v", chain, table, err)
+			}
+			delete(r.rules, ruleKey)
 		}
 	}
 	return nil
 }
 
-func (i *routerManager) createContainers() error {
-	if i.rules[Ipv4Forwarding] != nil {
+func (r *router) addNatRule(pair firewall.RouterPair) error {
+	ruleKey := firewall.GenKey(firewall.NatFormat, pair)
+
+	if rule, exists := r.rules[ruleKey]; exists {
+		if err := r.iptablesClient.DeleteIfExists(tableMangle, chainRTPRE, rule...); err != nil {
+			return fmt.Errorf("error while removing existing marking rule for %s: %v", pair.Destination, err)
+		}
+		delete(r.rules, ruleKey)
+	}
+
+	markValue := nbnet.PreroutingFwmarkMasquerade
+	if pair.Inverse {
+		markValue = nbnet.PreroutingFwmarkMasqueradeReturn
+	}
+
+	rule := []string{"-i", r.wgIface.Name()}
+	if pair.Inverse {
+		rule = []string{"!", "-i", r.wgIface.Name()}
+	}
+
+	rule = append(rule,
+		"-m", "conntrack",
+		"--ctstate", "NEW",
+		"-s", pair.Source.String(),
+		"-d", pair.Destination.String(),
+		"-j", "MARK", "--set-mark", fmt.Sprintf("%#x", markValue),
+	)
+
+	if err := r.iptablesClient.Append(tableMangle, chainRTPRE, rule...); err != nil {
+		return fmt.Errorf("error while adding marking rule for %s: %v", pair.Destination, err)
+	}
+
+	r.rules[ruleKey] = rule
+	return nil
+}
+
+func (r *router) removeNatRule(pair firewall.RouterPair) error {
+	ruleKey := firewall.GenKey(firewall.NatFormat, pair)
+
+	if rule, exists := r.rules[ruleKey]; exists {
+		if err := r.iptablesClient.DeleteIfExists(tableMangle, chainRTPRE, rule...); err != nil {
+			return fmt.Errorf("error while removing marking rule for %s: %v", pair.Destination, err)
+		}
+		delete(r.rules, ruleKey)
+	} else {
+		log.Debugf("marking rule %s not found", ruleKey)
+	}
+
+	return nil
+}
+
+func (r *router) updateState() {
+	if r.stateManager == nil {
+		return
+	}
+
+	var currentState *ShutdownState
+	if existing := r.stateManager.GetState(currentState); existing != nil {
+		if existingState, ok := existing.(*ShutdownState); ok {
+			currentState = existingState
+		}
+	}
+	if currentState == nil {
+		currentState = &ShutdownState{}
+	}
+
+	currentState.Lock()
+	defer currentState.Unlock()
+
+	currentState.RouteRules = r.rules
+	currentState.RouteIPsetCounter = r.ipsetCounter
+
+	if err := r.stateManager.UpdateState(currentState); err != nil {
+		log.Errorf("failed to update state: %v", err)
+	}
+}
+
+func genRouteFilteringRuleSpec(params routeFilteringRuleParams) []string {
+	var rule []string
+
+	if params.SetName != "" {
+		rule = append(rule, "-m", "set", matchSet, params.SetName, "src")
+	} else if len(params.Sources) > 0 {
+		source := params.Sources[0]
+		rule = append(rule, "-s", source.String())
+	}
+
+	rule = append(rule, "-d", params.Destination.String())
+
+	if params.Proto != firewall.ProtocolALL {
+		rule = append(rule, "-p", strings.ToLower(string(params.Proto)))
+		rule = append(rule, applyPort("--sport", params.SPort)...)
+		rule = append(rule, applyPort("--dport", params.DPort)...)
+	}
+
+	rule = append(rule, "-j", actionToStr(params.Action))
+
+	return rule
+}
+
+func applyPort(flag string, port *firewall.Port) []string {
+	if port == nil {
 		return nil
 	}
 
-	errMSGFormat := "failed creating chain %s,error: %v"
-	err := i.createChain(tableFilter, chainRTFWD)
-	if err != nil {
-		return fmt.Errorf(errMSGFormat, chainRTFWD, err)
+	if port.IsRange && len(port.Values) == 2 {
+		return []string{flag, fmt.Sprintf("%d:%d", port.Values[0], port.Values[1])}
 	}
 
-	err = i.createChain(tableNat, chainRTNAT)
-	if err != nil {
-		return fmt.Errorf(errMSGFormat, chainRTNAT, err)
+	if len(port.Values) > 1 {
+		portList := make([]string, len(port.Values))
+		for i, p := range port.Values {
+			portList[i] = strconv.Itoa(int(p))
+		}
+		return []string{"-m", "multiport", flag, strings.Join(portList, ",")}
 	}
 
-	err = i.addJumpRules()
-	if err != nil {
-		return fmt.Errorf("error while creating jump rules: %v", err)
-	}
-
-	return nil
-}
-
-// addJumpRules create jump rules to send packets to NetBird chains
-func (i *routerManager) addJumpRules() error {
-	rule := []string{"-j", chainRTFWD}
-	err := i.iptablesClient.Insert(tableFilter, chainFORWARD, 1, rule...)
-	if err != nil {
-		return err
-	}
-	i.rules[Ipv4Forwarding] = rule
-
-	rule = []string{"-j", chainRTNAT}
-	err = i.iptablesClient.Insert(tableNat, chainPOSTROUTING, 1, rule...)
-	if err != nil {
-		return err
-	}
-	i.rules[ipv4Nat] = rule
-
-	return nil
-}
-
-// cleanJumpRules cleans jump rules that was sending packets to NetBird chains
-func (i *routerManager) cleanJumpRules() error {
-	var err error
-	errMSGFormat := "failed cleaning rule from chain %s,err: %v"
-	rule, found := i.rules[Ipv4Forwarding]
-	if found {
-		err = i.iptablesClient.DeleteIfExists(tableFilter, chainFORWARD, rule...)
-		if err != nil {
-			return fmt.Errorf(errMSGFormat, chainFORWARD, err)
-		}
-	}
-	rule, found = i.rules[ipv4Nat]
-	if found {
-		err = i.iptablesClient.DeleteIfExists(tableNat, chainPOSTROUTING, rule...)
-		if err != nil {
-			return fmt.Errorf(errMSGFormat, chainPOSTROUTING, err)
-		}
-	}
-
-	rules, err := i.iptablesClient.List("nat", "POSTROUTING")
-	if err != nil {
-		return fmt.Errorf("failed to list rules: %s", err)
-	}
-
-	for _, ruleString := range rules {
-		if !strings.Contains(ruleString, "NETBIRD") {
-			continue
-		}
-		rule := strings.Fields(ruleString)
-		err := i.iptablesClient.DeleteIfExists("nat", "POSTROUTING", rule[2:]...)
-		if err != nil {
-			return fmt.Errorf("failed to delete postrouting jump rule: %s", err)
-		}
-	}
-
-	rules, err = i.iptablesClient.List(tableFilter, "FORWARD")
-	if err != nil {
-		return fmt.Errorf("failed to list rules in FORWARD chain: %s", err)
-	}
-
-	for _, ruleString := range rules {
-		if !strings.Contains(ruleString, "NETBIRD") {
-			continue
-		}
-		rule := strings.Fields(ruleString)
-		err := i.iptablesClient.DeleteIfExists(tableFilter, "FORWARD", rule[2:]...)
-		if err != nil {
-			return fmt.Errorf("failed to delete FORWARD jump rule: %s", err)
-		}
-	}
-	return nil
-}
-
-func (i *routerManager) createChain(table, newChain string) error {
-	chains, err := i.iptablesClient.ListChains(table)
-	if err != nil {
-		return fmt.Errorf("couldn't get %s table chains, error: %v", table, err)
-	}
-
-	shouldCreateChain := true
-	for _, chain := range chains {
-		if chain == newChain {
-			shouldCreateChain = false
-		}
-	}
-
-	if shouldCreateChain {
-		err = i.iptablesClient.NewChain(table, newChain)
-		if err != nil {
-			return fmt.Errorf("couldn't create chain %s in %s table, error: %v", newChain, table, err)
-		}
-
-		err = i.iptablesClient.Append(table, newChain, "-j", "RETURN")
-		if err != nil {
-			return fmt.Errorf("couldn't create chain %s default rule, error: %v", newChain, err)
-		}
-
-	}
-	return nil
-}
-
-// genRuleSpec generates rule specification with comment identifier
-func genRuleSpec(jump, id, source, destination string) []string {
-	return []string{"-s", source, "-d", destination, "-j", jump, "-m", "comment", "--comment", id}
-}
-
-func getIptablesRuleType(table string) string {
-	ruleType := "forwarding"
-	if table == tableNat {
-		ruleType = "nat"
-	}
-	return ruleType
+	return []string{flag, strconv.Itoa(int(port.Values[0]))}
 }

client/firewall/iptables/router_linux_test.go

@@ -3,16 +3,18 @@
 package iptables
 
 import (
-	"context"
+	"fmt"
+	"net/netip"
 	"os/exec"
 	"testing"
 
 	"github.com/coreos/go-iptables/iptables"
-	log "github.com/sirupsen/logrus"
+	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
 	"github.com/netbirdio/netbird/client/firewall/test"
+	nbnet "github.com/netbirdio/netbird/util/net"
 )
 
 func isIptablesSupported() bool {
@@ -28,47 +30,45 @@ func TestIptablesManager_RestoreOrCreateContainers(t *testing.T) {
 	iptablesClient, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
 	require.NoError(t, err, "failed to init iptables client")
 
-	manager, err := newRouterManager(context.TODO(), iptablesClient)
+	manager, err := newRouter(iptablesClient, ifaceMock)
 	require.NoError(t, err, "should return a valid iptables manager")
+	require.NoError(t, manager.init(nil))
 
 	defer func() {
-		_ = manager.Reset()
+		assert.NoError(t, manager.Reset(), "shouldn't return error")
 	}()
 
-	require.Len(t, manager.rules, 2, "should have created rules map")
+	// Now 5 rules:
+	// 1. established rule in forward chain
+	// 2. jump rule to NAT chain
+	// 3. jump rule to PRE chain
+	// 4. static outbound masquerade rule
+	// 5. static return masquerade rule
+	require.Len(t, manager.rules, 5, "should have created rules map")
 
-	exists, err := manager.iptablesClient.Exists(tableFilter, chainFORWARD, manager.rules[Ipv4Forwarding]...)
-	require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableFilter, chainFORWARD)
-	require.True(t, exists, "forwarding rule should exist")
-
-	exists, err = manager.iptablesClient.Exists(tableNat, chainPOSTROUTING, manager.rules[ipv4Nat]...)
+	exists, err := manager.iptablesClient.Exists(tableNat, chainPOSTROUTING, "-j", chainRTNAT)
 	require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableNat, chainPOSTROUTING)
-	require.True(t, exists, "postrouting rule should exist")
+	require.True(t, exists, "postrouting jump rule should exist")
+
+	exists, err = manager.iptablesClient.Exists(tableMangle, chainPREROUTING, "-j", chainRTPRE)
+	require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableMangle, chainPREROUTING)
+	require.True(t, exists, "prerouting jump rule should exist")
 
 	pair := firewall.RouterPair{
 		ID:          "abc",
-		Source:      "100.100.100.1/32",
-		Destination: "100.100.100.0/24",
+		Source:      netip.MustParsePrefix("100.100.100.1/32"),
+		Destination: netip.MustParsePrefix("100.100.100.0/24"),
 		Masquerade:  true,
 	}
-	forward4RuleKey := firewall.GenKey(firewall.ForwardingFormat, pair.ID)
-	forward4Rule := genRuleSpec(routingFinalForwardJump, forward4RuleKey, pair.Source, pair.Destination)
 
-	err = manager.iptablesClient.Insert(tableFilter, chainRTFWD, 1, forward4Rule...)
-	require.NoError(t, err, "inserting rule should not return error")
-
-	nat4RuleKey := firewall.GenKey(firewall.NatFormat, pair.ID)
-	nat4Rule := genRuleSpec(routingFinalNatJump, nat4RuleKey, pair.Source, pair.Destination)
-
-	err = manager.iptablesClient.Insert(tableNat, chainRTNAT, 1, nat4Rule...)
-	require.NoError(t, err, "inserting rule should not return error")
+	err = manager.AddNatRule(pair)
+	require.NoError(t, err, "adding NAT rule should not return error")
 
 	err = manager.Reset()
 	require.NoError(t, err, "shouldn't return error")
 }
 
-func TestIptablesManager_InsertRoutingRules(t *testing.T) {
-
+func TestIptablesManager_AddNatRule(t *testing.T) {
 	if !isIptablesSupported() {
 		t.SkipNow()
 	}
@@ -78,78 +78,71 @@ func TestIptablesManager_InsertRoutingRules(t *testing.T) {
 			iptablesClient, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
 			require.NoError(t, err, "failed to init iptables client")
 
-			manager, err := newRouterManager(context.TODO(), iptablesClient)
+			manager, err := newRouter(iptablesClient, ifaceMock)
 			require.NoError(t, err, "shouldn't return error")
+			require.NoError(t, manager.init(nil))
 
 			defer func() {
-				err := manager.Reset()
-				if err != nil {
-					log.Errorf("failed to reset iptables manager: %s", err)
-				}
+				assert.NoError(t, manager.Reset(), "shouldn't return error")
 			}()
 
-			err = manager.InsertRoutingRules(testCase.InputPair)
-			require.NoError(t, err, "forwarding pair should be inserted")
+			err = manager.AddNatRule(testCase.InputPair)
+			require.NoError(t, err, "marking rule should be inserted")
 
-			forwardRuleKey := firewall.GenKey(firewall.ForwardingFormat, testCase.InputPair.ID)
-			forwardRule := genRuleSpec(routingFinalForwardJump, forwardRuleKey, testCase.InputPair.Source, testCase.InputPair.Destination)
-
-			exists, err := iptablesClient.Exists(tableFilter, chainRTFWD, forwardRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableFilter, chainRTFWD)
-			require.True(t, exists, "forwarding rule should exist")
-
-			foundRule, found := manager.rules[forwardRuleKey]
-			require.True(t, found, "forwarding rule should exist in the manager map")
-			require.Equal(t, forwardRule[:4], foundRule[:4], "stored forwarding rule should match")
-
-			inForwardRuleKey := firewall.GenKey(firewall.InForwardingFormat, testCase.InputPair.ID)
-			inForwardRule := genRuleSpec(routingFinalForwardJump, inForwardRuleKey, firewall.GetInPair(testCase.InputPair).Source, firewall.GetInPair(testCase.InputPair).Destination)
-
-			exists, err = iptablesClient.Exists(tableFilter, chainRTFWD, inForwardRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableFilter, chainRTFWD)
-			require.True(t, exists, "income forwarding rule should exist")
-
-			foundRule, found = manager.rules[inForwardRuleKey]
-			require.True(t, found, "income forwarding rule should exist in the manager map")
-			require.Equal(t, inForwardRule[:4], foundRule[:4], "stored income forwarding rule should match")
-
-			natRuleKey := firewall.GenKey(firewall.NatFormat, testCase.InputPair.ID)
-			natRule := genRuleSpec(routingFinalNatJump, natRuleKey, testCase.InputPair.Source, testCase.InputPair.Destination)
-
-			exists, err = iptablesClient.Exists(tableNat, chainRTNAT, natRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableNat, chainRTNAT)
-			if testCase.InputPair.Masquerade {
-				require.True(t, exists, "nat rule should be created")
-				foundNatRule, foundNat := manager.rules[natRuleKey]
-				require.True(t, foundNat, "nat rule should exist in the map")
-				require.Equal(t, natRule[:4], foundNatRule[:4], "stored nat rule should match")
-			} else {
-				require.False(t, exists, "nat rule should not be created")
-				_, foundNat := manager.rules[natRuleKey]
-				require.False(t, foundNat, "nat rule should not exist in the map")
+			natRuleKey := firewall.GenKey(firewall.NatFormat, testCase.InputPair)
+			markingRule := []string{
+				"-i", ifaceMock.Name(),
+				"-m", "conntrack",
+				"--ctstate", "NEW",
+				"-s", testCase.InputPair.Source.String(),
+				"-d", testCase.InputPair.Destination.String(),
+				"-j", "MARK", "--set-mark",
+				fmt.Sprintf("%#x", nbnet.PreroutingFwmarkMasquerade),
 			}
 
-			inNatRuleKey := firewall.GenKey(firewall.InNatFormat, testCase.InputPair.ID)
-			inNatRule := genRuleSpec(routingFinalNatJump, inNatRuleKey, firewall.GetInPair(testCase.InputPair).Source, firewall.GetInPair(testCase.InputPair).Destination)
-
-			exists, err = iptablesClient.Exists(tableNat, chainRTNAT, inNatRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableNat, chainRTNAT)
+			exists, err := iptablesClient.Exists(tableMangle, chainRTPRE, markingRule...)
+			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableMangle, chainRTPRE)
 			if testCase.InputPair.Masquerade {
-				require.True(t, exists, "income nat rule should be created")
-				foundNatRule, foundNat := manager.rules[inNatRuleKey]
-				require.True(t, foundNat, "income nat rule should exist in the map")
-				require.Equal(t, inNatRule[:4], foundNatRule[:4], "stored income nat rule should match")
+				require.True(t, exists, "marking rule should be created")
+				foundRule, found := manager.rules[natRuleKey]
+				require.True(t, found, "marking rule should exist in the map")
+				require.Equal(t, markingRule, foundRule, "stored marking rule should match")
 			} else {
-				require.False(t, exists, "nat rule should not be created")
-				_, foundNat := manager.rules[inNatRuleKey]
-				require.False(t, foundNat, "income nat rule should not exist in the map")
+				require.False(t, exists, "marking rule should not be created")
+				_, found := manager.rules[natRuleKey]
+				require.False(t, found, "marking rule should not exist in the map")
+			}
+
+			// Check inverse rule
+			inversePair := firewall.GetInversePair(testCase.InputPair)
+			inverseRuleKey := firewall.GenKey(firewall.NatFormat, inversePair)
+			inverseMarkingRule := []string{
+				"!", "-i", ifaceMock.Name(),
+				"-m", "conntrack",
+				"--ctstate", "NEW",
+				"-s", inversePair.Source.String(),
+				"-d", inversePair.Destination.String(),
+				"-j", "MARK", "--set-mark",
+				fmt.Sprintf("%#x", nbnet.PreroutingFwmarkMasqueradeReturn),
+			}
+
+			exists, err = iptablesClient.Exists(tableMangle, chainRTPRE, inverseMarkingRule...)
+			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableMangle, chainRTPRE)
+			if testCase.InputPair.Masquerade {
+				require.True(t, exists, "inverse marking rule should be created")
+				foundRule, found := manager.rules[inverseRuleKey]
+				require.True(t, found, "inverse marking rule should exist in the map")
+				require.Equal(t, inverseMarkingRule, foundRule, "stored inverse marking rule should match")
+			} else {
+				require.False(t, exists, "inverse marking rule should not be created")
+				_, found := manager.rules[inverseRuleKey]
+				require.False(t, found, "inverse marking rule should not exist in the map")
 			}
 		})
 	}
 }
 
-func TestIptablesManager_RemoveRoutingRules(t *testing.T) {
-
+func TestIptablesManager_RemoveNatRule(t *testing.T) {
 	if !isIptablesSupported() {
 		t.SkipNow()
 	}
@@ -158,72 +151,226 @@ func TestIptablesManager_RemoveRoutingRules(t *testing.T) {
 		t.Run(testCase.Name, func(t *testing.T) {
 			iptablesClient, _ := iptables.NewWithProtocol(iptables.ProtocolIPv4)
 
-			manager, err := newRouterManager(context.TODO(), iptablesClient)
+			manager, err := newRouter(iptablesClient, ifaceMock)
 			require.NoError(t, err, "shouldn't return error")
+			require.NoError(t, manager.init(nil))
 			defer func() {
-				_ = manager.Reset()
+				assert.NoError(t, manager.Reset(), "shouldn't return error")
 			}()
 
+			err = manager.AddNatRule(testCase.InputPair)
+			require.NoError(t, err, "should add NAT rule without error")
+
+			err = manager.RemoveNatRule(testCase.InputPair)
 			require.NoError(t, err, "shouldn't return error")
 
-			forwardRuleKey := firewall.GenKey(firewall.ForwardingFormat, testCase.InputPair.ID)
-			forwardRule := genRuleSpec(routingFinalForwardJump, forwardRuleKey, testCase.InputPair.Source, testCase.InputPair.Destination)
+			natRuleKey := firewall.GenKey(firewall.NatFormat, testCase.InputPair)
+			markingRule := []string{
+				"-i", ifaceMock.Name(),
+				"-m", "conntrack",
+				"--ctstate", "NEW",
+				"-s", testCase.InputPair.Source.String(),
+				"-d", testCase.InputPair.Destination.String(),
+				"-j", "MARK", "--set-mark",
+				fmt.Sprintf("%#x", nbnet.PreroutingFwmarkMasquerade),
+			}
 
-			err = iptablesClient.Insert(tableFilter, chainRTFWD, 1, forwardRule...)
-			require.NoError(t, err, "inserting rule should not return error")
+			exists, err := iptablesClient.Exists(tableMangle, chainRTPRE, markingRule...)
+			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableMangle, chainRTPRE)
+			require.False(t, exists, "marking rule should not exist")
 
-			inForwardRuleKey := firewall.GenKey(firewall.InForwardingFormat, testCase.InputPair.ID)
-			inForwardRule := genRuleSpec(routingFinalForwardJump, inForwardRuleKey, firewall.GetInPair(testCase.InputPair).Source, firewall.GetInPair(testCase.InputPair).Destination)
+			_, found := manager.rules[natRuleKey]
+			require.False(t, found, "marking rule should not exist in the manager map")
 
-			err = iptablesClient.Insert(tableFilter, chainRTFWD, 1, inForwardRule...)
-			require.NoError(t, err, "inserting rule should not return error")
+			// Check inverse rule removal
+			inversePair := firewall.GetInversePair(testCase.InputPair)
+			inverseRuleKey := firewall.GenKey(firewall.NatFormat, inversePair)
+			inverseMarkingRule := []string{
+				"!", "-i", ifaceMock.Name(),
+				"-m", "conntrack",
+				"--ctstate", "NEW",
+				"-s", inversePair.Source.String(),
+				"-d", inversePair.Destination.String(),
+				"-j", "MARK", "--set-mark",
+				fmt.Sprintf("%#x", nbnet.PreroutingFwmarkMasqueradeReturn),
+			}
 
-			natRuleKey := firewall.GenKey(firewall.NatFormat, testCase.InputPair.ID)
-			natRule := genRuleSpec(routingFinalNatJump, natRuleKey, testCase.InputPair.Source, testCase.InputPair.Destination)
-
-			err = iptablesClient.Insert(tableNat, chainRTNAT, 1, natRule...)
-			require.NoError(t, err, "inserting rule should not return error")
-
-			inNatRuleKey := firewall.GenKey(firewall.InNatFormat, testCase.InputPair.ID)
-			inNatRule := genRuleSpec(routingFinalNatJump, inNatRuleKey, firewall.GetInPair(testCase.InputPair).Source, firewall.GetInPair(testCase.InputPair).Destination)
-
-			err = iptablesClient.Insert(tableNat, chainRTNAT, 1, inNatRule...)
-			require.NoError(t, err, "inserting rule should not return error")
-
-			err = manager.Reset()
-			require.NoError(t, err, "shouldn't return error")
-
-			err = manager.RemoveRoutingRules(testCase.InputPair)
-			require.NoError(t, err, "shouldn't return error")
-
-			exists, err := iptablesClient.Exists(tableFilter, chainRTFWD, forwardRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableFilter, chainRTFWD)
-			require.False(t, exists, "forwarding rule should not exist")
-
-			_, found := manager.rules[forwardRuleKey]
-			require.False(t, found, "forwarding rule should exist in the manager map")
-
-			exists, err = iptablesClient.Exists(tableFilter, chainRTFWD, inForwardRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableFilter, chainRTFWD)
-			require.False(t, exists, "income forwarding rule should not exist")
-
-			_, found = manager.rules[inForwardRuleKey]
-			require.False(t, found, "income forwarding rule should exist in the manager map")
-
-			exists, err = iptablesClient.Exists(tableNat, chainRTNAT, natRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableNat, chainRTNAT)
-			require.False(t, exists, "nat rule should not exist")
-
-			_, found = manager.rules[natRuleKey]
-			require.False(t, found, "nat rule should exist in the manager map")
-
-			exists, err = iptablesClient.Exists(tableNat, chainRTNAT, inNatRule...)
-			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableNat, chainRTNAT)
-			require.False(t, exists, "income nat rule should not exist")
-
-			_, found = manager.rules[inNatRuleKey]
-			require.False(t, found, "income nat rule should exist in the manager map")
+			exists, err = iptablesClient.Exists(tableMangle, chainRTPRE, inverseMarkingRule...)
+			require.NoError(t, err, "should be able to query the iptables %s table and %s chain", tableMangle, chainRTPRE)
+			require.False(t, exists, "inverse marking rule should not exist")
 
+			_, found = manager.rules[inverseRuleKey]
+			require.False(t, found, "inverse marking rule should not exist in the map")
+		})
+	}
+}
+
+func TestRouter_AddRouteFiltering(t *testing.T) {
+	if !isIptablesSupported() {
+		t.Skip("iptables not supported on this system")
+	}
+
+	iptablesClient, err := iptables.NewWithProtocol(iptables.ProtocolIPv4)
+	require.NoError(t, err, "Failed to create iptables client")
+
+	r, err := newRouter(iptablesClient, ifaceMock)
+	require.NoError(t, err, "Failed to create router manager")
+	require.NoError(t, r.init(nil))
+
+	defer func() {
+		err := r.Reset()
+		require.NoError(t, err, "Failed to reset router")
+	}()
+
+	tests := []struct {
+		name        string
+		sources     []netip.Prefix
+		destination netip.Prefix
+		proto       firewall.Protocol
+		sPort       *firewall.Port
+		dPort       *firewall.Port
+		direction   firewall.RuleDirection
+		action      firewall.Action
+		expectSet   bool
+	}{
+		{
+			name:        "Basic TCP rule with single source",
+			sources:     []netip.Prefix{netip.MustParsePrefix("192.168.1.0/24")},
+			destination: netip.MustParsePrefix("10.0.0.0/24"),
+			proto:       firewall.ProtocolTCP,
+			sPort:       nil,
+			dPort:       &firewall.Port{Values: []uint16{80}},
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name: "UDP rule with multiple sources",
+			sources: []netip.Prefix{
+				netip.MustParsePrefix("172.16.0.0/16"),
+				netip.MustParsePrefix("192.168.0.0/16"),
+			},
+			destination: netip.MustParsePrefix("10.0.0.0/8"),
+			proto:       firewall.ProtocolUDP,
+			sPort:       &firewall.Port{Values: []uint16{1024, 2048}, IsRange: true},
+			dPort:       nil,
+			direction:   firewall.RuleDirectionOUT,
+			action:      firewall.ActionDrop,
+			expectSet:   true,
+		},
+		{
+			name:        "All protocols rule",
+			sources:     []netip.Prefix{netip.MustParsePrefix("10.0.0.0/8")},
+			destination: netip.MustParsePrefix("0.0.0.0/0"),
+			proto:       firewall.ProtocolALL,
+			sPort:       nil,
+			dPort:       nil,
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "ICMP rule",
+			sources:     []netip.Prefix{netip.MustParsePrefix("192.168.0.0/16")},
+			destination: netip.MustParsePrefix("10.0.0.0/8"),
+			proto:       firewall.ProtocolICMP,
+			sPort:       nil,
+			dPort:       nil,
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "TCP rule with multiple source ports",
+			sources:     []netip.Prefix{netip.MustParsePrefix("172.16.0.0/12")},
+			destination: netip.MustParsePrefix("192.168.0.0/16"),
+			proto:       firewall.ProtocolTCP,
+			sPort:       &firewall.Port{Values: []uint16{80, 443, 8080}},
+			dPort:       nil,
+			direction:   firewall.RuleDirectionOUT,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "UDP rule with single IP and port range",
+			sources:     []netip.Prefix{netip.MustParsePrefix("192.168.1.1/32")},
+			destination: netip.MustParsePrefix("10.0.0.0/24"),
+			proto:       firewall.ProtocolUDP,
+			sPort:       nil,
+			dPort:       &firewall.Port{Values: []uint16{5000, 5100}, IsRange: true},
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionDrop,
+			expectSet:   false,
+		},
+		{
+			name:        "TCP rule with source and destination ports",
+			sources:     []netip.Prefix{netip.MustParsePrefix("10.0.0.0/24")},
+			destination: netip.MustParsePrefix("172.16.0.0/16"),
+			proto:       firewall.ProtocolTCP,
+			sPort:       &firewall.Port{Values: []uint16{1024, 65535}, IsRange: true},
+			dPort:       &firewall.Port{Values: []uint16{22}},
+			direction:   firewall.RuleDirectionOUT,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "Drop all incoming traffic",
+			sources:     []netip.Prefix{netip.MustParsePrefix("0.0.0.0/0")},
+			destination: netip.MustParsePrefix("192.168.0.0/24"),
+			proto:       firewall.ProtocolALL,
+			sPort:       nil,
+			dPort:       nil,
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionDrop,
+			expectSet:   false,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			ruleKey, err := r.AddRouteFiltering(tt.sources, tt.destination, tt.proto, tt.sPort, tt.dPort, tt.action)
+			require.NoError(t, err, "AddRouteFiltering failed")
+
+			// Check if the rule is in the internal map
+			rule, ok := r.rules[ruleKey.GetRuleID()]
+			assert.True(t, ok, "Rule not found in internal map")
+
+			// Log the internal rule
+			t.Logf("Internal rule: %v", rule)
+
+			// Check if the rule exists in iptables
+			exists, err := iptablesClient.Exists(tableFilter, chainRTFWD, rule...)
+			assert.NoError(t, err, "Failed to check rule existence")
+			assert.True(t, exists, "Rule not found in iptables")
+
+			// Verify rule content
+			params := routeFilteringRuleParams{
+				Sources:     tt.sources,
+				Destination: tt.destination,
+				Proto:       tt.proto,
+				SPort:       tt.sPort,
+				DPort:       tt.dPort,
+				Action:      tt.action,
+				SetName:     "",
+			}
+
+			expectedRule := genRouteFilteringRuleSpec(params)
+
+			if tt.expectSet {
+				setName := firewall.GenerateSetName(tt.sources)
+				params.SetName = setName
+				expectedRule = genRouteFilteringRuleSpec(params)
+
+				// Check if the set was created
+				_, exists := r.ipsetCounter.Get(setName)
+				assert.True(t, exists, "IPSet not created")
+			}
+
+			assert.Equal(t, expectedRule, rule, "Rule content mismatch")
+
+			// Clean up
+			err = r.DeleteRouteRule(ruleKey)
+			require.NoError(t, err, "Failed to delete rule")
 		})
 	}
 }

client/firewall/iptables/rule.go

@@ -5,9 +5,10 @@ type Rule struct {
 	ruleID    string
 	ipsetName string
 
-	specs []string
-	ip    string
-	chain string
+	specs       []string
+	mangleSpecs []string
+	ip          string
+	chain       string
 }
 
 // GetRuleID returns the rule id

client/firewall/iptables/rulestore_linux.go

@@ -1,14 +1,16 @@
 package iptables
 
+import "encoding/json"
+
 type ipList struct {
 	ips map[string]struct{}
 }
 
-func newIpList(ip string) ipList {
+func newIpList(ip string) *ipList {
 	ips := make(map[string]struct{})
 	ips[ip] = struct{}{}
 
-	return ipList{
+	return &ipList{
 		ips: ips,
 	}
 }
@@ -17,27 +19,52 @@ func (s *ipList) addIP(ip string) {
 	s.ips[ip] = struct{}{}
 }
 
+// MarshalJSON implements json.Marshaler
+func (s *ipList) MarshalJSON() ([]byte, error) {
+	return json.Marshal(struct {
+		IPs map[string]struct{} `json:"ips"`
+	}{
+		IPs: s.ips,
+	})
+}
+
+// UnmarshalJSON implements json.Unmarshaler
+func (s *ipList) UnmarshalJSON(data []byte) error {
+	temp := struct {
+		IPs map[string]struct{} `json:"ips"`
+	}{}
+	if err := json.Unmarshal(data, &temp); err != nil {
+		return err
+	}
+	s.ips = temp.IPs
+
+	if temp.IPs == nil {
+		temp.IPs = make(map[string]struct{})
+	}
+
+	return nil
+}
+
 type ipsetStore struct {
-	ipsets map[string]ipList // ipsetName -> ruleset
+	ipsets map[string]*ipList
 }
 
 func newIpsetStore() *ipsetStore {
 	return &ipsetStore{
-		ipsets: make(map[string]ipList),
+		ipsets: make(map[string]*ipList),
 	}
 }
 
-func (s *ipsetStore) ipset(ipsetName string) (ipList, bool) {
+func (s *ipsetStore) ipset(ipsetName string) (*ipList, bool) {
 	r, ok := s.ipsets[ipsetName]
 	return r, ok
 }
 
-func (s *ipsetStore) addIpList(ipsetName string, list ipList) {
+func (s *ipsetStore) addIpList(ipsetName string, list *ipList) {
 	s.ipsets[ipsetName] = list
 }
 
 func (s *ipsetStore) deleteIpset(ipsetName string) {
-	s.ipsets[ipsetName] = ipList{}
 	delete(s.ipsets, ipsetName)
 }
 
@@ -48,3 +75,29 @@ func (s *ipsetStore) ipsetNames() []string {
 	}
 	return names
 }
+
+// MarshalJSON implements json.Marshaler
+func (s *ipsetStore) MarshalJSON() ([]byte, error) {
+	return json.Marshal(struct {
+		IPSets map[string]*ipList `json:"ipsets"`
+	}{
+		IPSets: s.ipsets,
+	})
+}
+
+// UnmarshalJSON implements json.Unmarshaler
+func (s *ipsetStore) UnmarshalJSON(data []byte) error {
+	temp := struct {
+		IPSets map[string]*ipList `json:"ipsets"`
+	}{}
+	if err := json.Unmarshal(data, &temp); err != nil {
+		return err
+	}
+	s.ipsets = temp.IPSets
+
+	if temp.IPSets == nil {
+		temp.IPSets = make(map[string]*ipList)
+	}
+
+	return nil
+}

client/firewall/iptables/state_linux.go

@@ -0,0 +1,70 @@
+package iptables
+
+import (
+	"fmt"
+	"sync"
+
+	"github.com/netbirdio/netbird/client/iface"
+	"github.com/netbirdio/netbird/client/iface/device"
+)
+
+type InterfaceState struct {
+	NameStr       string          `json:"name"`
+	WGAddress     iface.WGAddress `json:"wg_address"`
+	UserspaceBind bool            `json:"userspace_bind"`
+}
+
+func (i *InterfaceState) Name() string {
+	return i.NameStr
+}
+
+func (i *InterfaceState) Address() device.WGAddress {
+	return i.WGAddress
+}
+
+func (i *InterfaceState) IsUserspaceBind() bool {
+	return i.UserspaceBind
+}
+
+type ShutdownState struct {
+	sync.Mutex
+
+	InterfaceState *InterfaceState `json:"interface_state,omitempty"`
+
+	RouteRules        routeRules    `json:"route_rules,omitempty"`
+	RouteIPsetCounter *ipsetCounter `json:"route_ipset_counter,omitempty"`
+
+	ACLEntries    aclEntries  `json:"acl_entries,omitempty"`
+	ACLIPsetStore *ipsetStore `json:"acl_ipset_store,omitempty"`
+}
+
+func (s *ShutdownState) Name() string {
+	return "iptables_state"
+}
+
+func (s *ShutdownState) Cleanup() error {
+	ipt, err := Create(s.InterfaceState)
+	if err != nil {
+		return fmt.Errorf("create iptables manager: %w", err)
+	}
+
+	if s.RouteRules != nil {
+		ipt.router.rules = s.RouteRules
+	}
+	if s.RouteIPsetCounter != nil {
+		ipt.router.ipsetCounter.LoadData(s.RouteIPsetCounter)
+	}
+
+	if s.ACLEntries != nil {
+		ipt.aclMgr.entries = s.ACLEntries
+	}
+	if s.ACLIPsetStore != nil {
+		ipt.aclMgr.ipsetStore = s.ACLIPsetStore
+	}
+
+	if err := ipt.Reset(nil); err != nil {
+		return fmt.Errorf("reset iptables manager: %w", err)
+	}
+
+	return nil
+}

client/firewall/manager/firewall.go

@@ -1,15 +1,24 @@
 package manager
 
 import (
+	"crypto/sha256"
+	"encoding/hex"
 	"fmt"
 	"net"
+	"net/netip"
+	"sort"
+	"strings"
+
+	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/internal/statemanager"
 )
 
 const (
-	NatFormat          = "netbird-nat-%s"
-	ForwardingFormat   = "netbird-fwd-%s"
-	InNatFormat        = "netbird-nat-in-%s"
-	InForwardingFormat = "netbird-fwd-in-%s"
+	ForwardingFormatPrefix = "netbird-fwd-"
+	ForwardingFormat       = "netbird-fwd-%s-%t"
+	PreroutingFormat       = "netbird-prerouting-%s-%t"
+	NatFormat              = "netbird-nat-%s-%t"
 )
 
 // Rule abstraction should be implemented by each firewall manager
@@ -46,43 +55,137 @@ const (
 // It declares methods which handle actions required by the
 // Netbird client for ACL and routing functionality
 type Manager interface {
+	Init(stateManager *statemanager.Manager) error
+
 	// AllowNetbird allows netbird interface traffic
 	AllowNetbird() error
 
-	// AddFiltering rule to the firewall
+	// AddPeerFiltering adds a rule to the firewall
 	//
 	// If comment argument is empty firewall manager should set
 	// rule ID as comment for the rule
-	AddFiltering(
+	AddPeerFiltering(
 		ip net.IP,
 		proto Protocol,
 		sPort *Port,
 		dPort *Port,
-		direction RuleDirection,
 		action Action,
 		ipsetName string,
 		comment string,
 	) ([]Rule, error)
 
-	// DeleteRule from the firewall by rule definition
-	DeleteRule(rule Rule) error
+	// DeletePeerRule from the firewall by rule definition
+	DeletePeerRule(rule Rule) error
 
 	// IsServerRouteSupported returns true if the firewall supports server side routing operations
 	IsServerRouteSupported() bool
 
-	// InsertRoutingRules inserts a routing firewall rule
-	InsertRoutingRules(pair RouterPair) error
+	AddRouteFiltering(source []netip.Prefix, destination netip.Prefix, proto Protocol, sPort *Port, dPort *Port, action Action) (Rule, error)
 
-	// RemoveRoutingRules removes a routing firewall rule
-	RemoveRoutingRules(pair RouterPair) error
+	// DeleteRouteRule deletes a routing rule
+	DeleteRouteRule(rule Rule) error
+
+	// AddNatRule inserts a routing NAT rule
+	AddNatRule(pair RouterPair) error
+
+	// RemoveNatRule removes a routing NAT rule
+	RemoveNatRule(pair RouterPair) error
+
+	// SetLegacyManagement sets the legacy management mode
+	SetLegacyManagement(legacy bool) error
 
 	// Reset firewall to the default state
-	Reset() error
+	Reset(stateManager *statemanager.Manager) error
 
 	// Flush the changes to firewall controller
 	Flush() error
+
+	SetLogLevel(log.Level)
 }
 
-func GenKey(format string, input string) string {
-	return fmt.Sprintf(format, input)
+func GenKey(format string, pair RouterPair) string {
+	return fmt.Sprintf(format, pair.ID, pair.Inverse)
+}
+
+// LegacyManager defines the interface for legacy management operations
+type LegacyManager interface {
+	RemoveAllLegacyRouteRules() error
+	GetLegacyManagement() bool
+	SetLegacyManagement(bool)
+}
+
+// SetLegacyManagement sets the route manager to use legacy management
+func SetLegacyManagement(router LegacyManager, isLegacy bool) error {
+	oldLegacy := router.GetLegacyManagement()
+
+	if oldLegacy != isLegacy {
+		router.SetLegacyManagement(isLegacy)
+		log.Debugf("Set legacy management to %v", isLegacy)
+	}
+
+	// client reconnected to a newer mgmt, we need to clean up the legacy rules
+	if !isLegacy && oldLegacy {
+		if err := router.RemoveAllLegacyRouteRules(); err != nil {
+			return fmt.Errorf("remove legacy routing rules: %v", err)
+		}
+
+		log.Debugf("Legacy routing rules removed")
+	}
+
+	return nil
+}
+
+// GenerateSetName generates a unique name for an ipset based on the given sources.
+func GenerateSetName(sources []netip.Prefix) string {
+	// sort for consistent naming
+	SortPrefixes(sources)
+
+	var sourcesStr strings.Builder
+	for _, src := range sources {
+		sourcesStr.WriteString(src.String())
+	}
+
+	hash := sha256.Sum256([]byte(sourcesStr.String()))
+	shortHash := hex.EncodeToString(hash[:])[:8]
+
+	return fmt.Sprintf("nb-%s", shortHash)
+}
+
+// MergeIPRanges merges overlapping IP ranges and returns a slice of non-overlapping netip.Prefix
+func MergeIPRanges(prefixes []netip.Prefix) []netip.Prefix {
+	if len(prefixes) == 0 {
+		return prefixes
+	}
+
+	merged := []netip.Prefix{prefixes[0]}
+	for _, prefix := range prefixes[1:] {
+		last := merged[len(merged)-1]
+		if last.Contains(prefix.Addr()) {
+			// If the current prefix is contained within the last merged prefix, skip it
+			continue
+		}
+		if prefix.Contains(last.Addr()) {
+			// If the current prefix contains the last merged prefix, replace it
+			merged[len(merged)-1] = prefix
+		} else {
+			// Otherwise, add the current prefix to the merged list
+			merged = append(merged, prefix)
+		}
+	}
+
+	return merged
+}
+
+// SortPrefixes sorts the given slice of netip.Prefix in place.
+// It sorts first by IP address, then by prefix length (most specific to least specific).
+func SortPrefixes(prefixes []netip.Prefix) {
+	sort.Slice(prefixes, func(i, j int) bool {
+		addrCmp := prefixes[i].Addr().Compare(prefixes[j].Addr())
+		if addrCmp != 0 {
+			return addrCmp < 0
+		}
+
+		// If IP addresses are the same, compare prefix lengths (longer prefixes first)
+		return prefixes[i].Bits() > prefixes[j].Bits()
+	})
 }

client/firewall/manager/firewall_test.go

@@ -0,0 +1,192 @@
+package manager_test
+
+import (
+	"net/netip"
+	"reflect"
+	"regexp"
+	"testing"
+
+	"github.com/netbirdio/netbird/client/firewall/manager"
+)
+
+func TestGenerateSetName(t *testing.T) {
+	t.Run("Different orders result in same hash", func(t *testing.T) {
+		prefixes1 := []netip.Prefix{
+			netip.MustParsePrefix("192.168.1.0/24"),
+			netip.MustParsePrefix("10.0.0.0/8"),
+		}
+		prefixes2 := []netip.Prefix{
+			netip.MustParsePrefix("10.0.0.0/8"),
+			netip.MustParsePrefix("192.168.1.0/24"),
+		}
+
+		result1 := manager.GenerateSetName(prefixes1)
+		result2 := manager.GenerateSetName(prefixes2)
+
+		if result1 != result2 {
+			t.Errorf("Different orders produced different hashes: %s != %s", result1, result2)
+		}
+	})
+
+	t.Run("Result format is correct", func(t *testing.T) {
+		prefixes := []netip.Prefix{
+			netip.MustParsePrefix("192.168.1.0/24"),
+			netip.MustParsePrefix("10.0.0.0/8"),
+		}
+
+		result := manager.GenerateSetName(prefixes)
+
+		matched, err := regexp.MatchString(`^nb-[0-9a-f]{8}$`, result)
+		if err != nil {
+			t.Fatalf("Error matching regex: %v", err)
+		}
+		if !matched {
+			t.Errorf("Result format is incorrect: %s", result)
+		}
+	})
+
+	t.Run("Empty input produces consistent result", func(t *testing.T) {
+		result1 := manager.GenerateSetName([]netip.Prefix{})
+		result2 := manager.GenerateSetName([]netip.Prefix{})
+
+		if result1 != result2 {
+			t.Errorf("Empty input produced inconsistent results: %s != %s", result1, result2)
+		}
+	})
+
+	t.Run("IPv4 and IPv6 mixing", func(t *testing.T) {
+		prefixes1 := []netip.Prefix{
+			netip.MustParsePrefix("192.168.1.0/24"),
+			netip.MustParsePrefix("2001:db8::/32"),
+		}
+		prefixes2 := []netip.Prefix{
+			netip.MustParsePrefix("2001:db8::/32"),
+			netip.MustParsePrefix("192.168.1.0/24"),
+		}
+
+		result1 := manager.GenerateSetName(prefixes1)
+		result2 := manager.GenerateSetName(prefixes2)
+
+		if result1 != result2 {
+			t.Errorf("Different orders of IPv4 and IPv6 produced different hashes: %s != %s", result1, result2)
+		}
+	})
+}
+
+func TestMergeIPRanges(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    []netip.Prefix
+		expected []netip.Prefix
+	}{
+		{
+			name:     "Empty input",
+			input:    []netip.Prefix{},
+			expected: []netip.Prefix{},
+		},
+		{
+			name: "Single range",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+			},
+		},
+		{
+			name: "Two non-overlapping ranges",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("10.0.0.0/8"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("10.0.0.0/8"),
+			},
+		},
+		{
+			name: "One range containing another",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.0.0/16"),
+				netip.MustParsePrefix("192.168.1.0/24"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.0.0/16"),
+			},
+		},
+		{
+			name: "One range containing another (different order)",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("192.168.0.0/16"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.0.0/16"),
+			},
+		},
+		{
+			name: "Overlapping ranges",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("192.168.1.128/25"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+			},
+		},
+		{
+			name: "Overlapping ranges (different order)",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.128/25"),
+				netip.MustParsePrefix("192.168.1.0/24"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.0/24"),
+			},
+		},
+		{
+			name: "Multiple overlapping ranges",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.0.0/16"),
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("192.168.2.0/24"),
+				netip.MustParsePrefix("192.168.1.128/25"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.0.0/16"),
+			},
+		},
+		{
+			name: "Partially overlapping ranges",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("192.168.0.0/23"),
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("192.168.2.0/25"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("192.168.0.0/23"),
+				netip.MustParsePrefix("192.168.2.0/25"),
+			},
+		},
+		{
+			name: "IPv6 ranges",
+			input: []netip.Prefix{
+				netip.MustParsePrefix("2001:db8::/32"),
+				netip.MustParsePrefix("2001:db8:1::/48"),
+				netip.MustParsePrefix("2001:db8:2::/48"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("2001:db8::/32"),
+			},
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := manager.MergeIPRanges(tt.input)
+			if !reflect.DeepEqual(result, tt.expected) {
+				t.Errorf("MergeIPRanges() = %v, want %v", result, tt.expected)
+			}
+		})
+	}
+}

client/firewall/manager/port.go

@@ -30,7 +30,7 @@ type Port struct {
 	IsRange bool
 
 	// Values contains one value for single port, multiple values for the list of ports, or two values for the range of ports
-	Values []int
+	Values []uint16
 }
 
 // String interface implementation
@@ -40,7 +40,11 @@ func (p *Port) String() string {
 		if ports != "" {
 			ports += ","
 		}
-		ports += strconv.Itoa(port)
+		ports += strconv.Itoa(int(port))
 	}
+	if p.IsRange {
+		ports = "range:" + ports
+	}
+
 	return ports
 }

client/firewall/manager/routerpair.go

@@ -1,18 +1,26 @@
 package manager
 
+import (
+	"net/netip"
+
+	"github.com/netbirdio/netbird/route"
+)
+
 type RouterPair struct {
-	ID          string
-	Source      string
-	Destination string
+	ID          route.ID
+	Source      netip.Prefix
+	Destination netip.Prefix
 	Masquerade  bool
+	Inverse     bool
 }
 
-func GetInPair(pair RouterPair) RouterPair {
+func GetInversePair(pair RouterPair) RouterPair {
 	return RouterPair{
 		ID: pair.ID,
 		// invert Source/Destination
 		Source:      pair.Destination,
 		Destination: pair.Source,
 		Masquerade:  pair.Masquerade,
+		Inverse:     true,
 	}
 }

client/firewall/nftables/manager_linux.go

@@ -5,20 +5,34 @@ import (
 	"context"
 	"fmt"
 	"net"
+	"net/netip"
 	"sync"
 
 	"github.com/google/nftables"
+	"github.com/google/nftables/binaryutil"
 	"github.com/google/nftables/expr"
 	log "github.com/sirupsen/logrus"
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
+	"github.com/netbirdio/netbird/client/iface"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
 )
 
 const (
-	// tableName is the name of the table that is used for filtering by the Netbird client
-	tableName = "netbird"
+	// tableNameNetbird is the name of the table that is used for filtering by the Netbird client
+	tableNameNetbird = "netbird"
+
+	tableNameFilter = "filter"
+	chainNameInput  = "INPUT"
 )
 
+// iFaceMapper defines subset methods of interface required for manager
+type iFaceMapper interface {
+	Name() string
+	Address() iface.WGAddress
+	IsUserspaceBind() bool
+}
+
 // Manager of iptables firewall
 type Manager struct {
 	mutex   sync.Mutex
@@ -30,40 +44,79 @@ type Manager struct {
 }
 
 // Create nftables firewall manager
-func Create(context context.Context, wgIface iFaceMapper) (*Manager, error) {
+func Create(wgIface iFaceMapper) (*Manager, error) {
 	m := &Manager{
 		rConn:   &nftables.Conn{},
 		wgIface: wgIface,
 	}
 
-	workTable, err := m.createWorkTable()
+	workTable := &nftables.Table{Name: tableNameNetbird, Family: nftables.TableFamilyIPv4}
+
+	var err error
+	m.router, err = newRouter(workTable, wgIface)
 	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("create router: %w", err)
 	}
 
-	m.router, err = newRouter(context, workTable)
+	m.aclManager, err = newAclManager(workTable, wgIface, chainNameRoutingFw)
 	if err != nil {
-		return nil, err
-	}
-
-	m.aclManager, err = newAclManager(workTable, wgIface, m.router.RouteingFwChainName())
-	if err != nil {
-		return nil, err
+		return nil, fmt.Errorf("create acl manager: %w", err)
 	}
 
 	return m, nil
 }
 
-// AddFiltering rule to the firewall
+// Init nftables firewall manager
+func (m *Manager) Init(stateManager *statemanager.Manager) error {
+	workTable, err := m.createWorkTable()
+	if err != nil {
+		return fmt.Errorf("create work table: %w", err)
+	}
+
+	if err := m.router.init(workTable); err != nil {
+		return fmt.Errorf("router init: %w", err)
+	}
+
+	if err := m.aclManager.init(workTable); err != nil {
+		// TODO: cleanup router
+		return fmt.Errorf("acl manager init: %w", err)
+	}
+
+	stateManager.RegisterState(&ShutdownState{})
+
+	// We only need to record minimal interface state for potential recreation.
+	// Unlike iptables, which requires tracking individual rules, nftables maintains
+	// a known state (our netbird table plus a few static rules). This allows for easy
+	// cleanup using Reset() without needing to store specific rules.
+	if err := stateManager.UpdateState(&ShutdownState{
+		InterfaceState: &InterfaceState{
+			NameStr:       m.wgIface.Name(),
+			WGAddress:     m.wgIface.Address(),
+			UserspaceBind: m.wgIface.IsUserspaceBind(),
+		},
+	}); err != nil {
+		log.Errorf("failed to update state: %v", err)
+	}
+
+	// persist early
+	go func() {
+		if err := stateManager.PersistState(context.Background()); err != nil {
+			log.Errorf("failed to persist state: %v", err)
+		}
+	}()
+
+	return nil
+}
+
+// AddPeerFiltering rule to the firewall
 //
 // If comment argument is empty firewall manager should set
 // rule ID as comment for the rule
-func (m *Manager) AddFiltering(
+func (m *Manager) AddPeerFiltering(
 	ip net.IP,
 	proto firewall.Protocol,
 	sPort *firewall.Port,
 	dPort *firewall.Port,
-	direction firewall.RuleDirection,
 	action firewall.Action,
 	ipsetName string,
 	comment string,
@@ -76,33 +129,59 @@ func (m *Manager) AddFiltering(
 		return nil, fmt.Errorf("unsupported IP version: %s", ip.String())
 	}
 
-	return m.aclManager.AddFiltering(ip, proto, sPort, dPort, direction, action, ipsetName, comment)
+	return m.aclManager.AddPeerFiltering(ip, proto, sPort, dPort, action, ipsetName, comment)
 }
 
-// DeleteRule from the firewall by rule definition
-func (m *Manager) DeleteRule(rule firewall.Rule) error {
+func (m *Manager) AddRouteFiltering(
+	sources []netip.Prefix,
+	destination netip.Prefix,
+	proto firewall.Protocol,
+	sPort *firewall.Port,
+	dPort *firewall.Port,
+	action firewall.Action,
+) (firewall.Rule, error) {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	return m.aclManager.DeleteRule(rule)
+	if !destination.Addr().Is4() {
+		return nil, fmt.Errorf("unsupported IP version: %s", destination.Addr().String())
+	}
+
+	return m.router.AddRouteFiltering(sources, destination, proto, sPort, dPort, action)
+}
+
+// DeletePeerRule from the firewall by rule definition
+func (m *Manager) DeletePeerRule(rule firewall.Rule) error {
+	m.mutex.Lock()
+	defer m.mutex.Unlock()
+
+	return m.aclManager.DeletePeerRule(rule)
+}
+
+// DeleteRouteRule deletes a routing rule
+func (m *Manager) DeleteRouteRule(rule firewall.Rule) error {
+	m.mutex.Lock()
+	defer m.mutex.Unlock()
+
+	return m.router.DeleteRouteRule(rule)
 }
 
 func (m *Manager) IsServerRouteSupported() bool {
 	return true
 }
 
-func (m *Manager) InsertRoutingRules(pair firewall.RouterPair) error {
+func (m *Manager) AddNatRule(pair firewall.RouterPair) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	return m.router.InsertRoutingRules(pair)
+	return m.router.AddNatRule(pair)
 }
 
-func (m *Manager) RemoveRoutingRules(pair firewall.RouterPair) error {
+func (m *Manager) RemoveNatRule(pair firewall.RouterPair) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	return m.router.RemoveRoutingRules(pair)
+	return m.router.RemoveNatRule(pair)
 }
 
 // AllowNetbird allows netbird interface traffic
@@ -126,7 +205,7 @@ func (m *Manager) AllowNetbird() error {
 
 	var chain *nftables.Chain
 	for _, c := range chains {
-		if c.Table.Name == "filter" && c.Name == "INPUT" {
+		if c.Table.Name == tableNameFilter && c.Name == chainNameInput {
 			chain = c
 			break
 		}
@@ -157,47 +236,91 @@ func (m *Manager) AllowNetbird() error {
 	return nil
 }
 
+// SetLegacyManagement sets the route manager to use legacy management
+func (m *Manager) SetLegacyManagement(isLegacy bool) error {
+	return firewall.SetLegacyManagement(m.router, isLegacy)
+}
+
 // Reset firewall to the default state
-func (m *Manager) Reset() error {
+func (m *Manager) Reset(stateManager *statemanager.Manager) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
-	chains, err := m.rConn.ListChains()
-	if err != nil {
-		return fmt.Errorf("list of chains: %w", err)
+	if err := m.resetNetbirdInputRules(); err != nil {
+		return fmt.Errorf("reset netbird input rules: %v", err)
 	}
 
+	if err := m.router.Reset(); err != nil {
+		return fmt.Errorf("reset router: %v", err)
+	}
+
+	if err := m.cleanupNetbirdTables(); err != nil {
+		return fmt.Errorf("cleanup netbird tables: %v", err)
+	}
+
+	if err := m.rConn.Flush(); err != nil {
+		return fmt.Errorf(flushError, err)
+	}
+
+	if err := stateManager.DeleteState(&ShutdownState{}); err != nil {
+		return fmt.Errorf("delete state: %v", err)
+	}
+
+	return nil
+}
+
+func (m *Manager) resetNetbirdInputRules() error {
+	chains, err := m.rConn.ListChains()
+	if err != nil {
+		return fmt.Errorf("list chains: %w", err)
+	}
+
+	m.deleteNetbirdInputRules(chains)
+
+	return nil
+}
+
+func (m *Manager) deleteNetbirdInputRules(chains []*nftables.Chain) {
 	for _, c := range chains {
-		// delete Netbird allow input traffic rule if it exists
-		if c.Table.Name == "filter" && c.Name == "INPUT" {
+		if c.Table.Name == tableNameFilter && c.Name == chainNameInput {
 			rules, err := m.rConn.GetRules(c.Table, c)
 			if err != nil {
 				log.Errorf("get rules for chain %q: %v", c.Name, err)
 				continue
 			}
-			for _, r := range rules {
-				if bytes.Equal(r.UserData, []byte(allowNetbirdInputRuleID)) {
-					if err := m.rConn.DelRule(r); err != nil {
-						log.Errorf("delete rule: %v", err)
-					}
-				}
+
+			m.deleteMatchingRules(rules)
+		}
+	}
+}
+
+func (m *Manager) deleteMatchingRules(rules []*nftables.Rule) {
+	for _, r := range rules {
+		if bytes.Equal(r.UserData, []byte(allowNetbirdInputRuleID)) {
+			if err := m.rConn.DelRule(r); err != nil {
+				log.Errorf("delete rule: %v", err)
 			}
 		}
 	}
+}
 
-	m.router.ResetForwardRules()
-
+func (m *Manager) cleanupNetbirdTables() error {
 	tables, err := m.rConn.ListTables()
 	if err != nil {
-		return fmt.Errorf("list of tables: %w", err)
+		return fmt.Errorf("list tables: %w", err)
 	}
+
 	for _, t := range tables {
-		if t.Name == tableName {
+		if t.Name == tableNameNetbird {
 			m.rConn.DelTable(t)
 		}
 	}
+	return nil
+}
 
-	return m.rConn.Flush()
+// SetLogLevel sets the log level for the firewall manager
+func (m *Manager) SetLogLevel(log.Level) {
+	// not supported
 }
 
 // Flush rule/chain/set operations from the buffer
@@ -218,12 +341,12 @@ func (m *Manager) createWorkTable() (*nftables.Table, error) {
 	}
 
 	for _, t := range tables {
-		if t.Name == tableName {
+		if t.Name == tableNameNetbird {
 			m.rConn.DelTable(t)
 		}
 	}
 
-	table := m.rConn.AddTable(&nftables.Table{Name: tableName, Family: nftables.TableFamilyIPv4})
+	table := m.rConn.AddTable(&nftables.Table{Name: tableNameNetbird, Family: nftables.TableFamilyIPv4})
 	err = m.rConn.Flush()
 	return table, err
 }
@@ -251,7 +374,7 @@ func (m *Manager) applyAllowNetbirdRules(chain *nftables.Chain) {
 func (m *Manager) detectAllowNetbirdRule(existedRules []*nftables.Rule) *nftables.Rule {
 	ifName := ifname(m.wgIface.Name())
 	for _, rule := range existedRules {
-		if rule.Table.Name == "filter" && rule.Chain.Name == "INPUT" {
+		if rule.Table.Name == tableNameFilter && rule.Chain.Name == chainNameInput {
 			if len(rule.Exprs) < 4 {
 				if e, ok := rule.Exprs[0].(*expr.Meta); !ok || e.Key != expr.MetaKeyIIFNAME {
 					continue
@@ -265,3 +388,38 @@ func (m *Manager) detectAllowNetbirdRule(existedRules []*nftables.Rule) *nftable
 	}
 	return nil
 }
+
+func insertReturnTrafficRule(conn *nftables.Conn, table *nftables.Table, chain *nftables.Chain) {
+	rule := &nftables.Rule{
+		Table: table,
+		Chain: chain,
+		Exprs: getEstablishedExprs(1),
+	}
+
+	conn.InsertRule(rule)
+}
+
+func getEstablishedExprs(register uint32) []expr.Any {
+	return []expr.Any{
+		&expr.Ct{
+			Key:      expr.CtKeySTATE,
+			Register: register,
+		},
+		&expr.Bitwise{
+			SourceRegister: register,
+			DestRegister:   register,
+			Len:            4,
+			Mask:           binaryutil.NativeEndian.PutUint32(expr.CtStateBitESTABLISHED | expr.CtStateBitRELATED),
+			Xor:            binaryutil.NativeEndian.PutUint32(0),
+		},
+		&expr.Cmp{
+			Op:       expr.CmpOpNeq,
+			Register: register,
+			Data:     []byte{0, 0, 0, 0},
+		},
+		&expr.Counter{},
+		&expr.Verdict{
+			Kind: expr.VerdictAccept,
+		},
+	}
+}

client/firewall/nftables/manager_linux_test.go

@@ -1,22 +1,39 @@
 package nftables
 
 import (
-	"context"
+	"bytes"
 	"fmt"
 	"net"
 	"net/netip"
+	"os/exec"
 	"testing"
 	"time"
 
 	"github.com/google/nftables"
+	"github.com/google/nftables/binaryutil"
 	"github.com/google/nftables/expr"
 	"github.com/stretchr/testify/require"
 	"golang.org/x/sys/unix"
 
 	fw "github.com/netbirdio/netbird/client/firewall/manager"
-	"github.com/netbirdio/netbird/iface"
+	"github.com/netbirdio/netbird/client/iface"
 )
 
+var ifaceMock = &iFaceMock{
+	NameFunc: func() string {
+		return "lo"
+	},
+	AddressFunc: func() iface.WGAddress {
+		return iface.WGAddress{
+			IP: net.ParseIP("100.96.0.1"),
+			Network: &net.IPNet{
+				IP:   net.ParseIP("100.96.0.0"),
+				Mask: net.IPv4Mask(255, 255, 255, 0),
+			},
+		}
+	},
+}
+
 // iFaceMapper defines subset methods of interface required for manager
 type iFaceMock struct {
 	NameFunc    func() string
@@ -40,28 +57,15 @@ func (i *iFaceMock) Address() iface.WGAddress {
 func (i *iFaceMock) IsUserspaceBind() bool { return false }
 
 func TestNftablesManager(t *testing.T) {
-	mock := &iFaceMock{
-		NameFunc: func() string {
-			return "lo"
-		},
-		AddressFunc: func() iface.WGAddress {
-			return iface.WGAddress{
-				IP: net.ParseIP("100.96.0.1"),
-				Network: &net.IPNet{
-					IP:   net.ParseIP("100.96.0.0"),
-					Mask: net.IPv4Mask(255, 255, 255, 0),
-				},
-			}
-		},
-	}
 
 	// just check on the local interface
-	manager, err := Create(context.Background(), mock)
+	manager, err := Create(ifaceMock)
 	require.NoError(t, err)
+	require.NoError(t, manager.Init(nil))
 	time.Sleep(time.Second * 3)
 
 	defer func() {
-		err = manager.Reset()
+		err = manager.Reset(nil)
 		require.NoError(t, err, "failed to reset")
 		time.Sleep(time.Second)
 	}()
@@ -70,16 +74,7 @@ func TestNftablesManager(t *testing.T) {
 
 	testClient := &nftables.Conn{}
 
-	rule, err := manager.AddFiltering(
-		ip,
-		fw.ProtocolTCP,
-		nil,
-		&fw.Port{Values: []int{53}},
-		fw.RuleDirectionIN,
-		fw.ActionDrop,
-		"",
-		"",
-	)
+	rule, err := manager.AddPeerFiltering(ip, fw.ProtocolTCP, nil, &fw.Port{Values: []uint16{53}}, fw.ActionDrop, "", "")
 	require.NoError(t, err, "failed to add rule")
 
 	err = manager.Flush()
@@ -88,17 +83,35 @@ func TestNftablesManager(t *testing.T) {
 	rules, err := testClient.GetRules(manager.aclManager.workTable, manager.aclManager.chainInputRules)
 	require.NoError(t, err, "failed to get rules")
 
-	require.Len(t, rules, 1, "expected 1 rules")
+	require.Len(t, rules, 2, "expected 2 rules")
+
+	expectedExprs1 := []expr.Any{
+		&expr.Ct{
+			Key:      expr.CtKeySTATE,
+			Register: 1,
+		},
+		&expr.Bitwise{
+			SourceRegister: 1,
+			DestRegister:   1,
+			Len:            4,
+			Mask:           binaryutil.NativeEndian.PutUint32(expr.CtStateBitESTABLISHED | expr.CtStateBitRELATED),
+			Xor:            binaryutil.NativeEndian.PutUint32(0),
+		},
+		&expr.Cmp{
+			Op:       expr.CmpOpNeq,
+			Register: 1,
+			Data:     []byte{0, 0, 0, 0},
+		},
+		&expr.Counter{},
+		&expr.Verdict{
+			Kind: expr.VerdictAccept,
+		},
+	}
+	compareExprsIgnoringCounters(t, rules[0].Exprs, expectedExprs1)
 
 	ipToAdd, _ := netip.AddrFromSlice(ip)
 	add := ipToAdd.Unmap()
-	expectedExprs := []expr.Any{
-		&expr.Meta{Key: expr.MetaKeyIIFNAME, Register: 1},
-		&expr.Cmp{
-			Op:       expr.CmpOpEq,
-			Register: 1,
-			Data:     ifname("lo"),
-		},
+	expectedExprs2 := []expr.Any{
 		&expr.Payload{
 			DestRegister: 1,
 			Base:         expr.PayloadBaseNetworkHeader,
@@ -134,10 +147,10 @@ func TestNftablesManager(t *testing.T) {
 		},
 		&expr.Verdict{Kind: expr.VerdictDrop},
 	}
-	require.ElementsMatch(t, rules[0].Exprs, expectedExprs, "expected the same expressions")
+	require.ElementsMatch(t, rules[1].Exprs, expectedExprs2, "expected the same expressions")
 
 	for _, r := range rule {
-		err = manager.DeleteRule(r)
+		err = manager.DeletePeerRule(r)
 		require.NoError(t, err, "failed to delete rule")
 	}
 
@@ -146,9 +159,10 @@ func TestNftablesManager(t *testing.T) {
 
 	rules, err = testClient.GetRules(manager.aclManager.workTable, manager.aclManager.chainInputRules)
 	require.NoError(t, err, "failed to get rules")
-	require.Len(t, rules, 0, "expected 0 rules after deletion")
+	// established rule remains
+	require.Len(t, rules, 1, "expected 1 rules after deletion")
 
-	err = manager.Reset()
+	err = manager.Reset(nil)
 	require.NoError(t, err, "failed to reset")
 }
 
@@ -171,12 +185,13 @@ func TestNFtablesCreatePerformance(t *testing.T) {
 	for _, testMax := range []int{10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000} {
 		t.Run(fmt.Sprintf("Testing %d rules", testMax), func(t *testing.T) {
 			// just check on the local interface
-			manager, err := Create(context.Background(), mock)
+			manager, err := Create(mock)
 			require.NoError(t, err)
+			require.NoError(t, manager.Init(nil))
 			time.Sleep(time.Second * 3)
 
 			defer func() {
-				if err := manager.Reset(); err != nil {
+				if err := manager.Reset(nil); err != nil {
 					t.Errorf("clear the manager state: %v", err)
 				}
 				time.Sleep(time.Second)
@@ -185,12 +200,8 @@ func TestNFtablesCreatePerformance(t *testing.T) {
 			ip := net.ParseIP("10.20.0.100")
 			start := time.Now()
 			for i := 0; i < testMax; i++ {
-				port := &fw.Port{Values: []int{1000 + i}}
-				if i%2 == 0 {
-					_, err = manager.AddFiltering(ip, "tcp", nil, port, fw.RuleDirectionOUT, fw.ActionAccept, "", "accept HTTP traffic")
-				} else {
-					_, err = manager.AddFiltering(ip, "tcp", nil, port, fw.RuleDirectionIN, fw.ActionAccept, "", "accept HTTP traffic")
-				}
+				port := &fw.Port{Values: []uint16{uint16(1000 + i)}}
+				_, err = manager.AddPeerFiltering(ip, "tcp", nil, port, fw.ActionAccept, "", "accept HTTP traffic")
 				require.NoError(t, err, "failed to add rule")
 
 				if i%100 == 0 {
@@ -203,3 +214,111 @@ func TestNFtablesCreatePerformance(t *testing.T) {
 		})
 	}
 }
+
+func runIptablesSave(t *testing.T) (string, string) {
+	t.Helper()
+	var stdout, stderr bytes.Buffer
+	cmd := exec.Command("iptables-save")
+	cmd.Stdout = &stdout
+	cmd.Stderr = &stderr
+
+	err := cmd.Run()
+	require.NoError(t, err, "iptables-save failed to run")
+
+	return stdout.String(), stderr.String()
+}
+
+func verifyIptablesOutput(t *testing.T, stdout, stderr string) {
+	t.Helper()
+	// Check for any incompatibility warnings
+	require.NotContains(t,
+		stderr,
+		"incompatible",
+		"iptables-save produced compatibility warning. Full stderr: %s",
+		stderr,
+	)
+
+	// Verify standard tables are present
+	expectedTables := []string{
+		"*filter",
+		"*nat",
+		"*mangle",
+	}
+
+	for _, table := range expectedTables {
+		require.Contains(t,
+			stdout,
+			table,
+			"iptables-save output missing expected table: %s\nFull stdout: %s",
+			table,
+			stdout,
+		)
+	}
+}
+
+func TestNftablesManagerCompatibilityWithIptables(t *testing.T) {
+	if check() != NFTABLES {
+		t.Skip("nftables not supported on this system")
+	}
+
+	if _, err := exec.LookPath("iptables-save"); err != nil {
+		t.Skipf("iptables-save not available on this system: %v", err)
+	}
+
+	// First ensure iptables-nft tables exist by running iptables-save
+	stdout, stderr := runIptablesSave(t)
+	verifyIptablesOutput(t, stdout, stderr)
+
+	manager, err := Create(ifaceMock)
+	require.NoError(t, err, "failed to create manager")
+	require.NoError(t, manager.Init(nil))
+
+	t.Cleanup(func() {
+		err := manager.Reset(nil)
+		require.NoError(t, err, "failed to reset manager state")
+
+		// Verify iptables output after reset
+		stdout, stderr := runIptablesSave(t)
+		verifyIptablesOutput(t, stdout, stderr)
+	})
+
+	ip := net.ParseIP("100.96.0.1")
+	_, err = manager.AddPeerFiltering(ip, fw.ProtocolTCP, nil, &fw.Port{Values: []uint16{80}}, fw.ActionAccept, "", "test rule")
+	require.NoError(t, err, "failed to add peer filtering rule")
+
+	_, err = manager.AddRouteFiltering(
+		[]netip.Prefix{netip.MustParsePrefix("192.168.2.0/24")},
+		netip.MustParsePrefix("10.1.0.0/24"),
+		fw.ProtocolTCP,
+		nil,
+		&fw.Port{Values: []uint16{443}},
+		fw.ActionAccept,
+	)
+	require.NoError(t, err, "failed to add route filtering rule")
+
+	pair := fw.RouterPair{
+		Source:      netip.MustParsePrefix("192.168.1.0/24"),
+		Destination: netip.MustParsePrefix("10.0.0.0/24"),
+		Masquerade:  true,
+	}
+	err = manager.AddNatRule(pair)
+	require.NoError(t, err, "failed to add NAT rule")
+
+	stdout, stderr = runIptablesSave(t)
+	verifyIptablesOutput(t, stdout, stderr)
+}
+
+func compareExprsIgnoringCounters(t *testing.T, got, want []expr.Any) {
+	t.Helper()
+	require.Equal(t, len(got), len(want), "expression count mismatch")
+
+	for i := range got {
+		if _, isCounter := got[i].(*expr.Counter); isCounter {
+			_, wantIsCounter := want[i].(*expr.Counter)
+			require.True(t, wantIsCounter, "expected Counter at index %d", i)
+			continue
+		}
+
+		require.Equal(t, got[i], want[i], "expression mismatch at index %d", i)
+	}
+}

client/firewall/nftables/route_linux.go

@@ -1,413 +0,0 @@
-package nftables
-
-import (
-	"bytes"
-	"context"
-	"errors"
-	"fmt"
-	"net"
-	"net/netip"
-
-	"github.com/google/nftables"
-	"github.com/google/nftables/binaryutil"
-	"github.com/google/nftables/expr"
-	log "github.com/sirupsen/logrus"
-
-	"github.com/netbirdio/netbird/client/firewall/manager"
-)
-
-const (
-	chainNameRouteingFw = "netbird-rt-fwd"
-	chainNameRoutingNat = "netbird-rt-nat"
-
-	userDataAcceptForwardRuleSrc = "frwacceptsrc"
-	userDataAcceptForwardRuleDst = "frwacceptdst"
-)
-
-// some presets for building nftable rules
-var (
-	zeroXor = binaryutil.NativeEndian.PutUint32(0)
-
-	exprCounterAccept = []expr.Any{
-		&expr.Counter{},
-		&expr.Verdict{
-			Kind: expr.VerdictAccept,
-		},
-	}
-
-	errFilterTableNotFound = fmt.Errorf("nftables: 'filter' table not found")
-)
-
-type router struct {
-	ctx         context.Context
-	stop        context.CancelFunc
-	conn        *nftables.Conn
-	workTable   *nftables.Table
-	filterTable *nftables.Table
-	chains      map[string]*nftables.Chain
-	// rules is useful to avoid duplicates and to get missing attributes that we don't have when adding new rules
-	rules                    map[string]*nftables.Rule
-	isDefaultFwdRulesEnabled bool
-}
-
-func newRouter(parentCtx context.Context, workTable *nftables.Table) (*router, error) {
-	ctx, cancel := context.WithCancel(parentCtx)
-
-	r := &router{
-		ctx:       ctx,
-		stop:      cancel,
-		conn:      &nftables.Conn{},
-		workTable: workTable,
-		chains:    make(map[string]*nftables.Chain),
-		rules:     make(map[string]*nftables.Rule),
-	}
-
-	var err error
-	r.filterTable, err = r.loadFilterTable()
-	if err != nil {
-		if errors.Is(err, errFilterTableNotFound) {
-			log.Warnf("table 'filter' not found for forward rules")
-		} else {
-			return nil, err
-		}
-	}
-
-	err = r.cleanUpDefaultForwardRules()
-	if err != nil {
-		log.Errorf("failed to clean up rules from FORWARD chain: %s", err)
-	}
-
-	err = r.createContainers()
-	if err != nil {
-		log.Errorf("failed to create containers for route: %s", err)
-	}
-	return r, err
-}
-
-func (r *router) RouteingFwChainName() string {
-	return chainNameRouteingFw
-}
-
-// ResetForwardRules cleans existing nftables default forward rules from the system
-func (r *router) ResetForwardRules() {
-	err := r.cleanUpDefaultForwardRules()
-	if err != nil {
-		log.Errorf("failed to reset forward rules: %s", err)
-	}
-}
-
-func (r *router) loadFilterTable() (*nftables.Table, error) {
-	tables, err := r.conn.ListTablesOfFamily(nftables.TableFamilyIPv4)
-	if err != nil {
-		return nil, fmt.Errorf("nftables: unable to list tables: %v", err)
-	}
-
-	for _, table := range tables {
-		if table.Name == "filter" {
-			return table, nil
-		}
-	}
-
-	return nil, errFilterTableNotFound
-}
-
-func (r *router) createContainers() error {
-
-	r.chains[chainNameRouteingFw] = r.conn.AddChain(&nftables.Chain{
-		Name:  chainNameRouteingFw,
-		Table: r.workTable,
-	})
-
-	r.chains[chainNameRoutingNat] = r.conn.AddChain(&nftables.Chain{
-		Name:     chainNameRoutingNat,
-		Table:    r.workTable,
-		Hooknum:  nftables.ChainHookPostrouting,
-		Priority: nftables.ChainPriorityNATSource - 1,
-		Type:     nftables.ChainTypeNAT,
-	})
-
-	err := r.refreshRulesMap()
-	if err != nil {
-		log.Errorf("failed to clean up rules from FORWARD chain: %s", err)
-	}
-
-	err = r.conn.Flush()
-	if err != nil {
-		return fmt.Errorf("nftables: unable to initialize table: %v", err)
-	}
-	return nil
-}
-
-// InsertRoutingRules inserts a nftable rule pair to the forwarding chain and if enabled, to the nat chain
-func (r *router) InsertRoutingRules(pair manager.RouterPair) error {
-	err := r.refreshRulesMap()
-	if err != nil {
-		return err
-	}
-
-	err = r.insertRoutingRule(manager.ForwardingFormat, chainNameRouteingFw, pair, false)
-	if err != nil {
-		return err
-	}
-	err = r.insertRoutingRule(manager.InForwardingFormat, chainNameRouteingFw, manager.GetInPair(pair), false)
-	if err != nil {
-		return err
-	}
-
-	if pair.Masquerade {
-		err = r.insertRoutingRule(manager.NatFormat, chainNameRoutingNat, pair, true)
-		if err != nil {
-			return err
-		}
-		err = r.insertRoutingRule(manager.InNatFormat, chainNameRoutingNat, manager.GetInPair(pair), true)
-		if err != nil {
-			return err
-		}
-	}
-
-	if r.filterTable != nil && !r.isDefaultFwdRulesEnabled {
-		log.Debugf("add default accept forward rule")
-		r.acceptForwardRule(pair.Source)
-	}
-
-	err = r.conn.Flush()
-	if err != nil {
-		return fmt.Errorf("nftables: unable to insert rules for %s: %v", pair.Destination, err)
-	}
-	return nil
-}
-
-// insertRoutingRule inserts a nftable rule to the conn client flush queue
-func (r *router) insertRoutingRule(format, chainName string, pair manager.RouterPair, isNat bool) error {
-	sourceExp := generateCIDRMatcherExpressions(true, pair.Source)
-	destExp := generateCIDRMatcherExpressions(false, pair.Destination)
-
-	var expression []expr.Any
-	if isNat {
-		expression = append(sourceExp, append(destExp, &expr.Counter{}, &expr.Masq{})...) // nolint:gocritic
-	} else {
-		expression = append(sourceExp, append(destExp, exprCounterAccept...)...) // nolint:gocritic
-	}
-
-	ruleKey := manager.GenKey(format, pair.ID)
-
-	_, exists := r.rules[ruleKey]
-	if exists {
-		err := r.removeRoutingRule(format, pair)
-		if err != nil {
-			return err
-		}
-	}
-
-	r.rules[ruleKey] = r.conn.InsertRule(&nftables.Rule{
-		Table:    r.workTable,
-		Chain:    r.chains[chainName],
-		Exprs:    expression,
-		UserData: []byte(ruleKey),
-	})
-	return nil
-}
-
-func (r *router) acceptForwardRule(sourceNetwork string) {
-	src := generateCIDRMatcherExpressions(true, sourceNetwork)
-	dst := generateCIDRMatcherExpressions(false, "0.0.0.0/0")
-
-	var exprs []expr.Any
-	exprs = append(src, append(dst, &expr.Verdict{ // nolint:gocritic
-		Kind: expr.VerdictAccept,
-	})...)
-
-	rule := &nftables.Rule{
-		Table: r.filterTable,
-		Chain: &nftables.Chain{
-			Name:     "FORWARD",
-			Table:    r.filterTable,
-			Type:     nftables.ChainTypeFilter,
-			Hooknum:  nftables.ChainHookForward,
-			Priority: nftables.ChainPriorityFilter,
-		},
-		Exprs:    exprs,
-		UserData: []byte(userDataAcceptForwardRuleSrc),
-	}
-
-	r.conn.AddRule(rule)
-
-	src = generateCIDRMatcherExpressions(true, "0.0.0.0/0")
-	dst = generateCIDRMatcherExpressions(false, sourceNetwork)
-
-	exprs = append(src, append(dst, &expr.Verdict{ //nolint:gocritic
-		Kind: expr.VerdictAccept,
-	})...)
-
-	rule = &nftables.Rule{
-		Table: r.filterTable,
-		Chain: &nftables.Chain{
-			Name:     "FORWARD",
-			Table:    r.filterTable,
-			Type:     nftables.ChainTypeFilter,
-			Hooknum:  nftables.ChainHookForward,
-			Priority: nftables.ChainPriorityFilter,
-		},
-		Exprs:    exprs,
-		UserData: []byte(userDataAcceptForwardRuleDst),
-	}
-	r.conn.AddRule(rule)
-	r.isDefaultFwdRulesEnabled = true
-}
-
-// RemoveRoutingRules removes a nftable rule pair from forwarding and nat chains
-func (r *router) RemoveRoutingRules(pair manager.RouterPair) error {
-	err := r.refreshRulesMap()
-	if err != nil {
-		return err
-	}
-
-	err = r.removeRoutingRule(manager.ForwardingFormat, pair)
-	if err != nil {
-		return err
-	}
-
-	err = r.removeRoutingRule(manager.InForwardingFormat, manager.GetInPair(pair))
-	if err != nil {
-		return err
-	}
-
-	err = r.removeRoutingRule(manager.NatFormat, pair)
-	if err != nil {
-		return err
-	}
-
-	err = r.removeRoutingRule(manager.InNatFormat, manager.GetInPair(pair))
-	if err != nil {
-		return err
-	}
-
-	if len(r.rules) == 0 {
-		err := r.cleanUpDefaultForwardRules()
-		if err != nil {
-			log.Errorf("failed to clean up rules from FORWARD chain: %s", err)
-		}
-	}
-
-	err = r.conn.Flush()
-	if err != nil {
-		return fmt.Errorf("nftables: received error while applying rule removal for %s: %v", pair.Destination, err)
-	}
-	log.Debugf("nftables: removed rules for %s", pair.Destination)
-	return nil
-}
-
-// removeRoutingRule add a nftable rule to the removal queue and delete from rules map
-func (r *router) removeRoutingRule(format string, pair manager.RouterPair) error {
-	ruleKey := manager.GenKey(format, pair.ID)
-
-	rule, found := r.rules[ruleKey]
-	if found {
-		ruleType := "forwarding"
-		if rule.Chain.Type == nftables.ChainTypeNAT {
-			ruleType = "nat"
-		}
-
-		err := r.conn.DelRule(rule)
-		if err != nil {
-			return fmt.Errorf("nftables: unable to remove %s rule for %s: %v", ruleType, pair.Destination, err)
-		}
-
-		log.Debugf("nftables: removing %s rule for %s", ruleType, pair.Destination)
-
-		delete(r.rules, ruleKey)
-	}
-	return nil
-}
-
-// refreshRulesMap refreshes the rule map with the latest rules. this is useful to avoid
-// duplicates and to get missing attributes that we don't have when adding new rules
-func (r *router) refreshRulesMap() error {
-	for _, chain := range r.chains {
-		rules, err := r.conn.GetRules(chain.Table, chain)
-		if err != nil {
-			return fmt.Errorf("nftables: unable to list rules: %v", err)
-		}
-		for _, rule := range rules {
-			if len(rule.UserData) > 0 {
-				r.rules[string(rule.UserData)] = rule
-			}
-		}
-	}
-	return nil
-}
-
-func (r *router) cleanUpDefaultForwardRules() error {
-	if r.filterTable == nil {
-		r.isDefaultFwdRulesEnabled = false
-		return nil
-	}
-
-	chains, err := r.conn.ListChainsOfTableFamily(nftables.TableFamilyIPv4)
-	if err != nil {
-		return err
-	}
-
-	var rules []*nftables.Rule
-	for _, chain := range chains {
-		if chain.Table.Name != r.filterTable.Name {
-			continue
-		}
-		if chain.Name != "FORWARD" {
-			continue
-		}
-
-		rules, err = r.conn.GetRules(r.filterTable, chain)
-		if err != nil {
-			return err
-		}
-	}
-
-	for _, rule := range rules {
-		if bytes.Equal(rule.UserData, []byte(userDataAcceptForwardRuleSrc)) || bytes.Equal(rule.UserData, []byte(userDataAcceptForwardRuleDst)) {
-			err := r.conn.DelRule(rule)
-			if err != nil {
-				return err
-			}
-		}
-	}
-	r.isDefaultFwdRulesEnabled = false
-	return r.conn.Flush()
-}
-
-// generateCIDRMatcherExpressions generates nftables expressions that matches a CIDR
-func generateCIDRMatcherExpressions(source bool, cidr string) []expr.Any {
-	ip, network, _ := net.ParseCIDR(cidr)
-	ipToAdd, _ := netip.AddrFromSlice(ip)
-	add := ipToAdd.Unmap()
-
-	var offSet uint32
-	if source {
-		offSet = 12 // src offset
-	} else {
-		offSet = 16 // dst offset
-	}
-
-	return []expr.Any{
-		// fetch src add
-		&expr.Payload{
-			DestRegister: 1,
-			Base:         expr.PayloadBaseNetworkHeader,
-			Offset:       offSet,
-			Len:          4,
-		},
-		// net mask
-		&expr.Bitwise{
-			DestRegister:   1,
-			SourceRegister: 1,
-			Len:            4,
-			Mask:           network.Mask,
-			Xor:            zeroXor,
-		},
-		// net address
-		&expr.Cmp{
-			Register: 1,
-			Data:     add.AsSlice(),
-		},
-	}
-}

client/firewall/nftables/router_linux.go

@@ -0,0 +1,995 @@
+package nftables
+
+import (
+	"bytes"
+	"encoding/binary"
+	"errors"
+	"fmt"
+	"net"
+	"net/netip"
+	"strings"
+
+	"github.com/coreos/go-iptables/iptables"
+	"github.com/davecgh/go-spew/spew"
+	"github.com/google/nftables"
+	"github.com/google/nftables/binaryutil"
+	"github.com/google/nftables/expr"
+	"github.com/hashicorp/go-multierror"
+	log "github.com/sirupsen/logrus"
+
+	nberrors "github.com/netbirdio/netbird/client/errors"
+	firewall "github.com/netbirdio/netbird/client/firewall/manager"
+	"github.com/netbirdio/netbird/client/internal/acl/id"
+	"github.com/netbirdio/netbird/client/internal/routemanager/refcounter"
+	nbnet "github.com/netbirdio/netbird/util/net"
+)
+
+const (
+	chainNameRoutingFw  = "netbird-rt-fwd"
+	chainNameRoutingNat = "netbird-rt-postrouting"
+	chainNameForward    = "FORWARD"
+
+	userDataAcceptForwardRuleIif = "frwacceptiif"
+	userDataAcceptForwardRuleOif = "frwacceptoif"
+)
+
+const refreshRulesMapError = "refresh rules map: %w"
+
+var (
+	errFilterTableNotFound = fmt.Errorf("nftables: 'filter' table not found")
+)
+
+type router struct {
+	conn        *nftables.Conn
+	workTable   *nftables.Table
+	filterTable *nftables.Table
+	chains      map[string]*nftables.Chain
+	// rules is useful to avoid duplicates and to get missing attributes that we don't have when adding new rules
+	rules        map[string]*nftables.Rule
+	ipsetCounter *refcounter.Counter[string, []netip.Prefix, *nftables.Set]
+
+	wgIface          iFaceMapper
+	legacyManagement bool
+}
+
+func newRouter(workTable *nftables.Table, wgIface iFaceMapper) (*router, error) {
+	r := &router{
+		conn:      &nftables.Conn{},
+		workTable: workTable,
+		chains:    make(map[string]*nftables.Chain),
+		rules:     make(map[string]*nftables.Rule),
+		wgIface:   wgIface,
+	}
+
+	r.ipsetCounter = refcounter.New(
+		r.createIpSet,
+		r.deleteIpSet,
+	)
+
+	var err error
+	r.filterTable, err = r.loadFilterTable()
+	if err != nil {
+		if errors.Is(err, errFilterTableNotFound) {
+			log.Warnf("table 'filter' not found for forward rules")
+		} else {
+			return nil, fmt.Errorf("load filter table: %w", err)
+		}
+	}
+
+	return r, nil
+}
+
+func (r *router) init(workTable *nftables.Table) error {
+	r.workTable = workTable
+
+	if err := r.removeAcceptForwardRules(); err != nil {
+		log.Errorf("failed to clean up rules from FORWARD chain: %s", err)
+	}
+
+	if err := r.createContainers(); err != nil {
+		return fmt.Errorf("create containers: %w", err)
+	}
+
+	return nil
+}
+
+// Reset cleans existing nftables default forward rules from the system
+func (r *router) Reset() error {
+	// clear without deleting the ipsets, the nf table will be deleted by the caller
+	r.ipsetCounter.Clear()
+
+	return r.removeAcceptForwardRules()
+}
+
+func (r *router) loadFilterTable() (*nftables.Table, error) {
+	tables, err := r.conn.ListTablesOfFamily(nftables.TableFamilyIPv4)
+	if err != nil {
+		return nil, fmt.Errorf("nftables: unable to list tables: %v", err)
+	}
+
+	for _, table := range tables {
+		if table.Name == "filter" {
+			return table, nil
+		}
+	}
+
+	return nil, errFilterTableNotFound
+}
+
+func (r *router) createContainers() error {
+	r.chains[chainNameRoutingFw] = r.conn.AddChain(&nftables.Chain{
+		Name:  chainNameRoutingFw,
+		Table: r.workTable,
+	})
+
+	insertReturnTrafficRule(r.conn, r.workTable, r.chains[chainNameRoutingFw])
+
+	prio := *nftables.ChainPriorityNATSource - 1
+	r.chains[chainNameRoutingNat] = r.conn.AddChain(&nftables.Chain{
+		Name:     chainNameRoutingNat,
+		Table:    r.workTable,
+		Hooknum:  nftables.ChainHookPostrouting,
+		Priority: &prio,
+		Type:     nftables.ChainTypeNAT,
+	})
+
+	// Chain is created by acl manager
+	// TODO: move creation to a common place
+	r.chains[chainNamePrerouting] = &nftables.Chain{
+		Name:     chainNamePrerouting,
+		Table:    r.workTable,
+		Type:     nftables.ChainTypeFilter,
+		Hooknum:  nftables.ChainHookPrerouting,
+		Priority: nftables.ChainPriorityMangle,
+	}
+
+	// Add the single NAT rule that matches on mark
+	if err := r.addPostroutingRules(); err != nil {
+		return fmt.Errorf("add single nat rule: %v", err)
+	}
+
+	if err := r.acceptForwardRules(); err != nil {
+		log.Errorf("failed to add accept rules for the forward chain: %s", err)
+	}
+
+	if err := r.refreshRulesMap(); err != nil {
+		log.Errorf("failed to clean up rules from FORWARD chain: %s", err)
+	}
+
+	if err := r.conn.Flush(); err != nil {
+		return fmt.Errorf("nftables: unable to initialize table: %v", err)
+	}
+
+	return nil
+}
+
+// AddRouteFiltering appends a nftables rule to the routing chain
+func (r *router) AddRouteFiltering(
+	sources []netip.Prefix,
+	destination netip.Prefix,
+	proto firewall.Protocol,
+	sPort *firewall.Port,
+	dPort *firewall.Port,
+	action firewall.Action,
+) (firewall.Rule, error) {
+
+	ruleKey := id.GenerateRouteRuleKey(sources, destination, proto, sPort, dPort, action)
+	if _, ok := r.rules[string(ruleKey)]; ok {
+		return ruleKey, nil
+	}
+
+	chain := r.chains[chainNameRoutingFw]
+	var exprs []expr.Any
+
+	switch {
+	case len(sources) == 1 && sources[0].Bits() == 0:
+		// If it's 0.0.0.0/0, we don't need to add any source matching
+	case len(sources) == 1:
+		// If there's only one source, we can use it directly
+		exprs = append(exprs, generateCIDRMatcherExpressions(true, sources[0])...)
+	default:
+		// If there are multiple sources, create or get an ipset
+		var err error
+		exprs, err = r.getIpSetExprs(sources, exprs)
+		if err != nil {
+			return nil, fmt.Errorf("get ipset expressions: %w", err)
+		}
+	}
+
+	// Handle destination
+	exprs = append(exprs, generateCIDRMatcherExpressions(false, destination)...)
+
+	// Handle protocol
+	if proto != firewall.ProtocolALL {
+		protoNum, err := protoToInt(proto)
+		if err != nil {
+			return nil, fmt.Errorf("convert protocol to number: %w", err)
+		}
+		exprs = append(exprs, &expr.Meta{Key: expr.MetaKeyL4PROTO, Register: 1})
+		exprs = append(exprs, &expr.Cmp{
+			Op:       expr.CmpOpEq,
+			Register: 1,
+			Data:     []byte{protoNum},
+		})
+
+		exprs = append(exprs, applyPort(sPort, true)...)
+		exprs = append(exprs, applyPort(dPort, false)...)
+	}
+
+	exprs = append(exprs, &expr.Counter{})
+
+	var verdict expr.VerdictKind
+	if action == firewall.ActionAccept {
+		verdict = expr.VerdictAccept
+	} else {
+		verdict = expr.VerdictDrop
+	}
+	exprs = append(exprs, &expr.Verdict{Kind: verdict})
+
+	rule := &nftables.Rule{
+		Table:    r.workTable,
+		Chain:    chain,
+		Exprs:    exprs,
+		UserData: []byte(ruleKey),
+	}
+
+	// Insert DROP rules at the beginning, append ACCEPT rules at the end
+	if action == firewall.ActionDrop {
+		// TODO: Insert after the established rule
+		rule = r.conn.InsertRule(rule)
+	} else {
+		rule = r.conn.AddRule(rule)
+	}
+
+	log.Tracef("Adding route rule %s", spew.Sdump(rule))
+	if err := r.conn.Flush(); err != nil {
+		return nil, fmt.Errorf(flushError, err)
+	}
+
+	r.rules[string(ruleKey)] = rule
+
+	log.Debugf("nftables: added route rule: sources=%v, destination=%v, proto=%v, sPort=%v, dPort=%v, action=%v", sources, destination, proto, sPort, dPort, action)
+
+	return ruleKey, nil
+}
+
+func (r *router) getIpSetExprs(sources []netip.Prefix, exprs []expr.Any) ([]expr.Any, error) {
+	setName := firewall.GenerateSetName(sources)
+	ref, err := r.ipsetCounter.Increment(setName, sources)
+	if err != nil {
+		return nil, fmt.Errorf("create or get ipset for sources: %w", err)
+	}
+
+	exprs = append(exprs,
+		&expr.Payload{
+			DestRegister: 1,
+			Base:         expr.PayloadBaseNetworkHeader,
+			Offset:       12,
+			Len:          4,
+		},
+		&expr.Lookup{
+			SourceRegister: 1,
+			SetName:        ref.Out.Name,
+			SetID:          ref.Out.ID,
+		},
+	)
+	return exprs, nil
+}
+
+func (r *router) DeleteRouteRule(rule firewall.Rule) error {
+	if err := r.refreshRulesMap(); err != nil {
+		return fmt.Errorf(refreshRulesMapError, err)
+	}
+
+	ruleKey := rule.GetRuleID()
+	nftRule, exists := r.rules[ruleKey]
+	if !exists {
+		log.Debugf("route rule %s not found", ruleKey)
+		return nil
+	}
+
+	if nftRule.Handle == 0 {
+		return fmt.Errorf("route rule %s has no handle", ruleKey)
+	}
+
+	setName := r.findSetNameInRule(nftRule)
+
+	if err := r.deleteNftRule(nftRule, ruleKey); err != nil {
+		return fmt.Errorf("delete: %w", err)
+	}
+
+	if setName != "" {
+		if _, err := r.ipsetCounter.Decrement(setName); err != nil {
+			return fmt.Errorf("decrement ipset reference: %w", err)
+		}
+	}
+
+	if err := r.conn.Flush(); err != nil {
+		return fmt.Errorf(flushError, err)
+	}
+
+	return nil
+}
+
+func (r *router) createIpSet(setName string, sources []netip.Prefix) (*nftables.Set, error) {
+	// overlapping prefixes will result in an error, so we need to merge them
+	sources = firewall.MergeIPRanges(sources)
+
+	set := &nftables.Set{
+		Name:  setName,
+		Table: r.workTable,
+		// required for prefixes
+		Interval: true,
+		KeyType:  nftables.TypeIPAddr,
+	}
+
+	var elements []nftables.SetElement
+	for _, prefix := range sources {
+		// TODO: Implement IPv6 support
+		if prefix.Addr().Is6() {
+			log.Printf("Skipping IPv6 prefix %s: IPv6 support not yet implemented", prefix)
+			continue
+		}
+
+		// nftables needs half-open intervals [firstIP, lastIP) for prefixes
+		// e.g. 10.0.0.0/24 becomes [10.0.0.0, 10.0.1.0), 10.1.1.1/32 becomes [10.1.1.1, 10.1.1.2) etc
+		firstIP := prefix.Addr()
+		lastIP := calculateLastIP(prefix).Next()
+
+		elements = append(elements,
+			// the nft tool also adds a line like this, see https://github.com/google/nftables/issues/247
+			// nftables.SetElement{Key: []byte{0, 0, 0, 0}, IntervalEnd: true},
+			nftables.SetElement{Key: firstIP.AsSlice()},
+			nftables.SetElement{Key: lastIP.AsSlice(), IntervalEnd: true},
+		)
+	}
+
+	if err := r.conn.AddSet(set, elements); err != nil {
+		return nil, fmt.Errorf("error adding elements to set %s: %w", setName, err)
+	}
+
+	if err := r.conn.Flush(); err != nil {
+		return nil, fmt.Errorf("flush error: %w", err)
+	}
+
+	log.Printf("Created new ipset: %s with %d elements", setName, len(elements)/2)
+
+	return set, nil
+}
+
+// calculateLastIP determines the last IP in a given prefix.
+func calculateLastIP(prefix netip.Prefix) netip.Addr {
+	hostMask := ^uint32(0) >> prefix.Masked().Bits()
+	lastIP := uint32FromNetipAddr(prefix.Addr()) | hostMask
+
+	return netip.AddrFrom4(uint32ToBytes(lastIP))
+}
+
+// Utility function to convert netip.Addr to uint32.
+func uint32FromNetipAddr(addr netip.Addr) uint32 {
+	b := addr.As4()
+	return binary.BigEndian.Uint32(b[:])
+}
+
+// Utility function to convert uint32 to a netip-compatible byte slice.
+func uint32ToBytes(ip uint32) [4]byte {
+	var b [4]byte
+	binary.BigEndian.PutUint32(b[:], ip)
+	return b
+}
+
+func (r *router) deleteIpSet(setName string, set *nftables.Set) error {
+	r.conn.DelSet(set)
+	if err := r.conn.Flush(); err != nil {
+		return fmt.Errorf(flushError, err)
+	}
+
+	log.Debugf("Deleted unused ipset %s", setName)
+	return nil
+}
+
+func (r *router) findSetNameInRule(rule *nftables.Rule) string {
+	for _, e := range rule.Exprs {
+		if lookup, ok := e.(*expr.Lookup); ok {
+			return lookup.SetName
+		}
+	}
+	return ""
+}
+
+func (r *router) deleteNftRule(rule *nftables.Rule, ruleKey string) error {
+	if err := r.conn.DelRule(rule); err != nil {
+		return fmt.Errorf("delete rule %s: %w", ruleKey, err)
+	}
+	delete(r.rules, ruleKey)
+
+	log.Debugf("removed route rule %s", ruleKey)
+
+	return nil
+}
+
+// AddNatRule appends a nftables rule pair to the nat chain
+func (r *router) AddNatRule(pair firewall.RouterPair) error {
+	if err := r.refreshRulesMap(); err != nil {
+		return fmt.Errorf(refreshRulesMapError, err)
+	}
+
+	if r.legacyManagement {
+		log.Warnf("This peer is connected to a NetBird Management service with an older version. Allowing all traffic for %s", pair.Destination)
+		if err := r.addLegacyRouteRule(pair); err != nil {
+			return fmt.Errorf("add legacy routing rule: %w", err)
+		}
+	}
+
+	if pair.Masquerade {
+		if err := r.addNatRule(pair); err != nil {
+			return fmt.Errorf("add nat rule: %w", err)
+		}
+
+		if err := r.addNatRule(firewall.GetInversePair(pair)); err != nil {
+			return fmt.Errorf("add inverse nat rule: %w", err)
+		}
+	}
+
+	if err := r.conn.Flush(); err != nil {
+		return fmt.Errorf("nftables: insert rules for %s: %v", pair.Destination, err)
+	}
+
+	return nil
+}
+
+// addNatRule inserts a nftables rule to the conn client flush queue
+func (r *router) addNatRule(pair firewall.RouterPair) error {
+	sourceExp := generateCIDRMatcherExpressions(true, pair.Source)
+	destExp := generateCIDRMatcherExpressions(false, pair.Destination)
+
+	op := expr.CmpOpEq
+	if pair.Inverse {
+		op = expr.CmpOpNeq
+	}
+
+	exprs := []expr.Any{
+		// We only care about NEW connections to mark them and later identify them in the postrouting chain for masquerading.
+		// Masquerading will take care of the conntrack state, which means we won't need to mark established connections.
+		&expr.Ct{
+			Key:      expr.CtKeySTATE,
+			Register: 1,
+		},
+		&expr.Bitwise{
+			SourceRegister: 1,
+			DestRegister:   1,
+			Len:            4,
+			Mask:           binaryutil.NativeEndian.PutUint32(expr.CtStateBitNEW),
+			Xor:            binaryutil.NativeEndian.PutUint32(0),
+		},
+		&expr.Cmp{
+			Op:       expr.CmpOpNeq,
+			Register: 1,
+			Data:     []byte{0, 0, 0, 0},
+		},
+
+		// interface matching
+		&expr.Meta{
+			Key:      expr.MetaKeyIIFNAME,
+			Register: 1,
+		},
+		&expr.Cmp{
+			Op:       op,
+			Register: 1,
+			Data:     ifname(r.wgIface.Name()),
+		},
+	}
+
+	exprs = append(exprs, sourceExp...)
+	exprs = append(exprs, destExp...)
+
+	var markValue uint32 = nbnet.PreroutingFwmarkMasquerade
+	if pair.Inverse {
+		markValue = nbnet.PreroutingFwmarkMasqueradeReturn
+	}
+
+	exprs = append(exprs,
+		&expr.Immediate{
+			Register: 1,
+			Data:     binaryutil.NativeEndian.PutUint32(markValue),
+		},
+		&expr.Meta{
+			Key:            expr.MetaKeyMARK,
+			SourceRegister: true,
+			Register:       1,
+		},
+	)
+
+	ruleKey := firewall.GenKey(firewall.PreroutingFormat, pair)
+
+	if _, exists := r.rules[ruleKey]; exists {
+		if err := r.removeNatRule(pair); err != nil {
+			return fmt.Errorf("remove prerouting rule: %w", err)
+		}
+	}
+
+	r.rules[ruleKey] = r.conn.AddRule(&nftables.Rule{
+		Table:    r.workTable,
+		Chain:    r.chains[chainNamePrerouting],
+		Exprs:    exprs,
+		UserData: []byte(ruleKey),
+	})
+
+	return nil
+}
+
+// addPostroutingRules adds the masquerade rules
+func (r *router) addPostroutingRules() error {
+	// First masquerade rule for traffic coming in from WireGuard interface
+	exprs := []expr.Any{
+		// Match on the first fwmark
+		&expr.Meta{
+			Key:      expr.MetaKeyMARK,
+			Register: 1,
+		},
+		&expr.Cmp{
+			Op:       expr.CmpOpEq,
+			Register: 1,
+			Data:     binaryutil.NativeEndian.PutUint32(nbnet.PreroutingFwmarkMasquerade),
+		},
+
+		// We need to exclude the loopback interface as this changes the ebpf proxy port
+		&expr.Meta{
+			Key:      expr.MetaKeyOIFNAME,
+			Register: 1,
+		},
+		&expr.Cmp{
+			Op:       expr.CmpOpNeq,
+			Register: 1,
+			Data:     ifname("lo"),
+		},
+		&expr.Counter{},
+		&expr.Masq{},
+	}
+
+	r.conn.AddRule(&nftables.Rule{
+		Table: r.workTable,
+		Chain: r.chains[chainNameRoutingNat],
+		Exprs: exprs,
+	})
+
+	// Second masquerade rule for traffic going out through WireGuard interface
+	exprs2 := []expr.Any{
+		// Match on the second fwmark
+		&expr.Meta{
+			Key:      expr.MetaKeyMARK,
+			Register: 1,
+		},
+		&expr.Cmp{
+			Op:       expr.CmpOpEq,
+			Register: 1,
+			Data:     binaryutil.NativeEndian.PutUint32(nbnet.PreroutingFwmarkMasqueradeReturn),
+		},
+
+		// Match WireGuard interface
+		&expr.Meta{
+			Key:      expr.MetaKeyOIFNAME,
+			Register: 1,
+		},
+		&expr.Cmp{
+			Op:       expr.CmpOpEq,
+			Register: 1,
+			Data:     ifname(r.wgIface.Name()),
+		},
+		&expr.Counter{},
+		&expr.Masq{},
+	}
+
+	r.conn.AddRule(&nftables.Rule{
+		Table: r.workTable,
+		Chain: r.chains[chainNameRoutingNat],
+		Exprs: exprs2,
+	})
+
+	return nil
+}
+
+// addLegacyRouteRule adds a legacy routing rule for mgmt servers pre route acls
+func (r *router) addLegacyRouteRule(pair firewall.RouterPair) error {
+	sourceExp := generateCIDRMatcherExpressions(true, pair.Source)
+	destExp := generateCIDRMatcherExpressions(false, pair.Destination)
+
+	exprs := []expr.Any{
+		&expr.Counter{},
+		&expr.Verdict{
+			Kind: expr.VerdictAccept,
+		},
+	}
+
+	expression := append(sourceExp, append(destExp, exprs...)...) // nolint:gocritic
+
+	ruleKey := firewall.GenKey(firewall.ForwardingFormat, pair)
+
+	if _, exists := r.rules[ruleKey]; exists {
+		if err := r.removeLegacyRouteRule(pair); err != nil {
+			return fmt.Errorf("remove legacy routing rule: %w", err)
+		}
+	}
+
+	r.rules[ruleKey] = r.conn.AddRule(&nftables.Rule{
+		Table:    r.workTable,
+		Chain:    r.chains[chainNameRoutingFw],
+		Exprs:    expression,
+		UserData: []byte(ruleKey),
+	})
+	return nil
+}
+
+// removeLegacyRouteRule removes a legacy routing rule for mgmt servers pre route acls
+func (r *router) removeLegacyRouteRule(pair firewall.RouterPair) error {
+	ruleKey := firewall.GenKey(firewall.ForwardingFormat, pair)
+
+	if rule, exists := r.rules[ruleKey]; exists {
+		if err := r.conn.DelRule(rule); err != nil {
+			return fmt.Errorf("remove legacy forwarding rule %s -> %s: %v", pair.Source, pair.Destination, err)
+		}
+
+		log.Debugf("nftables: removed legacy forwarding rule %s -> %s", pair.Source, pair.Destination)
+
+		delete(r.rules, ruleKey)
+	} else {
+		log.Debugf("nftables: legacy forwarding rule %s not found", ruleKey)
+	}
+
+	return nil
+}
+
+// GetLegacyManagement returns the route manager's legacy management mode
+func (r *router) GetLegacyManagement() bool {
+	return r.legacyManagement
+}
+
+// SetLegacyManagement sets the route manager to use legacy management mode
+func (r *router) SetLegacyManagement(isLegacy bool) {
+	r.legacyManagement = isLegacy
+}
+
+// RemoveAllLegacyRouteRules removes all legacy routing rules for mgmt servers pre route acls
+func (r *router) RemoveAllLegacyRouteRules() error {
+	if err := r.refreshRulesMap(); err != nil {
+		return fmt.Errorf(refreshRulesMapError, err)
+	}
+
+	var merr *multierror.Error
+	for k, rule := range r.rules {
+		if !strings.HasPrefix(k, firewall.ForwardingFormatPrefix) {
+			continue
+		}
+		if err := r.conn.DelRule(rule); err != nil {
+			merr = multierror.Append(merr, fmt.Errorf("remove legacy forwarding rule: %v", err))
+		} else {
+			delete(r.rules, k)
+		}
+
+	}
+	return nberrors.FormatErrorOrNil(merr)
+}
+
+// acceptForwardRules adds iif/oif rules in the filter table/forward chain to make sure
+// that our traffic is not dropped by existing rules there.
+// The existing FORWARD rules/policies decide outbound traffic towards our interface.
+// In case the FORWARD policy is set to "drop", we add an established/related rule to allow return traffic for the inbound rule.
+func (r *router) acceptForwardRules() error {
+	if r.filterTable == nil {
+		log.Debugf("table 'filter' not found for forward rules, skipping accept rules")
+		return nil
+	}
+
+	fw := "iptables"
+
+	defer func() {
+		log.Debugf("Used %s to add accept forward rules", fw)
+	}()
+
+	// Try iptables first and fallback to nftables if iptables is not available
+	ipt, err := iptables.New()
+	if err != nil {
+		// filter table exists but iptables is not
+		log.Warnf("Will use nftables to manipulate the filter table because iptables is not available: %v", err)
+
+		fw = "nftables"
+		return r.acceptForwardRulesNftables()
+	}
+
+	return r.acceptForwardRulesIptables(ipt)
+}
+
+func (r *router) acceptForwardRulesIptables(ipt *iptables.IPTables) error {
+	var merr *multierror.Error
+	for _, rule := range r.getAcceptForwardRules() {
+		if err := ipt.Insert("filter", chainNameForward, 1, rule...); err != nil {
+			merr = multierror.Append(err, fmt.Errorf("add iptables rule: %v", err))
+		} else {
+			log.Debugf("added iptables rule: %v", rule)
+		}
+	}
+
+	return nberrors.FormatErrorOrNil(merr)
+}
+
+func (r *router) getAcceptForwardRules() [][]string {
+	intf := r.wgIface.Name()
+	return [][]string{
+		{"-i", intf, "-j", "ACCEPT"},
+		{"-o", intf, "-m", "conntrack", "--ctstate", "RELATED,ESTABLISHED", "-j", "ACCEPT"},
+	}
+}
+
+func (r *router) acceptForwardRulesNftables() error {
+	intf := ifname(r.wgIface.Name())
+
+	// Rule for incoming interface (iif) with counter
+	iifRule := &nftables.Rule{
+		Table: r.filterTable,
+		Chain: &nftables.Chain{
+			Name:     chainNameForward,
+			Table:    r.filterTable,
+			Type:     nftables.ChainTypeFilter,
+			Hooknum:  nftables.ChainHookForward,
+			Priority: nftables.ChainPriorityFilter,
+		},
+		Exprs: []expr.Any{
+			&expr.Meta{Key: expr.MetaKeyIIFNAME, Register: 1},
+			&expr.Cmp{
+				Op:       expr.CmpOpEq,
+				Register: 1,
+				Data:     intf,
+			},
+			&expr.Counter{},
+			&expr.Verdict{Kind: expr.VerdictAccept},
+		},
+		UserData: []byte(userDataAcceptForwardRuleIif),
+	}
+	r.conn.InsertRule(iifRule)
+
+	oifExprs := []expr.Any{
+		&expr.Meta{Key: expr.MetaKeyOIFNAME, Register: 1},
+		&expr.Cmp{
+			Op:       expr.CmpOpEq,
+			Register: 1,
+			Data:     intf,
+		},
+	}
+
+	// Rule for outgoing interface (oif) with counter
+	oifRule := &nftables.Rule{
+		Table: r.filterTable,
+		Chain: &nftables.Chain{
+			Name:     "FORWARD",
+			Table:    r.filterTable,
+			Type:     nftables.ChainTypeFilter,
+			Hooknum:  nftables.ChainHookForward,
+			Priority: nftables.ChainPriorityFilter,
+		},
+		Exprs:    append(oifExprs, getEstablishedExprs(2)...),
+		UserData: []byte(userDataAcceptForwardRuleOif),
+	}
+
+	r.conn.InsertRule(oifRule)
+
+	return nil
+}
+
+func (r *router) removeAcceptForwardRules() error {
+	if r.filterTable == nil {
+		return nil
+	}
+
+	// Try iptables first and fallback to nftables if iptables is not available
+	ipt, err := iptables.New()
+	if err != nil {
+		log.Warnf("Will use nftables to manipulate the filter table because iptables is not available: %v", err)
+		return r.removeAcceptForwardRulesNftables()
+	}
+
+	return r.removeAcceptForwardRulesIptables(ipt)
+}
+
+func (r *router) removeAcceptForwardRulesNftables() error {
+	chains, err := r.conn.ListChainsOfTableFamily(nftables.TableFamilyIPv4)
+	if err != nil {
+		return fmt.Errorf("list chains: %v", err)
+	}
+
+	for _, chain := range chains {
+		if chain.Table.Name != r.filterTable.Name || chain.Name != chainNameForward {
+			continue
+		}
+
+		rules, err := r.conn.GetRules(r.filterTable, chain)
+		if err != nil {
+			return fmt.Errorf("get rules: %v", err)
+		}
+
+		for _, rule := range rules {
+			if bytes.Equal(rule.UserData, []byte(userDataAcceptForwardRuleIif)) ||
+				bytes.Equal(rule.UserData, []byte(userDataAcceptForwardRuleOif)) {
+				if err := r.conn.DelRule(rule); err != nil {
+					return fmt.Errorf("delete rule: %v", err)
+				}
+			}
+		}
+	}
+
+	if err := r.conn.Flush(); err != nil {
+		return fmt.Errorf(flushError, err)
+	}
+
+	return nil
+}
+
+func (r *router) removeAcceptForwardRulesIptables(ipt *iptables.IPTables) error {
+	var merr *multierror.Error
+	for _, rule := range r.getAcceptForwardRules() {
+		if err := ipt.DeleteIfExists("filter", chainNameForward, rule...); err != nil {
+			merr = multierror.Append(err, fmt.Errorf("remove iptables rule: %v", err))
+		}
+	}
+
+	return nberrors.FormatErrorOrNil(merr)
+}
+
+// RemoveNatRule removes the prerouting mark rule
+func (r *router) RemoveNatRule(pair firewall.RouterPair) error {
+	if err := r.refreshRulesMap(); err != nil {
+		return fmt.Errorf(refreshRulesMapError, err)
+	}
+
+	if err := r.removeNatRule(pair); err != nil {
+		return fmt.Errorf("remove prerouting rule: %w", err)
+	}
+
+	if err := r.removeNatRule(firewall.GetInversePair(pair)); err != nil {
+		return fmt.Errorf("remove inverse prerouting rule: %w", err)
+	}
+
+	if err := r.removeLegacyRouteRule(pair); err != nil {
+		return fmt.Errorf("remove legacy routing rule: %w", err)
+	}
+
+	if err := r.conn.Flush(); err != nil {
+		return fmt.Errorf("nftables: received error while applying rule removal for %s: %v", pair.Destination, err)
+	}
+
+	log.Debugf("nftables: removed nat rules for %s", pair.Destination)
+	return nil
+}
+
+func (r *router) removeNatRule(pair firewall.RouterPair) error {
+	ruleKey := firewall.GenKey(firewall.PreroutingFormat, pair)
+
+	if rule, exists := r.rules[ruleKey]; exists {
+		err := r.conn.DelRule(rule)
+		if err != nil {
+			return fmt.Errorf("remove prerouting rule %s -> %s: %v", pair.Source, pair.Destination, err)
+		}
+
+		log.Debugf("nftables: removed prerouting rule %s -> %s", pair.Source, pair.Destination)
+
+		delete(r.rules, ruleKey)
+	} else {
+		log.Debugf("nftables: prerouting rule %s not found", ruleKey)
+	}
+
+	return nil
+}
+
+// refreshRulesMap refreshes the rule map with the latest rules. this is useful to avoid
+// duplicates and to get missing attributes that we don't have when adding new rules
+func (r *router) refreshRulesMap() error {
+	for _, chain := range r.chains {
+		rules, err := r.conn.GetRules(chain.Table, chain)
+		if err != nil {
+			return fmt.Errorf("nftables: unable to list rules: %v", err)
+		}
+		for _, rule := range rules {
+			if len(rule.UserData) > 0 {
+				r.rules[string(rule.UserData)] = rule
+			}
+		}
+	}
+	return nil
+}
+
+// generateCIDRMatcherExpressions generates nftables expressions that matches a CIDR
+func generateCIDRMatcherExpressions(source bool, prefix netip.Prefix) []expr.Any {
+	var offset uint32
+	if source {
+		offset = 12 // src offset
+	} else {
+		offset = 16 // dst offset
+	}
+
+	ones := prefix.Bits()
+	// 0.0.0.0/0 doesn't need extra expressions
+	if ones == 0 {
+		return nil
+	}
+
+	mask := net.CIDRMask(ones, 32)
+
+	return []expr.Any{
+		&expr.Payload{
+			DestRegister: 1,
+			Base:         expr.PayloadBaseNetworkHeader,
+			Offset:       offset,
+			Len:          4,
+		},
+		// netmask
+		&expr.Bitwise{
+			DestRegister:   1,
+			SourceRegister: 1,
+			Len:            4,
+			Mask:           mask,
+			Xor:            []byte{0, 0, 0, 0},
+		},
+		// net address
+		&expr.Cmp{
+			Op:       expr.CmpOpEq,
+			Register: 1,
+			Data:     prefix.Masked().Addr().AsSlice(),
+		},
+	}
+}
+
+func applyPort(port *firewall.Port, isSource bool) []expr.Any {
+	if port == nil {
+		return nil
+	}
+
+	var exprs []expr.Any
+
+	offset := uint32(2) // Default offset for destination port
+	if isSource {
+		offset = 0 // Offset for source port
+	}
+
+	exprs = append(exprs, &expr.Payload{
+		DestRegister: 1,
+		Base:         expr.PayloadBaseTransportHeader,
+		Offset:       offset,
+		Len:          2,
+	})
+
+	if port.IsRange && len(port.Values) == 2 {
+		// Handle port range
+		exprs = append(exprs,
+			&expr.Cmp{
+				Op:       expr.CmpOpGte,
+				Register: 1,
+				Data:     binaryutil.BigEndian.PutUint16(port.Values[0]),
+			},
+			&expr.Cmp{
+				Op:       expr.CmpOpLte,
+				Register: 1,
+				Data:     binaryutil.BigEndian.PutUint16(port.Values[1]),
+			},
+		)
+	} else {
+		// Handle single port or multiple ports
+		for i, p := range port.Values {
+			if i > 0 {
+				// Add a bitwise OR operation between port checks
+				exprs = append(exprs, &expr.Bitwise{
+					SourceRegister: 1,
+					DestRegister:   1,
+					Len:            4,
+					Mask:           []byte{0x00, 0x00, 0xff, 0xff},
+					Xor:            []byte{0x00, 0x00, 0x00, 0x00},
+				})
+			}
+			exprs = append(exprs, &expr.Cmp{
+				Op:       expr.CmpOpEq,
+				Register: 1,
+				Data:     binaryutil.BigEndian.PutUint16(p),
+			})
+		}
+	}
+
+	return exprs
+}

client/firewall/nftables/router_linux_test.go

@@ -3,12 +3,16 @@
 package nftables
 
 import (
-	"context"
+	"encoding/binary"
+	"net/netip"
+	"os/exec"
 	"testing"
 
 	"github.com/coreos/go-iptables/iptables"
 	"github.com/google/nftables"
+	"github.com/google/nftables/binaryutil"
 	"github.com/google/nftables/expr"
+	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
@@ -24,195 +28,627 @@ const (
 	NFTABLES
 )
 
-func TestNftablesManager_InsertRoutingRules(t *testing.T) {
+func TestNftablesManager_AddNatRule(t *testing.T) {
 	if check() != NFTABLES {
 		t.Skip("nftables not supported on this OS")
 	}
 
-	table, err := createWorkTable()
-	if err != nil {
-		t.Fatal(err)
-	}
-
-	defer deleteWorkTable()
-
 	for _, testCase := range test.InsertRuleTestCases {
 		t.Run(testCase.Name, func(t *testing.T) {
-			manager, err := newRouter(context.TODO(), table)
-			require.NoError(t, err, "failed to create router")
+			// need fw manager to init both acl mgr and router for all chains to be present
+			manager, err := Create(ifaceMock)
+			t.Cleanup(func() {
+				require.NoError(t, manager.Reset(nil))
+			})
+			require.NoError(t, err)
+			require.NoError(t, manager.Init(nil))
 
 			nftablesTestingClient := &nftables.Conn{}
 
-			defer manager.ResetForwardRules()
+			rtr := manager.router
+			err = rtr.AddNatRule(testCase.InputPair)
+			require.NoError(t, err, "pair should be inserted")
 
-			require.NoError(t, err, "shouldn't return error")
-
-			err = manager.InsertRoutingRules(testCase.InputPair)
-			defer func() {
-				_ = manager.RemoveRoutingRules(testCase.InputPair)
-			}()
-			require.NoError(t, err, "forwarding pair should be inserted")
-
-			sourceExp := generateCIDRMatcherExpressions(true, testCase.InputPair.Source)
-			destExp := generateCIDRMatcherExpressions(false, testCase.InputPair.Destination)
-			testingExpression := append(sourceExp, destExp...) //nolint:gocritic
-			fwdRuleKey := firewall.GenKey(firewall.ForwardingFormat, testCase.InputPair.ID)
-
-			found := 0
-			for _, chain := range manager.chains {
-				rules, err := nftablesTestingClient.GetRules(chain.Table, chain)
-				require.NoError(t, err, "should list rules for %s table and %s chain", chain.Table.Name, chain.Name)
-				for _, rule := range rules {
-					if len(rule.UserData) > 0 && string(rule.UserData) == fwdRuleKey {
-						require.ElementsMatchf(t, rule.Exprs[:len(testingExpression)], testingExpression, "forwarding rule elements should match")
-						found = 1
-					}
-				}
-			}
-
-			require.Equal(t, 1, found, "should find at least 1 rule to test")
+			t.Cleanup(func() {
+				require.NoError(t, rtr.RemoveNatRule(testCase.InputPair), "failed to remove rule")
+			})
 
 			if testCase.InputPair.Masquerade {
-				natRuleKey := firewall.GenKey(firewall.NatFormat, testCase.InputPair.ID)
+				// Build expected expressions for connection tracking
+				conntrackExprs := []expr.Any{
+					&expr.Ct{
+						Key:      expr.CtKeySTATE,
+						Register: 1,
+					},
+					&expr.Bitwise{
+						SourceRegister: 1,
+						DestRegister:   1,
+						Len:            4,
+						Mask:           binaryutil.NativeEndian.PutUint32(expr.CtStateBitNEW),
+						Xor:            binaryutil.NativeEndian.PutUint32(0),
+					},
+					&expr.Cmp{
+						Op:       expr.CmpOpNeq,
+						Register: 1,
+						Data:     []byte{0, 0, 0, 0},
+					},
+				}
+
+				// Build interface matching expression
+				ifaceExprs := []expr.Any{
+					&expr.Meta{
+						Key:      expr.MetaKeyIIFNAME,
+						Register: 1,
+					},
+					&expr.Cmp{
+						Op:       expr.CmpOpEq,
+						Register: 1,
+						Data:     ifname(ifaceMock.Name()),
+					},
+				}
+
+				// Build CIDR matching expressions
+				sourceExp := generateCIDRMatcherExpressions(true, testCase.InputPair.Source)
+				destExp := generateCIDRMatcherExpressions(false, testCase.InputPair.Destination)
+
+				// Combine all expressions in the correct order
+				// nolint:gocritic
+				testingExpression := append(conntrackExprs, ifaceExprs...)
+				testingExpression = append(testingExpression, sourceExp...)
+				testingExpression = append(testingExpression, destExp...)
+
+				natRuleKey := firewall.GenKey(firewall.PreroutingFormat, testCase.InputPair)
 				found := 0
-				for _, chain := range manager.chains {
-					rules, err := nftablesTestingClient.GetRules(chain.Table, chain)
-					require.NoError(t, err, "should list rules for %s table and %s chain", chain.Table.Name, chain.Name)
-					for _, rule := range rules {
-						if len(rule.UserData) > 0 && string(rule.UserData) == natRuleKey {
-							require.ElementsMatchf(t, rule.Exprs[:len(testingExpression)], testingExpression, "nat rule elements should match")
-							found = 1
+				for _, chain := range rtr.chains {
+					if chain.Name == chainNamePrerouting {
+						rules, err := nftablesTestingClient.GetRules(chain.Table, chain)
+						require.NoError(t, err, "should list rules for %s table and %s chain", chain.Table.Name, chain.Name)
+						for _, rule := range rules {
+							if len(rule.UserData) > 0 && string(rule.UserData) == natRuleKey {
+								// Compare expressions up to the mark setting expressions
+								require.ElementsMatchf(t, rule.Exprs[:len(testingExpression)], testingExpression, "prerouting nat rule elements should match")
+								found = 1
+							}
 						}
 					}
 				}
-				require.Equal(t, 1, found, "should find at least 1 rule to test")
-			}
-
-			sourceExp = generateCIDRMatcherExpressions(true, firewall.GetInPair(testCase.InputPair).Source)
-			destExp = generateCIDRMatcherExpressions(false, firewall.GetInPair(testCase.InputPair).Destination)
-			testingExpression = append(sourceExp, destExp...) //nolint:gocritic
-			inFwdRuleKey := firewall.GenKey(firewall.InForwardingFormat, testCase.InputPair.ID)
-
-			found = 0
-			for _, chain := range manager.chains {
-				rules, err := nftablesTestingClient.GetRules(chain.Table, chain)
-				require.NoError(t, err, "should list rules for %s table and %s chain", chain.Table.Name, chain.Name)
-				for _, rule := range rules {
-					if len(rule.UserData) > 0 && string(rule.UserData) == inFwdRuleKey {
-						require.ElementsMatchf(t, rule.Exprs[:len(testingExpression)], testingExpression, "income forwarding rule elements should match")
-						found = 1
-					}
-				}
-			}
-
-			require.Equal(t, 1, found, "should find at least 1 rule to test")
-
-			if testCase.InputPair.Masquerade {
-				inNatRuleKey := firewall.GenKey(firewall.InNatFormat, testCase.InputPair.ID)
-				found := 0
-				for _, chain := range manager.chains {
-					rules, err := nftablesTestingClient.GetRules(chain.Table, chain)
-					require.NoError(t, err, "should list rules for %s table and %s chain", chain.Table.Name, chain.Name)
-					for _, rule := range rules {
-						if len(rule.UserData) > 0 && string(rule.UserData) == inNatRuleKey {
-							require.ElementsMatchf(t, rule.Exprs[:len(testingExpression)], testingExpression, "income nat rule elements should match")
-							found = 1
-						}
-					}
-				}
-				require.Equal(t, 1, found, "should find at least 1 rule to test")
+				require.Equal(t, 1, found, "should find at least 1 rule in prerouting chain")
 			}
 		})
 	}
 }
 
-func TestNftablesManager_RemoveRoutingRules(t *testing.T) {
+func TestNftablesManager_RemoveNatRule(t *testing.T) {
 	if check() != NFTABLES {
 		t.Skip("nftables not supported on this OS")
 	}
 
-	table, err := createWorkTable()
-	if err != nil {
-		t.Fatal(err)
+	for _, testCase := range test.RemoveRuleTestCases {
+		t.Run(testCase.Name, func(t *testing.T) {
+			manager, err := Create(ifaceMock)
+			t.Cleanup(func() {
+				require.NoError(t, manager.Reset(nil))
+			})
+			require.NoError(t, err)
+			require.NoError(t, manager.Init(nil))
+
+			rtr := manager.router
+
+			// First add the NAT rule using the router's method
+			err = rtr.AddNatRule(testCase.InputPair)
+			require.NoError(t, err, "should add NAT rule")
+
+			// Verify the rule was added
+			natRuleKey := firewall.GenKey(firewall.PreroutingFormat, testCase.InputPair)
+			found := false
+			rules, err := rtr.conn.GetRules(rtr.workTable, rtr.chains[chainNamePrerouting])
+			require.NoError(t, err, "should list rules")
+			for _, rule := range rules {
+				if len(rule.UserData) > 0 && string(rule.UserData) == natRuleKey {
+					found = true
+					break
+				}
+			}
+			require.True(t, found, "NAT rule should exist before removal")
+
+			// Now remove the rule
+			err = rtr.RemoveNatRule(testCase.InputPair)
+			require.NoError(t, err, "shouldn't return error when removing rule")
+
+			// Verify the rule was removed
+			found = false
+			rules, err = rtr.conn.GetRules(rtr.workTable, rtr.chains[chainNamePrerouting])
+			require.NoError(t, err, "should list rules after removal")
+			for _, rule := range rules {
+				if len(rule.UserData) > 0 && string(rule.UserData) == natRuleKey {
+					found = true
+					break
+				}
+			}
+			require.False(t, found, "NAT rule should not exist after removal")
+
+			// Verify the static postrouting rules still exist
+			rules, err = rtr.conn.GetRules(rtr.workTable, rtr.chains[chainNameRoutingNat])
+			require.NoError(t, err, "should list postrouting rules")
+			foundCounter := false
+			for _, rule := range rules {
+				for _, e := range rule.Exprs {
+					if _, ok := e.(*expr.Counter); ok {
+						foundCounter = true
+						break
+					}
+				}
+				if foundCounter {
+					break
+				}
+			}
+			require.True(t, foundCounter, "static postrouting rule should remain")
+		})
 	}
+}
+
+func TestRouter_AddRouteFiltering(t *testing.T) {
+	if check() != NFTABLES {
+		t.Skip("nftables not supported on this system")
+	}
+
+	workTable, err := createWorkTable()
+	require.NoError(t, err, "Failed to create work table")
 
 	defer deleteWorkTable()
 
-	for _, testCase := range test.RemoveRuleTestCases {
-		t.Run(testCase.Name, func(t *testing.T) {
-			manager, err := newRouter(context.TODO(), table)
-			require.NoError(t, err, "failed to create router")
+	r, err := newRouter(workTable, ifaceMock)
+	require.NoError(t, err, "Failed to create router")
+	require.NoError(t, r.init(workTable))
 
-			nftablesTestingClient := &nftables.Conn{}
+	defer func(r *router) {
+		require.NoError(t, r.Reset(), "Failed to reset rules")
+	}(r)
 
-			defer manager.ResetForwardRules()
+	tests := []struct {
+		name        string
+		sources     []netip.Prefix
+		destination netip.Prefix
+		proto       firewall.Protocol
+		sPort       *firewall.Port
+		dPort       *firewall.Port
+		direction   firewall.RuleDirection
+		action      firewall.Action
+		expectSet   bool
+	}{
+		{
+			name:        "Basic TCP rule with single source",
+			sources:     []netip.Prefix{netip.MustParsePrefix("192.168.1.0/24")},
+			destination: netip.MustParsePrefix("10.0.0.0/24"),
+			proto:       firewall.ProtocolTCP,
+			sPort:       nil,
+			dPort:       &firewall.Port{Values: []uint16{80}},
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name: "UDP rule with multiple sources",
+			sources: []netip.Prefix{
+				netip.MustParsePrefix("172.16.0.0/16"),
+				netip.MustParsePrefix("192.168.0.0/16"),
+			},
+			destination: netip.MustParsePrefix("10.0.0.0/8"),
+			proto:       firewall.ProtocolUDP,
+			sPort:       &firewall.Port{Values: []uint16{1024, 2048}, IsRange: true},
+			dPort:       nil,
+			direction:   firewall.RuleDirectionOUT,
+			action:      firewall.ActionDrop,
+			expectSet:   true,
+		},
+		{
+			name:        "All protocols rule",
+			sources:     []netip.Prefix{netip.MustParsePrefix("10.0.0.0/8")},
+			destination: netip.MustParsePrefix("0.0.0.0/0"),
+			proto:       firewall.ProtocolALL,
+			sPort:       nil,
+			dPort:       nil,
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "ICMP rule",
+			sources:     []netip.Prefix{netip.MustParsePrefix("192.168.0.0/16")},
+			destination: netip.MustParsePrefix("10.0.0.0/8"),
+			proto:       firewall.ProtocolICMP,
+			sPort:       nil,
+			dPort:       nil,
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "TCP rule with multiple source ports",
+			sources:     []netip.Prefix{netip.MustParsePrefix("172.16.0.0/12")},
+			destination: netip.MustParsePrefix("192.168.0.0/16"),
+			proto:       firewall.ProtocolTCP,
+			sPort:       &firewall.Port{Values: []uint16{80, 443, 8080}},
+			dPort:       nil,
+			direction:   firewall.RuleDirectionOUT,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "UDP rule with single IP and port range",
+			sources:     []netip.Prefix{netip.MustParsePrefix("192.168.1.1/32")},
+			destination: netip.MustParsePrefix("10.0.0.0/24"),
+			proto:       firewall.ProtocolUDP,
+			sPort:       nil,
+			dPort:       &firewall.Port{Values: []uint16{5000, 5100}, IsRange: true},
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionDrop,
+			expectSet:   false,
+		},
+		{
+			name:        "TCP rule with source and destination ports",
+			sources:     []netip.Prefix{netip.MustParsePrefix("10.0.0.0/24")},
+			destination: netip.MustParsePrefix("172.16.0.0/16"),
+			proto:       firewall.ProtocolTCP,
+			sPort:       &firewall.Port{Values: []uint16{1024, 65535}, IsRange: true},
+			dPort:       &firewall.Port{Values: []uint16{22}},
+			direction:   firewall.RuleDirectionOUT,
+			action:      firewall.ActionAccept,
+			expectSet:   false,
+		},
+		{
+			name:        "Drop all incoming traffic",
+			sources:     []netip.Prefix{netip.MustParsePrefix("0.0.0.0/0")},
+			destination: netip.MustParsePrefix("192.168.0.0/24"),
+			proto:       firewall.ProtocolALL,
+			sPort:       nil,
+			dPort:       nil,
+			direction:   firewall.RuleDirectionIN,
+			action:      firewall.ActionDrop,
+			expectSet:   false,
+		},
+	}
 
-			sourceExp := generateCIDRMatcherExpressions(true, testCase.InputPair.Source)
-			destExp := generateCIDRMatcherExpressions(false, testCase.InputPair.Destination)
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			ruleKey, err := r.AddRouteFiltering(tt.sources, tt.destination, tt.proto, tt.sPort, tt.dPort, tt.action)
+			require.NoError(t, err, "AddRouteFiltering failed")
 
-			forwardExp := append(sourceExp, append(destExp, exprCounterAccept...)...) //nolint:gocritic
-			forwardRuleKey := firewall.GenKey(firewall.ForwardingFormat, testCase.InputPair.ID)
-			insertedForwarding := nftablesTestingClient.InsertRule(&nftables.Rule{
-				Table:    manager.workTable,
-				Chain:    manager.chains[chainNameRouteingFw],
-				Exprs:    forwardExp,
-				UserData: []byte(forwardRuleKey),
+			t.Cleanup(func() {
+				require.NoError(t, r.DeleteRouteRule(ruleKey), "Failed to delete rule")
 			})
 
-			natExp := append(sourceExp, append(destExp, &expr.Counter{}, &expr.Masq{})...) //nolint:gocritic
-			natRuleKey := firewall.GenKey(firewall.NatFormat, testCase.InputPair.ID)
+			// Check if the rule is in the internal map
+			rule, ok := r.rules[ruleKey.GetRuleID()]
+			assert.True(t, ok, "Rule not found in internal map")
 
-			insertedNat := nftablesTestingClient.InsertRule(&nftables.Rule{
-				Table:    manager.workTable,
-				Chain:    manager.chains[chainNameRoutingNat],
-				Exprs:    natExp,
-				UserData: []byte(natRuleKey),
-			})
+			t.Log("Internal rule expressions:")
+			for i, expr := range rule.Exprs {
+				t.Logf("  [%d] %T: %+v", i, expr, expr)
+			}
 
-			sourceExp = generateCIDRMatcherExpressions(true, firewall.GetInPair(testCase.InputPair).Source)
-			destExp = generateCIDRMatcherExpressions(false, firewall.GetInPair(testCase.InputPair).Destination)
+			// Verify internal rule content
+			verifyRule(t, rule, tt.sources, tt.destination, tt.proto, tt.sPort, tt.dPort, tt.direction, tt.action, tt.expectSet)
 
-			forwardExp = append(sourceExp, append(destExp, exprCounterAccept...)...) //nolint:gocritic
-			inForwardRuleKey := firewall.GenKey(firewall.InForwardingFormat, testCase.InputPair.ID)
-			insertedInForwarding := nftablesTestingClient.InsertRule(&nftables.Rule{
-				Table:    manager.workTable,
-				Chain:    manager.chains[chainNameRouteingFw],
-				Exprs:    forwardExp,
-				UserData: []byte(inForwardRuleKey),
-			})
+			// Check if the rule exists in nftables and verify its content
+			rules, err := r.conn.GetRules(r.workTable, r.chains[chainNameRoutingFw])
+			require.NoError(t, err, "Failed to get rules from nftables")
 
-			natExp = append(sourceExp, append(destExp, &expr.Counter{}, &expr.Masq{})...) //nolint:gocritic
-			inNatRuleKey := firewall.GenKey(firewall.InNatFormat, testCase.InputPair.ID)
+			var nftRule *nftables.Rule
+			for _, rule := range rules {
+				if string(rule.UserData) == ruleKey.GetRuleID() {
+					nftRule = rule
+					break
+				}
+			}
 
-			insertedInNat := nftablesTestingClient.InsertRule(&nftables.Rule{
-				Table:    manager.workTable,
-				Chain:    manager.chains[chainNameRoutingNat],
-				Exprs:    natExp,
-				UserData: []byte(inNatRuleKey),
-			})
+			require.NotNil(t, nftRule, "Rule not found in nftables")
+			t.Log("Actual nftables rule expressions:")
+			for i, expr := range nftRule.Exprs {
+				t.Logf("  [%d] %T: %+v", i, expr, expr)
+			}
 
-			err = nftablesTestingClient.Flush()
-			require.NoError(t, err, "shouldn't return error")
+			// Verify actual nftables rule content
+			verifyRule(t, nftRule, tt.sources, tt.destination, tt.proto, tt.sPort, tt.dPort, tt.direction, tt.action, tt.expectSet)
+		})
+	}
+}
 
-			manager.ResetForwardRules()
+func TestNftablesCreateIpSet(t *testing.T) {
+	if check() != NFTABLES {
+		t.Skip("nftables not supported on this system")
+	}
 
-			err = manager.RemoveRoutingRules(testCase.InputPair)
-			require.NoError(t, err, "shouldn't return error")
+	workTable, err := createWorkTable()
+	require.NoError(t, err, "Failed to create work table")
 
-			for _, chain := range manager.chains {
-				rules, err := nftablesTestingClient.GetRules(chain.Table, chain)
-				require.NoError(t, err, "should list rules for %s table and %s chain", chain.Table.Name, chain.Name)
-				for _, rule := range rules {
-					if len(rule.UserData) > 0 {
-						require.NotEqual(t, insertedForwarding.UserData, rule.UserData, "forwarding rule should not exist")
-						require.NotEqual(t, insertedNat.UserData, rule.UserData, "nat rule should not exist")
-						require.NotEqual(t, insertedInForwarding.UserData, rule.UserData, "income forwarding rule should not exist")
-						require.NotEqual(t, insertedInNat.UserData, rule.UserData, "income nat rule should not exist")
+	defer deleteWorkTable()
+
+	r, err := newRouter(workTable, ifaceMock)
+	require.NoError(t, err, "Failed to create router")
+	require.NoError(t, r.init(workTable))
+
+	defer func() {
+		require.NoError(t, r.Reset(), "Failed to reset router")
+	}()
+
+	tests := []struct {
+		name     string
+		sources  []netip.Prefix
+		expected []netip.Prefix
+	}{
+		{
+			name:    "Single IP",
+			sources: []netip.Prefix{netip.MustParsePrefix("192.168.1.1/32")},
+		},
+		{
+			name: "Multiple IPs",
+			sources: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.1/32"),
+				netip.MustParsePrefix("10.0.0.1/32"),
+				netip.MustParsePrefix("172.16.0.1/32"),
+			},
+		},
+		{
+			name:    "Single Subnet",
+			sources: []netip.Prefix{netip.MustParsePrefix("192.168.0.0/24")},
+		},
+		{
+			name: "Multiple Subnets with Various Prefix Lengths",
+			sources: []netip.Prefix{
+				netip.MustParsePrefix("10.0.0.0/8"),
+				netip.MustParsePrefix("172.16.0.0/16"),
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("203.0.113.0/26"),
+			},
+		},
+		{
+			name: "Mix of Single IPs and Subnets in Different Positions",
+			sources: []netip.Prefix{
+				netip.MustParsePrefix("192.168.1.1/32"),
+				netip.MustParsePrefix("10.0.0.0/16"),
+				netip.MustParsePrefix("172.16.0.1/32"),
+				netip.MustParsePrefix("203.0.113.0/24"),
+			},
+		},
+		{
+			name: "Overlapping IPs/Subnets",
+			sources: []netip.Prefix{
+				netip.MustParsePrefix("10.0.0.0/8"),
+				netip.MustParsePrefix("10.0.0.0/16"),
+				netip.MustParsePrefix("10.0.0.1/32"),
+				netip.MustParsePrefix("192.168.0.0/16"),
+				netip.MustParsePrefix("192.168.1.0/24"),
+				netip.MustParsePrefix("192.168.1.1/32"),
+			},
+			expected: []netip.Prefix{
+				netip.MustParsePrefix("10.0.0.0/8"),
+				netip.MustParsePrefix("192.168.0.0/16"),
+			},
+		},
+	}
+
+	// Add this helper function inside TestNftablesCreateIpSet
+	printNftSets := func() {
+		cmd := exec.Command("nft", "list", "sets")
+		output, err := cmd.CombinedOutput()
+		if err != nil {
+			t.Logf("Failed to run 'nft list sets': %v", err)
+		} else {
+			t.Logf("Current nft sets:\n%s", output)
+		}
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			setName := firewall.GenerateSetName(tt.sources)
+			set, err := r.createIpSet(setName, tt.sources)
+			if err != nil {
+				t.Logf("Failed to create IP set: %v", err)
+				printNftSets()
+				require.NoError(t, err, "Failed to create IP set")
+			}
+			require.NotNil(t, set, "Created set is nil")
+
+			// Verify set properties
+			assert.Equal(t, setName, set.Name, "Set name mismatch")
+			assert.Equal(t, r.workTable, set.Table, "Set table mismatch")
+			assert.True(t, set.Interval, "Set interval property should be true")
+			assert.Equal(t, nftables.TypeIPAddr, set.KeyType, "Set key type mismatch")
+
+			// Fetch the created set from nftables
+			fetchedSet, err := r.conn.GetSetByName(r.workTable, setName)
+			require.NoError(t, err, "Failed to fetch created set")
+			require.NotNil(t, fetchedSet, "Fetched set is nil")
+
+			// Verify set elements
+			elements, err := r.conn.GetSetElements(fetchedSet)
+			require.NoError(t, err, "Failed to get set elements")
+
+			// Count the number of unique prefixes (excluding interval end markers)
+			uniquePrefixes := make(map[string]bool)
+			for _, elem := range elements {
+				if !elem.IntervalEnd {
+					ip := netip.AddrFrom4(*(*[4]byte)(elem.Key))
+					uniquePrefixes[ip.String()] = true
+				}
+			}
+
+			// Check against expected merged prefixes
+			expectedCount := len(tt.expected)
+			if expectedCount == 0 {
+				expectedCount = len(tt.sources)
+			}
+			assert.Equal(t, expectedCount, len(uniquePrefixes), "Number of unique prefixes in set doesn't match expected")
+
+			// Verify each expected prefix is in the set
+			for _, expected := range tt.expected {
+				found := false
+				for _, elem := range elements {
+					if !elem.IntervalEnd {
+						ip := netip.AddrFrom4(*(*[4]byte)(elem.Key))
+						if expected.Contains(ip) {
+							found = true
+							break
+						}
+					}
+				}
+				assert.True(t, found, "Expected prefix %s not found in set", expected)
+			}
+
+			r.conn.DelSet(set)
+			if err := r.conn.Flush(); err != nil {
+				t.Logf("Failed to delete set: %v", err)
+				printNftSets()
+			}
+			require.NoError(t, err, "Failed to delete set")
+		})
+	}
+}
+
+func verifyRule(t *testing.T, rule *nftables.Rule, sources []netip.Prefix, destination netip.Prefix, proto firewall.Protocol, sPort, dPort *firewall.Port, direction firewall.RuleDirection, action firewall.Action, expectSet bool) {
+	t.Helper()
+
+	assert.NotNil(t, rule, "Rule should not be nil")
+
+	// Verify sources and destination
+	if expectSet {
+		assert.True(t, containsSetLookup(rule.Exprs), "Rule should contain set lookup for multiple sources")
+	} else if len(sources) == 1 && sources[0].Bits() != 0 {
+		if direction == firewall.RuleDirectionIN {
+			assert.True(t, containsCIDRMatcher(rule.Exprs, sources[0], true), "Rule should contain source CIDR matcher for %s", sources[0])
+		} else {
+			assert.True(t, containsCIDRMatcher(rule.Exprs, sources[0], false), "Rule should contain destination CIDR matcher for %s", sources[0])
+		}
+	}
+
+	if direction == firewall.RuleDirectionIN {
+		assert.True(t, containsCIDRMatcher(rule.Exprs, destination, false), "Rule should contain destination CIDR matcher for %s", destination)
+	} else {
+		assert.True(t, containsCIDRMatcher(rule.Exprs, destination, true), "Rule should contain source CIDR matcher for %s", destination)
+	}
+
+	// Verify protocol
+	if proto != firewall.ProtocolALL {
+		assert.True(t, containsProtocol(rule.Exprs, proto), "Rule should contain protocol matcher for %s", proto)
+	}
+
+	// Verify ports
+	if sPort != nil {
+		assert.True(t, containsPort(rule.Exprs, sPort, true), "Rule should contain source port matcher for %v", sPort)
+	}
+	if dPort != nil {
+		assert.True(t, containsPort(rule.Exprs, dPort, false), "Rule should contain destination port matcher for %v", dPort)
+	}
+
+	// Verify action
+	assert.True(t, containsAction(rule.Exprs, action), "Rule should contain correct action: %s", action)
+}
+
+func containsSetLookup(exprs []expr.Any) bool {
+	for _, e := range exprs {
+		if _, ok := e.(*expr.Lookup); ok {
+			return true
+		}
+	}
+	return false
+}
+
+func containsCIDRMatcher(exprs []expr.Any, prefix netip.Prefix, isSource bool) bool {
+	var offset uint32
+	if isSource {
+		offset = 12 // src offset
+	} else {
+		offset = 16 // dst offset
+	}
+
+	var payloadFound, bitwiseFound, cmpFound bool
+	for _, e := range exprs {
+		switch ex := e.(type) {
+		case *expr.Payload:
+			if ex.Base == expr.PayloadBaseNetworkHeader && ex.Offset == offset && ex.Len == 4 {
+				payloadFound = true
+			}
+		case *expr.Bitwise:
+			if ex.Len == 4 && len(ex.Mask) == 4 && len(ex.Xor) == 4 {
+				bitwiseFound = true
+			}
+		case *expr.Cmp:
+			if ex.Op == expr.CmpOpEq && len(ex.Data) == 4 {
+				cmpFound = true
+			}
+		}
+	}
+	return (payloadFound && bitwiseFound && cmpFound) || prefix.Bits() == 0
+}
+
+func containsPort(exprs []expr.Any, port *firewall.Port, isSource bool) bool {
+	var offset uint32 = 2 // Default offset for destination port
+	if isSource {
+		offset = 0 // Offset for source port
+	}
+
+	var payloadFound, portMatchFound bool
+	for _, e := range exprs {
+		switch ex := e.(type) {
+		case *expr.Payload:
+			if ex.Base == expr.PayloadBaseTransportHeader && ex.Offset == offset && ex.Len == 2 {
+				payloadFound = true
+			}
+		case *expr.Cmp:
+			if port.IsRange {
+				if ex.Op == expr.CmpOpGte || ex.Op == expr.CmpOpLte {
+					portMatchFound = true
+				}
+			} else {
+				if ex.Op == expr.CmpOpEq && len(ex.Data) == 2 {
+					portValue := binary.BigEndian.Uint16(ex.Data)
+					for _, p := range port.Values {
+						if uint16(p) == portValue {
+							portMatchFound = true
+							break
+						}
 					}
 				}
 			}
-		})
+		}
+		if payloadFound && portMatchFound {
+			return true
+		}
 	}
+	return false
+}
+
+func containsProtocol(exprs []expr.Any, proto firewall.Protocol) bool {
+	var metaFound, cmpFound bool
+	expectedProto, _ := protoToInt(proto)
+	for _, e := range exprs {
+		switch ex := e.(type) {
+		case *expr.Meta:
+			if ex.Key == expr.MetaKeyL4PROTO {
+				metaFound = true
+			}
+		case *expr.Cmp:
+			if ex.Op == expr.CmpOpEq && len(ex.Data) == 1 && ex.Data[0] == expectedProto {
+				cmpFound = true
+			}
+		}
+	}
+	return metaFound && cmpFound
+}
+
+func containsAction(exprs []expr.Any, action firewall.Action) bool {
+	for _, e := range exprs {
+		if verdict, ok := e.(*expr.Verdict); ok {
+			switch action {
+			case firewall.ActionAccept:
+				return verdict.Kind == expr.VerdictAccept
+			case firewall.ActionDrop:
+				return verdict.Kind == expr.VerdictDrop
+			}
+		}
+	}
+	return false
 }
 
 // check returns the firewall type based on common lib checks. It returns UNKNOWN if no firewall is found.
@@ -250,12 +686,12 @@ func createWorkTable() (*nftables.Table, error) {
 	}
 
 	for _, t := range tables {
-		if t.Name == tableName {
+		if t.Name == tableNameNetbird {
 			sConn.DelTable(t)
 		}
 	}
 
-	table := sConn.AddTable(&nftables.Table{Name: tableName, Family: nftables.TableFamilyIPv4})
+	table := sConn.AddTable(&nftables.Table{Name: tableNameNetbird, Family: nftables.TableFamilyIPv4})
 	err = sConn.Flush()
 
 	return table, err
@@ -273,7 +709,7 @@ func deleteWorkTable() {
 	}
 
 	for _, t := range tables {
-		if t.Name == tableName {
+		if t.Name == tableNameNetbird {
 			sConn.DelTable(t)
 		}
 	}

client/firewall/nftables/rule_linux.go

@@ -8,10 +8,11 @@ import (
 
 // Rule to handle management of rules
 type Rule struct {
-	nftRule *nftables.Rule
-	nftSet  *nftables.Set
-	ruleID  string
-	ip      net.IP
+	nftRule    *nftables.Rule
+	mangleRule *nftables.Rule
+	nftSet     *nftables.Set
+	ruleID     string
+	ip         net.IP
 }
 
 // GetRuleID returns the rule id

client/firewall/nftables/state_linux.go

@@ -0,0 +1,47 @@
+package nftables
+
+import (
+	"fmt"
+
+	"github.com/netbirdio/netbird/client/iface"
+	"github.com/netbirdio/netbird/client/iface/device"
+)
+
+type InterfaceState struct {
+	NameStr       string          `json:"name"`
+	WGAddress     iface.WGAddress `json:"wg_address"`
+	UserspaceBind bool            `json:"userspace_bind"`
+}
+
+func (i *InterfaceState) Name() string {
+	return i.NameStr
+}
+
+func (i *InterfaceState) Address() device.WGAddress {
+	return i.WGAddress
+}
+
+func (i *InterfaceState) IsUserspaceBind() bool {
+	return i.UserspaceBind
+}
+
+type ShutdownState struct {
+	InterfaceState *InterfaceState `json:"interface_state,omitempty"`
+}
+
+func (s *ShutdownState) Name() string {
+	return "nftables_state"
+}
+
+func (s *ShutdownState) Cleanup() error {
+	nft, err := Create(s.InterfaceState)
+	if err != nil {
+		return fmt.Errorf("create nftables manager: %w", err)
+	}
+
+	if err := nft.Reset(nil); err != nil {
+		return fmt.Errorf("reset nftables manager: %w", err)
+	}
+
+	return nil
+}

client/firewall/test/cases_linux.go

@@ -1,8 +1,10 @@
-//go:build !android
-
 package test
 
-import firewall "github.com/netbirdio/netbird/client/firewall/manager"
+import (
+	"net/netip"
+
+	firewall "github.com/netbirdio/netbird/client/firewall/manager"
+)
 
 var (
 	InsertRuleTestCases = []struct {
@@ -13,8 +15,8 @@ var (
 			Name: "Insert Forwarding IPV4 Rule",
 			InputPair: firewall.RouterPair{
 				ID:          "zxa",
-				Source:      "100.100.100.1/32",
-				Destination: "100.100.200.0/24",
+				Source:      netip.MustParsePrefix("100.100.100.1/32"),
+				Destination: netip.MustParsePrefix("100.100.200.0/24"),
 				Masquerade:  false,
 			},
 		},
@@ -22,8 +24,8 @@ var (
 			Name: "Insert Forwarding And Nat IPV4 Rules",
 			InputPair: firewall.RouterPair{
 				ID:          "zxa",
-				Source:      "100.100.100.1/32",
-				Destination: "100.100.200.0/24",
+				Source:      netip.MustParsePrefix("100.100.100.1/32"),
+				Destination: netip.MustParsePrefix("100.100.200.0/24"),
 				Masquerade:  true,
 			},
 		},
@@ -38,8 +40,8 @@ var (
 			Name: "Remove Forwarding And Nat IPV4 Rules",
 			InputPair: firewall.RouterPair{
 				ID:          "zxa",
-				Source:      "100.100.100.1/32",
-				Destination: "100.100.200.0/24",
+				Source:      netip.MustParsePrefix("100.100.100.1/32"),
+				Destination: netip.MustParsePrefix("100.100.200.0/24"),
 				Masquerade:  true,
 			},
 		},

client/firewall/uspfilter/allow_netbird.go

@@ -2,16 +2,53 @@
 
 package uspfilter
 
+import (
+	"context"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/firewall/uspfilter/conntrack"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
+)
+
 // Reset firewall to the default state
-func (m *Manager) Reset() error {
+func (m *Manager) Reset(stateManager *statemanager.Manager) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
 	m.outgoingRules = make(map[string]RuleSet)
 	m.incomingRules = make(map[string]RuleSet)
 
+	if m.udpTracker != nil {
+		m.udpTracker.Close()
+		m.udpTracker = conntrack.NewUDPTracker(conntrack.DefaultUDPTimeout, m.logger)
+	}
+
+	if m.icmpTracker != nil {
+		m.icmpTracker.Close()
+		m.icmpTracker = conntrack.NewICMPTracker(conntrack.DefaultICMPTimeout, m.logger)
+	}
+
+	if m.tcpTracker != nil {
+		m.tcpTracker.Close()
+		m.tcpTracker = conntrack.NewTCPTracker(conntrack.DefaultTCPTimeout, m.logger)
+	}
+
+	if m.forwarder != nil {
+		m.forwarder.Stop()
+	}
+
+	if m.logger != nil {
+		ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
+		defer cancel()
+		if err := m.logger.Stop(ctx); err != nil {
+			log.Errorf("failed to shutdown logger: %v", err)
+		}
+	}
+
 	if m.nativeFirewall != nil {
-		return m.nativeFirewall.Reset()
+		return m.nativeFirewall.Reset(stateManager)
 	}
 	return nil
 }

client/firewall/uspfilter/allow_netbird_windows.go

@@ -1,11 +1,16 @@
 package uspfilter
 
 import (
+	"context"
 	"fmt"
 	"os/exec"
 	"syscall"
+	"time"
 
 	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/firewall/uspfilter/conntrack"
+	"github.com/netbirdio/netbird/client/internal/statemanager"
 )
 
 type action string
@@ -17,13 +22,40 @@ const (
 )
 
 // Reset firewall to the default state
-func (m *Manager) Reset() error {
+func (m *Manager) Reset(*statemanager.Manager) error {
 	m.mutex.Lock()
 	defer m.mutex.Unlock()
 
 	m.outgoingRules = make(map[string]RuleSet)
 	m.incomingRules = make(map[string]RuleSet)
 
+	if m.udpTracker != nil {
+		m.udpTracker.Close()
+		m.udpTracker = conntrack.NewUDPTracker(conntrack.DefaultUDPTimeout, m.logger)
+	}
+
+	if m.icmpTracker != nil {
+		m.icmpTracker.Close()
+		m.icmpTracker = conntrack.NewICMPTracker(conntrack.DefaultICMPTimeout, m.logger)
+	}
+
+	if m.tcpTracker != nil {
+		m.tcpTracker.Close()
+		m.tcpTracker = conntrack.NewTCPTracker(conntrack.DefaultTCPTimeout, m.logger)
+	}
+
+	if m.forwarder != nil {
+		m.forwarder.Stop()
+	}
+
+	if m.logger != nil {
+		ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
+		defer cancel()
+		if err := m.logger.Stop(ctx); err != nil {
+			log.Errorf("failed to shutdown logger: %v", err)
+		}
+	}
+
 	if !isWindowsFirewallReachable() {
 		return nil
 	}
@@ -64,15 +96,18 @@ func manageFirewallRule(ruleName string, action action, extraArgs ...string) err
 	if action == addRule {
 		args = append(args, extraArgs...)
 	}
-
-	cmd := exec.Command("netsh", args...)
+	netshCmd := GetSystem32Command("netsh")
+	cmd := exec.Command(netshCmd, args...)
 	cmd.SysProcAttr = &syscall.SysProcAttr{HideWindow: true}
 	return cmd.Run()
 }
 
 func isWindowsFirewallReachable() bool {
 	args := []string{"advfirewall", "show", "allprofiles", "state"}
-	cmd := exec.Command("netsh", args...)
+
+	netshCmd := GetSystem32Command("netsh")
+
+	cmd := exec.Command(netshCmd, args...)
 	cmd.SysProcAttr = &syscall.SysProcAttr{HideWindow: true}
 
 	_, err := cmd.Output()
@@ -87,8 +122,23 @@ func isWindowsFirewallReachable() bool {
 func isFirewallRuleActive(ruleName string) bool {
 	args := []string{"advfirewall", "firewall", "show", "rule", "name=" + ruleName}
 
-	cmd := exec.Command("netsh", args...)
+	netshCmd := GetSystem32Command("netsh")
+
+	cmd := exec.Command(netshCmd, args...)
 	cmd.SysProcAttr = &syscall.SysProcAttr{HideWindow: true}
 	_, err := cmd.Output()
 	return err == nil
 }
+
+// GetSystem32Command checks if a command can be found in the system path and returns it. In case it can't find it
+// in the path it will return the full path of a command assuming C:\windows\system32 as the base path.
+func GetSystem32Command(command string) string {
+	_, err := exec.LookPath(command)
+	if err == nil {
+		return command
+	}
+
+	log.Tracef("Command %s not found in PATH, using C:\\windows\\system32\\%s.exe path", command, command)
+
+	return "C:\\windows\\system32\\" + command + ".exe"
+}

client/firewall/uspfilter/common/iface.go

@@ -0,0 +1,16 @@
+package common
+
+import (
+	wgdevice "golang.zx2c4.com/wireguard/device"
+
+	"github.com/netbirdio/netbird/client/iface"
+	"github.com/netbirdio/netbird/client/iface/device"
+)
+
+// IFaceMapper defines subset methods of interface required for manager
+type IFaceMapper interface {
+	SetFilter(device.PacketFilter) error
+	Address() iface.WGAddress
+	GetWGDevice() *wgdevice.Device
+	GetDevice() *device.FilteredDevice
+}

client/firewall/uspfilter/conntrack/common.go

@@ -0,0 +1,126 @@
+// common.go
+package conntrack
+
+import (
+	"net"
+	"sync"
+	"sync/atomic"
+	"time"
+)
+
+// BaseConnTrack provides common fields and locking for all connection types
+type BaseConnTrack struct {
+	SourceIP   net.IP
+	DestIP     net.IP
+	SourcePort uint16
+	DestPort   uint16
+	lastSeen   atomic.Int64 // Unix nano for atomic access
+}
+
+// these small methods will be inlined by the compiler
+
+// UpdateLastSeen safely updates the last seen timestamp
+func (b *BaseConnTrack) UpdateLastSeen() {
+	b.lastSeen.Store(time.Now().UnixNano())
+}
+
+// GetLastSeen safely gets the last seen timestamp
+func (b *BaseConnTrack) GetLastSeen() time.Time {
+	return time.Unix(0, b.lastSeen.Load())
+}
+
+// timeoutExceeded checks if the connection has exceeded the given timeout
+func (b *BaseConnTrack) timeoutExceeded(timeout time.Duration) bool {
+	lastSeen := time.Unix(0, b.lastSeen.Load())
+	return time.Since(lastSeen) > timeout
+}
+
+// IPAddr is a fixed-size IP address to avoid allocations
+type IPAddr [16]byte
+
+// MakeIPAddr creates an IPAddr from net.IP
+func MakeIPAddr(ip net.IP) (addr IPAddr) {
+	// Optimization: check for v4 first as it's more common
+	if ip4 := ip.To4(); ip4 != nil {
+		copy(addr[12:], ip4)
+	} else {
+		copy(addr[:], ip.To16())
+	}
+	return addr
+}
+
+// ConnKey uniquely identifies a connection
+type ConnKey struct {
+	SrcIP   IPAddr
+	DstIP   IPAddr
+	SrcPort uint16
+	DstPort uint16
+}
+
+// makeConnKey creates a connection key
+func makeConnKey(srcIP net.IP, dstIP net.IP, srcPort uint16, dstPort uint16) ConnKey {
+	return ConnKey{
+		SrcIP:   MakeIPAddr(srcIP),
+		DstIP:   MakeIPAddr(dstIP),
+		SrcPort: srcPort,
+		DstPort: dstPort,
+	}
+}
+
+// ValidateIPs checks if IPs match without allocation
+func ValidateIPs(connIP IPAddr, pktIP net.IP) bool {
+	if ip4 := pktIP.To4(); ip4 != nil {
+		// Compare IPv4 addresses (last 4 bytes)
+		for i := 0; i < 4; i++ {
+			if connIP[12+i] != ip4[i] {
+				return false
+			}
+		}
+		return true
+	}
+	// Compare full IPv6 addresses
+	ip6 := pktIP.To16()
+	for i := 0; i < 16; i++ {
+		if connIP[i] != ip6[i] {
+			return false
+		}
+	}
+	return true
+}
+
+// PreallocatedIPs is a pool of IP byte slices to reduce allocations
+type PreallocatedIPs struct {
+	sync.Pool
+}
+
+// NewPreallocatedIPs creates a new IP pool
+func NewPreallocatedIPs() *PreallocatedIPs {
+	return &PreallocatedIPs{
+		Pool: sync.Pool{
+			New: func() interface{} {
+				ip := make(net.IP, 16)
+				return &ip
+			},
+		},
+	}
+}
+
+// Get retrieves an IP from the pool
+func (p *PreallocatedIPs) Get() net.IP {
+	return *p.Pool.Get().(*net.IP)
+}
+
+// Put returns an IP to the pool
+func (p *PreallocatedIPs) Put(ip net.IP) {
+	p.Pool.Put(&ip)
+}
+
+// copyIP copies an IP address efficiently
+func copyIP(dst, src net.IP) {
+	if len(src) == 16 {
+		copy(dst, src)
+	} else {
+		// Handle IPv4
+		copy(dst[12:], src.To4())
+	}
+}

client/firewall/uspfilter/conntrack/common_test.go

@@ -0,0 +1,95 @@
+package conntrack
+
+import (
+	"net"
+	"testing"
+
+	"github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/firewall/uspfilter/log"
+)
+
+var logger = log.NewFromLogrus(logrus.StandardLogger())
+
+func BenchmarkIPOperations(b *testing.B) {
+	b.Run("MakeIPAddr", func(b *testing.B) {
+		ip := net.ParseIP("192.168.1.1")
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			_ = MakeIPAddr(ip)
+		}
+	})
+
+	b.Run("ValidateIPs", func(b *testing.B) {
+		ip1 := net.ParseIP("192.168.1.1")
+		ip2 := net.ParseIP("192.168.1.1")
+		addr := MakeIPAddr(ip1)
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			_ = ValidateIPs(addr, ip2)
+		}
+	})
+
+	b.Run("IPPool", func(b *testing.B) {
+		pool := NewPreallocatedIPs()
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			ip := pool.Get()
+			pool.Put(ip)
+		}
+	})
+
+}
+
+// Memory pressure tests
+func BenchmarkMemoryPressure(b *testing.B) {
+	b.Run("TCPHighLoad", func(b *testing.B) {
+		tracker := NewTCPTracker(DefaultTCPTimeout, logger)
+		defer tracker.Close()
+
+		// Generate different IPs
+		srcIPs := make([]net.IP, 100)
+		dstIPs := make([]net.IP, 100)
+		for i := 0; i < 100; i++ {
+			srcIPs[i] = net.IPv4(192, 168, byte(i/256), byte(i%256))
+			dstIPs[i] = net.IPv4(10, 0, byte(i/256), byte(i%256))
+		}
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			srcIdx := i % len(srcIPs)
+			dstIdx := (i + 1) % len(dstIPs)
+			tracker.TrackOutbound(srcIPs[srcIdx], dstIPs[dstIdx], uint16(i%65535), 80, TCPSyn)
+
+			// Simulate some valid inbound packets
+			if i%3 == 0 {
+				tracker.IsValidInbound(dstIPs[dstIdx], srcIPs[srcIdx], 80, uint16(i%65535), TCPAck)
+			}
+		}
+	})
+
+	b.Run("UDPHighLoad", func(b *testing.B) {
+		tracker := NewUDPTracker(DefaultUDPTimeout, logger)
+		defer tracker.Close()
+
+		// Generate different IPs
+		srcIPs := make([]net.IP, 100)
+		dstIPs := make([]net.IP, 100)
+		for i := 0; i < 100; i++ {
+			srcIPs[i] = net.IPv4(192, 168, byte(i/256), byte(i%256))
+			dstIPs[i] = net.IPv4(10, 0, byte(i/256), byte(i%256))
+		}
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			srcIdx := i % len(srcIPs)
+			dstIdx := (i + 1) % len(dstIPs)
+			tracker.TrackOutbound(srcIPs[srcIdx], dstIPs[dstIdx], uint16(i%65535), 80)
+
+			// Simulate some valid inbound packets
+			if i%3 == 0 {
+				tracker.IsValidInbound(dstIPs[dstIdx], srcIPs[srcIdx], 80, uint16(i%65535))
+			}
+		}
+	})
+}

client/firewall/uspfilter/conntrack/icmp.go

@@ -0,0 +1,169 @@
+package conntrack
+
+import (
+	"net"
+	"sync"
+	"time"
+
+	"github.com/google/gopacket/layers"
+
+	nblog "github.com/netbirdio/netbird/client/firewall/uspfilter/log"
+)
+
+const (
+	// DefaultICMPTimeout is the default timeout for ICMP connections
+	DefaultICMPTimeout = 30 * time.Second
+	// ICMPCleanupInterval is how often we check for stale ICMP connections
+	ICMPCleanupInterval = 15 * time.Second
+)
+
+// ICMPConnKey uniquely identifies an ICMP connection
+type ICMPConnKey struct {
+	// Supports both IPv4 and IPv6
+	SrcIP    [16]byte
+	DstIP    [16]byte
+	Sequence uint16 // ICMP sequence number
+	ID       uint16 // ICMP identifier
+}
+
+// ICMPConnTrack represents an ICMP connection state
+type ICMPConnTrack struct {
+	BaseConnTrack
+	Sequence uint16
+	ID       uint16
+}
+
+// ICMPTracker manages ICMP connection states
+type ICMPTracker struct {
+	logger        *nblog.Logger
+	connections   map[ICMPConnKey]*ICMPConnTrack
+	timeout       time.Duration
+	cleanupTicker *time.Ticker
+	mutex         sync.RWMutex
+	done          chan struct{}
+	ipPool        *PreallocatedIPs
+}
+
+// NewICMPTracker creates a new ICMP connection tracker
+func NewICMPTracker(timeout time.Duration, logger *nblog.Logger) *ICMPTracker {
+	if timeout == 0 {
+		timeout = DefaultICMPTimeout
+	}
+
+	tracker := &ICMPTracker{
+		logger:        logger,
+		connections:   make(map[ICMPConnKey]*ICMPConnTrack),
+		timeout:       timeout,
+		cleanupTicker: time.NewTicker(ICMPCleanupInterval),
+		done:          make(chan struct{}),
+		ipPool:        NewPreallocatedIPs(),
+	}
+
+	go tracker.cleanupRoutine()
+	return tracker
+}
+
+// TrackOutbound records an outbound ICMP Echo Request
+func (t *ICMPTracker) TrackOutbound(srcIP net.IP, dstIP net.IP, id uint16, seq uint16) {
+	key := makeICMPKey(srcIP, dstIP, id, seq)
+
+	t.mutex.Lock()
+	conn, exists := t.connections[key]
+	if !exists {
+		srcIPCopy := t.ipPool.Get()
+		dstIPCopy := t.ipPool.Get()
+		copyIP(srcIPCopy, srcIP)
+		copyIP(dstIPCopy, dstIP)
+
+		conn = &ICMPConnTrack{
+			BaseConnTrack: BaseConnTrack{
+				SourceIP: srcIPCopy,
+				DestIP:   dstIPCopy,
+			},
+			ID:       id,
+			Sequence: seq,
+		}
+		conn.UpdateLastSeen()
+		t.connections[key] = conn
+
+		t.logger.Trace("New ICMP connection %v", key)
+	}
+	t.mutex.Unlock()
+
+	conn.UpdateLastSeen()
+}
+
+// IsValidInbound checks if an inbound ICMP Echo Reply matches a tracked request
+func (t *ICMPTracker) IsValidInbound(srcIP net.IP, dstIP net.IP, id uint16, seq uint16, icmpType uint8) bool {
+	if icmpType != uint8(layers.ICMPv4TypeEchoReply) {
+		return false
+	}
+
+	key := makeICMPKey(dstIP, srcIP, id, seq)
+
+	t.mutex.RLock()
+	conn, exists := t.connections[key]
+	t.mutex.RUnlock()
+
+	if !exists {
+		return false
+	}
+
+	if conn.timeoutExceeded(t.timeout) {
+		return false
+	}
+
+	return ValidateIPs(MakeIPAddr(srcIP), conn.DestIP) &&
+		ValidateIPs(MakeIPAddr(dstIP), conn.SourceIP) &&
+		conn.ID == id &&
+		conn.Sequence == seq
+}
+
+func (t *ICMPTracker) cleanupRoutine() {
+	for {
+		select {
+		case <-t.cleanupTicker.C:
+			t.cleanup()
+		case <-t.done:
+			return
+		}
+	}
+}
+func (t *ICMPTracker) cleanup() {
+	t.mutex.Lock()
+	defer t.mutex.Unlock()
+
+	for key, conn := range t.connections {
+		if conn.timeoutExceeded(t.timeout) {
+			t.ipPool.Put(conn.SourceIP)
+			t.ipPool.Put(conn.DestIP)
+			delete(t.connections, key)
+
+			t.logger.Debug("Removed ICMP connection %v (timeout)", key)
+		}
+	}
+}
+
+// Close stops the cleanup routine and releases resources
+func (t *ICMPTracker) Close() {
+	t.cleanupTicker.Stop()
+	close(t.done)
+
+	t.mutex.Lock()
+	for _, conn := range t.connections {
+		t.ipPool.Put(conn.SourceIP)
+		t.ipPool.Put(conn.DestIP)
+	}
+	t.connections = nil
+	t.mutex.Unlock()
+}
+
+// makeICMPKey creates an ICMP connection key
+func makeICMPKey(srcIP net.IP, dstIP net.IP, id uint16, seq uint16) ICMPConnKey {
+	return ICMPConnKey{
+		SrcIP:    MakeIPAddr(srcIP),
+		DstIP:    MakeIPAddr(dstIP),
+		ID:       id,
+		Sequence: seq,
+	}
+}

client/firewall/uspfilter/conntrack/icmp_test.go

@@ -0,0 +1,39 @@
+package conntrack
+
+import (
+	"net"
+	"testing"
+)
+
+func BenchmarkICMPTracker(b *testing.B) {
+	b.Run("TrackOutbound", func(b *testing.B) {
+		tracker := NewICMPTracker(DefaultICMPTimeout, logger)
+		defer tracker.Close()
+
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			tracker.TrackOutbound(srcIP, dstIP, uint16(i%65535), uint16(i%65535))
+		}
+	})
+
+	b.Run("IsValidInbound", func(b *testing.B) {
+		tracker := NewICMPTracker(DefaultICMPTimeout, logger)
+		defer tracker.Close()
+
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+
+		// Pre-populate some connections
+		for i := 0; i < 1000; i++ {
+			tracker.TrackOutbound(srcIP, dstIP, uint16(i), uint16(i))
+		}
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			tracker.IsValidInbound(dstIP, srcIP, uint16(i%1000), uint16(i%1000), 0)
+		}
+	})
+}

client/firewall/uspfilter/conntrack/tcp.go

@@ -0,0 +1,383 @@
+package conntrack
+
+// TODO: Send RST packets for invalid/timed-out connections
+
+import (
+	"net"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	nblog "github.com/netbirdio/netbird/client/firewall/uspfilter/log"
+)
+
+const (
+	// MSL (Maximum Segment Lifetime) is typically 2 minutes
+	MSL = 2 * time.Minute
+	// TimeWaitTimeout (TIME-WAIT) should last 2*MSL
+	TimeWaitTimeout = 2 * MSL
+)
+
+const (
+	TCPSyn  uint8 = 0x02
+	TCPAck  uint8 = 0x10
+	TCPFin  uint8 = 0x01
+	TCPRst  uint8 = 0x04
+	TCPPush uint8 = 0x08
+	TCPUrg  uint8 = 0x20
+)
+
+const (
+	// DefaultTCPTimeout is the default timeout for established TCP connections
+	DefaultTCPTimeout = 3 * time.Hour
+	// TCPHandshakeTimeout is timeout for TCP handshake completion
+	TCPHandshakeTimeout = 60 * time.Second
+	// TCPCleanupInterval is how often we check for stale connections
+	TCPCleanupInterval = 5 * time.Minute
+)
+
+// TCPState represents the state of a TCP connection
+type TCPState int
+
+const (
+	TCPStateNew TCPState = iota
+	TCPStateSynSent
+	TCPStateSynReceived
+	TCPStateEstablished
+	TCPStateFinWait1
+	TCPStateFinWait2
+	TCPStateClosing
+	TCPStateTimeWait
+	TCPStateCloseWait
+	TCPStateLastAck
+	TCPStateClosed
+)
+
+// TCPConnKey uniquely identifies a TCP connection
+type TCPConnKey struct {
+	SrcIP   [16]byte
+	DstIP   [16]byte
+	SrcPort uint16
+	DstPort uint16
+}
+
+// TCPConnTrack represents a TCP connection state
+type TCPConnTrack struct {
+	BaseConnTrack
+	State       TCPState
+	established atomic.Bool
+	sync.RWMutex
+}
+
+// IsEstablished safely checks if connection is established
+func (t *TCPConnTrack) IsEstablished() bool {
+	return t.established.Load()
+}
+
+// SetEstablished safely sets the established state
+func (t *TCPConnTrack) SetEstablished(state bool) {
+	t.established.Store(state)
+}
+
+// TCPTracker manages TCP connection states
+type TCPTracker struct {
+	logger        *nblog.Logger
+	connections   map[ConnKey]*TCPConnTrack
+	mutex         sync.RWMutex
+	cleanupTicker *time.Ticker
+	done          chan struct{}
+	timeout       time.Duration
+	ipPool        *PreallocatedIPs
+}
+
+// NewTCPTracker creates a new TCP connection tracker
+func NewTCPTracker(timeout time.Duration, logger *nblog.Logger) *TCPTracker {
+	tracker := &TCPTracker{
+		logger:        logger,
+		connections:   make(map[ConnKey]*TCPConnTrack),
+		cleanupTicker: time.NewTicker(TCPCleanupInterval),
+		done:          make(chan struct{}),
+		timeout:       timeout,
+		ipPool:        NewPreallocatedIPs(),
+	}
+
+	go tracker.cleanupRoutine()
+	return tracker
+}
+
+// TrackOutbound processes an outbound TCP packet and updates connection state
+func (t *TCPTracker) TrackOutbound(srcIP net.IP, dstIP net.IP, srcPort uint16, dstPort uint16, flags uint8) {
+	// Create key before lock
+	key := makeConnKey(srcIP, dstIP, srcPort, dstPort)
+
+	t.mutex.Lock()
+	conn, exists := t.connections[key]
+	if !exists {
+		// Use preallocated IPs
+		srcIPCopy := t.ipPool.Get()
+		dstIPCopy := t.ipPool.Get()
+		copyIP(srcIPCopy, srcIP)
+		copyIP(dstIPCopy, dstIP)
+
+		conn = &TCPConnTrack{
+			BaseConnTrack: BaseConnTrack{
+				SourceIP:   srcIPCopy,
+				DestIP:     dstIPCopy,
+				SourcePort: srcPort,
+				DestPort:   dstPort,
+			},
+			State: TCPStateNew,
+		}
+		conn.UpdateLastSeen()
+		conn.established.Store(false)
+		t.connections[key] = conn
+
+		t.logger.Trace("New TCP connection: %s:%d -> %s:%d", srcIP, srcPort, dstIP, dstPort)
+	}
+	t.mutex.Unlock()
+
+	// Lock individual connection for state update
+	conn.Lock()
+	t.updateState(conn, flags, true)
+	conn.Unlock()
+	conn.UpdateLastSeen()
+}
+
+// IsValidInbound checks if an inbound TCP packet matches a tracked connection
+func (t *TCPTracker) IsValidInbound(srcIP net.IP, dstIP net.IP, srcPort uint16, dstPort uint16, flags uint8) bool {
+	if !isValidFlagCombination(flags) {
+		return false
+	}
+
+	key := makeConnKey(dstIP, srcIP, dstPort, srcPort)
+
+	t.mutex.RLock()
+	conn, exists := t.connections[key]
+	t.mutex.RUnlock()
+
+	if !exists {
+		return false
+	}
+
+	// Handle RST packets
+	if flags&TCPRst != 0 {
+		conn.Lock()
+		if conn.IsEstablished() || conn.State == TCPStateSynSent || conn.State == TCPStateSynReceived {
+			conn.State = TCPStateClosed
+			conn.SetEstablished(false)
+			conn.Unlock()
+			return true
+		}
+		conn.Unlock()
+		return false
+	}
+
+	conn.Lock()
+	t.updateState(conn, flags, false)
+	conn.UpdateLastSeen()
+	isEstablished := conn.IsEstablished()
+	isValidState := t.isValidStateForFlags(conn.State, flags)
+	conn.Unlock()
+
+	return isEstablished || isValidState
+}
+
+// updateState updates the TCP connection state based on flags
+func (t *TCPTracker) updateState(conn *TCPConnTrack, flags uint8, isOutbound bool) {
+	// Handle RST flag specially - it always causes transition to closed
+	if flags&TCPRst != 0 {
+		conn.State = TCPStateClosed
+		conn.SetEstablished(false)
+
+		t.logger.Trace("TCP connection reset: %s:%d -> %s:%d",
+			conn.SourceIP, conn.SourcePort, conn.DestIP, conn.DestPort)
+		return
+	}
+
+	switch conn.State {
+	case TCPStateNew:
+		if flags&TCPSyn != 0 && flags&TCPAck == 0 {
+			conn.State = TCPStateSynSent
+		}
+
+	case TCPStateSynSent:
+		if flags&TCPSyn != 0 && flags&TCPAck != 0 {
+			if isOutbound {
+				conn.State = TCPStateSynReceived
+			} else {
+				// Simultaneous open
+				conn.State = TCPStateEstablished
+				conn.SetEstablished(true)
+			}
+		}
+
+	case TCPStateSynReceived:
+		if flags&TCPAck != 0 && flags&TCPSyn == 0 {
+			conn.State = TCPStateEstablished
+			conn.SetEstablished(true)
+		}
+
+	case TCPStateEstablished:
+		if flags&TCPFin != 0 {
+			if isOutbound {
+				conn.State = TCPStateFinWait1
+			} else {
+				conn.State = TCPStateCloseWait
+			}
+			conn.SetEstablished(false)
+		}
+
+	case TCPStateFinWait1:
+		switch {
+		case flags&TCPFin != 0 && flags&TCPAck != 0:
+			// Simultaneous close - both sides sent FIN
+			conn.State = TCPStateClosing
+		case flags&TCPFin != 0:
+			conn.State = TCPStateFinWait2
+		case flags&TCPAck != 0:
+			conn.State = TCPStateFinWait2
+		}
+
+	case TCPStateFinWait2:
+		if flags&TCPFin != 0 {
+			conn.State = TCPStateTimeWait
+		}
+
+	case TCPStateClosing:
+		if flags&TCPAck != 0 {
+			conn.State = TCPStateTimeWait
+			// Keep established = false from previous state
+
+			t.logger.Trace("TCP connection closed (simultaneous) - %s:%d -> %s:%d",
+				conn.SourceIP, conn.SourcePort, conn.DestIP, conn.DestPort)
+		}
+
+	case TCPStateCloseWait:
+		if flags&TCPFin != 0 {
+			conn.State = TCPStateLastAck
+		}
+
+	case TCPStateLastAck:
+		if flags&TCPAck != 0 {
+			conn.State = TCPStateClosed
+
+			t.logger.Trace("TCP connection gracefully closed: %s:%d -> %s:%d",
+				conn.SourceIP, conn.SourcePort, conn.DestIP, conn.DestPort)
+		}
+
+	case TCPStateTimeWait:
+		// Stay in TIME-WAIT for 2MSL before transitioning to closed
+		// This is handled by the cleanup routine
+
+		t.logger.Trace("TCP connection completed - %s:%d -> %s:%d",
+			conn.SourceIP, conn.SourcePort, conn.DestIP, conn.DestPort)
+	}
+}
+
+// isValidStateForFlags checks if the TCP flags are valid for the current connection state
+func (t *TCPTracker) isValidStateForFlags(state TCPState, flags uint8) bool {
+	if !isValidFlagCombination(flags) {
+		return false
+	}
+
+	switch state {
+	case TCPStateNew:
+		return flags&TCPSyn != 0 && flags&TCPAck == 0
+	case TCPStateSynSent:
+		return flags&TCPSyn != 0 && flags&TCPAck != 0
+	case TCPStateSynReceived:
+		return flags&TCPAck != 0
+	case TCPStateEstablished:
+		if flags&TCPRst != 0 {
+			return true
+		}
+		return flags&TCPAck != 0
+	case TCPStateFinWait1:
+		return flags&TCPFin != 0 || flags&TCPAck != 0
+	case TCPStateFinWait2:
+		return flags&TCPFin != 0 || flags&TCPAck != 0
+	case TCPStateClosing:
+		// In CLOSING state, we should accept the final ACK
+		return flags&TCPAck != 0
+	case TCPStateTimeWait:
+		// In TIME_WAIT, we might see retransmissions
+		return flags&TCPAck != 0
+	case TCPStateCloseWait:
+		return flags&TCPFin != 0 || flags&TCPAck != 0
+	case TCPStateLastAck:
+		return flags&TCPAck != 0
+	case TCPStateClosed:
+		// Accept retransmitted ACKs in closed state
+		// This is important because the final ACK might be lost
+		// and the peer will retransmit their FIN-ACK
+		return flags&TCPAck != 0
+	}
+	return false
+}
+
+func (t *TCPTracker) cleanupRoutine() {
+	for {
+		select {
+		case <-t.cleanupTicker.C:
+			t.cleanup()
+		case <-t.done:
+			return
+		}
+	}
+}
+
+func (t *TCPTracker) cleanup() {
+	t.mutex.Lock()
+	defer t.mutex.Unlock()
+
+	for key, conn := range t.connections {
+		var timeout time.Duration
+		switch {
+		case conn.State == TCPStateTimeWait:
+			timeout = TimeWaitTimeout
+		case conn.IsEstablished():
+			timeout = t.timeout
+		default:
+			timeout = TCPHandshakeTimeout
+		}
+
+		lastSeen := conn.GetLastSeen()
+		if time.Since(lastSeen) > timeout {
+			// Return IPs to pool
+			t.ipPool.Put(conn.SourceIP)
+			t.ipPool.Put(conn.DestIP)
+			delete(t.connections, key)
+
+			t.logger.Trace("Cleaned up TCP connection: %s:%d -> %s:%d", conn.SourceIP, conn.SourcePort, conn.DestIP, conn.DestPort)
+		}
+	}
+}
+
+// Close stops the cleanup routine and releases resources
+func (t *TCPTracker) Close() {
+	t.cleanupTicker.Stop()
+	close(t.done)
+
+	// Clean up all remaining IPs
+	t.mutex.Lock()
+	for _, conn := range t.connections {
+		t.ipPool.Put(conn.SourceIP)
+		t.ipPool.Put(conn.DestIP)
+	}
+	t.connections = nil
+	t.mutex.Unlock()
+}
+
+func isValidFlagCombination(flags uint8) bool {
+	// Invalid: SYN+FIN
+	if flags&TCPSyn != 0 && flags&TCPFin != 0 {
+		return false
+	}
+
+	// Invalid: RST with SYN or FIN
+	if flags&TCPRst != 0 && (flags&TCPSyn != 0 || flags&TCPFin != 0) {
+		return false
+	}
+
+	return true
+}

client/firewall/uspfilter/conntrack/tcp_test.go

@@ -0,0 +1,308 @@
+package conntrack
+
+import (
+	"net"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/require"
+)
+
+func TestTCPStateMachine(t *testing.T) {
+	tracker := NewTCPTracker(DefaultTCPTimeout, logger)
+	defer tracker.Close()
+
+	srcIP := net.ParseIP("100.64.0.1")
+	dstIP := net.ParseIP("100.64.0.2")
+	srcPort := uint16(12345)
+	dstPort := uint16(80)
+
+	t.Run("Security Tests", func(t *testing.T) {
+		tests := []struct {
+			name     string
+			flags    uint8
+			wantDrop bool
+			desc     string
+		}{
+			{
+				name:     "Block unsolicited SYN-ACK",
+				flags:    TCPSyn | TCPAck,
+				wantDrop: true,
+				desc:     "Should block SYN-ACK without prior SYN",
+			},
+			{
+				name:     "Block invalid SYN-FIN",
+				flags:    TCPSyn | TCPFin,
+				wantDrop: true,
+				desc:     "Should block invalid SYN-FIN combination",
+			},
+			{
+				name:     "Block unsolicited RST",
+				flags:    TCPRst,
+				wantDrop: true,
+				desc:     "Should block RST without connection",
+			},
+			{
+				name:     "Block unsolicited ACK",
+				flags:    TCPAck,
+				wantDrop: true,
+				desc:     "Should block ACK without connection",
+			},
+			{
+				name:     "Block data without connection",
+				flags:    TCPAck | TCPPush,
+				wantDrop: true,
+				desc:     "Should block data without established connection",
+			},
+		}
+
+		for _, tt := range tests {
+			t.Run(tt.name, func(t *testing.T) {
+				isValid := tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, tt.flags)
+				require.Equal(t, !tt.wantDrop, isValid, tt.desc)
+			})
+		}
+	})
+
+	t.Run("Connection Flow Tests", func(t *testing.T) {
+		tests := []struct {
+			name string
+			test func(*testing.T)
+			desc string
+		}{
+			{
+				name: "Normal Handshake",
+				test: func(t *testing.T) {
+					t.Helper()
+
+					// Send initial SYN
+					tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPSyn)
+
+					// Receive SYN-ACK
+					valid := tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPSyn|TCPAck)
+					require.True(t, valid, "SYN-ACK should be allowed")
+
+					// Send ACK
+					tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPAck)
+
+					// Test data transfer
+					valid = tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPPush|TCPAck)
+					require.True(t, valid, "Data should be allowed after handshake")
+				},
+			},
+			{
+				name: "Normal Close",
+				test: func(t *testing.T) {
+					t.Helper()
+
+					// First establish connection
+					establishConnection(t, tracker, srcIP, dstIP, srcPort, dstPort)
+
+					// Send FIN
+					tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPFin|TCPAck)
+
+					// Receive ACK for FIN
+					valid := tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPAck)
+					require.True(t, valid, "ACK for FIN should be allowed")
+
+					// Receive FIN from other side
+					valid = tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPFin|TCPAck)
+					require.True(t, valid, "FIN should be allowed")
+
+					// Send final ACK
+					tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPAck)
+				},
+			},
+			{
+				name: "RST During Connection",
+				test: func(t *testing.T) {
+					t.Helper()
+
+					// First establish connection
+					establishConnection(t, tracker, srcIP, dstIP, srcPort, dstPort)
+
+					// Receive RST
+					valid := tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPRst)
+					require.True(t, valid, "RST should be allowed for established connection")
+
+					// Connection is logically dead but we don't enforce blocking subsequent packets
+					// The connection will be cleaned up by timeout
+				},
+			},
+			{
+				name: "Simultaneous Close",
+				test: func(t *testing.T) {
+					t.Helper()
+
+					// First establish connection
+					establishConnection(t, tracker, srcIP, dstIP, srcPort, dstPort)
+
+					// Both sides send FIN+ACK
+					tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPFin|TCPAck)
+					valid := tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPFin|TCPAck)
+					require.True(t, valid, "Simultaneous FIN should be allowed")
+
+					// Both sides send final ACK
+					tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPAck)
+					valid = tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPAck)
+					require.True(t, valid, "Final ACKs should be allowed")
+				},
+			},
+		}
+
+		for _, tt := range tests {
+			t.Run(tt.name, func(t *testing.T) {
+				t.Helper()
+
+				tracker = NewTCPTracker(DefaultTCPTimeout, logger)
+				tt.test(t)
+			})
+		}
+	})
+}
+
+func TestRSTHandling(t *testing.T) {
+	tracker := NewTCPTracker(DefaultTCPTimeout, logger)
+	defer tracker.Close()
+
+	srcIP := net.ParseIP("100.64.0.1")
+	dstIP := net.ParseIP("100.64.0.2")
+	srcPort := uint16(12345)
+	dstPort := uint16(80)
+
+	tests := []struct {
+		name       string
+		setupState func()
+		sendRST    func()
+		wantValid  bool
+		desc       string
+	}{
+		{
+			name: "RST in established",
+			setupState: func() {
+				// Establish connection first
+				tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPSyn)
+				tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPSyn|TCPAck)
+				tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPAck)
+			},
+			sendRST: func() {
+				tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPRst)
+			},
+			wantValid: true,
+			desc:      "Should accept RST for established connection",
+		},
+		{
+			name:       "RST without connection",
+			setupState: func() {},
+			sendRST: func() {
+				tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPRst)
+			},
+			wantValid: false,
+			desc:      "Should reject RST without connection",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tt.setupState()
+			tt.sendRST()
+
+			// Verify connection state is as expected
+			key := makeConnKey(srcIP, dstIP, srcPort, dstPort)
+			conn := tracker.connections[key]
+			if tt.wantValid {
+				require.NotNil(t, conn)
+				require.Equal(t, TCPStateClosed, conn.State)
+				require.False(t, conn.IsEstablished())
+			}
+		})
+	}
+}
+
+// Helper to establish a TCP connection
+func establishConnection(t *testing.T, tracker *TCPTracker, srcIP, dstIP net.IP, srcPort, dstPort uint16) {
+	t.Helper()
+
+	tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPSyn)
+
+	valid := tracker.IsValidInbound(dstIP, srcIP, dstPort, srcPort, TCPSyn|TCPAck)
+	require.True(t, valid, "SYN-ACK should be allowed")
+
+	tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort, TCPAck)
+}
+
+func BenchmarkTCPTracker(b *testing.B) {
+	b.Run("TrackOutbound", func(b *testing.B) {
+		tracker := NewTCPTracker(DefaultTCPTimeout, logger)
+		defer tracker.Close()
+
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			tracker.TrackOutbound(srcIP, dstIP, uint16(i%65535), 80, TCPSyn)
+		}
+	})
+
+	b.Run("IsValidInbound", func(b *testing.B) {
+		tracker := NewTCPTracker(DefaultTCPTimeout, logger)
+		defer tracker.Close()
+
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+
+		// Pre-populate some connections
+		for i := 0; i < 1000; i++ {
+			tracker.TrackOutbound(srcIP, dstIP, uint16(i), 80, TCPSyn)
+		}
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			tracker.IsValidInbound(dstIP, srcIP, 80, uint16(i%1000), TCPAck)
+		}
+	})
+
+	b.Run("ConcurrentAccess", func(b *testing.B) {
+		tracker := NewTCPTracker(DefaultTCPTimeout, logger)
+		defer tracker.Close()
+
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+
+		b.RunParallel(func(pb *testing.PB) {
+			i := 0
+			for pb.Next() {
+				if i%2 == 0 {
+					tracker.TrackOutbound(srcIP, dstIP, uint16(i%65535), 80, TCPSyn)
+				} else {
+					tracker.IsValidInbound(dstIP, srcIP, 80, uint16(i%65535), TCPAck)
+				}
+				i++
+			}
+		})
+	})
+}
+
+// Benchmark connection cleanup
+func BenchmarkCleanup(b *testing.B) {
+	b.Run("TCPCleanup", func(b *testing.B) {
+		tracker := NewTCPTracker(100*time.Millisecond, logger) // Short timeout for testing
+		defer tracker.Close()
+
+		// Pre-populate with expired connections
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+		for i := 0; i < 10000; i++ {
+			tracker.TrackOutbound(srcIP, dstIP, uint16(i), 80, TCPSyn)
+		}
+
+		// Wait for connections to expire
+		time.Sleep(200 * time.Millisecond)
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			tracker.cleanup()
+		}
+	})
+}

client/firewall/uspfilter/conntrack/udp.go

@@ -0,0 +1,163 @@
+package conntrack
+
+import (
+	"net"
+	"sync"
+	"time"
+
+	nblog "github.com/netbirdio/netbird/client/firewall/uspfilter/log"
+)
+
+const (
+	// DefaultUDPTimeout is the default timeout for UDP connections
+	DefaultUDPTimeout = 30 * time.Second
+	// UDPCleanupInterval is how often we check for stale connections
+	UDPCleanupInterval = 15 * time.Second
+)
+
+// UDPConnTrack represents a UDP connection state
+type UDPConnTrack struct {
+	BaseConnTrack
+}
+
+// UDPTracker manages UDP connection states
+type UDPTracker struct {
+	logger        *nblog.Logger
+	connections   map[ConnKey]*UDPConnTrack
+	timeout       time.Duration
+	cleanupTicker *time.Ticker
+	mutex         sync.RWMutex
+	done          chan struct{}
+	ipPool        *PreallocatedIPs
+}
+
+// NewUDPTracker creates a new UDP connection tracker
+func NewUDPTracker(timeout time.Duration, logger *nblog.Logger) *UDPTracker {
+	if timeout == 0 {
+		timeout = DefaultUDPTimeout
+	}
+
+	tracker := &UDPTracker{
+		logger:        logger,
+		connections:   make(map[ConnKey]*UDPConnTrack),
+		timeout:       timeout,
+		cleanupTicker: time.NewTicker(UDPCleanupInterval),
+		done:          make(chan struct{}),
+		ipPool:        NewPreallocatedIPs(),
+	}
+
+	go tracker.cleanupRoutine()
+	return tracker
+}
+
+// TrackOutbound records an outbound UDP connection
+func (t *UDPTracker) TrackOutbound(srcIP net.IP, dstIP net.IP, srcPort uint16, dstPort uint16) {
+	key := makeConnKey(srcIP, dstIP, srcPort, dstPort)
+
+	t.mutex.Lock()
+	conn, exists := t.connections[key]
+	if !exists {
+		srcIPCopy := t.ipPool.Get()
+		dstIPCopy := t.ipPool.Get()
+		copyIP(srcIPCopy, srcIP)
+		copyIP(dstIPCopy, dstIP)
+
+		conn = &UDPConnTrack{
+			BaseConnTrack: BaseConnTrack{
+				SourceIP:   srcIPCopy,
+				DestIP:     dstIPCopy,
+				SourcePort: srcPort,
+				DestPort:   dstPort,
+			},
+		}
+		conn.UpdateLastSeen()
+		t.connections[key] = conn
+
+		t.logger.Trace("New UDP connection: %v", conn)
+	}
+	t.mutex.Unlock()
+
+	conn.UpdateLastSeen()
+}
+
+// IsValidInbound checks if an inbound packet matches a tracked connection
+func (t *UDPTracker) IsValidInbound(srcIP net.IP, dstIP net.IP, srcPort uint16, dstPort uint16) bool {
+	key := makeConnKey(dstIP, srcIP, dstPort, srcPort)
+
+	t.mutex.RLock()
+	conn, exists := t.connections[key]
+	t.mutex.RUnlock()
+
+	if !exists {
+		return false
+	}
+
+	if conn.timeoutExceeded(t.timeout) {
+		return false
+	}
+
+	return ValidateIPs(MakeIPAddr(srcIP), conn.DestIP) &&
+		ValidateIPs(MakeIPAddr(dstIP), conn.SourceIP) &&
+		conn.DestPort == srcPort &&
+		conn.SourcePort == dstPort
+}
+
+// cleanupRoutine periodically removes stale connections
+func (t *UDPTracker) cleanupRoutine() {
+	for {
+		select {
+		case <-t.cleanupTicker.C:
+			t.cleanup()
+		case <-t.done:
+			return
+		}
+	}
+}
+
+func (t *UDPTracker) cleanup() {
+	t.mutex.Lock()
+	defer t.mutex.Unlock()
+
+	for key, conn := range t.connections {
+		if conn.timeoutExceeded(t.timeout) {
+			t.ipPool.Put(conn.SourceIP)
+			t.ipPool.Put(conn.DestIP)
+			delete(t.connections, key)
+
+			t.logger.Trace("Removed UDP connection %v (timeout)", conn)
+		}
+	}
+}
+
+// Close stops the cleanup routine and releases resources
+func (t *UDPTracker) Close() {
+	t.cleanupTicker.Stop()
+	close(t.done)
+
+	t.mutex.Lock()
+	for _, conn := range t.connections {
+		t.ipPool.Put(conn.SourceIP)
+		t.ipPool.Put(conn.DestIP)
+	}
+	t.connections = nil
+	t.mutex.Unlock()
+}
+
+// GetConnection safely retrieves a connection state
+func (t *UDPTracker) GetConnection(srcIP net.IP, srcPort uint16, dstIP net.IP, dstPort uint16) (*UDPConnTrack, bool) {
+	t.mutex.RLock()
+	defer t.mutex.RUnlock()
+
+	key := makeConnKey(srcIP, dstIP, srcPort, dstPort)
+	conn, exists := t.connections[key]
+	if !exists {
+		return nil, false
+	}
+
+	return conn, true
+}
+
+// Timeout returns the configured timeout duration for the tracker
+func (t *UDPTracker) Timeout() time.Duration {
+	return t.timeout
+}

client/firewall/uspfilter/conntrack/udp_test.go

@@ -0,0 +1,243 @@
+package conntrack
+
+import (
+	"net"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestNewUDPTracker(t *testing.T) {
+	tests := []struct {
+		name        string
+		timeout     time.Duration
+		wantTimeout time.Duration
+	}{
+		{
+			name:        "with custom timeout",
+			timeout:     1 * time.Minute,
+			wantTimeout: 1 * time.Minute,
+		},
+		{
+			name:        "with zero timeout uses default",
+			timeout:     0,
+			wantTimeout: DefaultUDPTimeout,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			tracker := NewUDPTracker(tt.timeout, logger)
+			assert.NotNil(t, tracker)
+			assert.Equal(t, tt.wantTimeout, tracker.timeout)
+			assert.NotNil(t, tracker.connections)
+			assert.NotNil(t, tracker.cleanupTicker)
+			assert.NotNil(t, tracker.done)
+		})
+	}
+}
+
+func TestUDPTracker_TrackOutbound(t *testing.T) {
+	tracker := NewUDPTracker(DefaultUDPTimeout, logger)
+	defer tracker.Close()
+
+	srcIP := net.ParseIP("192.168.1.2")
+	dstIP := net.ParseIP("192.168.1.3")
+	srcPort := uint16(12345)
+	dstPort := uint16(53)
+
+	tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort)
+
+	// Verify connection was tracked
+	key := makeConnKey(srcIP, dstIP, srcPort, dstPort)
+	conn, exists := tracker.connections[key]
+	require.True(t, exists)
+	assert.True(t, conn.SourceIP.Equal(srcIP))
+	assert.True(t, conn.DestIP.Equal(dstIP))
+	assert.Equal(t, srcPort, conn.SourcePort)
+	assert.Equal(t, dstPort, conn.DestPort)
+	assert.WithinDuration(t, time.Now(), conn.GetLastSeen(), 1*time.Second)
+}
+
+func TestUDPTracker_IsValidInbound(t *testing.T) {
+	tracker := NewUDPTracker(1*time.Second, logger)
+	defer tracker.Close()
+
+	srcIP := net.ParseIP("192.168.1.2")
+	dstIP := net.ParseIP("192.168.1.3")
+	srcPort := uint16(12345)
+	dstPort := uint16(53)
+
+	// Track outbound connection
+	tracker.TrackOutbound(srcIP, dstIP, srcPort, dstPort)
+
+	tests := []struct {
+		name    string
+		srcIP   net.IP
+		dstIP   net.IP
+		srcPort uint16
+		dstPort uint16
+		sleep   time.Duration
+		want    bool
+	}{
+		{
+			name:    "valid inbound response",
+			srcIP:   dstIP,   // Original destination is now source
+			dstIP:   srcIP,   // Original source is now destination
+			srcPort: dstPort, // Original destination port is now source
+			dstPort: srcPort, // Original source port is now destination
+			sleep:   0,
+			want:    true,
+		},
+		{
+			name:    "invalid source IP",
+			srcIP:   net.ParseIP("192.168.1.4"),
+			dstIP:   srcIP,
+			srcPort: dstPort,
+			dstPort: srcPort,
+			sleep:   0,
+			want:    false,
+		},
+		{
+			name:    "invalid destination IP",
+			srcIP:   dstIP,
+			dstIP:   net.ParseIP("192.168.1.4"),
+			srcPort: dstPort,
+			dstPort: srcPort,
+			sleep:   0,
+			want:    false,
+		},
+		{
+			name:    "invalid source port",
+			srcIP:   dstIP,
+			dstIP:   srcIP,
+			srcPort: 54321,
+			dstPort: srcPort,
+			sleep:   0,
+			want:    false,
+		},
+		{
+			name:    "invalid destination port",
+			srcIP:   dstIP,
+			dstIP:   srcIP,
+			srcPort: dstPort,
+			dstPort: 54321,
+			sleep:   0,
+			want:    false,
+		},
+		{
+			name:    "expired connection",
+			srcIP:   dstIP,
+			dstIP:   srcIP,
+			srcPort: dstPort,
+			dstPort: srcPort,
+			sleep:   2 * time.Second, // Longer than tracker timeout
+			want:    false,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if tt.sleep > 0 {
+				time.Sleep(tt.sleep)
+			}
+			got := tracker.IsValidInbound(tt.srcIP, tt.dstIP, tt.srcPort, tt.dstPort)
+			assert.Equal(t, tt.want, got)
+		})
+	}
+}
+
+func TestUDPTracker_Cleanup(t *testing.T) {
+	// Use shorter intervals for testing
+	timeout := 50 * time.Millisecond
+	cleanupInterval := 25 * time.Millisecond
+
+	// Create tracker with custom cleanup interval
+	tracker := &UDPTracker{
+		connections:   make(map[ConnKey]*UDPConnTrack),
+		timeout:       timeout,
+		cleanupTicker: time.NewTicker(cleanupInterval),
+		done:          make(chan struct{}),
+		ipPool:        NewPreallocatedIPs(),
+		logger:        logger,
+	}
+
+	// Start cleanup routine
+	go tracker.cleanupRoutine()
+
+	// Add some connections
+	connections := []struct {
+		srcIP   net.IP
+		dstIP   net.IP
+		srcPort uint16
+		dstPort uint16
+	}{
+		{
+			srcIP:   net.ParseIP("192.168.1.2"),
+			dstIP:   net.ParseIP("192.168.1.3"),
+			srcPort: 12345,
+			dstPort: 53,
+		},
+		{
+			srcIP:   net.ParseIP("192.168.1.4"),
+			dstIP:   net.ParseIP("192.168.1.5"),
+			srcPort: 12346,
+			dstPort: 53,
+		},
+	}
+
+	for _, conn := range connections {
+		tracker.TrackOutbound(conn.srcIP, conn.dstIP, conn.srcPort, conn.dstPort)
+	}
+
+	// Verify initial connections
+	assert.Len(t, tracker.connections, 2)
+
+	// Wait for connection timeout and cleanup interval
+	time.Sleep(timeout + 2*cleanupInterval)
+
+	tracker.mutex.RLock()
+	connCount := len(tracker.connections)
+	tracker.mutex.RUnlock()
+
+	// Verify connections were cleaned up
+	assert.Equal(t, 0, connCount, "Expected all connections to be cleaned up")
+
+	// Properly close the tracker
+	tracker.Close()
+}
+
+func BenchmarkUDPTracker(b *testing.B) {
+	b.Run("TrackOutbound", func(b *testing.B) {
+		tracker := NewUDPTracker(DefaultUDPTimeout, logger)
+		defer tracker.Close()
+
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			tracker.TrackOutbound(srcIP, dstIP, uint16(i%65535), 80)
+		}
+	})
+
+	b.Run("IsValidInbound", func(b *testing.B) {
+		tracker := NewUDPTracker(DefaultUDPTimeout, logger)
+		defer tracker.Close()
+
+		srcIP := net.ParseIP("192.168.1.1")
+		dstIP := net.ParseIP("192.168.1.2")
+
+		// Pre-populate some connections
+		for i := 0; i < 1000; i++ {
+			tracker.TrackOutbound(srcIP, dstIP, uint16(i), 80)
+		}
+
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			tracker.IsValidInbound(dstIP, srcIP, 80, uint16(i%1000))
+		}
+	})
+}

client/firewall/uspfilter/forwarder/endpoint.go

@@ -0,0 +1,81 @@
+package forwarder
+
+import (
+	wgdevice "golang.zx2c4.com/wireguard/device"
+	"gvisor.dev/gvisor/pkg/tcpip"
+	"gvisor.dev/gvisor/pkg/tcpip/header"
+	"gvisor.dev/gvisor/pkg/tcpip/stack"
+
+	nblog "github.com/netbirdio/netbird/client/firewall/uspfilter/log"
+)
+
+// endpoint implements stack.LinkEndpoint and handles integration with the wireguard device
+type endpoint struct {
+	logger     *nblog.Logger
+	dispatcher stack.NetworkDispatcher
+	device     *wgdevice.Device
+	mtu        uint32
+}
+
+func (e *endpoint) Attach(dispatcher stack.NetworkDispatcher) {
+	e.dispatcher = dispatcher
+}
+
+func (e *endpoint) IsAttached() bool {
+	return e.dispatcher != nil
+}
+
+func (e *endpoint) MTU() uint32 {
+	return e.mtu
+}
+
+func (e *endpoint) Capabilities() stack.LinkEndpointCapabilities {
+	return stack.CapabilityNone
+}
+
+func (e *endpoint) MaxHeaderLength() uint16 {
+	return 0
+}
+
+func (e *endpoint) LinkAddress() tcpip.LinkAddress {
+	return ""
+}
+
+func (e *endpoint) WritePackets(pkts stack.PacketBufferList) (int, tcpip.Error) {
+	var written int
+	for _, pkt := range pkts.AsSlice() {
+		netHeader := header.IPv4(pkt.NetworkHeader().View().AsSlice())
+
+		data := stack.PayloadSince(pkt.NetworkHeader())
+		if data == nil {
+			continue
+		}
+
+		// Send the packet through WireGuard
+		address := netHeader.DestinationAddress()
+		err := e.device.CreateOutboundPacket(data.AsSlice(), address.AsSlice())
+		if err != nil {
+			e.logger.Error("CreateOutboundPacket: %v", err)
+			continue
+		}
+		written++
+	}
+
+	return written, nil
+}
+
+func (e *endpoint) Wait() {
+	// not required
+}
+
+func (e *endpoint) ARPHardwareType() header.ARPHardwareType {
+	return header.ARPHardwareNone
+}
+
+func (e *endpoint) AddHeader(*stack.PacketBuffer) {
+	// not required
+}
+
+func (e *endpoint) ParseHeader(*stack.PacketBuffer) bool {
+	return true
+}

client/firewall/uspfilter/forwarder/forwarder.go

@@ -0,0 +1,166 @@
+package forwarder
+
+import (
+	"context"
+	"fmt"
+	"net"
+	"runtime"
+
+	log "github.com/sirupsen/logrus"
+	"gvisor.dev/gvisor/pkg/buffer"
+	"gvisor.dev/gvisor/pkg/tcpip"
+	"gvisor.dev/gvisor/pkg/tcpip/header"
+	"gvisor.dev/gvisor/pkg/tcpip/network/ipv4"
+	"gvisor.dev/gvisor/pkg/tcpip/stack"
+	"gvisor.dev/gvisor/pkg/tcpip/transport/icmp"
+	"gvisor.dev/gvisor/pkg/tcpip/transport/tcp"
+	"gvisor.dev/gvisor/pkg/tcpip/transport/udp"
+
+	"github.com/netbirdio/netbird/client/firewall/uspfilter/common"
+	nblog "github.com/netbirdio/netbird/client/firewall/uspfilter/log"
+)
+
+const (
+	defaultReceiveWindow = 32768
+	defaultMaxInFlight   = 1024
+	iosReceiveWindow     = 16384
+	iosMaxInFlight       = 256
+)
+
+type Forwarder struct {
+	logger       *nblog.Logger
+	stack        *stack.Stack
+	endpoint     *endpoint
+	udpForwarder *udpForwarder
+	ctx          context.Context
+	cancel       context.CancelFunc
+	ip           net.IP
+	netstack     bool
+}
+
+func New(iface common.IFaceMapper, logger *nblog.Logger, netstack bool) (*Forwarder, error) {
+	s := stack.New(stack.Options{
+		NetworkProtocols: []stack.NetworkProtocolFactory{ipv4.NewProtocol},
+		TransportProtocols: []stack.TransportProtocolFactory{
+			tcp.NewProtocol,
+			udp.NewProtocol,
+			icmp.NewProtocol4,
+		},
+		HandleLocal: false,
+	})
+
+	mtu, err := iface.GetDevice().MTU()
+	if err != nil {
+		return nil, fmt.Errorf("get MTU: %w", err)
+	}
+	nicID := tcpip.NICID(1)
+	endpoint := &endpoint{
+		logger: logger,
+		device: iface.GetWGDevice(),
+		mtu:    uint32(mtu),
+	}
+
+	if err := s.CreateNIC(nicID, endpoint); err != nil {
+		return nil, fmt.Errorf("failed to create NIC: %v", err)
+	}
+
+	ones, _ := iface.Address().Network.Mask.Size()
+	protoAddr := tcpip.ProtocolAddress{
+		Protocol: ipv4.ProtocolNumber,
+		AddressWithPrefix: tcpip.AddressWithPrefix{
+			Address:   tcpip.AddrFromSlice(iface.Address().IP.To4()),
+			PrefixLen: ones,
+		},
+	}
+
+	if err := s.AddProtocolAddress(nicID, protoAddr, stack.AddressProperties{}); err != nil {
+		return nil, fmt.Errorf("failed to add protocol address: %s", err)
+	}
+
+	defaultSubnet, err := tcpip.NewSubnet(
+		tcpip.AddrFrom4([4]byte{0, 0, 0, 0}),
+		tcpip.MaskFromBytes([]byte{0, 0, 0, 0}),
+	)
+	if err != nil {
+		return nil, fmt.Errorf("creating default subnet: %w", err)
+	}
+
+	if err := s.SetPromiscuousMode(nicID, true); err != nil {
+		return nil, fmt.Errorf("set promiscuous mode: %s", err)
+	}
+	if err := s.SetSpoofing(nicID, true); err != nil {
+		return nil, fmt.Errorf("set spoofing: %s", err)
+	}
+
+	s.SetRouteTable([]tcpip.Route{
+		{
+			Destination: defaultSubnet,
+			NIC:         nicID,
+		},
+	})
+
+	ctx, cancel := context.WithCancel(context.Background())
+	f := &Forwarder{
+		logger:       logger,
+		stack:        s,
+		endpoint:     endpoint,
+		udpForwarder: newUDPForwarder(mtu, logger),
+		ctx:          ctx,
+		cancel:       cancel,
+		netstack:     netstack,
+		ip:           iface.Address().IP,
+	}
+
+	receiveWindow := defaultReceiveWindow
+	maxInFlight := defaultMaxInFlight
+	if runtime.GOOS == "ios" {
+		receiveWindow = iosReceiveWindow
+		maxInFlight = iosMaxInFlight
+	}
+
+	tcpForwarder := tcp.NewForwarder(s, receiveWindow, maxInFlight, f.handleTCP)
+	s.SetTransportProtocolHandler(tcp.ProtocolNumber, tcpForwarder.HandlePacket)
+
+	udpForwarder := udp.NewForwarder(s, f.handleUDP)
+	s.SetTransportProtocolHandler(udp.ProtocolNumber, udpForwarder.HandlePacket)
+
+	s.SetTransportProtocolHandler(icmp.ProtocolNumber4, f.handleICMP)
+
+	log.Debugf("forwarder: Initialization complete with NIC %d", nicID)
+	return f, nil
+}
+
+func (f *Forwarder) InjectIncomingPacket(payload []byte) error {
+	if len(payload) < header.IPv4MinimumSize {
+		return fmt.Errorf("packet too small: %d bytes", len(payload))
+	}
+
+	pkt := stack.NewPacketBuffer(stack.PacketBufferOptions{
+		Payload: buffer.MakeWithData(payload),
+	})
+	defer pkt.DecRef()
+
+	if f.endpoint.dispatcher != nil {
+		f.endpoint.dispatcher.DeliverNetworkPacket(ipv4.ProtocolNumber, pkt)
+	}
+	return nil
+}
+
+// Stop gracefully shuts down the forwarder
+func (f *Forwarder) Stop() {
+	f.cancel()
+
+	if f.udpForwarder != nil {
+		f.udpForwarder.Stop()
+	}
+
+	f.stack.Close()
+	f.stack.Wait()
+}
+
+func (f *Forwarder) determineDialAddr(addr tcpip.Address) net.IP {
+	if f.netstack && f.ip.Equal(addr.AsSlice()) {
+		return net.IPv4(127, 0, 0, 1)
+	}
+	return addr.AsSlice()
+}

client/firewall/uspfilter/forwarder/icmp.go

@@ -0,0 +1,109 @@
+package forwarder
+
+import (
+	"context"
+	"net"
+	"time"
+
+	"gvisor.dev/gvisor/pkg/tcpip/header"
+	"gvisor.dev/gvisor/pkg/tcpip/stack"
+)
+
+// handleICMP handles ICMP packets from the network stack
+func (f *Forwarder) handleICMP(id stack.TransportEndpointID, pkt stack.PacketBufferPtr) bool {
+	ctx, cancel := context.WithTimeout(f.ctx, 5*time.Second)
+	defer cancel()
+
+	lc := net.ListenConfig{}
+	// TODO: support non-root
+	conn, err := lc.ListenPacket(ctx, "ip4:icmp", "0.0.0.0")
+	if err != nil {
+		f.logger.Error("Failed to create ICMP socket for %v: %v", id, err)
+
+		// This will make netstack reply on behalf of the original destination, that's ok for now
+		return false
+	}
+	defer func() {
+		if err := conn.Close(); err != nil {
+			f.logger.Debug("Failed to close ICMP socket: %v", err)
+		}
+	}()
+
+	dstIP := f.determineDialAddr(id.LocalAddress)
+	dst := &net.IPAddr{IP: dstIP}
+
+	// Get the complete ICMP message (header + data)
+	fullPacket := stack.PayloadSince(pkt.TransportHeader())
+	payload := fullPacket.AsSlice()
+
+	icmpHdr := header.ICMPv4(pkt.TransportHeader().View().AsSlice())
+
+	// For Echo Requests, send and handle response
+	switch icmpHdr.Type() {
+	case header.ICMPv4Echo:
+		return f.handleEchoResponse(icmpHdr, payload, dst, conn, id)
+	case header.ICMPv4EchoReply:
+		// dont process our own replies
+		return true
+	default:
+	}
+
+	// For other ICMP types (Time Exceeded, Destination Unreachable, etc)
+	_, err = conn.WriteTo(payload, dst)
+	if err != nil {
+		f.logger.Error("Failed to write ICMP packet for %v: %v", id, err)
+		return true
+	}
+
+	f.logger.Trace("Forwarded ICMP packet %v type=%v code=%v",
+		id, icmpHdr.Type(), icmpHdr.Code())
+
+	return true
+}
+
+func (f *Forwarder) handleEchoResponse(icmpHdr header.ICMPv4, payload []byte, dst *net.IPAddr, conn net.PacketConn, id stack.TransportEndpointID) bool {
+	if _, err := conn.WriteTo(payload, dst); err != nil {
+		f.logger.Error("Failed to write ICMP packet for %v: %v", id, err)
+		return true
+	}
+
+	f.logger.Trace("Forwarded ICMP packet %v type=%v code=%v",
+		id, icmpHdr.Type(), icmpHdr.Code())
+
+	if err := conn.SetReadDeadline(time.Now().Add(5 * time.Second)); err != nil {
+		f.logger.Error("Failed to set read deadline for ICMP response: %v", err)
+		return true
+	}
+
+	response := make([]byte, f.endpoint.mtu)
+	n, _, err := conn.ReadFrom(response)
+	if err != nil {
+		if !isTimeout(err) {
+			f.logger.Error("Failed to read ICMP response: %v", err)
+		}
+		return true
+	}
+
+	ipHdr := make([]byte, header.IPv4MinimumSize)
+	ip := header.IPv4(ipHdr)
+	ip.Encode(&header.IPv4Fields{
+		TotalLength: uint16(header.IPv4MinimumSize + n),
+		TTL:         64,
+		Protocol:    uint8(header.ICMPv4ProtocolNumber),
+		SrcAddr:     id.LocalAddress,
+		DstAddr:     id.RemoteAddress,
+	})
+	ip.SetChecksum(^ip.CalculateChecksum())
+
+	fullPacket := make([]byte, 0, len(ipHdr)+n)
+	fullPacket = append(fullPacket, ipHdr...)
+	fullPacket = append(fullPacket, response[:n]...)
+
+	if err := f.InjectIncomingPacket(fullPacket); err != nil {
+		f.logger.Error("Failed to inject ICMP response: %v", err)
+		return true
+	}
+
+	f.logger.Trace("Forwarded ICMP echo reply for %v", id)
+	return true
+}

client/firewall/uspfilter/forwarder/tcp.go

@@ -0,0 +1,90 @@
+package forwarder
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"net"
+
+	"gvisor.dev/gvisor/pkg/tcpip"
+	"gvisor.dev/gvisor/pkg/tcpip/adapters/gonet"
+	"gvisor.dev/gvisor/pkg/tcpip/stack"
+	"gvisor.dev/gvisor/pkg/tcpip/transport/tcp"
+	"gvisor.dev/gvisor/pkg/waiter"
+)
+
+// handleTCP is called by the TCP forwarder for new connections.
+func (f *Forwarder) handleTCP(r *tcp.ForwarderRequest) {
+	id := r.ID()
+
+	dialAddr := fmt.Sprintf("%s:%d", f.determineDialAddr(id.LocalAddress), id.LocalPort)
+
+	outConn, err := (&net.Dialer{}).DialContext(f.ctx, "tcp", dialAddr)
+	if err != nil {
+		r.Complete(true)
+		f.logger.Trace("forwarder: dial error for %v: %v", id, err)
+		return
+	}
+
+	// Create wait queue for blocking syscalls
+	wq := waiter.Queue{}
+
+	ep, epErr := r.CreateEndpoint(&wq)
+	if epErr != nil {
+		f.logger.Error("forwarder: failed to create TCP endpoint: %v", epErr)
+		if err := outConn.Close(); err != nil {
+			f.logger.Debug("forwarder: outConn close error: %v", err)
+		}
+		r.Complete(true)
+		return
+	}
+
+	// Complete the handshake
+	r.Complete(false)
+
+	inConn := gonet.NewTCPConn(&wq, ep)
+
+	f.logger.Trace("forwarder: established TCP connection %v", id)
+
+	go f.proxyTCP(id, inConn, outConn, ep)
+}
+
+func (f *Forwarder) proxyTCP(id stack.TransportEndpointID, inConn *gonet.TCPConn, outConn net.Conn, ep tcpip.Endpoint) {
+	defer func() {
+		if err := inConn.Close(); err != nil {
+			f.logger.Debug("forwarder: inConn close error: %v", err)
+		}
+		if err := outConn.Close(); err != nil {
+			f.logger.Debug("forwarder: outConn close error: %v", err)
+		}
+		ep.Close()
+	}()
+
+	// Create context for managing the proxy goroutines
+	ctx, cancel := context.WithCancel(f.ctx)
+	defer cancel()
+
+	errChan := make(chan error, 2)
+
+	go func() {
+		_, err := io.Copy(outConn, inConn)
+		errChan <- err
+	}()
+
+	go func() {
+		_, err := io.Copy(inConn, outConn)
+		errChan <- err
+	}()
+
+	select {
+	case <-ctx.Done():
+		f.logger.Trace("forwarder: tearing down TCP connection %v due to context done", id)
+		return
+	case err := <-errChan:
+		if err != nil && !isClosedError(err) {
+			f.logger.Error("proxyTCP: copy error: %v", err)
+		}
+		f.logger.Trace("forwarder: tearing down TCP connection %v", id)
+		return
+	}
+}

client/firewall/uspfilter/forwarder/udp.go

@@ -0,0 +1,288 @@
+package forwarder
+
+import (
+	"context"
+	"errors"
+	"fmt"
+	"net"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	"gvisor.dev/gvisor/pkg/tcpip"
+	"gvisor.dev/gvisor/pkg/tcpip/adapters/gonet"
+	"gvisor.dev/gvisor/pkg/tcpip/stack"
+	"gvisor.dev/gvisor/pkg/tcpip/transport/udp"
+	"gvisor.dev/gvisor/pkg/waiter"
+
+	nblog "github.com/netbirdio/netbird/client/firewall/uspfilter/log"
+)
+
+const (
+	udpTimeout = 30 * time.Second
+)
+
+type udpPacketConn struct {
+	conn     *gonet.UDPConn
+	outConn  net.Conn
+	lastSeen atomic.Int64
+	cancel   context.CancelFunc
+	ep       tcpip.Endpoint
+}
+
+type udpForwarder struct {
+	sync.RWMutex
+	logger  *nblog.Logger
+	conns   map[stack.TransportEndpointID]*udpPacketConn
+	bufPool sync.Pool
+	ctx     context.Context
+	cancel  context.CancelFunc
+}
+
+type idleConn struct {
+	id   stack.TransportEndpointID
+	conn *udpPacketConn
+}
+
+func newUDPForwarder(mtu int, logger *nblog.Logger) *udpForwarder {
+	ctx, cancel := context.WithCancel(context.Background())
+	f := &udpForwarder{
+		logger: logger,
+		conns:  make(map[stack.TransportEndpointID]*udpPacketConn),
+		ctx:    ctx,
+		cancel: cancel,
+		bufPool: sync.Pool{
+			New: func() any {
+				b := make([]byte, mtu)
+				return &b
+			},
+		},
+	}
+	go f.cleanup()
+	return f
+}
+
+// Stop stops the UDP forwarder and all active connections
+func (f *udpForwarder) Stop() {
+	f.cancel()
+
+	f.Lock()
+	defer f.Unlock()
+
+	for id, conn := range f.conns {
+		conn.cancel()
+		if err := conn.conn.Close(); err != nil {
+			f.logger.Debug("forwarder: UDP conn close error for %v: %v", id, err)
+		}
+		if err := conn.outConn.Close(); err != nil {
+			f.logger.Debug("forwarder: UDP outConn close error for %v: %v", id, err)
+		}
+
+		conn.ep.Close()
+		delete(f.conns, id)
+	}
+}
+
+// cleanup periodically removes idle UDP connections
+func (f *udpForwarder) cleanup() {
+	ticker := time.NewTicker(time.Minute)
+	defer ticker.Stop()
+
+	for {
+		select {
+		case <-f.ctx.Done():
+			return
+		case <-ticker.C:
+			var idleConns []idleConn
+
+			f.RLock()
+			for id, conn := range f.conns {
+				if conn.getIdleDuration() > udpTimeout {
+					idleConns = append(idleConns, idleConn{id, conn})
+				}
+			}
+			f.RUnlock()
+
+			for _, idle := range idleConns {
+				idle.conn.cancel()
+				if err := idle.conn.conn.Close(); err != nil {
+					f.logger.Debug("forwarder: UDP conn close error for %v: %v", idle.id, err)
+				}
+				if err := idle.conn.outConn.Close(); err != nil {
+					f.logger.Debug("forwarder: UDP outConn close error for %v: %v", idle.id, err)
+				}
+
+				idle.conn.ep.Close()
+
+				f.Lock()
+				delete(f.conns, idle.id)
+				f.Unlock()
+
+				f.logger.Trace("forwarder: cleaned up idle UDP connection %v", idle.id)
+			}
+		}
+	}
+}
+
+// handleUDP is called by the UDP forwarder for new packets
+func (f *Forwarder) handleUDP(r *udp.ForwarderRequest) {
+	if f.ctx.Err() != nil {
+		f.logger.Trace("forwarder: context done, dropping UDP packet")
+		return
+	}
+
+	id := r.ID()
+
+	f.udpForwarder.RLock()
+	_, exists := f.udpForwarder.conns[id]
+	f.udpForwarder.RUnlock()
+	if exists {
+		f.logger.Trace("forwarder: existing UDP connection for %v", id)
+		return
+	}
+
+	dstAddr := fmt.Sprintf("%s:%d", f.determineDialAddr(id.LocalAddress), id.LocalPort)
+	outConn, err := (&net.Dialer{}).DialContext(f.ctx, "udp", dstAddr)
+	if err != nil {
+		f.logger.Debug("forwarder: UDP dial error for %v: %v", id, err)
+		// TODO: Send ICMP error message
+		return
+	}
+
+	// Create wait queue for blocking syscalls
+	wq := waiter.Queue{}
+	ep, epErr := r.CreateEndpoint(&wq)
+	if epErr != nil {
+		f.logger.Debug("forwarder: failed to create UDP endpoint: %v", epErr)
+		if err := outConn.Close(); err != nil {
+			f.logger.Debug("forwarder: UDP outConn close error for %v: %v", id, err)
+		}
+		return
+	}
+
+	inConn := gonet.NewUDPConn(f.stack, &wq, ep)
+	connCtx, connCancel := context.WithCancel(f.ctx)
+
+	pConn := &udpPacketConn{
+		conn:    inConn,
+		outConn: outConn,
+		cancel:  connCancel,
+		ep:      ep,
+	}
+	pConn.updateLastSeen()
+
+	f.udpForwarder.Lock()
+	// Double-check no connection was created while we were setting up
+	if _, exists := f.udpForwarder.conns[id]; exists {
+		f.udpForwarder.Unlock()
+		pConn.cancel()
+		if err := inConn.Close(); err != nil {
+			f.logger.Debug("forwarder: UDP inConn close error for %v: %v", id, err)
+		}
+		if err := outConn.Close(); err != nil {
+			f.logger.Debug("forwarder: UDP outConn close error for %v: %v", id, err)
+		}
+		return
+	}
+	f.udpForwarder.conns[id] = pConn
+	f.udpForwarder.Unlock()
+
+	f.logger.Trace("forwarder: established UDP connection to %v", id)
+	go f.proxyUDP(connCtx, pConn, id, ep)
+}
+
+func (f *Forwarder) proxyUDP(ctx context.Context, pConn *udpPacketConn, id stack.TransportEndpointID, ep tcpip.Endpoint) {
+	defer func() {
+		pConn.cancel()
+		if err := pConn.conn.Close(); err != nil {
+			f.logger.Debug("forwarder: UDP inConn close error for %v: %v", id, err)
+		}
+		if err := pConn.outConn.Close(); err != nil {
+			f.logger.Debug("forwarder: UDP outConn close error for %v: %v", id, err)
+		}
+
+		ep.Close()
+
+		f.udpForwarder.Lock()
+		delete(f.udpForwarder.conns, id)
+		f.udpForwarder.Unlock()
+	}()
+
+	errChan := make(chan error, 2)
+
+	go func() {
+		errChan <- pConn.copy(ctx, pConn.conn, pConn.outConn, &f.udpForwarder.bufPool, "outbound->inbound")
+	}()
+
+	go func() {
+		errChan <- pConn.copy(ctx, pConn.outConn, pConn.conn, &f.udpForwarder.bufPool, "inbound->outbound")
+	}()
+
+	select {
+	case <-ctx.Done():
+		f.logger.Trace("forwarder: tearing down UDP connection %v due to context done", id)
+		return
+	case err := <-errChan:
+		if err != nil && !isClosedError(err) {
+			f.logger.Error("proxyUDP: copy error: %v", err)
+		}
+		f.logger.Trace("forwarder: tearing down UDP connection %v", id)
+		return
+	}
+}
+
+func (c *udpPacketConn) updateLastSeen() {
+	c.lastSeen.Store(time.Now().UnixNano())
+}
+
+func (c *udpPacketConn) getIdleDuration() time.Duration {
+	lastSeen := time.Unix(0, c.lastSeen.Load())
+	return time.Since(lastSeen)
+}
+
+func (c *udpPacketConn) copy(ctx context.Context, dst net.Conn, src net.Conn, bufPool *sync.Pool, direction string) error {
+	bufp := bufPool.Get().(*[]byte)
+	defer bufPool.Put(bufp)
+	buffer := *bufp
+
+	if err := src.SetReadDeadline(time.Now().Add(udpTimeout)); err != nil {
+		return fmt.Errorf("set read deadline: %w", err)
+	}
+	if err := src.SetWriteDeadline(time.Now().Add(udpTimeout)); err != nil {
+		return fmt.Errorf("set write deadline: %w", err)
+	}
+
+	for {
+		select {
+		case <-ctx.Done():
+			return ctx.Err()
+		default:
+			n, err := src.Read(buffer)
+			if err != nil {
+				if isTimeout(err) {
+					continue
+				}
+				return fmt.Errorf("read from %s: %w", direction, err)
+			}
+
+			_, err = dst.Write(buffer[:n])
+			if err != nil {
+				return fmt.Errorf("write to %s: %w", direction, err)
+			}
+
+			c.updateLastSeen()
+		}
+	}
+}
+
+func isClosedError(err error) bool {
+	return errors.Is(err, net.ErrClosed) || errors.Is(err, context.Canceled)
+}
+
+func isTimeout(err error) bool {
+	var netErr net.Error
+	if errors.As(err, &netErr) {
+		return netErr.Timeout()
+	}
+	return false
+}

client/firewall/uspfilter/localip.go

@@ -0,0 +1,134 @@
+package uspfilter
+
+import (
+	"fmt"
+	"net"
+	"sync"
+
+	log "github.com/sirupsen/logrus"
+
+	"github.com/netbirdio/netbird/client/firewall/uspfilter/common"
+)
+
+type localIPManager struct {
+	mu sync.RWMutex
+
+	// Use bitmap for IPv4 (32 bits * 2^16 = 256KB memory)
+	ipv4Bitmap [1 << 16]uint32
+}
+
+func newLocalIPManager() *localIPManager {
+	return &localIPManager{}
+}
+
+func (m *localIPManager) setBitmapBit(ip net.IP) {
+	ipv4 := ip.To4()
+	if ipv4 == nil {
+		return
+	}
+	high := (uint16(ipv4[0]) << 8) | uint16(ipv4[1])
+	low := (uint16(ipv4[2]) << 8) | uint16(ipv4[3])
+	m.ipv4Bitmap[high] |= 1 << (low % 32)
+}
+
+func (m *localIPManager) checkBitmapBit(ip net.IP) bool {
+	ipv4 := ip.To4()
+	if ipv4 == nil {
+		return false
+	}
+	high := (uint16(ipv4[0]) << 8) | uint16(ipv4[1])
+	low := (uint16(ipv4[2]) << 8) | uint16(ipv4[3])
+	return (m.ipv4Bitmap[high] & (1 << (low % 32))) != 0
+}
+
+func (m *localIPManager) processIP(ip net.IP, newIPv4Bitmap *[1 << 16]uint32, ipv4Set map[string]struct{}, ipv4Addresses *[]string) error {
+	if ipv4 := ip.To4(); ipv4 != nil {
+		high := (uint16(ipv4[0]) << 8) | uint16(ipv4[1])
+		low := (uint16(ipv4[2]) << 8) | uint16(ipv4[3])
+		if int(high) >= len(*newIPv4Bitmap) {
+			return fmt.Errorf("invalid IPv4 address: %s", ip)
+		}
+		ipStr := ip.String()
+		if _, exists := ipv4Set[ipStr]; !exists {
+			ipv4Set[ipStr] = struct{}{}
+			*ipv4Addresses = append(*ipv4Addresses, ipStr)
+			newIPv4Bitmap[high] |= 1 << (low % 32)
+		}
+	}
+	return nil
+}
+
+func (m *localIPManager) processInterface(iface net.Interface, newIPv4Bitmap *[1 << 16]uint32, ipv4Set map[string]struct{}, ipv4Addresses *[]string) {
+	addrs, err := iface.Addrs()
+	if err != nil {
+		log.Debugf("get addresses for interface %s failed: %v", iface.Name, err)
+		return
+	}
+
+	for _, addr := range addrs {
+		var ip net.IP
+		switch v := addr.(type) {
+		case *net.IPNet:
+			ip = v.IP
+		case *net.IPAddr:
+			ip = v.IP
+		default:
+			continue
+		}
+
+		if err := m.processIP(ip, newIPv4Bitmap, ipv4Set, ipv4Addresses); err != nil {
+			log.Debugf("process IP failed: %v", err)
+		}
+	}
+}
+
+func (m *localIPManager) UpdateLocalIPs(iface common.IFaceMapper) (err error) {
+	defer func() {
+		if r := recover(); r != nil {
+			err = fmt.Errorf("panic: %v", r)
+		}
+	}()
+
+	var newIPv4Bitmap [1 << 16]uint32
+	ipv4Set := make(map[string]struct{})
+	var ipv4Addresses []string
+
+	// 127.0.0.0/8
+	high := uint16(127) << 8
+	for i := uint16(0); i < 256; i++ {
+		newIPv4Bitmap[high|i] = 0xffffffff
+	}
+
+	if iface != nil {
+		if err := m.processIP(iface.Address().IP, &newIPv4Bitmap, ipv4Set, &ipv4Addresses); err != nil {
+			return err
+		}
+	}
+
+	interfaces, err := net.Interfaces()
+	if err != nil {
+		log.Warnf("failed to get interfaces: %v", err)
+	} else {
+		for _, intf := range interfaces {
+			m.processInterface(intf, &newIPv4Bitmap, ipv4Set, &ipv4Addresses)
+		}
+	}
+
+	m.mu.Lock()
+	m.ipv4Bitmap = newIPv4Bitmap
+	m.mu.Unlock()
+
+	log.Debugf("Local IPv4 addresses: %v", ipv4Addresses)
+	return nil
+}
+
+func (m *localIPManager) IsLocalIP(ip net.IP) bool {
+	m.mu.RLock()
+	defer m.mu.RUnlock()
+
+	if ipv4 := ip.To4(); ipv4 != nil {
+		return m.checkBitmapBit(ipv4)
+	}
+
+	return false
+}

client/firewall/uspfilter/localip_test.go

@@ -0,0 +1,270 @@
+package uspfilter
+
+import (
+	"net"
+	"testing"
+
+	"github.com/stretchr/testify/require"
+
+	"github.com/netbirdio/netbird/client/iface"
+)
+
+func TestLocalIPManager(t *testing.T) {
+	tests := []struct {
+		name      string
+		setupAddr iface.WGAddress
+		testIP    net.IP
+		expected  bool
+	}{
+		{
+			name: "Localhost range",
+			setupAddr: iface.WGAddress{
+				IP: net.ParseIP("192.168.1.1"),
+				Network: &net.IPNet{
+					IP:   net.ParseIP("192.168.1.0"),
+					Mask: net.CIDRMask(24, 32),
+				},
+			},
+			testIP:   net.ParseIP("127.0.0.2"),
+			expected: true,
+		},
+		{
+			name: "Localhost standard address",
+			setupAddr: iface.WGAddress{
+				IP: net.ParseIP("192.168.1.1"),
+				Network: &net.IPNet{
+					IP:   net.ParseIP("192.168.1.0"),
+					Mask: net.CIDRMask(24, 32),
+				},
+			},
+			testIP:   net.ParseIP("127.0.0.1"),
+			expected: true,
+		},
+		{
+			name: "Localhost range edge",
+			setupAddr: iface.WGAddress{
+				IP: net.ParseIP("192.168.1.1"),
+				Network: &net.IPNet{
+					IP:   net.ParseIP("192.168.1.0"),
+					Mask: net.CIDRMask(24, 32),
+				},
+			},
+			testIP:   net.ParseIP("127.255.255.255"),
+			expected: true,
+		},
+		{
+			name: "Local IP matches",
+			setupAddr: iface.WGAddress{
+				IP: net.ParseIP("192.168.1.1"),
+				Network: &net.IPNet{
+					IP:   net.ParseIP("192.168.1.0"),
+					Mask: net.CIDRMask(24, 32),
+				},
+			},
+			testIP:   net.ParseIP("192.168.1.1"),
+			expected: true,
+		},
+		{
+			name: "Local IP doesn't match",
+			setupAddr: iface.WGAddress{
+				IP: net.ParseIP("192.168.1.1"),
+				Network: &net.IPNet{
+					IP:   net.ParseIP("192.168.1.0"),
+					Mask: net.CIDRMask(24, 32),
+				},
+			},
+			testIP:   net.ParseIP("192.168.1.2"),
+			expected: false,
+		},
+		{
+			name: "IPv6 address",
+			setupAddr: iface.WGAddress{
+				IP: net.ParseIP("fe80::1"),
+				Network: &net.IPNet{
+					IP:   net.ParseIP("fe80::"),
+					Mask: net.CIDRMask(64, 128),
+				},
+			},
+			testIP:   net.ParseIP("fe80::1"),
+			expected: false,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			manager := newLocalIPManager()
+
+			mock := &IFaceMock{
+				AddressFunc: func() iface.WGAddress {
+					return tt.setupAddr
+				},
+			}
+
+			err := manager.UpdateLocalIPs(mock)
+			require.NoError(t, err)
+
+			result := manager.IsLocalIP(tt.testIP)
+			require.Equal(t, tt.expected, result)
+		})
+	}
+}
+
+func TestLocalIPManager_AllInterfaces(t *testing.T) {
+	manager := newLocalIPManager()
+	mock := &IFaceMock{}
+
+	// Get actual local interfaces
+	interfaces, err := net.Interfaces()
+	require.NoError(t, err)
+
+	var tests []struct {
+		ip       string
+		expected bool
+	}
+
+	// Add all local interface IPs to test cases
+	for _, iface := range interfaces {
+		addrs, err := iface.Addrs()
+		require.NoError(t, err)
+
+		for _, addr := range addrs {
+			var ip net.IP
+			switch v := addr.(type) {
+			case *net.IPNet:
+				ip = v.IP
+			case *net.IPAddr:
+				ip = v.IP
+			default:
+				continue
+			}
+
+			if ip4 := ip.To4(); ip4 != nil {
+				tests = append(tests, struct {
+					ip       string
+					expected bool
+				}{
+					ip:       ip4.String(),
+					expected: true,
+				})
+			}
+		}
+	}
+
+	// Add some external IPs as negative test cases
+	externalIPs := []string{
+		"8.8.8.8",
+		"1.1.1.1",
+		"208.67.222.222",
+	}
+	for _, ip := range externalIPs {
+		tests = append(tests, struct {
+			ip       string
+			expected bool
+		}{
+			ip:       ip,
+			expected: false,
+		})
+	}
+
+	require.NotEmpty(t, tests, "No test cases generated")
+
+	err = manager.UpdateLocalIPs(mock)
+	require.NoError(t, err)
+
+	t.Logf("Testing %d IPs", len(tests))
+	for _, tt := range tests {
+		t.Run(tt.ip, func(t *testing.T) {
+			result := manager.IsLocalIP(net.ParseIP(tt.ip))
+			require.Equal(t, tt.expected, result, "IP: %s", tt.ip)
+		})
+	}
+}
+
+// MapImplementation is a version using map[string]struct{}
+type MapImplementation struct {
+	localIPs map[string]struct{}
+}
+
+func BenchmarkIPChecks(b *testing.B) {
+	interfaces := make([]net.IP, 16)
+	for i := range interfaces {
+		interfaces[i] = net.IPv4(10, 0, byte(i>>8), byte(i))
+	}
+
+	// Setup bitmap version
+	bitmapManager := &localIPManager{
+		ipv4Bitmap: [1 << 16]uint32{},
+	}
+	for _, ip := range interfaces[:8] { // Add half of IPs
+		bitmapManager.setBitmapBit(ip)
+	}
+
+	// Setup map version
+	mapManager := &MapImplementation{
+		localIPs: make(map[string]struct{}),
+	}
+	for _, ip := range interfaces[:8] {
+		mapManager.localIPs[ip.String()] = struct{}{}
+	}
+
+	b.Run("Bitmap_Hit", func(b *testing.B) {
+		ip := interfaces[4]
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			bitmapManager.checkBitmapBit(ip)
+		}
+	})
+
+	b.Run("Bitmap_Miss", func(b *testing.B) {
+		ip := interfaces[12]
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			bitmapManager.checkBitmapBit(ip)
+		}
+	})
+
+	b.Run("Map_Hit", func(b *testing.B) {
+		ip := interfaces[4]
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			// nolint:gosimple
+			_, _ = mapManager.localIPs[ip.String()]
+		}
+	})
+
+	b.Run("Map_Miss", func(b *testing.B) {
+		ip := interfaces[12]
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			// nolint:gosimple
+			_, _ = mapManager.localIPs[ip.String()]
+		}
+	})
+}
+
+func BenchmarkWGPosition(b *testing.B) {
+	wgIP := net.ParseIP("10.10.0.1")
+
+	// Create two managers - one checks WG IP first, other checks it last
+	b.Run("WG_First", func(b *testing.B) {
+		bm := &localIPManager{ipv4Bitmap: [1 << 16]uint32{}}
+		bm.setBitmapBit(wgIP)
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			bm.checkBitmapBit(wgIP)
+		}
+	})
+
+	b.Run("WG_Last", func(b *testing.B) {
+		bm := &localIPManager{ipv4Bitmap: [1 << 16]uint32{}}
+		// Fill with other IPs first
+		for i := 0; i < 15; i++ {
+			bm.setBitmapBit(net.IPv4(10, 0, byte(i>>8), byte(i)))
+		}
+		bm.setBitmapBit(wgIP) // Add WG IP last
+		b.ResetTimer()
+		for i := 0; i < b.N; i++ {
+			bm.checkBitmapBit(wgIP)
+		}
+	})
+}

client/firewall/uspfilter/log/log.go

@@ -0,0 +1,196 @@
+// Package logger provides a high-performance, non-blocking logger for userspace networking
+package log
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	log "github.com/sirupsen/logrus"
+)
+
+const (
+	maxBatchSize         = 1024 * 16  // 16KB max batch size
+	maxMessageSize       = 1024 * 2   // 2KB per message
+	bufferSize           = 1024 * 256 // 256KB ring buffer
+	defaultFlushInterval = 2 * time.Second
+)
+
+// Level represents log severity
+type Level uint32
+
+const (
+	LevelPanic Level = iota
+	LevelFatal
+	LevelError
+	LevelWarn
+	LevelInfo
+	LevelDebug
+	LevelTrace
+)
+
+var levelStrings = map[Level]string{
+	LevelPanic: "PANC",
+	LevelFatal: "FATL",
+	LevelError: "ERRO",
+	LevelWarn:  "WARN",
+	LevelInfo:  "INFO",
+	LevelDebug: "DEBG",
+	LevelTrace: "TRAC",
+}
+
+// Logger is a high-performance, non-blocking logger
+type Logger struct {
+	output    io.Writer
+	level     atomic.Uint32
+	buffer    *ringBuffer
+	shutdown  chan struct{}
+	closeOnce sync.Once
+	wg        sync.WaitGroup
+
+	// Reusable buffer pool for formatting messages
+	bufPool sync.Pool
+}
+
+func NewFromLogrus(logrusLogger *log.Logger) *Logger {
+	l := &Logger{
+		output:   logrusLogger.Out,
+		buffer:   newRingBuffer(bufferSize),
+		shutdown: make(chan struct{}),
+		bufPool: sync.Pool{
+			New: func() interface{} {
+				// Pre-allocate buffer for message formatting
+				b := make([]byte, 0, maxMessageSize)
+				return &b
+			},
+		},
+	}
+	logrusLevel := logrusLogger.GetLevel()
+	l.level.Store(uint32(logrusLevel))
+	level := levelStrings[Level(logrusLevel)]
+	log.Debugf("New uspfilter logger created with loglevel %v", level)
+
+	l.wg.Add(1)
+	go l.worker()
+
+	return l
+}
+
+func (l *Logger) SetLevel(level Level) {
+	l.level.Store(uint32(level))
+
+	log.Debugf("Set uspfilter logger loglevel to %v", levelStrings[level])
+}
+
+func (l *Logger) formatMessage(buf *[]byte, level Level, format string, args ...interface{}) {
+	*buf = (*buf)[:0]
+
+	// Timestamp
+	*buf = time.Now().AppendFormat(*buf, "2006-01-02T15:04:05-07:00")
+	*buf = append(*buf, ' ')
+
+	// Level
+	*buf = append(*buf, levelStrings[level]...)
+	*buf = append(*buf, ' ')
+
+	// Message
+	if len(args) > 0 {
+		*buf = append(*buf, fmt.Sprintf(format, args...)...)
+	} else {
+		*buf = append(*buf, format...)
+	}
+
+	*buf = append(*buf, '\n')
+}
+
+func (l *Logger) log(level Level, format string, args ...interface{}) {
+	bufp := l.bufPool.Get().(*[]byte)
+	l.formatMessage(bufp, level, format, args...)
+
+	if len(*bufp) > maxMessageSize {
+		*bufp = (*bufp)[:maxMessageSize]
+	}
+	_, _ = l.buffer.Write(*bufp)
+
+	l.bufPool.Put(bufp)
+}
+
+func (l *Logger) Error(format string, args ...interface{}) {
+	if l.level.Load() >= uint32(LevelError) {
+		l.log(LevelError, format, args...)
+	}
+}
+
+func (l *Logger) Warn(format string, args ...interface{}) {
+	if l.level.Load() >= uint32(LevelWarn) {
+		l.log(LevelWarn, format, args...)
+	}
+}
+
+func (l *Logger) Info(format string, args ...interface{}) {
+	if l.level.Load() >= uint32(LevelInfo) {
+		l.log(LevelInfo, format, args...)
+	}
+}
+
+func (l *Logger) Debug(format string, args ...interface{}) {
+	if l.level.Load() >= uint32(LevelDebug) {
+		l.log(LevelDebug, format, args...)
+	}
+}
+
+func (l *Logger) Trace(format string, args ...interface{}) {
+	if l.level.Load() >= uint32(LevelTrace) {
+		l.log(LevelTrace, format, args...)
+	}
+}
+
+// worker periodically flushes the buffer
+func (l *Logger) worker() {
+	defer l.wg.Done()
+
+	ticker := time.NewTicker(defaultFlushInterval)
+	defer ticker.Stop()
+
+	buf := make([]byte, 0, maxBatchSize)
+
+	for {
+		select {
+		case <-l.shutdown:
+			return
+		case <-ticker.C:
+			// Read accumulated messages
+			n, _ := l.buffer.Read(buf[:cap(buf)])
+			if n == 0 {
+				continue
+			}
+
+			// Write batch
+			_, _ = l.output.Write(buf[:n])
+		}
+	}
+}
+
+// Stop gracefully shuts down the logger
+func (l *Logger) Stop(ctx context.Context) error {
+	done := make(chan struct{})
+
+	l.closeOnce.Do(func() {
+		close(l.shutdown)
+	})
+
+	go func() {
+		l.wg.Wait()
+		close(done)
+	}()
+
+	select {
+	case <-ctx.Done():
+		return ctx.Err()
+	case <-done:
+		return nil
+	}
+}

client/firewall/uspfilter/log/ringbuffer.go

@@ -0,0 +1,85 @@
+package log
+
+import "sync"
+
+// ringBuffer is a simple ring buffer implementation
+type ringBuffer struct {
+	buf  []byte
+	size int
+	r, w int64 // Read and write positions
+	mu   sync.Mutex
+}
+
+func newRingBuffer(size int) *ringBuffer {
+	return &ringBuffer{
+		buf:  make([]byte, size),
+		size: size,
+	}
+}
+
+func (r *ringBuffer) Write(p []byte) (n int, err error) {
+	if len(p) == 0 {
+		return 0, nil
+	}
+
+	r.mu.Lock()
+	defer r.mu.Unlock()
+
+	if len(p) > r.size {
+		p = p[:r.size]
+	}
+
+	n = len(p)
+
+	// Write data, handling wrap-around
+	pos := int(r.w % int64(r.size))
+	writeLen := min(len(p), r.size-pos)
+	copy(r.buf[pos:], p[:writeLen])
+
+	// If we have more data and need to wrap around
+	if writeLen < len(p) {
+		copy(r.buf, p[writeLen:])
+	}
+
+	// Update write position
+	r.w += int64(n)
+
+	return n, nil
+}
+
+func (r *ringBuffer) Read(p []byte) (n int, err error) {
+	r.mu.Lock()
+	defer r.mu.Unlock()
+
+	if r.w == r.r {
+		return 0, nil
+	}
+
+	// Calculate available data accounting for wraparound
+	available := int(r.w - r.r)
+	if available < 0 {
+		available += r.size
+	}
+	available = min(available, r.size)
+
+	// Limit read to buffer size
+	toRead := min(available, len(p))
+	if toRead == 0 {
+		return 0, nil
+	}
+
+	// Read data, handling wrap-around
+	pos := int(r.r % int64(r.size))
+	readLen := min(toRead, r.size-pos)
+	n = copy(p, r.buf[pos:pos+readLen])
+
+	// If we need more data and need to wrap around
+	if readLen < toRead {
+		n += copy(p[readLen:toRead], r.buf[:toRead-readLen])
+	}
+
+	// Update read position
+	r.r += int64(n)
+
+	return n, nil
+}

client/firewall/uspfilter/rule.go

@@ -2,22 +2,22 @@ package uspfilter
 
 import (
 	"net"
+	"net/netip"
 
 	"github.com/google/gopacket"
 
 	firewall "github.com/netbirdio/netbird/client/firewall/manager"
 )
 
-// Rule to handle management of rules
-type Rule struct {
+// PeerRule to handle management of rules
+type PeerRule struct {
 	id         string
 	ip         net.IP
 	ipLayer    gopacket.LayerType
 	matchByIP  bool
 	protoLayer gopacket.LayerType
-	direction  firewall.RuleDirection
-	sPort      uint16
-	dPort      uint16
+	sPort      *firewall.Port
+	dPort      *firewall.Port
 	drop       bool
 	comment    string
 
@@ -25,6 +25,21 @@ type Rule struct {
 }
 
 // GetRuleID returns the rule id
-func (r *Rule) GetRuleID() string {
+func (r *PeerRule) GetRuleID() string {
+	return r.id
+}
+
+type RouteRule struct {
+	id          string
+	sources     []netip.Prefix
+	destination netip.Prefix
+	proto       firewall.Protocol
+	srcPort     *firewall.Port
+	dstPort     *firewall.Port
+	action      firewall.Action
+}
+
+// GetRuleID returns the rule id
+func (r *RouteRule) GetRuleID() string {
 	return r.id
 }

client/firewall/uspfilter/tracer.go

@@ -0,0 +1,390 @@
+package uspfilter
+
+import (
+	"fmt"
+	"net"
+	"time"
+
+	"github.com/google/gopacket"
+	"github.com/google/gopacket/layers"
+
+	fw "github.com/netbirdio/netbird/client/firewall/manager"
+	"github.com/netbirdio/netbird/client/firewall/uspfilter/conntrack"
+)
+
+type PacketStage int
+
+const (
+	StageReceived PacketStage = iota
+	StageConntrack
+	StagePeerACL
+	StageRouting
+	StageRouteACL
+	StageForwarding
+	StageCompleted
+)
+
+const msgProcessingCompleted = "Processing completed"
+
+func (s PacketStage) String() string {
+	return map[PacketStage]string{
+		StageReceived:   "Received",
+		StageConntrack:  "Connection Tracking",
+		StagePeerACL:    "Peer ACL",
+		StageRouting:    "Routing",
+		StageRouteACL:   "Route ACL",
+		StageForwarding: "Forwarding",
+		StageCompleted:  "Completed",
+	}[s]
+}
+
+type ForwarderAction struct {
+	Action     string
+	RemoteAddr string
+	Error      error
+}
+
+type TraceResult struct {
+	Timestamp       time.Time
+	Stage           PacketStage
+	Message         string
+	Allowed         bool
+	ForwarderAction *ForwarderAction
+}
+
+type PacketTrace struct {
+	SourceIP        net.IP
+	DestinationIP   net.IP
+	Protocol        string
+	SourcePort      uint16
+	DestinationPort uint16
+	Direction       fw.RuleDirection
+	Results         []TraceResult
+}
+
+type TCPState struct {
+	SYN bool
+	ACK bool
+	FIN bool
+	RST bool
+	PSH bool
+	URG bool
+}
+
+type PacketBuilder struct {
+	SrcIP       net.IP
+	DstIP       net.IP
+	Protocol    fw.Protocol
+	SrcPort     uint16
+	DstPort     uint16
+	ICMPType    uint8
+	ICMPCode    uint8
+	Direction   fw.RuleDirection
+	PayloadSize int
+	TCPState    *TCPState
+}
+
+func (t *PacketTrace) AddResult(stage PacketStage, message string, allowed bool) {
+	t.Results = append(t.Results, TraceResult{
+		Timestamp: time.Now(),
+		Stage:     stage,
+		Message:   message,
+		Allowed:   allowed,
+	})
+}
+
+func (t *PacketTrace) AddResultWithForwarder(stage PacketStage, message string, allowed bool, action *ForwarderAction) {
+	t.Results = append(t.Results, TraceResult{
+		Timestamp:       time.Now(),
+		Stage:           stage,
+		Message:         message,
+		Allowed:         allowed,
+		ForwarderAction: action,
+	})
+}
+
+func (p *PacketBuilder) Build() ([]byte, error) {
+	ip := p.buildIPLayer()
+	pktLayers := []gopacket.SerializableLayer{ip}
+
+	transportLayer, err := p.buildTransportLayer(ip)
+	if err != nil {
+		return nil, err
+	}
+	pktLayers = append(pktLayers, transportLayer...)
+
+	if p.PayloadSize > 0 {
+		payload := make([]byte, p.PayloadSize)
+		pktLayers = append(pktLayers, gopacket.Payload(payload))
+	}
+
+	return serializePacket(pktLayers)
+}
+
+func (p *PacketBuilder) buildIPLayer() *layers.IPv4 {
+	return &layers.IPv4{
+		Version:  4,
+		TTL:      64,
+		Protocol: layers.IPProtocol(getIPProtocolNumber(p.Protocol)),
+		SrcIP:    p.SrcIP,
+		DstIP:    p.DstIP,
+	}
+}
+
+func (p *PacketBuilder) buildTransportLayer(ip *layers.IPv4) ([]gopacket.SerializableLayer, error) {
+	switch p.Protocol {
+	case "tcp":
+		return p.buildTCPLayer(ip)
+	case "udp":
+		return p.buildUDPLayer(ip)
+	case "icmp":
+		return p.buildICMPLayer()
+	default:
+		return nil, fmt.Errorf("unsupported protocol: %s", p.Protocol)
+	}
+}
+
+func (p *PacketBuilder) buildTCPLayer(ip *layers.IPv4) ([]gopacket.SerializableLayer, error) {
+	tcp := &layers.TCP{
+		SrcPort: layers.TCPPort(p.SrcPort),
+		DstPort: layers.TCPPort(p.DstPort),
+		Window:  65535,
+		SYN:     p.TCPState != nil && p.TCPState.SYN,
+		ACK:     p.TCPState != nil && p.TCPState.ACK,
+		FIN:     p.TCPState != nil && p.TCPState.FIN,
+		RST:     p.TCPState != nil && p.TCPState.RST,
+		PSH:     p.TCPState != nil && p.TCPState.PSH,
+		URG:     p.TCPState != nil && p.TCPState.URG,
+	}
+	if err := tcp.SetNetworkLayerForChecksum(ip); err != nil {
+		return nil, fmt.Errorf("set network layer for TCP checksum: %w", err)
+	}
+	return []gopacket.SerializableLayer{tcp}, nil
+}
+
+func (p *PacketBuilder) buildUDPLayer(ip *layers.IPv4) ([]gopacket.SerializableLayer, error) {
+	udp := &layers.UDP{
+		SrcPort: layers.UDPPort(p.SrcPort),
+		DstPort: layers.UDPPort(p.DstPort),
+	}
+	if err := udp.SetNetworkLayerForChecksum(ip); err != nil {
+		return nil, fmt.Errorf("set network layer for UDP checksum: %w", err)
+	}
+	return []gopacket.SerializableLayer{udp}, nil
+}
+
+func (p *PacketBuilder) buildICMPLayer() ([]gopacket.SerializableLayer, error) {
+	icmp := &layers.ICMPv4{
+		TypeCode: layers.CreateICMPv4TypeCode(p.ICMPType, p.ICMPCode),
+	}
+	if p.ICMPType == layers.ICMPv4TypeEchoRequest || p.ICMPType == layers.ICMPv4TypeEchoReply {
+		icmp.Id = uint16(1)
+		icmp.Seq = uint16(1)
+	}
+	return []gopacket.SerializableLayer{icmp}, nil
+}
+
+func serializePacket(layers []gopacket.SerializableLayer) ([]byte, error) {
+	buf := gopacket.NewSerializeBuffer()
+	opts := gopacket.SerializeOptions{
+		ComputeChecksums: true,
+		FixLengths:       true,
+	}
+	if err := gopacket.SerializeLayers(buf, opts, layers...); err != nil {
+		return nil, fmt.Errorf("serialize packet: %w", err)
+	}
+	return buf.Bytes(), nil
+}
+
+func getIPProtocolNumber(protocol fw.Protocol) int {
+	switch protocol {
+	case fw.ProtocolTCP:
+		return int(layers.IPProtocolTCP)
+	case fw.ProtocolUDP:
+		return int(layers.IPProtocolUDP)
+	case fw.ProtocolICMP:
+		return int(layers.IPProtocolICMPv4)
+	default:
+		return 0
+	}
+}
+
+func (m *Manager) TracePacketFromBuilder(builder *PacketBuilder) (*PacketTrace, error) {
+	packetData, err := builder.Build()
+	if err != nil {
+		return nil, fmt.Errorf("build packet: %w", err)
+	}
+
+	return m.TracePacket(packetData, builder.Direction), nil
+}
+
+func (m *Manager) TracePacket(packetData []byte, direction fw.RuleDirection) *PacketTrace {
+
+	d := m.decoders.Get().(*decoder)
+	defer m.decoders.Put(d)
+
+	trace := &PacketTrace{Direction: direction}
+
+	// Initial packet decoding
+	if err := d.parser.DecodeLayers(packetData, &d.decoded); err != nil {
+		trace.AddResult(StageReceived, fmt.Sprintf("Failed to decode packet: %v", err), false)
+		return trace
+	}
+
+	// Extract base packet info
+	srcIP, dstIP := m.extractIPs(d)
+	trace.SourceIP = srcIP
+	trace.DestinationIP = dstIP
+
+	// Determine protocol and ports
+	switch d.decoded[1] {
+	case layers.LayerTypeTCP:
+		trace.Protocol = "TCP"
+		trace.SourcePort = uint16(d.tcp.SrcPort)
+		trace.DestinationPort = uint16(d.tcp.DstPort)
+	case layers.LayerTypeUDP:
+		trace.Protocol = "UDP"
+		trace.SourcePort = uint16(d.udp.SrcPort)
+		trace.DestinationPort = uint16(d.udp.DstPort)
+	case layers.LayerTypeICMPv4:
+		trace.Protocol = "ICMP"
+	}
+
+	trace.AddResult(StageReceived, fmt.Sprintf("Received %s packet: %s:%d -> %s:%d",
+		trace.Protocol, srcIP, trace.SourcePort, dstIP, trace.DestinationPort), true)
+
+	if direction == fw.RuleDirectionOUT {
+		return m.traceOutbound(packetData, trace)
+	}
+
+	return m.traceInbound(packetData, trace, d, srcIP, dstIP)
+}
+
+func (m *Manager) traceInbound(packetData []byte, trace *PacketTrace, d *decoder, srcIP net.IP, dstIP net.IP) *PacketTrace {
+	if m.stateful && m.handleConntrackState(trace, d, srcIP, dstIP) {
+		return trace
+	}
+
+	if m.handleLocalDelivery(trace, packetData, d, srcIP, dstIP) {
+		return trace
+	}
+
+	if !m.handleRouting(trace) {
+		return trace
+	}
+
+	if m.nativeRouter {
+		return m.handleNativeRouter(trace)
+	}
+
+	return m.handleRouteACLs(trace, d, srcIP, dstIP)
+}
+
+func (m *Manager) handleConntrackState(trace *PacketTrace, d *decoder, srcIP, dstIP net.IP) bool {
+	allowed := m.isValidTrackedConnection(d, srcIP, dstIP)
+	msg := "No existing connection found"
+	if allowed {
+		msg = m.buildConntrackStateMessage(d)
+		trace.AddResult(StageConntrack, msg, true)
+		trace.AddResult(StageCompleted, "Packet allowed by connection tracking", true)
+		return true
+	}
+	trace.AddResult(StageConntrack, msg, false)
+	return false
+}
+
+func (m *Manager) buildConntrackStateMessage(d *decoder) string {
+	msg := "Matched existing connection state"
+	switch d.decoded[1] {
+	case layers.LayerTypeTCP:
+		flags := getTCPFlags(&d.tcp)
+		msg += fmt.Sprintf(" (TCP Flags: SYN=%v ACK=%v RST=%v FIN=%v)",
+			flags&conntrack.TCPSyn != 0,
+			flags&conntrack.TCPAck != 0,
+			flags&conntrack.TCPRst != 0,
+			flags&conntrack.TCPFin != 0)
+	case layers.LayerTypeICMPv4:
+		msg += fmt.Sprintf(" (ICMP ID=%d, Seq=%d)", d.icmp4.Id, d.icmp4.Seq)
+	}
+	return msg
+}
+
+func (m *Manager) handleLocalDelivery(trace *PacketTrace, packetData []byte, d *decoder, srcIP, dstIP net.IP) bool {
+	if !m.localForwarding {
+		trace.AddResult(StageRouting, "Local forwarding disabled", false)
+		trace.AddResult(StageCompleted, "Packet dropped - local forwarding disabled", false)
+		return true
+	}
+
+	trace.AddResult(StageRouting, "Packet destined for local delivery", true)
+	blocked := m.peerACLsBlock(srcIP, packetData, m.incomingRules, d)
+
+	msg := "Allowed by peer ACL rules"
+	if blocked {
+		msg = "Blocked by peer ACL rules"
+	}
+	trace.AddResult(StagePeerACL, msg, !blocked)
+
+	if m.netstack {
+		m.addForwardingResult(trace, "proxy-local", "127.0.0.1", !blocked)
+	}
+
+	trace.AddResult(StageCompleted, msgProcessingCompleted, !blocked)
+	return true
+}
+
+func (m *Manager) handleRouting(trace *PacketTrace) bool {
+	if !m.routingEnabled {
+		trace.AddResult(StageRouting, "Routing disabled", false)
+		trace.AddResult(StageCompleted, "Packet dropped - routing disabled", false)
+		return false
+	}
+	trace.AddResult(StageRouting, "Routing enabled, checking ACLs", true)
+	return true
+}
+
+func (m *Manager) handleNativeRouter(trace *PacketTrace) *PacketTrace {
+	trace.AddResult(StageRouteACL, "Using native router, skipping ACL checks", true)
+	trace.AddResult(StageForwarding, "Forwarding via native router", true)
+	trace.AddResult(StageCompleted, msgProcessingCompleted, true)
+	return trace
+}
+
+func (m *Manager) handleRouteACLs(trace *PacketTrace, d *decoder, srcIP, dstIP net.IP) *PacketTrace {
+	proto := getProtocolFromPacket(d)
+	srcPort, dstPort := getPortsFromPacket(d)
+	allowed := m.routeACLsPass(srcIP, dstIP, proto, srcPort, dstPort)
+
+	msg := "Allowed by route ACLs"
+	if !allowed {
+		msg = "Blocked by route ACLs"
+	}
+	trace.AddResult(StageRouteACL, msg, allowed)
+
+	if allowed && m.forwarder != nil {
+		m.addForwardingResult(trace, "proxy-remote", fmt.Sprintf("%s:%d", dstIP, dstPort), true)
+	}
+
+	trace.AddResult(StageCompleted, msgProcessingCompleted, allowed)
+	return trace
+}
+
+func (m *Manager) addForwardingResult(trace *PacketTrace, action, remoteAddr string, allowed bool) {
+	fwdAction := &ForwarderAction{
+		Action:     action,
+		RemoteAddr: remoteAddr,
+	}
+	trace.AddResultWithForwarder(StageForwarding,
+		fmt.Sprintf("Forwarding to %s", fwdAction.Action), allowed, fwdAction)
+}
+
+func (m *Manager) traceOutbound(packetData []byte, trace *PacketTrace) *PacketTrace {
+	// will create or update the connection state
+	dropped := m.processOutgoingHooks(packetData)
+	if dropped {
+		trace.AddResult(StageCompleted, "Packet dropped by outgoing hook", false)
+	} else {
+		trace.AddResult(StageCompleted, "Packet allowed (outgoing)", true)
+	}
+	return trace
+}